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 * Command Settings:: How to change default behavior of commands
1565 * Completion:: Command completion
1566 * Command Options:: Command options
1567 * Help:: How to ask @value{GDBN} for help
1570 @node Command Syntax
1571 @section Command Syntax
1573 A @value{GDBN} command is a single line of input. There is no limit on
1574 how long it can be. It starts with a command name, which is followed by
1575 arguments whose meaning depends on the command name. For example, the
1576 command @code{step} accepts an argument which is the number of times to
1577 step, as in @samp{step 5}. You can also use the @code{step} command
1578 with no arguments. Some commands do not allow any arguments.
1580 @cindex abbreviation
1581 @value{GDBN} command names may always be truncated if that abbreviation is
1582 unambiguous. Other possible command abbreviations are listed in the
1583 documentation for individual commands. In some cases, even ambiguous
1584 abbreviations are allowed; for example, @code{s} is specially defined as
1585 equivalent to @code{step} even though there are other commands whose
1586 names start with @code{s}. You can test abbreviations by using them as
1587 arguments to the @code{help} command.
1589 @cindex repeating commands
1590 @kindex RET @r{(repeat last command)}
1591 A blank line as input to @value{GDBN} (typing just @key{RET}) means to
1592 repeat the previous command. Certain commands (for example, @code{run})
1593 will not repeat this way; these are commands whose unintentional
1594 repetition might cause trouble and which you are unlikely to want to
1595 repeat. User-defined commands can disable this feature; see
1596 @ref{Define, dont-repeat}.
1598 The @code{list} and @code{x} commands, when you repeat them with
1599 @key{RET}, construct new arguments rather than repeating
1600 exactly as typed. This permits easy scanning of source or memory.
1602 @value{GDBN} can also use @key{RET} in another way: to partition lengthy
1603 output, in a way similar to the common utility @code{more}
1604 (@pxref{Screen Size,,Screen Size}). Since it is easy to press one
1605 @key{RET} too many in this situation, @value{GDBN} disables command
1606 repetition after any command that generates this sort of display.
1608 @kindex # @r{(a comment)}
1610 Any text from a @kbd{#} to the end of the line is a comment; it does
1611 nothing. This is useful mainly in command files (@pxref{Command
1612 Files,,Command Files}).
1614 @cindex repeating command sequences
1615 @kindex Ctrl-o @r{(operate-and-get-next)}
1616 The @kbd{Ctrl-o} binding is useful for repeating a complex sequence of
1617 commands. This command accepts the current line, like @key{RET}, and
1618 then fetches the next line relative to the current line from the history
1622 @node Command Settings
1623 @section Command Settings
1624 @cindex default behavior of commands, changing
1625 @cindex default settings, changing
1627 Many commands change their behavior according to command-specific
1628 variables or settings. These settings can be changed with the
1629 @code{set} subcommands. For example, the @code{print} command
1630 (@pxref{Data, ,Examining Data}) prints arrays differently depending on
1631 settings changeable with the commands @code{set print elements
1632 NUMBER-OF-ELEMENTS} and @code{set print array-indexes}, among others.
1634 You can change these settings to your preference in the gdbinit files
1635 loaded at @value{GDBN} startup. @xref{Startup}.
1637 The settings can also be changed interactively during the debugging
1638 session. For example, to change the limit of array elements to print,
1639 you can do the following:
1641 (@value{GDBN}) set print elements 10
1642 (@value{GDBN}) print some_array
1643 $1 = @{0, 10, 20, 30, 40, 50, 60, 70, 80, 90...@}
1646 The above @code{set print elements 10} command changes the number of
1647 elements to print from the default of 200 to 10. If you only intend
1648 this limit of 10 to be used for printing @code{some_array}, then you
1649 must restore the limit back to 200, with @code{set print elements
1652 Some commands allow overriding settings with command options. For
1653 example, the @code{print} command supports a number of options that
1654 allow overriding relevant global print settings as set by @code{set
1655 print} subcommands. @xref{print options}. The example above could be
1658 (@value{GDBN}) print -elements 10 -- some_array
1659 $1 = @{0, 10, 20, 30, 40, 50, 60, 70, 80, 90...@}
1662 Alternatively, you can use the @code{with} command to change a setting
1663 temporarily, for the duration of a command invocation.
1666 @kindex with command
1667 @kindex w @r{(@code{with})}
1669 @cindex temporarily change settings
1670 @item with @var{setting} [@var{value}] [-- @var{command}]
1671 @itemx w @var{setting} [@var{value}] [-- @var{command}]
1672 Temporarily set @var{setting} to @var{value} for the duration of
1675 @var{setting} is any setting you can change with the @code{set}
1676 subcommands. @var{value} is the value to assign to @code{setting}
1677 while running @code{command}.
1679 If no @var{command} is provided, the last command executed is
1682 If a @var{command} is provided, it must be preceded by a double dash
1683 (@code{--}) separator. This is required because some settings accept
1684 free-form arguments, such as expressions or filenames.
1686 For example, the command
1688 (@value{GDBN}) with print array on -- print some_array
1691 is equivalent to the following 3 commands:
1693 (@value{GDBN}) set print array on
1694 (@value{GDBN}) print some_array
1695 (@value{GDBN}) set print array off
1698 The @code{with} command is particularly useful when you want to
1699 override a setting while running user-defined commands, or commands
1700 defined in Python or Guile. @xref{Extending GDB,, Extending GDB}.
1703 (@value{GDBN}) with print pretty on -- my_complex_command
1706 To change several settings for the same command, you can nest
1707 @code{with} commands. For example, @code{with language ada -- with
1708 print elements 10} temporarily changes the language to Ada and sets a
1709 limit of 10 elements to print for arrays and strings.
1714 @section Command Completion
1717 @cindex word completion
1718 @value{GDBN} can fill in the rest of a word in a command for you, if there is
1719 only one possibility; it can also show you what the valid possibilities
1720 are for the next word in a command, at any time. This works for @value{GDBN}
1721 commands, @value{GDBN} subcommands, command options, and the names of symbols
1724 Press the @key{TAB} key whenever you want @value{GDBN} to fill out the rest
1725 of a word. If there is only one possibility, @value{GDBN} fills in the
1726 word, and waits for you to finish the command (or press @key{RET} to
1727 enter it). For example, if you type
1729 @c FIXME "@key" does not distinguish its argument sufficiently to permit
1730 @c complete accuracy in these examples; space introduced for clarity.
1731 @c If texinfo enhancements make it unnecessary, it would be nice to
1732 @c replace " @key" by "@key" in the following...
1734 (@value{GDBP}) info bre @key{TAB}
1738 @value{GDBN} fills in the rest of the word @samp{breakpoints}, since that is
1739 the only @code{info} subcommand beginning with @samp{bre}:
1742 (@value{GDBP}) info breakpoints
1746 You can either press @key{RET} at this point, to run the @code{info
1747 breakpoints} command, or backspace and enter something else, if
1748 @samp{breakpoints} does not look like the command you expected. (If you
1749 were sure you wanted @code{info breakpoints} in the first place, you
1750 might as well just type @key{RET} immediately after @samp{info bre},
1751 to exploit command abbreviations rather than command completion).
1753 If there is more than one possibility for the next word when you press
1754 @key{TAB}, @value{GDBN} sounds a bell. You can either supply more
1755 characters and try again, or just press @key{TAB} a second time;
1756 @value{GDBN} displays all the possible completions for that word. For
1757 example, you might want to set a breakpoint on a subroutine whose name
1758 begins with @samp{make_}, but when you type @kbd{b make_@key{TAB}} @value{GDBN}
1759 just sounds the bell. Typing @key{TAB} again displays all the
1760 function names in your program that begin with those characters, for
1764 (@value{GDBP}) b make_ @key{TAB}
1765 @exdent @value{GDBN} sounds bell; press @key{TAB} again, to see:
1766 make_a_section_from_file make_environ
1767 make_abs_section make_function_type
1768 make_blockvector make_pointer_type
1769 make_cleanup make_reference_type
1770 make_command make_symbol_completion_list
1771 (@value{GDBP}) b make_
1775 After displaying the available possibilities, @value{GDBN} copies your
1776 partial input (@samp{b make_} in the example) so you can finish the
1779 If you just want to see the list of alternatives in the first place, you
1780 can press @kbd{M-?} rather than pressing @key{TAB} twice. @kbd{M-?}
1781 means @kbd{@key{META} ?}. You can type this either by holding down a
1782 key designated as the @key{META} shift on your keyboard (if there is
1783 one) while typing @kbd{?}, or as @key{ESC} followed by @kbd{?}.
1785 If the number of possible completions is large, @value{GDBN} will
1786 print as much of the list as it has collected, as well as a message
1787 indicating that the list may be truncated.
1790 (@value{GDBP}) b m@key{TAB}@key{TAB}
1792 <... the rest of the possible completions ...>
1793 *** List may be truncated, max-completions reached. ***
1798 This behavior can be controlled with the following commands:
1801 @kindex set max-completions
1802 @item set max-completions @var{limit}
1803 @itemx set max-completions unlimited
1804 Set the maximum number of completion candidates. @value{GDBN} will
1805 stop looking for more completions once it collects this many candidates.
1806 This is useful when completing on things like function names as collecting
1807 all the possible candidates can be time consuming.
1808 The default value is 200. A value of zero disables tab-completion.
1809 Note that setting either no limit or a very large limit can make
1811 @kindex show max-completions
1812 @item show max-completions
1813 Show the maximum number of candidates that @value{GDBN} will collect and show
1817 @cindex quotes in commands
1818 @cindex completion of quoted strings
1819 Sometimes the string you need, while logically a ``word'', may contain
1820 parentheses or other characters that @value{GDBN} normally excludes from
1821 its notion of a word. To permit word completion to work in this
1822 situation, you may enclose words in @code{'} (single quote marks) in
1823 @value{GDBN} commands.
1825 A likely situation where you might need this is in typing an
1826 expression that involves a C@t{++} symbol name with template
1827 parameters. This is because when completing expressions, GDB treats
1828 the @samp{<} character as word delimiter, assuming that it's the
1829 less-than comparison operator (@pxref{C Operators, , C and C@t{++}
1832 For example, when you want to call a C@t{++} template function
1833 interactively using the @code{print} or @code{call} commands, you may
1834 need to distinguish whether you mean the version of @code{name} that
1835 was specialized for @code{int}, @code{name<int>()}, or the version
1836 that was specialized for @code{float}, @code{name<float>()}. To use
1837 the word-completion facilities in this situation, type a single quote
1838 @code{'} at the beginning of the function name. This alerts
1839 @value{GDBN} that it may need to consider more information than usual
1840 when you press @key{TAB} or @kbd{M-?} to request word completion:
1843 (@value{GDBP}) p 'func< @kbd{M-?}
1844 func<int>() func<float>()
1845 (@value{GDBP}) p 'func<
1848 When setting breakpoints however (@pxref{Specify Location}), you don't
1849 usually need to type a quote before the function name, because
1850 @value{GDBN} understands that you want to set a breakpoint on a
1854 (@value{GDBP}) b func< @kbd{M-?}
1855 func<int>() func<float>()
1856 (@value{GDBP}) b func<
1859 This is true even in the case of typing the name of C@t{++} overloaded
1860 functions (multiple definitions of the same function, distinguished by
1861 argument type). For example, when you want to set a breakpoint you
1862 don't need to distinguish whether you mean the version of @code{name}
1863 that takes an @code{int} parameter, @code{name(int)}, or the version
1864 that takes a @code{float} parameter, @code{name(float)}.
1867 (@value{GDBP}) b bubble( @kbd{M-?}
1868 bubble(int) bubble(double)
1869 (@value{GDBP}) b bubble(dou @kbd{M-?}
1873 See @ref{quoting names} for a description of other scenarios that
1876 For more information about overloaded functions, see @ref{C Plus Plus
1877 Expressions, ,C@t{++} Expressions}. You can use the command @code{set
1878 overload-resolution off} to disable overload resolution;
1879 see @ref{Debugging C Plus Plus, ,@value{GDBN} Features for C@t{++}}.
1881 @cindex completion of structure field names
1882 @cindex structure field name completion
1883 @cindex completion of union field names
1884 @cindex union field name completion
1885 When completing in an expression which looks up a field in a
1886 structure, @value{GDBN} also tries@footnote{The completer can be
1887 confused by certain kinds of invalid expressions. Also, it only
1888 examines the static type of the expression, not the dynamic type.} to
1889 limit completions to the field names available in the type of the
1893 (@value{GDBP}) p gdb_stdout.@kbd{M-?}
1894 magic to_fputs to_rewind
1895 to_data to_isatty to_write
1896 to_delete to_put to_write_async_safe
1901 This is because the @code{gdb_stdout} is a variable of the type
1902 @code{struct ui_file} that is defined in @value{GDBN} sources as
1909 ui_file_flush_ftype *to_flush;
1910 ui_file_write_ftype *to_write;
1911 ui_file_write_async_safe_ftype *to_write_async_safe;
1912 ui_file_fputs_ftype *to_fputs;
1913 ui_file_read_ftype *to_read;
1914 ui_file_delete_ftype *to_delete;
1915 ui_file_isatty_ftype *to_isatty;
1916 ui_file_rewind_ftype *to_rewind;
1917 ui_file_put_ftype *to_put;
1922 @node Command Options
1923 @section Command options
1925 @cindex command options
1926 Some commands accept options starting with a leading dash. For
1927 example, @code{print -pretty}. Similarly to command names, you can
1928 abbreviate a @value{GDBN} option to the first few letters of the
1929 option name, if that abbreviation is unambiguous, and you can also use
1930 the @key{TAB} key to get @value{GDBN} to fill out the rest of a word
1931 in an option (or to show you the alternatives available, if there is
1932 more than one possibility).
1934 @cindex command options, raw input
1935 Some commands take raw input as argument. For example, the print
1936 command processes arbitrary expressions in any of the languages
1937 supported by @value{GDBN}. With such commands, because raw input may
1938 start with a leading dash that would be confused with an option or any
1939 of its abbreviations, e.g.@: @code{print -r} (short for @code{print
1940 -raw} or printing negative @code{r}?), if you specify any command
1941 option, then you must use a double-dash (@code{--}) delimiter to
1942 indicate the end of options.
1944 @cindex command options, boolean
1946 Some options are described as accepting an argument which can be
1947 either @code{on} or @code{off}. These are known as @dfn{boolean
1948 options}. Similarly to boolean settings commands---@code{on} and
1949 @code{off} are the typical values, but any of @code{1}, @code{yes} and
1950 @code{enable} can also be used as ``true'' value, and any of @code{0},
1951 @code{no} and @code{disable} can also be used as ``false'' value. You
1952 can also omit a ``true'' value, as it is implied by default.
1954 For example, these are equivalent:
1957 (@value{GDBP}) print -object on -pretty off -element unlimited -- *myptr
1958 (@value{GDBP}) p -o -p 0 -e u -- *myptr
1961 You can discover the set of options some command accepts by completing
1962 on @code{-} after the command name. For example:
1965 (@value{GDBP}) print -@key{TAB}@key{TAB}
1966 -address -max-depth -repeats -vtbl
1967 -array -null-stop -static-members
1968 -array-indexes -object -symbol
1969 -elements -pretty -union
1972 Completion will in some cases guide you with a suggestion of what kind
1973 of argument an option expects. For example:
1976 (@value{GDBP}) print -elements @key{TAB}@key{TAB}
1980 Here, the option expects a number (e.g., @code{100}), not literal
1981 @code{NUMBER}. Such metasyntactical arguments are always presented in
1984 (For more on using the @code{print} command, see @ref{Data, ,Examining
1988 @section Getting Help
1989 @cindex online documentation
1992 You can always ask @value{GDBN} itself for information on its commands,
1993 using the command @code{help}.
1996 @kindex h @r{(@code{help})}
1999 You can use @code{help} (abbreviated @code{h}) with no arguments to
2000 display a short list of named classes of commands:
2004 List of classes of commands:
2006 aliases -- Aliases of other commands
2007 breakpoints -- Making program stop at certain points
2008 data -- Examining data
2009 files -- Specifying and examining files
2010 internals -- Maintenance commands
2011 obscure -- Obscure features
2012 running -- Running the program
2013 stack -- Examining the stack
2014 status -- Status inquiries
2015 support -- Support facilities
2016 tracepoints -- Tracing of program execution without
2017 stopping the program
2018 user-defined -- User-defined commands
2020 Type "help" followed by a class name for a list of
2021 commands in that class.
2022 Type "help" followed by command name for full
2024 Command name abbreviations are allowed if unambiguous.
2027 @c the above line break eliminates huge line overfull...
2029 @item help @var{class}
2030 Using one of the general help classes as an argument, you can get a
2031 list of the individual commands in that class. For example, here is the
2032 help display for the class @code{status}:
2035 (@value{GDBP}) help status
2040 @c Line break in "show" line falsifies real output, but needed
2041 @c to fit in smallbook page size.
2042 info -- Generic command for showing things
2043 about the program being debugged
2044 show -- Generic command for showing things
2047 Type "help" followed by command name for full
2049 Command name abbreviations are allowed if unambiguous.
2053 @item help @var{command}
2054 With a command name as @code{help} argument, @value{GDBN} displays a
2055 short paragraph on how to use that command.
2058 @item apropos [-v] @var{regexp}
2059 The @code{apropos} command searches through all of the @value{GDBN}
2060 commands, and their documentation, for the regular expression specified in
2061 @var{args}. It prints out all matches found. The optional flag @samp{-v},
2062 which stands for @samp{verbose}, indicates to output the full documentation
2063 of the matching commands and highlight the parts of the documentation
2064 matching @var{regexp}. For example:
2075 alias -- Define a new command that is an alias of an existing command
2076 aliases -- Aliases of other commands
2077 d -- Delete some breakpoints or auto-display expressions
2078 del -- Delete some breakpoints or auto-display expressions
2079 delete -- Delete some breakpoints or auto-display expressions
2087 apropos -v cut.*thread apply
2091 results in the below output, where @samp{cut for 'thread apply}
2092 is highlighted if styling is enabled.
2096 taas -- Apply a command to all threads (ignoring errors
2099 shortcut for 'thread apply all -s COMMAND'
2101 tfaas -- Apply a command to all frames of all threads
2102 (ignoring errors and empty output).
2103 Usage: tfaas COMMAND
2104 shortcut for 'thread apply all -s frame apply all -s COMMAND'
2109 @item complete @var{args}
2110 The @code{complete @var{args}} command lists all the possible completions
2111 for the beginning of a command. Use @var{args} to specify the beginning of the
2112 command you want completed. For example:
2118 @noindent results in:
2129 @noindent This is intended for use by @sc{gnu} Emacs.
2132 In addition to @code{help}, you can use the @value{GDBN} commands @code{info}
2133 and @code{show} to inquire about the state of your program, or the state
2134 of @value{GDBN} itself. Each command supports many topics of inquiry; this
2135 manual introduces each of them in the appropriate context. The listings
2136 under @code{info} and under @code{show} in the Command, Variable, and
2137 Function Index point to all the sub-commands. @xref{Command and Variable
2143 @kindex i @r{(@code{info})}
2145 This command (abbreviated @code{i}) is for describing the state of your
2146 program. For example, you can show the arguments passed to a function
2147 with @code{info args}, list the registers currently in use with @code{info
2148 registers}, or list the breakpoints you have set with @code{info breakpoints}.
2149 You can get a complete list of the @code{info} sub-commands with
2150 @w{@code{help info}}.
2154 You can assign the result of an expression to an environment variable with
2155 @code{set}. For example, you can set the @value{GDBN} prompt to a $-sign with
2156 @code{set prompt $}.
2160 In contrast to @code{info}, @code{show} is for describing the state of
2161 @value{GDBN} itself.
2162 You can change most of the things you can @code{show}, by using the
2163 related command @code{set}; for example, you can control what number
2164 system is used for displays with @code{set radix}, or simply inquire
2165 which is currently in use with @code{show radix}.
2168 To display all the settable parameters and their current
2169 values, you can use @code{show} with no arguments; you may also use
2170 @code{info set}. Both commands produce the same display.
2171 @c FIXME: "info set" violates the rule that "info" is for state of
2172 @c FIXME...program. Ck w/ GNU: "info set" to be called something else,
2173 @c FIXME...or change desc of rule---eg "state of prog and debugging session"?
2177 Here are several miscellaneous @code{show} subcommands, all of which are
2178 exceptional in lacking corresponding @code{set} commands:
2181 @kindex show version
2182 @cindex @value{GDBN} version number
2184 Show what version of @value{GDBN} is running. You should include this
2185 information in @value{GDBN} bug-reports. If multiple versions of
2186 @value{GDBN} are in use at your site, you may need to determine which
2187 version of @value{GDBN} you are running; as @value{GDBN} evolves, new
2188 commands are introduced, and old ones may wither away. Also, many
2189 system vendors ship variant versions of @value{GDBN}, and there are
2190 variant versions of @value{GDBN} in @sc{gnu}/Linux distributions as well.
2191 The version number is the same as the one announced when you start
2194 @kindex show copying
2195 @kindex info copying
2196 @cindex display @value{GDBN} copyright
2199 Display information about permission for copying @value{GDBN}.
2201 @kindex show warranty
2202 @kindex info warranty
2204 @itemx info warranty
2205 Display the @sc{gnu} ``NO WARRANTY'' statement, or a warranty,
2206 if your version of @value{GDBN} comes with one.
2208 @kindex show configuration
2209 @item show configuration
2210 Display detailed information about the way @value{GDBN} was configured
2211 when it was built. This displays the optional arguments passed to the
2212 @file{configure} script and also configuration parameters detected
2213 automatically by @command{configure}. When reporting a @value{GDBN}
2214 bug (@pxref{GDB Bugs}), it is important to include this information in
2220 @chapter Running Programs Under @value{GDBN}
2222 When you run a program under @value{GDBN}, you must first generate
2223 debugging information when you compile it.
2225 You may start @value{GDBN} with its arguments, if any, in an environment
2226 of your choice. If you are doing native debugging, you may redirect
2227 your program's input and output, debug an already running process, or
2228 kill a child process.
2231 * Compilation:: Compiling for debugging
2232 * Starting:: Starting your program
2233 * Arguments:: Your program's arguments
2234 * Environment:: Your program's environment
2236 * Working Directory:: Your program's working directory
2237 * Input/Output:: Your program's input and output
2238 * Attach:: Debugging an already-running process
2239 * Kill Process:: Killing the child process
2241 * Inferiors and Programs:: Debugging multiple inferiors and programs
2242 * Threads:: Debugging programs with multiple threads
2243 * Forks:: Debugging forks
2244 * Checkpoint/Restart:: Setting a @emph{bookmark} to return to later
2248 @section Compiling for Debugging
2250 In order to debug a program effectively, you need to generate
2251 debugging information when you compile it. This debugging information
2252 is stored in the object file; it describes the data type of each
2253 variable or function and the correspondence between source line numbers
2254 and addresses in the executable code.
2256 To request debugging information, specify the @samp{-g} option when you run
2259 Programs that are to be shipped to your customers are compiled with
2260 optimizations, using the @samp{-O} compiler option. However, some
2261 compilers are unable to handle the @samp{-g} and @samp{-O} options
2262 together. Using those compilers, you cannot generate optimized
2263 executables containing debugging information.
2265 @value{NGCC}, the @sc{gnu} C/C@t{++} compiler, supports @samp{-g} with or
2266 without @samp{-O}, making it possible to debug optimized code. We
2267 recommend that you @emph{always} use @samp{-g} whenever you compile a
2268 program. You may think your program is correct, but there is no sense
2269 in pushing your luck. For more information, see @ref{Optimized Code}.
2271 Older versions of the @sc{gnu} C compiler permitted a variant option
2272 @w{@samp{-gg}} for debugging information. @value{GDBN} no longer supports this
2273 format; if your @sc{gnu} C compiler has this option, do not use it.
2275 @value{GDBN} knows about preprocessor macros and can show you their
2276 expansion (@pxref{Macros}). Most compilers do not include information
2277 about preprocessor macros in the debugging information if you specify
2278 the @option{-g} flag alone. Version 3.1 and later of @value{NGCC},
2279 the @sc{gnu} C compiler, provides macro information if you are using
2280 the DWARF debugging format, and specify the option @option{-g3}.
2282 @xref{Debugging Options,,Options for Debugging Your Program or GCC,
2283 gcc, Using the @sc{gnu} Compiler Collection (GCC)}, for more
2284 information on @value{NGCC} options affecting debug information.
2286 You will have the best debugging experience if you use the latest
2287 version of the DWARF debugging format that your compiler supports.
2288 DWARF is currently the most expressive and best supported debugging
2289 format in @value{GDBN}.
2293 @section Starting your Program
2299 @kindex r @r{(@code{run})}
2302 Use the @code{run} command to start your program under @value{GDBN}.
2303 You must first specify the program name with an argument to
2304 @value{GDBN} (@pxref{Invocation, ,Getting In and Out of
2305 @value{GDBN}}), or by using the @code{file} or @code{exec-file}
2306 command (@pxref{Files, ,Commands to Specify Files}).
2310 If you are running your program in an execution environment that
2311 supports processes, @code{run} creates an inferior process and makes
2312 that process run your program. In some environments without processes,
2313 @code{run} jumps to the start of your program. Other targets,
2314 like @samp{remote}, are always running. If you get an error
2315 message like this one:
2318 The "remote" target does not support "run".
2319 Try "help target" or "continue".
2323 then use @code{continue} to run your program. You may need @code{load}
2324 first (@pxref{load}).
2326 The execution of a program is affected by certain information it
2327 receives from its superior. @value{GDBN} provides ways to specify this
2328 information, which you must do @emph{before} starting your program. (You
2329 can change it after starting your program, but such changes only affect
2330 your program the next time you start it.) This information may be
2331 divided into four categories:
2334 @item The @emph{arguments.}
2335 Specify the arguments to give your program as the arguments of the
2336 @code{run} command. If a shell is available on your target, the shell
2337 is used to pass the arguments, so that you may use normal conventions
2338 (such as wildcard expansion or variable substitution) in describing
2340 In Unix systems, you can control which shell is used with the
2341 @code{SHELL} environment variable. If you do not define @code{SHELL},
2342 @value{GDBN} uses the default shell (@file{/bin/sh}). You can disable
2343 use of any shell with the @code{set startup-with-shell} command (see
2346 @item The @emph{environment.}
2347 Your program normally inherits its environment from @value{GDBN}, but you can
2348 use the @value{GDBN} commands @code{set environment} and @code{unset
2349 environment} to change parts of the environment that affect
2350 your program. @xref{Environment, ,Your Program's Environment}.
2352 @item The @emph{working directory.}
2353 You can set your program's working directory with the command
2354 @kbd{set cwd}. If you do not set any working directory with this
2355 command, your program will inherit @value{GDBN}'s working directory if
2356 native debugging, or the remote server's working directory if remote
2357 debugging. @xref{Working Directory, ,Your Program's Working
2360 @item The @emph{standard input and output.}
2361 Your program normally uses the same device for standard input and
2362 standard output as @value{GDBN} is using. You can redirect input and output
2363 in the @code{run} command line, or you can use the @code{tty} command to
2364 set a different device for your program.
2365 @xref{Input/Output, ,Your Program's Input and Output}.
2368 @emph{Warning:} While input and output redirection work, you cannot use
2369 pipes to pass the output of the program you are debugging to another
2370 program; if you attempt this, @value{GDBN} is likely to wind up debugging the
2374 When you issue the @code{run} command, your program begins to execute
2375 immediately. @xref{Stopping, ,Stopping and Continuing}, for discussion
2376 of how to arrange for your program to stop. Once your program has
2377 stopped, you may call functions in your program, using the @code{print}
2378 or @code{call} commands. @xref{Data, ,Examining Data}.
2380 If the modification time of your symbol file has changed since the last
2381 time @value{GDBN} read its symbols, @value{GDBN} discards its symbol
2382 table, and reads it again. When it does this, @value{GDBN} tries to retain
2383 your current breakpoints.
2388 @cindex run to main procedure
2389 The name of the main procedure can vary from language to language.
2390 With C or C@t{++}, the main procedure name is always @code{main}, but
2391 other languages such as Ada do not require a specific name for their
2392 main procedure. The debugger provides a convenient way to start the
2393 execution of the program and to stop at the beginning of the main
2394 procedure, depending on the language used.
2396 The @samp{start} command does the equivalent of setting a temporary
2397 breakpoint at the beginning of the main procedure and then invoking
2398 the @samp{run} command.
2400 @cindex elaboration phase
2401 Some programs contain an @dfn{elaboration} phase where some startup code is
2402 executed before the main procedure is called. This depends on the
2403 languages used to write your program. In C@t{++}, for instance,
2404 constructors for static and global objects are executed before
2405 @code{main} is called. It is therefore possible that the debugger stops
2406 before reaching the main procedure. However, the temporary breakpoint
2407 will remain to halt execution.
2409 Specify the arguments to give to your program as arguments to the
2410 @samp{start} command. These arguments will be given verbatim to the
2411 underlying @samp{run} command. Note that the same arguments will be
2412 reused if no argument is provided during subsequent calls to
2413 @samp{start} or @samp{run}.
2415 It is sometimes necessary to debug the program during elaboration. In
2416 these cases, using the @code{start} command would stop the execution
2417 of your program too late, as the program would have already completed
2418 the elaboration phase. Under these circumstances, either insert
2419 breakpoints in your elaboration code before running your program or
2420 use the @code{starti} command.
2424 @cindex run to first instruction
2425 The @samp{starti} command does the equivalent of setting a temporary
2426 breakpoint at the first instruction of a program's execution and then
2427 invoking the @samp{run} command. For programs containing an
2428 elaboration phase, the @code{starti} command will stop execution at
2429 the start of the elaboration phase.
2431 @anchor{set exec-wrapper}
2432 @kindex set exec-wrapper
2433 @item set exec-wrapper @var{wrapper}
2434 @itemx show exec-wrapper
2435 @itemx unset exec-wrapper
2436 When @samp{exec-wrapper} is set, the specified wrapper is used to
2437 launch programs for debugging. @value{GDBN} starts your program
2438 with a shell command of the form @kbd{exec @var{wrapper}
2439 @var{program}}. Quoting is added to @var{program} and its
2440 arguments, but not to @var{wrapper}, so you should add quotes if
2441 appropriate for your shell. The wrapper runs until it executes
2442 your program, and then @value{GDBN} takes control.
2444 You can use any program that eventually calls @code{execve} with
2445 its arguments as a wrapper. Several standard Unix utilities do
2446 this, e.g.@: @code{env} and @code{nohup}. Any Unix shell script ending
2447 with @code{exec "$@@"} will also work.
2449 For example, you can use @code{env} to pass an environment variable to
2450 the debugged program, without setting the variable in your shell's
2454 (@value{GDBP}) set exec-wrapper env 'LD_PRELOAD=libtest.so'
2458 This command is available when debugging locally on most targets, excluding
2459 @sc{djgpp}, Cygwin, MS Windows, and QNX Neutrino.
2461 @kindex set startup-with-shell
2462 @anchor{set startup-with-shell}
2463 @item set startup-with-shell
2464 @itemx set startup-with-shell on
2465 @itemx set startup-with-shell off
2466 @itemx show startup-with-shell
2467 On Unix systems, by default, if a shell is available on your target,
2468 @value{GDBN}) uses it to start your program. Arguments of the
2469 @code{run} command are passed to the shell, which does variable
2470 substitution, expands wildcard characters and performs redirection of
2471 I/O. In some circumstances, it may be useful to disable such use of a
2472 shell, for example, when debugging the shell itself or diagnosing
2473 startup failures such as:
2477 Starting program: ./a.out
2478 During startup program terminated with signal SIGSEGV, Segmentation fault.
2482 which indicates the shell or the wrapper specified with
2483 @samp{exec-wrapper} crashed, not your program. Most often, this is
2484 caused by something odd in your shell's non-interactive mode
2485 initialization file---such as @file{.cshrc} for C-shell,
2486 $@file{.zshenv} for the Z shell, or the file specified in the
2487 @samp{BASH_ENV} environment variable for BASH.
2489 @anchor{set auto-connect-native-target}
2490 @kindex set auto-connect-native-target
2491 @item set auto-connect-native-target
2492 @itemx set auto-connect-native-target on
2493 @itemx set auto-connect-native-target off
2494 @itemx show auto-connect-native-target
2496 By default, if not connected to any target yet (e.g., with
2497 @code{target remote}), the @code{run} command starts your program as a
2498 native process under @value{GDBN}, on your local machine. If you're
2499 sure you don't want to debug programs on your local machine, you can
2500 tell @value{GDBN} to not connect to the native target automatically
2501 with the @code{set auto-connect-native-target off} command.
2503 If @code{on}, which is the default, and if @value{GDBN} is not
2504 connected to a target already, the @code{run} command automaticaly
2505 connects to the native target, if one is available.
2507 If @code{off}, and if @value{GDBN} is not connected to a target
2508 already, the @code{run} command fails with an error:
2512 Don't know how to run. Try "help target".
2515 If @value{GDBN} is already connected to a target, @value{GDBN} always
2516 uses it with the @code{run} command.
2518 In any case, you can explicitly connect to the native target with the
2519 @code{target native} command. For example,
2522 (@value{GDBP}) set auto-connect-native-target off
2524 Don't know how to run. Try "help target".
2525 (@value{GDBP}) target native
2527 Starting program: ./a.out
2528 [Inferior 1 (process 10421) exited normally]
2531 In case you connected explicitly to the @code{native} target,
2532 @value{GDBN} remains connected even if all inferiors exit, ready for
2533 the next @code{run} command. Use the @code{disconnect} command to
2536 Examples of other commands that likewise respect the
2537 @code{auto-connect-native-target} setting: @code{attach}, @code{info
2538 proc}, @code{info os}.
2540 @kindex set disable-randomization
2541 @item set disable-randomization
2542 @itemx set disable-randomization on
2543 This option (enabled by default in @value{GDBN}) will turn off the native
2544 randomization of the virtual address space of the started program. This option
2545 is useful for multiple debugging sessions to make the execution better
2546 reproducible and memory addresses reusable across debugging sessions.
2548 This feature is implemented only on certain targets, including @sc{gnu}/Linux.
2549 On @sc{gnu}/Linux you can get the same behavior using
2552 (@value{GDBP}) set exec-wrapper setarch `uname -m` -R
2555 @item set disable-randomization off
2556 Leave the behavior of the started executable unchanged. Some bugs rear their
2557 ugly heads only when the program is loaded at certain addresses. If your bug
2558 disappears when you run the program under @value{GDBN}, that might be because
2559 @value{GDBN} by default disables the address randomization on platforms, such
2560 as @sc{gnu}/Linux, which do that for stand-alone programs. Use @kbd{set
2561 disable-randomization off} to try to reproduce such elusive bugs.
2563 On targets where it is available, virtual address space randomization
2564 protects the programs against certain kinds of security attacks. In these
2565 cases the attacker needs to know the exact location of a concrete executable
2566 code. Randomizing its location makes it impossible to inject jumps misusing
2567 a code at its expected addresses.
2569 Prelinking shared libraries provides a startup performance advantage but it
2570 makes addresses in these libraries predictable for privileged processes by
2571 having just unprivileged access at the target system. Reading the shared
2572 library binary gives enough information for assembling the malicious code
2573 misusing it. Still even a prelinked shared library can get loaded at a new
2574 random address just requiring the regular relocation process during the
2575 startup. Shared libraries not already prelinked are always loaded at
2576 a randomly chosen address.
2578 Position independent executables (PIE) contain position independent code
2579 similar to the shared libraries and therefore such executables get loaded at
2580 a randomly chosen address upon startup. PIE executables always load even
2581 already prelinked shared libraries at a random address. You can build such
2582 executable using @command{gcc -fPIE -pie}.
2584 Heap (malloc storage), stack and custom mmap areas are always placed randomly
2585 (as long as the randomization is enabled).
2587 @item show disable-randomization
2588 Show the current setting of the explicit disable of the native randomization of
2589 the virtual address space of the started program.
2594 @section Your Program's Arguments
2596 @cindex arguments (to your program)
2597 The arguments to your program can be specified by the arguments of the
2599 They are passed to a shell, which expands wildcard characters and
2600 performs redirection of I/O, and thence to your program. Your
2601 @code{SHELL} environment variable (if it exists) specifies what shell
2602 @value{GDBN} uses. If you do not define @code{SHELL}, @value{GDBN} uses
2603 the default shell (@file{/bin/sh} on Unix).
2605 On non-Unix systems, the program is usually invoked directly by
2606 @value{GDBN}, which emulates I/O redirection via the appropriate system
2607 calls, and the wildcard characters are expanded by the startup code of
2608 the program, not by the shell.
2610 @code{run} with no arguments uses the same arguments used by the previous
2611 @code{run}, or those set by the @code{set args} command.
2616 Specify the arguments to be used the next time your program is run. If
2617 @code{set args} has no arguments, @code{run} executes your program
2618 with no arguments. Once you have run your program with arguments,
2619 using @code{set args} before the next @code{run} is the only way to run
2620 it again without arguments.
2624 Show the arguments to give your program when it is started.
2628 @section Your Program's Environment
2630 @cindex environment (of your program)
2631 The @dfn{environment} consists of a set of environment variables and
2632 their values. Environment variables conventionally record such things as
2633 your user name, your home directory, your terminal type, and your search
2634 path for programs to run. Usually you set up environment variables with
2635 the shell and they are inherited by all the other programs you run. When
2636 debugging, it can be useful to try running your program with a modified
2637 environment without having to start @value{GDBN} over again.
2641 @item path @var{directory}
2642 Add @var{directory} to the front of the @code{PATH} environment variable
2643 (the search path for executables) that will be passed to your program.
2644 The value of @code{PATH} used by @value{GDBN} does not change.
2645 You may specify several directory names, separated by whitespace or by a
2646 system-dependent separator character (@samp{:} on Unix, @samp{;} on
2647 MS-DOS and MS-Windows). If @var{directory} is already in the path, it
2648 is moved to the front, so it is searched sooner.
2650 You can use the string @samp{$cwd} to refer to whatever is the current
2651 working directory at the time @value{GDBN} searches the path. If you
2652 use @samp{.} instead, it refers to the directory where you executed the
2653 @code{path} command. @value{GDBN} replaces @samp{.} in the
2654 @var{directory} argument (with the current path) before adding
2655 @var{directory} to the search path.
2656 @c 'path' is explicitly nonrepeatable, but RMS points out it is silly to
2657 @c document that, since repeating it would be a no-op.
2661 Display the list of search paths for executables (the @code{PATH}
2662 environment variable).
2664 @kindex show environment
2665 @item show environment @r{[}@var{varname}@r{]}
2666 Print the value of environment variable @var{varname} to be given to
2667 your program when it starts. If you do not supply @var{varname},
2668 print the names and values of all environment variables to be given to
2669 your program. You can abbreviate @code{environment} as @code{env}.
2671 @kindex set environment
2672 @anchor{set environment}
2673 @item set environment @var{varname} @r{[}=@var{value}@r{]}
2674 Set environment variable @var{varname} to @var{value}. The value
2675 changes for your program (and the shell @value{GDBN} uses to launch
2676 it), not for @value{GDBN} itself. The @var{value} may be any string; the
2677 values of environment variables are just strings, and any
2678 interpretation is supplied by your program itself. The @var{value}
2679 parameter is optional; if it is eliminated, the variable is set to a
2681 @c "any string" here does not include leading, trailing
2682 @c blanks. Gnu asks: does anyone care?
2684 For example, this command:
2691 tells the debugged program, when subsequently run, that its user is named
2692 @samp{foo}. (The spaces around @samp{=} are used for clarity here; they
2693 are not actually required.)
2695 Note that on Unix systems, @value{GDBN} runs your program via a shell,
2696 which also inherits the environment set with @code{set environment}.
2697 If necessary, you can avoid that by using the @samp{env} program as a
2698 wrapper instead of using @code{set environment}. @xref{set
2699 exec-wrapper}, for an example doing just that.
2701 Environment variables that are set by the user are also transmitted to
2702 @command{gdbserver} to be used when starting the remote inferior.
2703 @pxref{QEnvironmentHexEncoded}.
2705 @kindex unset environment
2706 @anchor{unset environment}
2707 @item unset environment @var{varname}
2708 Remove variable @var{varname} from the environment to be passed to your
2709 program. This is different from @samp{set env @var{varname} =};
2710 @code{unset environment} removes the variable from the environment,
2711 rather than assigning it an empty value.
2713 Environment variables that are unset by the user are also unset on
2714 @command{gdbserver} when starting the remote inferior.
2715 @pxref{QEnvironmentUnset}.
2718 @emph{Warning:} On Unix systems, @value{GDBN} runs your program using
2719 the shell indicated by your @code{SHELL} environment variable if it
2720 exists (or @code{/bin/sh} if not). If your @code{SHELL} variable
2721 names a shell that runs an initialization file when started
2722 non-interactively---such as @file{.cshrc} for C-shell, $@file{.zshenv}
2723 for the Z shell, or the file specified in the @samp{BASH_ENV}
2724 environment variable for BASH---any variables you set in that file
2725 affect your program. You may wish to move setting of environment
2726 variables to files that are only run when you sign on, such as
2727 @file{.login} or @file{.profile}.
2729 @node Working Directory
2730 @section Your Program's Working Directory
2732 @cindex working directory (of your program)
2733 Each time you start your program with @code{run}, the inferior will be
2734 initialized with the current working directory specified by the
2735 @kbd{set cwd} command. If no directory has been specified by this
2736 command, then the inferior will inherit @value{GDBN}'s current working
2737 directory as its working directory if native debugging, or it will
2738 inherit the remote server's current working directory if remote
2743 @cindex change inferior's working directory
2744 @anchor{set cwd command}
2745 @item set cwd @r{[}@var{directory}@r{]}
2746 Set the inferior's working directory to @var{directory}, which will be
2747 @code{glob}-expanded in order to resolve tildes (@file{~}). If no
2748 argument has been specified, the command clears the setting and resets
2749 it to an empty state. This setting has no effect on @value{GDBN}'s
2750 working directory, and it only takes effect the next time you start
2751 the inferior. The @file{~} in @var{directory} is a short for the
2752 @dfn{home directory}, usually pointed to by the @env{HOME} environment
2753 variable. On MS-Windows, if @env{HOME} is not defined, @value{GDBN}
2754 uses the concatenation of @env{HOMEDRIVE} and @env{HOMEPATH} as
2757 You can also change @value{GDBN}'s current working directory by using
2758 the @code{cd} command.
2762 @cindex show inferior's working directory
2764 Show the inferior's working directory. If no directory has been
2765 specified by @kbd{set cwd}, then the default inferior's working
2766 directory is the same as @value{GDBN}'s working directory.
2769 @cindex change @value{GDBN}'s working directory
2771 @item cd @r{[}@var{directory}@r{]}
2772 Set the @value{GDBN} working directory to @var{directory}. If not
2773 given, @var{directory} uses @file{'~'}.
2775 The @value{GDBN} working directory serves as a default for the
2776 commands that specify files for @value{GDBN} to operate on.
2777 @xref{Files, ,Commands to Specify Files}.
2778 @xref{set cwd command}.
2782 Print the @value{GDBN} working directory.
2785 It is generally impossible to find the current working directory of
2786 the process being debugged (since a program can change its directory
2787 during its run). If you work on a system where @value{GDBN} supports
2788 the @code{info proc} command (@pxref{Process Information}), you can
2789 use the @code{info proc} command to find out the
2790 current working directory of the debuggee.
2793 @section Your Program's Input and Output
2798 By default, the program you run under @value{GDBN} does input and output to
2799 the same terminal that @value{GDBN} uses. @value{GDBN} switches the terminal
2800 to its own terminal modes to interact with you, but it records the terminal
2801 modes your program was using and switches back to them when you continue
2802 running your program.
2805 @kindex info terminal
2807 Displays information recorded by @value{GDBN} about the terminal modes your
2811 You can redirect your program's input and/or output using shell
2812 redirection with the @code{run} command. For example,
2819 starts your program, diverting its output to the file @file{outfile}.
2822 @cindex controlling terminal
2823 Another way to specify where your program should do input and output is
2824 with the @code{tty} command. This command accepts a file name as
2825 argument, and causes this file to be the default for future @code{run}
2826 commands. It also resets the controlling terminal for the child
2827 process, for future @code{run} commands. For example,
2834 directs that processes started with subsequent @code{run} commands
2835 default to do input and output on the terminal @file{/dev/ttyb} and have
2836 that as their controlling terminal.
2838 An explicit redirection in @code{run} overrides the @code{tty} command's
2839 effect on the input/output device, but not its effect on the controlling
2842 When you use the @code{tty} command or redirect input in the @code{run}
2843 command, only the input @emph{for your program} is affected. The input
2844 for @value{GDBN} still comes from your terminal. @code{tty} is an alias
2845 for @code{set inferior-tty}.
2847 @cindex inferior tty
2848 @cindex set inferior controlling terminal
2849 You can use the @code{show inferior-tty} command to tell @value{GDBN} to
2850 display the name of the terminal that will be used for future runs of your
2854 @item set inferior-tty [ @var{tty} ]
2855 @kindex set inferior-tty
2856 Set the tty for the program being debugged to @var{tty}. Omitting @var{tty}
2857 restores the default behavior, which is to use the same terminal as
2860 @item show inferior-tty
2861 @kindex show inferior-tty
2862 Show the current tty for the program being debugged.
2866 @section Debugging an Already-running Process
2871 @item attach @var{process-id}
2872 This command attaches to a running process---one that was started
2873 outside @value{GDBN}. (@code{info files} shows your active
2874 targets.) The command takes as argument a process ID. The usual way to
2875 find out the @var{process-id} of a Unix process is with the @code{ps} utility,
2876 or with the @samp{jobs -l} shell command.
2878 @code{attach} does not repeat if you press @key{RET} a second time after
2879 executing the command.
2882 To use @code{attach}, your program must be running in an environment
2883 which supports processes; for example, @code{attach} does not work for
2884 programs on bare-board targets that lack an operating system. You must
2885 also have permission to send the process a signal.
2887 When you use @code{attach}, the debugger finds the program running in
2888 the process first by looking in the current working directory, then (if
2889 the program is not found) by using the source file search path
2890 (@pxref{Source Path, ,Specifying Source Directories}). You can also use
2891 the @code{file} command to load the program. @xref{Files, ,Commands to
2894 The first thing @value{GDBN} does after arranging to debug the specified
2895 process is to stop it. You can examine and modify an attached process
2896 with all the @value{GDBN} commands that are ordinarily available when
2897 you start processes with @code{run}. You can insert breakpoints; you
2898 can step and continue; you can modify storage. If you would rather the
2899 process continue running, you may use the @code{continue} command after
2900 attaching @value{GDBN} to the process.
2905 When you have finished debugging the attached process, you can use the
2906 @code{detach} command to release it from @value{GDBN} control. Detaching
2907 the process continues its execution. After the @code{detach} command,
2908 that process and @value{GDBN} become completely independent once more, and you
2909 are ready to @code{attach} another process or start one with @code{run}.
2910 @code{detach} does not repeat if you press @key{RET} again after
2911 executing the command.
2914 If you exit @value{GDBN} while you have an attached process, you detach
2915 that process. If you use the @code{run} command, you kill that process.
2916 By default, @value{GDBN} asks for confirmation if you try to do either of these
2917 things; you can control whether or not you need to confirm by using the
2918 @code{set confirm} command (@pxref{Messages/Warnings, ,Optional Warnings and
2922 @section Killing the Child Process
2927 Kill the child process in which your program is running under @value{GDBN}.
2930 This command is useful if you wish to debug a core dump instead of a
2931 running process. @value{GDBN} ignores any core dump file while your program
2934 On some operating systems, a program cannot be executed outside @value{GDBN}
2935 while you have breakpoints set on it inside @value{GDBN}. You can use the
2936 @code{kill} command in this situation to permit running your program
2937 outside the debugger.
2939 The @code{kill} command is also useful if you wish to recompile and
2940 relink your program, since on many systems it is impossible to modify an
2941 executable file while it is running in a process. In this case, when you
2942 next type @code{run}, @value{GDBN} notices that the file has changed, and
2943 reads the symbol table again (while trying to preserve your current
2944 breakpoint settings).
2946 @node Inferiors and Programs
2947 @section Debugging Multiple Inferiors and Programs
2949 @value{GDBN} lets you run and debug multiple programs in a single
2950 session. In addition, @value{GDBN} on some systems may let you run
2951 several programs simultaneously (otherwise you have to exit from one
2952 before starting another). In the most general case, you can have
2953 multiple threads of execution in each of multiple processes, launched
2954 from multiple executables.
2957 @value{GDBN} represents the state of each program execution with an
2958 object called an @dfn{inferior}. An inferior typically corresponds to
2959 a process, but is more general and applies also to targets that do not
2960 have processes. Inferiors may be created before a process runs, and
2961 may be retained after a process exits. Inferiors have unique
2962 identifiers that are different from process ids. Usually each
2963 inferior will also have its own distinct address space, although some
2964 embedded targets may have several inferiors running in different parts
2965 of a single address space. Each inferior may in turn have multiple
2966 threads running in it.
2968 To find out what inferiors exist at any moment, use @w{@code{info
2972 @kindex info inferiors [ @var{id}@dots{} ]
2973 @item info inferiors
2974 Print a list of all inferiors currently being managed by @value{GDBN}.
2975 By default all inferiors are printed, but the argument @var{id}@dots{}
2976 -- a space separated list of inferior numbers -- can be used to limit
2977 the display to just the requested inferiors.
2979 @value{GDBN} displays for each inferior (in this order):
2983 the inferior number assigned by @value{GDBN}
2986 the target system's inferior identifier
2989 the name of the executable the inferior is running.
2994 An asterisk @samp{*} preceding the @value{GDBN} inferior number
2995 indicates the current inferior.
2999 @c end table here to get a little more width for example
3002 (@value{GDBP}) info inferiors
3003 Num Description Executable
3004 2 process 2307 hello
3005 * 1 process 3401 goodbye
3008 To switch focus between inferiors, use the @code{inferior} command:
3011 @kindex inferior @var{infno}
3012 @item inferior @var{infno}
3013 Make inferior number @var{infno} the current inferior. The argument
3014 @var{infno} is the inferior number assigned by @value{GDBN}, as shown
3015 in the first field of the @samp{info inferiors} display.
3018 @vindex $_inferior@r{, convenience variable}
3019 The debugger convenience variable @samp{$_inferior} contains the
3020 number of the current inferior. You may find this useful in writing
3021 breakpoint conditional expressions, command scripts, and so forth.
3022 @xref{Convenience Vars,, Convenience Variables}, for general
3023 information on convenience variables.
3025 You can get multiple executables into a debugging session via the
3026 @code{add-inferior} and @w{@code{clone-inferior}} commands. On some
3027 systems @value{GDBN} can add inferiors to the debug session
3028 automatically by following calls to @code{fork} and @code{exec}. To
3029 remove inferiors from the debugging session use the
3030 @w{@code{remove-inferiors}} command.
3033 @kindex add-inferior
3034 @item add-inferior [ -copies @var{n} ] [ -exec @var{executable} ]
3035 Adds @var{n} inferiors to be run using @var{executable} as the
3036 executable; @var{n} defaults to 1. If no executable is specified,
3037 the inferiors begins empty, with no program. You can still assign or
3038 change the program assigned to the inferior at any time by using the
3039 @code{file} command with the executable name as its argument.
3041 @kindex clone-inferior
3042 @item clone-inferior [ -copies @var{n} ] [ @var{infno} ]
3043 Adds @var{n} inferiors ready to execute the same program as inferior
3044 @var{infno}; @var{n} defaults to 1, and @var{infno} defaults to the
3045 number of the current inferior. This is a convenient command when you
3046 want to run another instance of the inferior you are debugging.
3049 (@value{GDBP}) info inferiors
3050 Num Description Executable
3051 * 1 process 29964 helloworld
3052 (@value{GDBP}) clone-inferior
3055 (@value{GDBP}) info inferiors
3056 Num Description Executable
3058 * 1 process 29964 helloworld
3061 You can now simply switch focus to inferior 2 and run it.
3063 @kindex remove-inferiors
3064 @item remove-inferiors @var{infno}@dots{}
3065 Removes the inferior or inferiors @var{infno}@dots{}. It is not
3066 possible to remove an inferior that is running with this command. For
3067 those, use the @code{kill} or @code{detach} command first.
3071 To quit debugging one of the running inferiors that is not the current
3072 inferior, you can either detach from it by using the @w{@code{detach
3073 inferior}} command (allowing it to run independently), or kill it
3074 using the @w{@code{kill inferiors}} command:
3077 @kindex detach inferiors @var{infno}@dots{}
3078 @item detach inferior @var{infno}@dots{}
3079 Detach from the inferior or inferiors identified by @value{GDBN}
3080 inferior number(s) @var{infno}@dots{}. Note that the inferior's entry
3081 still stays on the list of inferiors shown by @code{info inferiors},
3082 but its Description will show @samp{<null>}.
3084 @kindex kill inferiors @var{infno}@dots{}
3085 @item kill inferiors @var{infno}@dots{}
3086 Kill the inferior or inferiors identified by @value{GDBN} inferior
3087 number(s) @var{infno}@dots{}. Note that the inferior's entry still
3088 stays on the list of inferiors shown by @code{info inferiors}, but its
3089 Description will show @samp{<null>}.
3092 After the successful completion of a command such as @code{detach},
3093 @code{detach inferiors}, @code{kill} or @code{kill inferiors}, or after
3094 a normal process exit, the inferior is still valid and listed with
3095 @code{info inferiors}, ready to be restarted.
3098 To be notified when inferiors are started or exit under @value{GDBN}'s
3099 control use @w{@code{set print inferior-events}}:
3102 @kindex set print inferior-events
3103 @cindex print messages on inferior start and exit
3104 @item set print inferior-events
3105 @itemx set print inferior-events on
3106 @itemx set print inferior-events off
3107 The @code{set print inferior-events} command allows you to enable or
3108 disable printing of messages when @value{GDBN} notices that new
3109 inferiors have started or that inferiors have exited or have been
3110 detached. By default, these messages will not be printed.
3112 @kindex show print inferior-events
3113 @item show print inferior-events
3114 Show whether messages will be printed when @value{GDBN} detects that
3115 inferiors have started, exited or have been detached.
3118 Many commands will work the same with multiple programs as with a
3119 single program: e.g., @code{print myglobal} will simply display the
3120 value of @code{myglobal} in the current inferior.
3123 Occasionaly, when debugging @value{GDBN} itself, it may be useful to
3124 get more info about the relationship of inferiors, programs, address
3125 spaces in a debug session. You can do that with the @w{@code{maint
3126 info program-spaces}} command.
3129 @kindex maint info program-spaces
3130 @item maint info program-spaces
3131 Print a list of all program spaces currently being managed by
3134 @value{GDBN} displays for each program space (in this order):
3138 the program space number assigned by @value{GDBN}
3141 the name of the executable loaded into the program space, with e.g.,
3142 the @code{file} command.
3147 An asterisk @samp{*} preceding the @value{GDBN} program space number
3148 indicates the current program space.
3150 In addition, below each program space line, @value{GDBN} prints extra
3151 information that isn't suitable to display in tabular form. For
3152 example, the list of inferiors bound to the program space.
3155 (@value{GDBP}) maint info program-spaces
3159 Bound inferiors: ID 1 (process 21561)
3162 Here we can see that no inferior is running the program @code{hello},
3163 while @code{process 21561} is running the program @code{goodbye}. On
3164 some targets, it is possible that multiple inferiors are bound to the
3165 same program space. The most common example is that of debugging both
3166 the parent and child processes of a @code{vfork} call. For example,
3169 (@value{GDBP}) maint info program-spaces
3172 Bound inferiors: ID 2 (process 18050), ID 1 (process 18045)
3175 Here, both inferior 2 and inferior 1 are running in the same program
3176 space as a result of inferior 1 having executed a @code{vfork} call.
3180 @section Debugging Programs with Multiple Threads
3182 @cindex threads of execution
3183 @cindex multiple threads
3184 @cindex switching threads
3185 In some operating systems, such as GNU/Linux and Solaris, a single program
3186 may have more than one @dfn{thread} of execution. The precise semantics
3187 of threads differ from one operating system to another, but in general
3188 the threads of a single program are akin to multiple processes---except
3189 that they share one address space (that is, they can all examine and
3190 modify the same variables). On the other hand, each thread has its own
3191 registers and execution stack, and perhaps private memory.
3193 @value{GDBN} provides these facilities for debugging multi-thread
3197 @item automatic notification of new threads
3198 @item @samp{thread @var{thread-id}}, a command to switch among threads
3199 @item @samp{info threads}, a command to inquire about existing threads
3200 @item @samp{thread apply [@var{thread-id-list} | all] @var{args}},
3201 a command to apply a command to a list of threads
3202 @item thread-specific breakpoints
3203 @item @samp{set print thread-events}, which controls printing of
3204 messages on thread start and exit.
3205 @item @samp{set libthread-db-search-path @var{path}}, which lets
3206 the user specify which @code{libthread_db} to use if the default choice
3207 isn't compatible with the program.
3210 @cindex focus of debugging
3211 @cindex current thread
3212 The @value{GDBN} thread debugging facility allows you to observe all
3213 threads while your program runs---but whenever @value{GDBN} takes
3214 control, one thread in particular is always the focus of debugging.
3215 This thread is called the @dfn{current thread}. Debugging commands show
3216 program information from the perspective of the current thread.
3218 @cindex @code{New} @var{systag} message
3219 @cindex thread identifier (system)
3220 @c FIXME-implementors!! It would be more helpful if the [New...] message
3221 @c included GDB's numeric thread handle, so you could just go to that
3222 @c thread without first checking `info threads'.
3223 Whenever @value{GDBN} detects a new thread in your program, it displays
3224 the target system's identification for the thread with a message in the
3225 form @samp{[New @var{systag}]}, where @var{systag} is a thread identifier
3226 whose form varies depending on the particular system. For example, on
3227 @sc{gnu}/Linux, you might see
3230 [New Thread 0x41e02940 (LWP 25582)]
3234 when @value{GDBN} notices a new thread. In contrast, on other systems,
3235 the @var{systag} is simply something like @samp{process 368}, with no
3238 @c FIXME!! (1) Does the [New...] message appear even for the very first
3239 @c thread of a program, or does it only appear for the
3240 @c second---i.e.@: when it becomes obvious we have a multithread
3242 @c (2) *Is* there necessarily a first thread always? Or do some
3243 @c multithread systems permit starting a program with multiple
3244 @c threads ab initio?
3246 @anchor{thread numbers}
3247 @cindex thread number, per inferior
3248 @cindex thread identifier (GDB)
3249 For debugging purposes, @value{GDBN} associates its own thread number
3250 ---always a single integer---with each thread of an inferior. This
3251 number is unique between all threads of an inferior, but not unique
3252 between threads of different inferiors.
3254 @cindex qualified thread ID
3255 You can refer to a given thread in an inferior using the qualified
3256 @var{inferior-num}.@var{thread-num} syntax, also known as
3257 @dfn{qualified thread ID}, with @var{inferior-num} being the inferior
3258 number and @var{thread-num} being the thread number of the given
3259 inferior. For example, thread @code{2.3} refers to thread number 3 of
3260 inferior 2. If you omit @var{inferior-num} (e.g., @code{thread 3}),
3261 then @value{GDBN} infers you're referring to a thread of the current
3264 Until you create a second inferior, @value{GDBN} does not show the
3265 @var{inferior-num} part of thread IDs, even though you can always use
3266 the full @var{inferior-num}.@var{thread-num} form to refer to threads
3267 of inferior 1, the initial inferior.
3269 @anchor{thread ID lists}
3270 @cindex thread ID lists
3271 Some commands accept a space-separated @dfn{thread ID list} as
3272 argument. A list element can be:
3276 A thread ID as shown in the first field of the @samp{info threads}
3277 display, with or without an inferior qualifier. E.g., @samp{2.1} or
3281 A range of thread numbers, again with or without an inferior
3282 qualifier, as in @var{inf}.@var{thr1}-@var{thr2} or
3283 @var{thr1}-@var{thr2}. E.g., @samp{1.2-4} or @samp{2-4}.
3286 All threads of an inferior, specified with a star wildcard, with or
3287 without an inferior qualifier, as in @var{inf}.@code{*} (e.g.,
3288 @samp{1.*}) or @code{*}. The former refers to all threads of the
3289 given inferior, and the latter form without an inferior qualifier
3290 refers to all threads of the current inferior.
3294 For example, if the current inferior is 1, and inferior 7 has one
3295 thread with ID 7.1, the thread list @samp{1 2-3 4.5 6.7-9 7.*}
3296 includes threads 1 to 3 of inferior 1, thread 5 of inferior 4, threads
3297 7 to 9 of inferior 6 and all threads of inferior 7. That is, in
3298 expanded qualified form, the same as @samp{1.1 1.2 1.3 4.5 6.7 6.8 6.9
3302 @anchor{global thread numbers}
3303 @cindex global thread number
3304 @cindex global thread identifier (GDB)
3305 In addition to a @emph{per-inferior} number, each thread is also
3306 assigned a unique @emph{global} number, also known as @dfn{global
3307 thread ID}, a single integer. Unlike the thread number component of
3308 the thread ID, no two threads have the same global ID, even when
3309 you're debugging multiple inferiors.
3311 From @value{GDBN}'s perspective, a process always has at least one
3312 thread. In other words, @value{GDBN} assigns a thread number to the
3313 program's ``main thread'' even if the program is not multi-threaded.
3315 @vindex $_thread@r{, convenience variable}
3316 @vindex $_gthread@r{, convenience variable}
3317 The debugger convenience variables @samp{$_thread} and
3318 @samp{$_gthread} contain, respectively, the per-inferior thread number
3319 and the global thread number of the current thread. You may find this
3320 useful in writing breakpoint conditional expressions, command scripts,
3321 and so forth. @xref{Convenience Vars,, Convenience Variables}, for
3322 general information on convenience variables.
3324 If @value{GDBN} detects the program is multi-threaded, it augments the
3325 usual message about stopping at a breakpoint with the ID and name of
3326 the thread that hit the breakpoint.
3329 Thread 2 "client" hit Breakpoint 1, send_message () at client.c:68
3332 Likewise when the program receives a signal:
3335 Thread 1 "main" received signal SIGINT, Interrupt.
3339 @kindex info threads
3340 @item info threads @r{[}@var{thread-id-list}@r{]}
3342 Display information about one or more threads. With no arguments
3343 displays information about all threads. You can specify the list of
3344 threads that you want to display using the thread ID list syntax
3345 (@pxref{thread ID lists}).
3347 @value{GDBN} displays for each thread (in this order):
3351 the per-inferior thread number assigned by @value{GDBN}
3354 the global thread number assigned by @value{GDBN}, if the @samp{-gid}
3355 option was specified
3358 the target system's thread identifier (@var{systag})
3361 the thread's name, if one is known. A thread can either be named by
3362 the user (see @code{thread name}, below), or, in some cases, by the
3366 the current stack frame summary for that thread
3370 An asterisk @samp{*} to the left of the @value{GDBN} thread number
3371 indicates the current thread.
3375 @c end table here to get a little more width for example
3378 (@value{GDBP}) info threads
3380 * 1 process 35 thread 13 main (argc=1, argv=0x7ffffff8)
3381 2 process 35 thread 23 0x34e5 in sigpause ()
3382 3 process 35 thread 27 0x34e5 in sigpause ()
3386 If you're debugging multiple inferiors, @value{GDBN} displays thread
3387 IDs using the qualified @var{inferior-num}.@var{thread-num} format.
3388 Otherwise, only @var{thread-num} is shown.
3390 If you specify the @samp{-gid} option, @value{GDBN} displays a column
3391 indicating each thread's global thread ID:
3394 (@value{GDBP}) info threads
3395 Id GId Target Id Frame
3396 1.1 1 process 35 thread 13 main (argc=1, argv=0x7ffffff8)
3397 1.2 3 process 35 thread 23 0x34e5 in sigpause ()
3398 1.3 4 process 35 thread 27 0x34e5 in sigpause ()
3399 * 2.1 2 process 65 thread 1 main (argc=1, argv=0x7ffffff8)
3402 On Solaris, you can display more information about user threads with a
3403 Solaris-specific command:
3406 @item maint info sol-threads
3407 @kindex maint info sol-threads
3408 @cindex thread info (Solaris)
3409 Display info on Solaris user threads.
3413 @kindex thread @var{thread-id}
3414 @item thread @var{thread-id}
3415 Make thread ID @var{thread-id} the current thread. The command
3416 argument @var{thread-id} is the @value{GDBN} thread ID, as shown in
3417 the first field of the @samp{info threads} display, with or without an
3418 inferior qualifier (e.g., @samp{2.1} or @samp{1}).
3420 @value{GDBN} responds by displaying the system identifier of the
3421 thread you selected, and its current stack frame summary:
3424 (@value{GDBP}) thread 2
3425 [Switching to thread 2 (Thread 0xb7fdab70 (LWP 12747))]
3426 #0 some_function (ignore=0x0) at example.c:8
3427 8 printf ("hello\n");
3431 As with the @samp{[New @dots{}]} message, the form of the text after
3432 @samp{Switching to} depends on your system's conventions for identifying
3435 @anchor{thread apply all}
3436 @kindex thread apply
3437 @cindex apply command to several threads
3438 @item thread apply [@var{thread-id-list} | all [-ascending]] [@var{flag}]@dots{} @var{command}
3439 The @code{thread apply} command allows you to apply the named
3440 @var{command} to one or more threads. Specify the threads that you
3441 want affected using the thread ID list syntax (@pxref{thread ID
3442 lists}), or specify @code{all} to apply to all threads. To apply a
3443 command to all threads in descending order, type @kbd{thread apply all
3444 @var{command}}. To apply a command to all threads in ascending order,
3445 type @kbd{thread apply all -ascending @var{command}}.
3447 The @var{flag} arguments control what output to produce and how to handle
3448 errors raised when applying @var{command} to a thread. @var{flag}
3449 must start with a @code{-} directly followed by one letter in
3450 @code{qcs}. If several flags are provided, they must be given
3451 individually, such as @code{-c -q}.
3453 By default, @value{GDBN} displays some thread information before the
3454 output produced by @var{command}, and an error raised during the
3455 execution of a @var{command} will abort @code{thread apply}. The
3456 following flags can be used to fine-tune this behavior:
3460 The flag @code{-c}, which stands for @samp{continue}, causes any
3461 errors in @var{command} to be displayed, and the execution of
3462 @code{thread apply} then continues.
3464 The flag @code{-s}, which stands for @samp{silent}, causes any errors
3465 or empty output produced by a @var{command} to be silently ignored.
3466 That is, the execution continues, but the thread information and errors
3469 The flag @code{-q} (@samp{quiet}) disables printing the thread
3473 Flags @code{-c} and @code{-s} cannot be used together.
3476 @cindex apply command to all threads (ignoring errors and empty output)
3477 @item taas [@var{option}]@dots{} @var{command}
3478 Shortcut for @code{thread apply all -s [@var{option}]@dots{} @var{command}}.
3479 Applies @var{command} on all threads, ignoring errors and empty output.
3481 The @code{taas} command accepts the same options as the @code{thread
3482 apply all} command. @xref{thread apply all}.
3485 @cindex apply a command to all frames of all threads (ignoring errors and empty output)
3486 @item tfaas [@var{option}]@dots{} @var{command}
3487 Shortcut for @code{thread apply all -s -- frame apply all -s [@var{option}]@dots{} @var{command}}.
3488 Applies @var{command} on all frames of all threads, ignoring errors
3489 and empty output. Note that the flag @code{-s} is specified twice:
3490 The first @code{-s} ensures that @code{thread apply} only shows the thread
3491 information of the threads for which @code{frame apply} produces
3492 some output. The second @code{-s} is needed to ensure that @code{frame
3493 apply} shows the frame information of a frame only if the
3494 @var{command} successfully produced some output.
3496 It can for example be used to print a local variable or a function
3497 argument without knowing the thread or frame where this variable or argument
3500 (@value{GDBP}) tfaas p some_local_var_i_do_not_remember_where_it_is
3503 The @code{tfaas} command accepts the same options as the @code{frame
3504 apply} command. @xref{frame apply}.
3507 @cindex name a thread
3508 @item thread name [@var{name}]
3509 This command assigns a name to the current thread. If no argument is
3510 given, any existing user-specified name is removed. The thread name
3511 appears in the @samp{info threads} display.
3513 On some systems, such as @sc{gnu}/Linux, @value{GDBN} is able to
3514 determine the name of the thread as given by the OS. On these
3515 systems, a name specified with @samp{thread name} will override the
3516 system-give name, and removing the user-specified name will cause
3517 @value{GDBN} to once again display the system-specified name.
3520 @cindex search for a thread
3521 @item thread find [@var{regexp}]
3522 Search for and display thread ids whose name or @var{systag}
3523 matches the supplied regular expression.
3525 As well as being the complement to the @samp{thread name} command,
3526 this command also allows you to identify a thread by its target
3527 @var{systag}. For instance, on @sc{gnu}/Linux, the target @var{systag}
3531 (@value{GDBN}) thread find 26688
3532 Thread 4 has target id 'Thread 0x41e02940 (LWP 26688)'
3533 (@value{GDBN}) info thread 4
3535 4 Thread 0x41e02940 (LWP 26688) 0x00000031ca6cd372 in select ()
3538 @kindex set print thread-events
3539 @cindex print messages on thread start and exit
3540 @item set print thread-events
3541 @itemx set print thread-events on
3542 @itemx set print thread-events off
3543 The @code{set print thread-events} command allows you to enable or
3544 disable printing of messages when @value{GDBN} notices that new threads have
3545 started or that threads have exited. By default, these messages will
3546 be printed if detection of these events is supported by the target.
3547 Note that these messages cannot be disabled on all targets.
3549 @kindex show print thread-events
3550 @item show print thread-events
3551 Show whether messages will be printed when @value{GDBN} detects that threads
3552 have started and exited.
3555 @xref{Thread Stops,,Stopping and Starting Multi-thread Programs}, for
3556 more information about how @value{GDBN} behaves when you stop and start
3557 programs with multiple threads.
3559 @xref{Set Watchpoints,,Setting Watchpoints}, for information about
3560 watchpoints in programs with multiple threads.
3562 @anchor{set libthread-db-search-path}
3564 @kindex set libthread-db-search-path
3565 @cindex search path for @code{libthread_db}
3566 @item set libthread-db-search-path @r{[}@var{path}@r{]}
3567 If this variable is set, @var{path} is a colon-separated list of
3568 directories @value{GDBN} will use to search for @code{libthread_db}.
3569 If you omit @var{path}, @samp{libthread-db-search-path} will be reset to
3570 its default value (@code{$sdir:$pdir} on @sc{gnu}/Linux and Solaris systems).
3571 Internally, the default value comes from the @code{LIBTHREAD_DB_SEARCH_PATH}
3574 On @sc{gnu}/Linux and Solaris systems, @value{GDBN} uses a ``helper''
3575 @code{libthread_db} library to obtain information about threads in the
3576 inferior process. @value{GDBN} will use @samp{libthread-db-search-path}
3577 to find @code{libthread_db}. @value{GDBN} also consults first if inferior
3578 specific thread debugging library loading is enabled
3579 by @samp{set auto-load libthread-db} (@pxref{libthread_db.so.1 file}).
3581 A special entry @samp{$sdir} for @samp{libthread-db-search-path}
3582 refers to the default system directories that are
3583 normally searched for loading shared libraries. The @samp{$sdir} entry
3584 is the only kind not needing to be enabled by @samp{set auto-load libthread-db}
3585 (@pxref{libthread_db.so.1 file}).
3587 A special entry @samp{$pdir} for @samp{libthread-db-search-path}
3588 refers to the directory from which @code{libpthread}
3589 was loaded in the inferior process.
3591 For any @code{libthread_db} library @value{GDBN} finds in above directories,
3592 @value{GDBN} attempts to initialize it with the current inferior process.
3593 If this initialization fails (which could happen because of a version
3594 mismatch between @code{libthread_db} and @code{libpthread}), @value{GDBN}
3595 will unload @code{libthread_db}, and continue with the next directory.
3596 If none of @code{libthread_db} libraries initialize successfully,
3597 @value{GDBN} will issue a warning and thread debugging will be disabled.
3599 Setting @code{libthread-db-search-path} is currently implemented
3600 only on some platforms.
3602 @kindex show libthread-db-search-path
3603 @item show libthread-db-search-path
3604 Display current libthread_db search path.
3606 @kindex set debug libthread-db
3607 @kindex show debug libthread-db
3608 @cindex debugging @code{libthread_db}
3609 @item set debug libthread-db
3610 @itemx show debug libthread-db
3611 Turns on or off display of @code{libthread_db}-related events.
3612 Use @code{1} to enable, @code{0} to disable.
3616 @section Debugging Forks
3618 @cindex fork, debugging programs which call
3619 @cindex multiple processes
3620 @cindex processes, multiple
3621 On most systems, @value{GDBN} has no special support for debugging
3622 programs which create additional processes using the @code{fork}
3623 function. When a program forks, @value{GDBN} will continue to debug the
3624 parent process and the child process will run unimpeded. If you have
3625 set a breakpoint in any code which the child then executes, the child
3626 will get a @code{SIGTRAP} signal which (unless it catches the signal)
3627 will cause it to terminate.
3629 However, if you want to debug the child process there is a workaround
3630 which isn't too painful. Put a call to @code{sleep} in the code which
3631 the child process executes after the fork. It may be useful to sleep
3632 only if a certain environment variable is set, or a certain file exists,
3633 so that the delay need not occur when you don't want to run @value{GDBN}
3634 on the child. While the child is sleeping, use the @code{ps} program to
3635 get its process ID. Then tell @value{GDBN} (a new invocation of
3636 @value{GDBN} if you are also debugging the parent process) to attach to
3637 the child process (@pxref{Attach}). From that point on you can debug
3638 the child process just like any other process which you attached to.
3640 On some systems, @value{GDBN} provides support for debugging programs
3641 that create additional processes using the @code{fork} or @code{vfork}
3642 functions. On @sc{gnu}/Linux platforms, this feature is supported
3643 with kernel version 2.5.46 and later.
3645 The fork debugging commands are supported in native mode and when
3646 connected to @code{gdbserver} in either @code{target remote} mode or
3647 @code{target extended-remote} mode.
3649 By default, when a program forks, @value{GDBN} will continue to debug
3650 the parent process and the child process will run unimpeded.
3652 If you want to follow the child process instead of the parent process,
3653 use the command @w{@code{set follow-fork-mode}}.
3656 @kindex set follow-fork-mode
3657 @item set follow-fork-mode @var{mode}
3658 Set the debugger response to a program call of @code{fork} or
3659 @code{vfork}. A call to @code{fork} or @code{vfork} creates a new
3660 process. The @var{mode} argument can be:
3664 The original process is debugged after a fork. The child process runs
3665 unimpeded. This is the default.
3668 The new process is debugged after a fork. The parent process runs
3673 @kindex show follow-fork-mode
3674 @item show follow-fork-mode
3675 Display the current debugger response to a @code{fork} or @code{vfork} call.
3678 @cindex debugging multiple processes
3679 On Linux, if you want to debug both the parent and child processes, use the
3680 command @w{@code{set detach-on-fork}}.
3683 @kindex set detach-on-fork
3684 @item set detach-on-fork @var{mode}
3685 Tells gdb whether to detach one of the processes after a fork, or
3686 retain debugger control over them both.
3690 The child process (or parent process, depending on the value of
3691 @code{follow-fork-mode}) will be detached and allowed to run
3692 independently. This is the default.
3695 Both processes will be held under the control of @value{GDBN}.
3696 One process (child or parent, depending on the value of
3697 @code{follow-fork-mode}) is debugged as usual, while the other
3702 @kindex show detach-on-fork
3703 @item show detach-on-fork
3704 Show whether detach-on-fork mode is on/off.
3707 If you choose to set @samp{detach-on-fork} mode off, then @value{GDBN}
3708 will retain control of all forked processes (including nested forks).
3709 You can list the forked processes under the control of @value{GDBN} by
3710 using the @w{@code{info inferiors}} command, and switch from one fork
3711 to another by using the @code{inferior} command (@pxref{Inferiors and
3712 Programs, ,Debugging Multiple Inferiors and Programs}).
3714 To quit debugging one of the forked processes, you can either detach
3715 from it by using the @w{@code{detach inferiors}} command (allowing it
3716 to run independently), or kill it using the @w{@code{kill inferiors}}
3717 command. @xref{Inferiors and Programs, ,Debugging Multiple Inferiors
3720 If you ask to debug a child process and a @code{vfork} is followed by an
3721 @code{exec}, @value{GDBN} executes the new target up to the first
3722 breakpoint in the new target. If you have a breakpoint set on
3723 @code{main} in your original program, the breakpoint will also be set on
3724 the child process's @code{main}.
3726 On some systems, when a child process is spawned by @code{vfork}, you
3727 cannot debug the child or parent until an @code{exec} call completes.
3729 If you issue a @code{run} command to @value{GDBN} after an @code{exec}
3730 call executes, the new target restarts. To restart the parent
3731 process, use the @code{file} command with the parent executable name
3732 as its argument. By default, after an @code{exec} call executes,
3733 @value{GDBN} discards the symbols of the previous executable image.
3734 You can change this behaviour with the @w{@code{set follow-exec-mode}}
3738 @kindex set follow-exec-mode
3739 @item set follow-exec-mode @var{mode}
3741 Set debugger response to a program call of @code{exec}. An
3742 @code{exec} call replaces the program image of a process.
3744 @code{follow-exec-mode} can be:
3748 @value{GDBN} creates a new inferior and rebinds the process to this
3749 new inferior. The program the process was running before the
3750 @code{exec} call can be restarted afterwards by restarting the
3756 (@value{GDBP}) info inferiors
3758 Id Description Executable
3761 process 12020 is executing new program: prog2
3762 Program exited normally.
3763 (@value{GDBP}) info inferiors
3764 Id Description Executable
3770 @value{GDBN} keeps the process bound to the same inferior. The new
3771 executable image replaces the previous executable loaded in the
3772 inferior. Restarting the inferior after the @code{exec} call, with
3773 e.g., the @code{run} command, restarts the executable the process was
3774 running after the @code{exec} call. This is the default mode.
3779 (@value{GDBP}) info inferiors
3780 Id Description Executable
3783 process 12020 is executing new program: prog2
3784 Program exited normally.
3785 (@value{GDBP}) info inferiors
3786 Id Description Executable
3793 @code{follow-exec-mode} is supported in native mode and
3794 @code{target extended-remote} mode.
3796 You can use the @code{catch} command to make @value{GDBN} stop whenever
3797 a @code{fork}, @code{vfork}, or @code{exec} call is made. @xref{Set
3798 Catchpoints, ,Setting Catchpoints}.
3800 @node Checkpoint/Restart
3801 @section Setting a @emph{Bookmark} to Return to Later
3806 @cindex snapshot of a process
3807 @cindex rewind program state
3809 On certain operating systems@footnote{Currently, only
3810 @sc{gnu}/Linux.}, @value{GDBN} is able to save a @dfn{snapshot} of a
3811 program's state, called a @dfn{checkpoint}, and come back to it
3814 Returning to a checkpoint effectively undoes everything that has
3815 happened in the program since the @code{checkpoint} was saved. This
3816 includes changes in memory, registers, and even (within some limits)
3817 system state. Effectively, it is like going back in time to the
3818 moment when the checkpoint was saved.
3820 Thus, if you're stepping thru a program and you think you're
3821 getting close to the point where things go wrong, you can save
3822 a checkpoint. Then, if you accidentally go too far and miss
3823 the critical statement, instead of having to restart your program
3824 from the beginning, you can just go back to the checkpoint and
3825 start again from there.
3827 This can be especially useful if it takes a lot of time or
3828 steps to reach the point where you think the bug occurs.
3830 To use the @code{checkpoint}/@code{restart} method of debugging:
3835 Save a snapshot of the debugged program's current execution state.
3836 The @code{checkpoint} command takes no arguments, but each checkpoint
3837 is assigned a small integer id, similar to a breakpoint id.
3839 @kindex info checkpoints
3840 @item info checkpoints
3841 List the checkpoints that have been saved in the current debugging
3842 session. For each checkpoint, the following information will be
3849 @item Source line, or label
3852 @kindex restart @var{checkpoint-id}
3853 @item restart @var{checkpoint-id}
3854 Restore the program state that was saved as checkpoint number
3855 @var{checkpoint-id}. All program variables, registers, stack frames
3856 etc.@: will be returned to the values that they had when the checkpoint
3857 was saved. In essence, gdb will ``wind back the clock'' to the point
3858 in time when the checkpoint was saved.
3860 Note that breakpoints, @value{GDBN} variables, command history etc.
3861 are not affected by restoring a checkpoint. In general, a checkpoint
3862 only restores things that reside in the program being debugged, not in
3865 @kindex delete checkpoint @var{checkpoint-id}
3866 @item delete checkpoint @var{checkpoint-id}
3867 Delete the previously-saved checkpoint identified by @var{checkpoint-id}.
3871 Returning to a previously saved checkpoint will restore the user state
3872 of the program being debugged, plus a significant subset of the system
3873 (OS) state, including file pointers. It won't ``un-write'' data from
3874 a file, but it will rewind the file pointer to the previous location,
3875 so that the previously written data can be overwritten. For files
3876 opened in read mode, the pointer will also be restored so that the
3877 previously read data can be read again.
3879 Of course, characters that have been sent to a printer (or other
3880 external device) cannot be ``snatched back'', and characters received
3881 from eg.@: a serial device can be removed from internal program buffers,
3882 but they cannot be ``pushed back'' into the serial pipeline, ready to
3883 be received again. Similarly, the actual contents of files that have
3884 been changed cannot be restored (at this time).
3886 However, within those constraints, you actually can ``rewind'' your
3887 program to a previously saved point in time, and begin debugging it
3888 again --- and you can change the course of events so as to debug a
3889 different execution path this time.
3891 @cindex checkpoints and process id
3892 Finally, there is one bit of internal program state that will be
3893 different when you return to a checkpoint --- the program's process
3894 id. Each checkpoint will have a unique process id (or @var{pid}),
3895 and each will be different from the program's original @var{pid}.
3896 If your program has saved a local copy of its process id, this could
3897 potentially pose a problem.
3899 @subsection A Non-obvious Benefit of Using Checkpoints
3901 On some systems such as @sc{gnu}/Linux, address space randomization
3902 is performed on new processes for security reasons. This makes it
3903 difficult or impossible to set a breakpoint, or watchpoint, on an
3904 absolute address if you have to restart the program, since the
3905 absolute location of a symbol will change from one execution to the
3908 A checkpoint, however, is an @emph{identical} copy of a process.
3909 Therefore if you create a checkpoint at (eg.@:) the start of main,
3910 and simply return to that checkpoint instead of restarting the
3911 process, you can avoid the effects of address randomization and
3912 your symbols will all stay in the same place.
3915 @chapter Stopping and Continuing
3917 The principal purposes of using a debugger are so that you can stop your
3918 program before it terminates; or so that, if your program runs into
3919 trouble, you can investigate and find out why.
3921 Inside @value{GDBN}, your program may stop for any of several reasons,
3922 such as a signal, a breakpoint, or reaching a new line after a
3923 @value{GDBN} command such as @code{step}. You may then examine and
3924 change variables, set new breakpoints or remove old ones, and then
3925 continue execution. Usually, the messages shown by @value{GDBN} provide
3926 ample explanation of the status of your program---but you can also
3927 explicitly request this information at any time.
3930 @kindex info program
3932 Display information about the status of your program: whether it is
3933 running or not, what process it is, and why it stopped.
3937 * Breakpoints:: Breakpoints, watchpoints, and catchpoints
3938 * Continuing and Stepping:: Resuming execution
3939 * Skipping Over Functions and Files::
3940 Skipping over functions and files
3942 * Thread Stops:: Stopping and starting multi-thread programs
3946 @section Breakpoints, Watchpoints, and Catchpoints
3949 A @dfn{breakpoint} makes your program stop whenever a certain point in
3950 the program is reached. For each breakpoint, you can add conditions to
3951 control in finer detail whether your program stops. You can set
3952 breakpoints with the @code{break} command and its variants (@pxref{Set
3953 Breaks, ,Setting Breakpoints}), to specify the place where your program
3954 should stop by line number, function name or exact address in the
3957 On some systems, you can set breakpoints in shared libraries before
3958 the executable is run.
3961 @cindex data breakpoints
3962 @cindex memory tracing
3963 @cindex breakpoint on memory address
3964 @cindex breakpoint on variable modification
3965 A @dfn{watchpoint} is a special breakpoint that stops your program
3966 when the value of an expression changes. The expression may be a value
3967 of a variable, or it could involve values of one or more variables
3968 combined by operators, such as @samp{a + b}. This is sometimes called
3969 @dfn{data breakpoints}. You must use a different command to set
3970 watchpoints (@pxref{Set Watchpoints, ,Setting Watchpoints}), but aside
3971 from that, you can manage a watchpoint like any other breakpoint: you
3972 enable, disable, and delete both breakpoints and watchpoints using the
3975 You can arrange to have values from your program displayed automatically
3976 whenever @value{GDBN} stops at a breakpoint. @xref{Auto Display,,
3980 @cindex breakpoint on events
3981 A @dfn{catchpoint} is another special breakpoint that stops your program
3982 when a certain kind of event occurs, such as the throwing of a C@t{++}
3983 exception or the loading of a library. As with watchpoints, you use a
3984 different command to set a catchpoint (@pxref{Set Catchpoints, ,Setting
3985 Catchpoints}), but aside from that, you can manage a catchpoint like any
3986 other breakpoint. (To stop when your program receives a signal, use the
3987 @code{handle} command; see @ref{Signals, ,Signals}.)
3989 @cindex breakpoint numbers
3990 @cindex numbers for breakpoints
3991 @value{GDBN} assigns a number to each breakpoint, watchpoint, or
3992 catchpoint when you create it; these numbers are successive integers
3993 starting with one. In many of the commands for controlling various
3994 features of breakpoints you use the breakpoint number to say which
3995 breakpoint you want to change. Each breakpoint may be @dfn{enabled} or
3996 @dfn{disabled}; if disabled, it has no effect on your program until you
3999 @cindex breakpoint ranges
4000 @cindex breakpoint lists
4001 @cindex ranges of breakpoints
4002 @cindex lists of breakpoints
4003 Some @value{GDBN} commands accept a space-separated list of breakpoints
4004 on which to operate. A list element can be either a single breakpoint number,
4005 like @samp{5}, or a range of such numbers, like @samp{5-7}.
4006 When a breakpoint list is given to a command, all breakpoints in that list
4010 * Set Breaks:: Setting breakpoints
4011 * Set Watchpoints:: Setting watchpoints
4012 * Set Catchpoints:: Setting catchpoints
4013 * Delete Breaks:: Deleting breakpoints
4014 * Disabling:: Disabling breakpoints
4015 * Conditions:: Break conditions
4016 * Break Commands:: Breakpoint command lists
4017 * Dynamic Printf:: Dynamic printf
4018 * Save Breakpoints:: How to save breakpoints in a file
4019 * Static Probe Points:: Listing static probe points
4020 * Error in Breakpoints:: ``Cannot insert breakpoints''
4021 * Breakpoint-related Warnings:: ``Breakpoint address adjusted...''
4025 @subsection Setting Breakpoints
4027 @c FIXME LMB what does GDB do if no code on line of breakpt?
4028 @c consider in particular declaration with/without initialization.
4030 @c FIXME 2 is there stuff on this already? break at fun start, already init?
4033 @kindex b @r{(@code{break})}
4034 @vindex $bpnum@r{, convenience variable}
4035 @cindex latest breakpoint
4036 Breakpoints are set with the @code{break} command (abbreviated
4037 @code{b}). The debugger convenience variable @samp{$bpnum} records the
4038 number of the breakpoint you've set most recently; see @ref{Convenience
4039 Vars,, Convenience Variables}, for a discussion of what you can do with
4040 convenience variables.
4043 @item break @var{location}
4044 Set a breakpoint at the given @var{location}, which can specify a
4045 function name, a line number, or an address of an instruction.
4046 (@xref{Specify Location}, for a list of all the possible ways to
4047 specify a @var{location}.) The breakpoint will stop your program just
4048 before it executes any of the code in the specified @var{location}.
4050 When using source languages that permit overloading of symbols, such as
4051 C@t{++}, a function name may refer to more than one possible place to break.
4052 @xref{Ambiguous Expressions,,Ambiguous Expressions}, for a discussion of
4055 It is also possible to insert a breakpoint that will stop the program
4056 only if a specific thread (@pxref{Thread-Specific Breakpoints})
4057 or a specific task (@pxref{Ada Tasks}) hits that breakpoint.
4060 When called without any arguments, @code{break} sets a breakpoint at
4061 the next instruction to be executed in the selected stack frame
4062 (@pxref{Stack, ,Examining the Stack}). In any selected frame but the
4063 innermost, this makes your program stop as soon as control
4064 returns to that frame. This is similar to the effect of a
4065 @code{finish} command in the frame inside the selected frame---except
4066 that @code{finish} does not leave an active breakpoint. If you use
4067 @code{break} without an argument in the innermost frame, @value{GDBN} stops
4068 the next time it reaches the current location; this may be useful
4071 @value{GDBN} normally ignores breakpoints when it resumes execution, until at
4072 least one instruction has been executed. If it did not do this, you
4073 would be unable to proceed past a breakpoint without first disabling the
4074 breakpoint. This rule applies whether or not the breakpoint already
4075 existed when your program stopped.
4077 @item break @dots{} if @var{cond}
4078 Set a breakpoint with condition @var{cond}; evaluate the expression
4079 @var{cond} each time the breakpoint is reached, and stop only if the
4080 value is nonzero---that is, if @var{cond} evaluates as true.
4081 @samp{@dots{}} stands for one of the possible arguments described
4082 above (or no argument) specifying where to break. @xref{Conditions,
4083 ,Break Conditions}, for more information on breakpoint conditions.
4086 @item tbreak @var{args}
4087 Set a breakpoint enabled only for one stop. The @var{args} are the
4088 same as for the @code{break} command, and the breakpoint is set in the same
4089 way, but the breakpoint is automatically deleted after the first time your
4090 program stops there. @xref{Disabling, ,Disabling Breakpoints}.
4093 @cindex hardware breakpoints
4094 @item hbreak @var{args}
4095 Set a hardware-assisted breakpoint. The @var{args} are the same as for the
4096 @code{break} command and the breakpoint is set in the same way, but the
4097 breakpoint requires hardware support and some target hardware may not
4098 have this support. The main purpose of this is EPROM/ROM code
4099 debugging, so you can set a breakpoint at an instruction without
4100 changing the instruction. This can be used with the new trap-generation
4101 provided by SPARClite DSU and most x86-based targets. These targets
4102 will generate traps when a program accesses some data or instruction
4103 address that is assigned to the debug registers. However the hardware
4104 breakpoint registers can take a limited number of breakpoints. For
4105 example, on the DSU, only two data breakpoints can be set at a time, and
4106 @value{GDBN} will reject this command if more than two are used. Delete
4107 or disable unused hardware breakpoints before setting new ones
4108 (@pxref{Disabling, ,Disabling Breakpoints}).
4109 @xref{Conditions, ,Break Conditions}.
4110 For remote targets, you can restrict the number of hardware
4111 breakpoints @value{GDBN} will use, see @ref{set remote
4112 hardware-breakpoint-limit}.
4115 @item thbreak @var{args}
4116 Set a hardware-assisted breakpoint enabled only for one stop. The @var{args}
4117 are the same as for the @code{hbreak} command and the breakpoint is set in
4118 the same way. However, like the @code{tbreak} command,
4119 the breakpoint is automatically deleted after the
4120 first time your program stops there. Also, like the @code{hbreak}
4121 command, the breakpoint requires hardware support and some target hardware
4122 may not have this support. @xref{Disabling, ,Disabling Breakpoints}.
4123 See also @ref{Conditions, ,Break Conditions}.
4126 @cindex regular expression
4127 @cindex breakpoints at functions matching a regexp
4128 @cindex set breakpoints in many functions
4129 @item rbreak @var{regex}
4130 Set breakpoints on all functions matching the regular expression
4131 @var{regex}. This command sets an unconditional breakpoint on all
4132 matches, printing a list of all breakpoints it set. Once these
4133 breakpoints are set, they are treated just like the breakpoints set with
4134 the @code{break} command. You can delete them, disable them, or make
4135 them conditional the same way as any other breakpoint.
4137 In programs using different languages, @value{GDBN} chooses the syntax
4138 to print the list of all breakpoints it sets according to the
4139 @samp{set language} value: using @samp{set language auto}
4140 (see @ref{Automatically, ,Set Language Automatically}) means to use the
4141 language of the breakpoint's function, other values mean to use
4142 the manually specified language (see @ref{Manually, ,Set Language Manually}).
4144 The syntax of the regular expression is the standard one used with tools
4145 like @file{grep}. Note that this is different from the syntax used by
4146 shells, so for instance @code{foo*} matches all functions that include
4147 an @code{fo} followed by zero or more @code{o}s. There is an implicit
4148 @code{.*} leading and trailing the regular expression you supply, so to
4149 match only functions that begin with @code{foo}, use @code{^foo}.
4151 @cindex non-member C@t{++} functions, set breakpoint in
4152 When debugging C@t{++} programs, @code{rbreak} is useful for setting
4153 breakpoints on overloaded functions that are not members of any special
4156 @cindex set breakpoints on all functions
4157 The @code{rbreak} command can be used to set breakpoints in
4158 @strong{all} the functions in a program, like this:
4161 (@value{GDBP}) rbreak .
4164 @item rbreak @var{file}:@var{regex}
4165 If @code{rbreak} is called with a filename qualification, it limits
4166 the search for functions matching the given regular expression to the
4167 specified @var{file}. This can be used, for example, to set breakpoints on
4168 every function in a given file:
4171 (@value{GDBP}) rbreak file.c:.
4174 The colon separating the filename qualifier from the regex may
4175 optionally be surrounded by spaces.
4177 @kindex info breakpoints
4178 @cindex @code{$_} and @code{info breakpoints}
4179 @item info breakpoints @r{[}@var{list}@dots{}@r{]}
4180 @itemx info break @r{[}@var{list}@dots{}@r{]}
4181 Print a table of all breakpoints, watchpoints, and catchpoints set and
4182 not deleted. Optional argument @var{n} means print information only
4183 about the specified breakpoint(s) (or watchpoint(s) or catchpoint(s)).
4184 For each breakpoint, following columns are printed:
4187 @item Breakpoint Numbers
4189 Breakpoint, watchpoint, or catchpoint.
4191 Whether the breakpoint is marked to be disabled or deleted when hit.
4192 @item Enabled or Disabled
4193 Enabled breakpoints are marked with @samp{y}. @samp{n} marks breakpoints
4194 that are not enabled.
4196 Where the breakpoint is in your program, as a memory address. For a
4197 pending breakpoint whose address is not yet known, this field will
4198 contain @samp{<PENDING>}. Such breakpoint won't fire until a shared
4199 library that has the symbol or line referred by breakpoint is loaded.
4200 See below for details. A breakpoint with several locations will
4201 have @samp{<MULTIPLE>} in this field---see below for details.
4203 Where the breakpoint is in the source for your program, as a file and
4204 line number. For a pending breakpoint, the original string passed to
4205 the breakpoint command will be listed as it cannot be resolved until
4206 the appropriate shared library is loaded in the future.
4210 If a breakpoint is conditional, there are two evaluation modes: ``host'' and
4211 ``target''. If mode is ``host'', breakpoint condition evaluation is done by
4212 @value{GDBN} on the host's side. If it is ``target'', then the condition
4213 is evaluated by the target. The @code{info break} command shows
4214 the condition on the line following the affected breakpoint, together with
4215 its condition evaluation mode in between parentheses.
4217 Breakpoint commands, if any, are listed after that. A pending breakpoint is
4218 allowed to have a condition specified for it. The condition is not parsed for
4219 validity until a shared library is loaded that allows the pending
4220 breakpoint to resolve to a valid location.
4223 @code{info break} with a breakpoint
4224 number @var{n} as argument lists only that breakpoint. The
4225 convenience variable @code{$_} and the default examining-address for
4226 the @code{x} command are set to the address of the last breakpoint
4227 listed (@pxref{Memory, ,Examining Memory}).
4230 @code{info break} displays a count of the number of times the breakpoint
4231 has been hit. This is especially useful in conjunction with the
4232 @code{ignore} command. You can ignore a large number of breakpoint
4233 hits, look at the breakpoint info to see how many times the breakpoint
4234 was hit, and then run again, ignoring one less than that number. This
4235 will get you quickly to the last hit of that breakpoint.
4238 For a breakpoints with an enable count (xref) greater than 1,
4239 @code{info break} also displays that count.
4243 @value{GDBN} allows you to set any number of breakpoints at the same place in
4244 your program. There is nothing silly or meaningless about this. When
4245 the breakpoints are conditional, this is even useful
4246 (@pxref{Conditions, ,Break Conditions}).
4248 @cindex multiple locations, breakpoints
4249 @cindex breakpoints, multiple locations
4250 It is possible that a breakpoint corresponds to several locations
4251 in your program. Examples of this situation are:
4255 Multiple functions in the program may have the same name.
4258 For a C@t{++} constructor, the @value{NGCC} compiler generates several
4259 instances of the function body, used in different cases.
4262 For a C@t{++} template function, a given line in the function can
4263 correspond to any number of instantiations.
4266 For an inlined function, a given source line can correspond to
4267 several places where that function is inlined.
4270 In all those cases, @value{GDBN} will insert a breakpoint at all
4271 the relevant locations.
4273 A breakpoint with multiple locations is displayed in the breakpoint
4274 table using several rows---one header row, followed by one row for
4275 each breakpoint location. The header row has @samp{<MULTIPLE>} in the
4276 address column. The rows for individual locations contain the actual
4277 addresses for locations, and show the functions to which those
4278 locations belong. The number column for a location is of the form
4279 @var{breakpoint-number}.@var{location-number}.
4284 Num Type Disp Enb Address What
4285 1 breakpoint keep y <MULTIPLE>
4287 breakpoint already hit 1 time
4288 1.1 y 0x080486a2 in void foo<int>() at t.cc:8
4289 1.2 y 0x080486ca in void foo<double>() at t.cc:8
4292 You cannot delete the individual locations from a breakpoint. However,
4293 each location can be individually enabled or disabled by passing
4294 @var{breakpoint-number}.@var{location-number} as argument to the
4295 @code{enable} and @code{disable} commands. It's also possible to
4296 @code{enable} and @code{disable} a range of @var{location-number}
4297 locations using a @var{breakpoint-number} and two @var{location-number}s,
4298 in increasing order, separated by a hyphen, like
4299 @kbd{@var{breakpoint-number}.@var{location-number1}-@var{location-number2}},
4300 in which case @value{GDBN} acts on all the locations in the range (inclusive).
4301 Disabling or enabling the parent breakpoint (@pxref{Disabling}) affects
4302 all of the locations that belong to that breakpoint.
4304 @cindex pending breakpoints
4305 It's quite common to have a breakpoint inside a shared library.
4306 Shared libraries can be loaded and unloaded explicitly,
4307 and possibly repeatedly, as the program is executed. To support
4308 this use case, @value{GDBN} updates breakpoint locations whenever
4309 any shared library is loaded or unloaded. Typically, you would
4310 set a breakpoint in a shared library at the beginning of your
4311 debugging session, when the library is not loaded, and when the
4312 symbols from the library are not available. When you try to set
4313 breakpoint, @value{GDBN} will ask you if you want to set
4314 a so called @dfn{pending breakpoint}---breakpoint whose address
4315 is not yet resolved.
4317 After the program is run, whenever a new shared library is loaded,
4318 @value{GDBN} reevaluates all the breakpoints. When a newly loaded
4319 shared library contains the symbol or line referred to by some
4320 pending breakpoint, that breakpoint is resolved and becomes an
4321 ordinary breakpoint. When a library is unloaded, all breakpoints
4322 that refer to its symbols or source lines become pending again.
4324 This logic works for breakpoints with multiple locations, too. For
4325 example, if you have a breakpoint in a C@t{++} template function, and
4326 a newly loaded shared library has an instantiation of that template,
4327 a new location is added to the list of locations for the breakpoint.
4329 Except for having unresolved address, pending breakpoints do not
4330 differ from regular breakpoints. You can set conditions or commands,
4331 enable and disable them and perform other breakpoint operations.
4333 @value{GDBN} provides some additional commands for controlling what
4334 happens when the @samp{break} command cannot resolve breakpoint
4335 address specification to an address:
4337 @kindex set breakpoint pending
4338 @kindex show breakpoint pending
4340 @item set breakpoint pending auto
4341 This is the default behavior. When @value{GDBN} cannot find the breakpoint
4342 location, it queries you whether a pending breakpoint should be created.
4344 @item set breakpoint pending on
4345 This indicates that an unrecognized breakpoint location should automatically
4346 result in a pending breakpoint being created.
4348 @item set breakpoint pending off
4349 This indicates that pending breakpoints are not to be created. Any
4350 unrecognized breakpoint location results in an error. This setting does
4351 not affect any pending breakpoints previously created.
4353 @item show breakpoint pending
4354 Show the current behavior setting for creating pending breakpoints.
4357 The settings above only affect the @code{break} command and its
4358 variants. Once breakpoint is set, it will be automatically updated
4359 as shared libraries are loaded and unloaded.
4361 @cindex automatic hardware breakpoints
4362 For some targets, @value{GDBN} can automatically decide if hardware or
4363 software breakpoints should be used, depending on whether the
4364 breakpoint address is read-only or read-write. This applies to
4365 breakpoints set with the @code{break} command as well as to internal
4366 breakpoints set by commands like @code{next} and @code{finish}. For
4367 breakpoints set with @code{hbreak}, @value{GDBN} will always use hardware
4370 You can control this automatic behaviour with the following commands:
4372 @kindex set breakpoint auto-hw
4373 @kindex show breakpoint auto-hw
4375 @item set breakpoint auto-hw on
4376 This is the default behavior. When @value{GDBN} sets a breakpoint, it
4377 will try to use the target memory map to decide if software or hardware
4378 breakpoint must be used.
4380 @item set breakpoint auto-hw off
4381 This indicates @value{GDBN} should not automatically select breakpoint
4382 type. If the target provides a memory map, @value{GDBN} will warn when
4383 trying to set software breakpoint at a read-only address.
4386 @value{GDBN} normally implements breakpoints by replacing the program code
4387 at the breakpoint address with a special instruction, which, when
4388 executed, given control to the debugger. By default, the program
4389 code is so modified only when the program is resumed. As soon as
4390 the program stops, @value{GDBN} restores the original instructions. This
4391 behaviour guards against leaving breakpoints inserted in the
4392 target should gdb abrubptly disconnect. However, with slow remote
4393 targets, inserting and removing breakpoint can reduce the performance.
4394 This behavior can be controlled with the following commands::
4396 @kindex set breakpoint always-inserted
4397 @kindex show breakpoint always-inserted
4399 @item set breakpoint always-inserted off
4400 All breakpoints, including newly added by the user, are inserted in
4401 the target only when the target is resumed. All breakpoints are
4402 removed from the target when it stops. This is the default mode.
4404 @item set breakpoint always-inserted on
4405 Causes all breakpoints to be inserted in the target at all times. If
4406 the user adds a new breakpoint, or changes an existing breakpoint, the
4407 breakpoints in the target are updated immediately. A breakpoint is
4408 removed from the target only when breakpoint itself is deleted.
4411 @value{GDBN} handles conditional breakpoints by evaluating these conditions
4412 when a breakpoint breaks. If the condition is true, then the process being
4413 debugged stops, otherwise the process is resumed.
4415 If the target supports evaluating conditions on its end, @value{GDBN} may
4416 download the breakpoint, together with its conditions, to it.
4418 This feature can be controlled via the following commands:
4420 @kindex set breakpoint condition-evaluation
4421 @kindex show breakpoint condition-evaluation
4423 @item set breakpoint condition-evaluation host
4424 This option commands @value{GDBN} to evaluate the breakpoint
4425 conditions on the host's side. Unconditional breakpoints are sent to
4426 the target which in turn receives the triggers and reports them back to GDB
4427 for condition evaluation. This is the standard evaluation mode.
4429 @item set breakpoint condition-evaluation target
4430 This option commands @value{GDBN} to download breakpoint conditions
4431 to the target at the moment of their insertion. The target
4432 is responsible for evaluating the conditional expression and reporting
4433 breakpoint stop events back to @value{GDBN} whenever the condition
4434 is true. Due to limitations of target-side evaluation, some conditions
4435 cannot be evaluated there, e.g., conditions that depend on local data
4436 that is only known to the host. Examples include
4437 conditional expressions involving convenience variables, complex types
4438 that cannot be handled by the agent expression parser and expressions
4439 that are too long to be sent over to the target, specially when the
4440 target is a remote system. In these cases, the conditions will be
4441 evaluated by @value{GDBN}.
4443 @item set breakpoint condition-evaluation auto
4444 This is the default mode. If the target supports evaluating breakpoint
4445 conditions on its end, @value{GDBN} will download breakpoint conditions to
4446 the target (limitations mentioned previously apply). If the target does
4447 not support breakpoint condition evaluation, then @value{GDBN} will fallback
4448 to evaluating all these conditions on the host's side.
4452 @cindex negative breakpoint numbers
4453 @cindex internal @value{GDBN} breakpoints
4454 @value{GDBN} itself sometimes sets breakpoints in your program for
4455 special purposes, such as proper handling of @code{longjmp} (in C
4456 programs). These internal breakpoints are assigned negative numbers,
4457 starting with @code{-1}; @samp{info breakpoints} does not display them.
4458 You can see these breakpoints with the @value{GDBN} maintenance command
4459 @samp{maint info breakpoints} (@pxref{maint info breakpoints}).
4462 @node Set Watchpoints
4463 @subsection Setting Watchpoints
4465 @cindex setting watchpoints
4466 You can use a watchpoint to stop execution whenever the value of an
4467 expression changes, without having to predict a particular place where
4468 this may happen. (This is sometimes called a @dfn{data breakpoint}.)
4469 The expression may be as simple as the value of a single variable, or
4470 as complex as many variables combined by operators. Examples include:
4474 A reference to the value of a single variable.
4477 An address cast to an appropriate data type. For example,
4478 @samp{*(int *)0x12345678} will watch a 4-byte region at the specified
4479 address (assuming an @code{int} occupies 4 bytes).
4482 An arbitrarily complex expression, such as @samp{a*b + c/d}. The
4483 expression can use any operators valid in the program's native
4484 language (@pxref{Languages}).
4487 You can set a watchpoint on an expression even if the expression can
4488 not be evaluated yet. For instance, you can set a watchpoint on
4489 @samp{*global_ptr} before @samp{global_ptr} is initialized.
4490 @value{GDBN} will stop when your program sets @samp{global_ptr} and
4491 the expression produces a valid value. If the expression becomes
4492 valid in some other way than changing a variable (e.g.@: if the memory
4493 pointed to by @samp{*global_ptr} becomes readable as the result of a
4494 @code{malloc} call), @value{GDBN} may not stop until the next time
4495 the expression changes.
4497 @cindex software watchpoints
4498 @cindex hardware watchpoints
4499 Depending on your system, watchpoints may be implemented in software or
4500 hardware. @value{GDBN} does software watchpointing by single-stepping your
4501 program and testing the variable's value each time, which is hundreds of
4502 times slower than normal execution. (But this may still be worth it, to
4503 catch errors where you have no clue what part of your program is the
4506 On some systems, such as most PowerPC or x86-based targets,
4507 @value{GDBN} includes support for hardware watchpoints, which do not
4508 slow down the running of your program.
4512 @item watch @r{[}-l@r{|}-location@r{]} @var{expr} @r{[}thread @var{thread-id}@r{]} @r{[}mask @var{maskvalue}@r{]}
4513 Set a watchpoint for an expression. @value{GDBN} will break when the
4514 expression @var{expr} is written into by the program and its value
4515 changes. The simplest (and the most popular) use of this command is
4516 to watch the value of a single variable:
4519 (@value{GDBP}) watch foo
4522 If the command includes a @code{@r{[}thread @var{thread-id}@r{]}}
4523 argument, @value{GDBN} breaks only when the thread identified by
4524 @var{thread-id} changes the value of @var{expr}. If any other threads
4525 change the value of @var{expr}, @value{GDBN} will not break. Note
4526 that watchpoints restricted to a single thread in this way only work
4527 with Hardware Watchpoints.
4529 Ordinarily a watchpoint respects the scope of variables in @var{expr}
4530 (see below). The @code{-location} argument tells @value{GDBN} to
4531 instead watch the memory referred to by @var{expr}. In this case,
4532 @value{GDBN} will evaluate @var{expr}, take the address of the result,
4533 and watch the memory at that address. The type of the result is used
4534 to determine the size of the watched memory. If the expression's
4535 result does not have an address, then @value{GDBN} will print an
4538 The @code{@r{[}mask @var{maskvalue}@r{]}} argument allows creation
4539 of masked watchpoints, if the current architecture supports this
4540 feature (e.g., PowerPC Embedded architecture, see @ref{PowerPC
4541 Embedded}.) A @dfn{masked watchpoint} specifies a mask in addition
4542 to an address to watch. The mask specifies that some bits of an address
4543 (the bits which are reset in the mask) should be ignored when matching
4544 the address accessed by the inferior against the watchpoint address.
4545 Thus, a masked watchpoint watches many addresses simultaneously---those
4546 addresses whose unmasked bits are identical to the unmasked bits in the
4547 watchpoint address. The @code{mask} argument implies @code{-location}.
4551 (@value{GDBP}) watch foo mask 0xffff00ff
4552 (@value{GDBP}) watch *0xdeadbeef mask 0xffffff00
4556 @item rwatch @r{[}-l@r{|}-location@r{]} @var{expr} @r{[}thread @var{thread-id}@r{]} @r{[}mask @var{maskvalue}@r{]}
4557 Set a watchpoint that will break when the value of @var{expr} is read
4561 @item awatch @r{[}-l@r{|}-location@r{]} @var{expr} @r{[}thread @var{thread-id}@r{]} @r{[}mask @var{maskvalue}@r{]}
4562 Set a watchpoint that will break when @var{expr} is either read from
4563 or written into by the program.
4565 @kindex info watchpoints @r{[}@var{list}@dots{}@r{]}
4566 @item info watchpoints @r{[}@var{list}@dots{}@r{]}
4567 This command prints a list of watchpoints, using the same format as
4568 @code{info break} (@pxref{Set Breaks}).
4571 If you watch for a change in a numerically entered address you need to
4572 dereference it, as the address itself is just a constant number which will
4573 never change. @value{GDBN} refuses to create a watchpoint that watches
4574 a never-changing value:
4577 (@value{GDBP}) watch 0x600850
4578 Cannot watch constant value 0x600850.
4579 (@value{GDBP}) watch *(int *) 0x600850
4580 Watchpoint 1: *(int *) 6293584
4583 @value{GDBN} sets a @dfn{hardware watchpoint} if possible. Hardware
4584 watchpoints execute very quickly, and the debugger reports a change in
4585 value at the exact instruction where the change occurs. If @value{GDBN}
4586 cannot set a hardware watchpoint, it sets a software watchpoint, which
4587 executes more slowly and reports the change in value at the next
4588 @emph{statement}, not the instruction, after the change occurs.
4590 @cindex use only software watchpoints
4591 You can force @value{GDBN} to use only software watchpoints with the
4592 @kbd{set can-use-hw-watchpoints 0} command. With this variable set to
4593 zero, @value{GDBN} will never try to use hardware watchpoints, even if
4594 the underlying system supports them. (Note that hardware-assisted
4595 watchpoints that were set @emph{before} setting
4596 @code{can-use-hw-watchpoints} to zero will still use the hardware
4597 mechanism of watching expression values.)
4600 @item set can-use-hw-watchpoints
4601 @kindex set can-use-hw-watchpoints
4602 Set whether or not to use hardware watchpoints.
4604 @item show can-use-hw-watchpoints
4605 @kindex show can-use-hw-watchpoints
4606 Show the current mode of using hardware watchpoints.
4609 For remote targets, you can restrict the number of hardware
4610 watchpoints @value{GDBN} will use, see @ref{set remote
4611 hardware-breakpoint-limit}.
4613 When you issue the @code{watch} command, @value{GDBN} reports
4616 Hardware watchpoint @var{num}: @var{expr}
4620 if it was able to set a hardware watchpoint.
4622 Currently, the @code{awatch} and @code{rwatch} commands can only set
4623 hardware watchpoints, because accesses to data that don't change the
4624 value of the watched expression cannot be detected without examining
4625 every instruction as it is being executed, and @value{GDBN} does not do
4626 that currently. If @value{GDBN} finds that it is unable to set a
4627 hardware breakpoint with the @code{awatch} or @code{rwatch} command, it
4628 will print a message like this:
4631 Expression cannot be implemented with read/access watchpoint.
4634 Sometimes, @value{GDBN} cannot set a hardware watchpoint because the
4635 data type of the watched expression is wider than what a hardware
4636 watchpoint on the target machine can handle. For example, some systems
4637 can only watch regions that are up to 4 bytes wide; on such systems you
4638 cannot set hardware watchpoints for an expression that yields a
4639 double-precision floating-point number (which is typically 8 bytes
4640 wide). As a work-around, it might be possible to break the large region
4641 into a series of smaller ones and watch them with separate watchpoints.
4643 If you set too many hardware watchpoints, @value{GDBN} might be unable
4644 to insert all of them when you resume the execution of your program.
4645 Since the precise number of active watchpoints is unknown until such
4646 time as the program is about to be resumed, @value{GDBN} might not be
4647 able to warn you about this when you set the watchpoints, and the
4648 warning will be printed only when the program is resumed:
4651 Hardware watchpoint @var{num}: Could not insert watchpoint
4655 If this happens, delete or disable some of the watchpoints.
4657 Watching complex expressions that reference many variables can also
4658 exhaust the resources available for hardware-assisted watchpoints.
4659 That's because @value{GDBN} needs to watch every variable in the
4660 expression with separately allocated resources.
4662 If you call a function interactively using @code{print} or @code{call},
4663 any watchpoints you have set will be inactive until @value{GDBN} reaches another
4664 kind of breakpoint or the call completes.
4666 @value{GDBN} automatically deletes watchpoints that watch local
4667 (automatic) variables, or expressions that involve such variables, when
4668 they go out of scope, that is, when the execution leaves the block in
4669 which these variables were defined. In particular, when the program
4670 being debugged terminates, @emph{all} local variables go out of scope,
4671 and so only watchpoints that watch global variables remain set. If you
4672 rerun the program, you will need to set all such watchpoints again. One
4673 way of doing that would be to set a code breakpoint at the entry to the
4674 @code{main} function and when it breaks, set all the watchpoints.
4676 @cindex watchpoints and threads
4677 @cindex threads and watchpoints
4678 In multi-threaded programs, watchpoints will detect changes to the
4679 watched expression from every thread.
4682 @emph{Warning:} In multi-threaded programs, software watchpoints
4683 have only limited usefulness. If @value{GDBN} creates a software
4684 watchpoint, it can only watch the value of an expression @emph{in a
4685 single thread}. If you are confident that the expression can only
4686 change due to the current thread's activity (and if you are also
4687 confident that no other thread can become current), then you can use
4688 software watchpoints as usual. However, @value{GDBN} may not notice
4689 when a non-current thread's activity changes the expression. (Hardware
4690 watchpoints, in contrast, watch an expression in all threads.)
4693 @xref{set remote hardware-watchpoint-limit}.
4695 @node Set Catchpoints
4696 @subsection Setting Catchpoints
4697 @cindex catchpoints, setting
4698 @cindex exception handlers
4699 @cindex event handling
4701 You can use @dfn{catchpoints} to cause the debugger to stop for certain
4702 kinds of program events, such as C@t{++} exceptions or the loading of a
4703 shared library. Use the @code{catch} command to set a catchpoint.
4707 @item catch @var{event}
4708 Stop when @var{event} occurs. The @var{event} can be any of the following:
4711 @item throw @r{[}@var{regexp}@r{]}
4712 @itemx rethrow @r{[}@var{regexp}@r{]}
4713 @itemx catch @r{[}@var{regexp}@r{]}
4715 @kindex catch rethrow
4717 @cindex stop on C@t{++} exceptions
4718 The throwing, re-throwing, or catching of a C@t{++} exception.
4720 If @var{regexp} is given, then only exceptions whose type matches the
4721 regular expression will be caught.
4723 @vindex $_exception@r{, convenience variable}
4724 The convenience variable @code{$_exception} is available at an
4725 exception-related catchpoint, on some systems. This holds the
4726 exception being thrown.
4728 There are currently some limitations to C@t{++} exception handling in
4733 The support for these commands is system-dependent. Currently, only
4734 systems using the @samp{gnu-v3} C@t{++} ABI (@pxref{ABI}) are
4738 The regular expression feature and the @code{$_exception} convenience
4739 variable rely on the presence of some SDT probes in @code{libstdc++}.
4740 If these probes are not present, then these features cannot be used.
4741 These probes were first available in the GCC 4.8 release, but whether
4742 or not they are available in your GCC also depends on how it was
4746 The @code{$_exception} convenience variable is only valid at the
4747 instruction at which an exception-related catchpoint is set.
4750 When an exception-related catchpoint is hit, @value{GDBN} stops at a
4751 location in the system library which implements runtime exception
4752 support for C@t{++}, usually @code{libstdc++}. You can use @code{up}
4753 (@pxref{Selection}) to get to your code.
4756 If you call a function interactively, @value{GDBN} normally returns
4757 control to you when the function has finished executing. If the call
4758 raises an exception, however, the call may bypass the mechanism that
4759 returns control to you and cause your program either to abort or to
4760 simply continue running until it hits a breakpoint, catches a signal
4761 that @value{GDBN} is listening for, or exits. This is the case even if
4762 you set a catchpoint for the exception; catchpoints on exceptions are
4763 disabled within interactive calls. @xref{Calling}, for information on
4764 controlling this with @code{set unwind-on-terminating-exception}.
4767 You cannot raise an exception interactively.
4770 You cannot install an exception handler interactively.
4773 @item exception @r{[}@var{name}@r{]}
4774 @kindex catch exception
4775 @cindex Ada exception catching
4776 @cindex catch Ada exceptions
4777 An Ada exception being raised. If an exception name is specified
4778 at the end of the command (eg @code{catch exception Program_Error}),
4779 the debugger will stop only when this specific exception is raised.
4780 Otherwise, the debugger stops execution when any Ada exception is raised.
4782 When inserting an exception catchpoint on a user-defined exception whose
4783 name is identical to one of the exceptions defined by the language, the
4784 fully qualified name must be used as the exception name. Otherwise,
4785 @value{GDBN} will assume that it should stop on the pre-defined exception
4786 rather than the user-defined one. For instance, assuming an exception
4787 called @code{Constraint_Error} is defined in package @code{Pck}, then
4788 the command to use to catch such exceptions is @kbd{catch exception
4789 Pck.Constraint_Error}.
4791 @item exception unhandled
4792 @kindex catch exception unhandled
4793 An exception that was raised but is not handled by the program.
4795 @item handlers @r{[}@var{name}@r{]}
4796 @kindex catch handlers
4797 @cindex Ada exception handlers catching
4798 @cindex catch Ada exceptions when handled
4799 An Ada exception being handled. If an exception name is
4800 specified at the end of the command
4801 (eg @kbd{catch handlers Program_Error}), the debugger will stop
4802 only when this specific exception is handled.
4803 Otherwise, the debugger stops execution when any Ada exception is handled.
4805 When inserting a handlers catchpoint on a user-defined
4806 exception whose name is identical to one of the exceptions
4807 defined by the language, the fully qualified name must be used
4808 as the exception name. Otherwise, @value{GDBN} will assume that it
4809 should stop on the pre-defined exception rather than the
4810 user-defined one. For instance, assuming an exception called
4811 @code{Constraint_Error} is defined in package @code{Pck}, then the
4812 command to use to catch such exceptions handling is
4813 @kbd{catch handlers Pck.Constraint_Error}.
4816 @kindex catch assert
4817 A failed Ada assertion.
4821 @cindex break on fork/exec
4822 A call to @code{exec}.
4824 @anchor{catch syscall}
4826 @itemx syscall @r{[}@var{name} @r{|} @var{number} @r{|} @r{group:}@var{groupname} @r{|} @r{g:}@var{groupname}@r{]} @dots{}
4827 @kindex catch syscall
4828 @cindex break on a system call.
4829 A call to or return from a system call, a.k.a.@: @dfn{syscall}. A
4830 syscall is a mechanism for application programs to request a service
4831 from the operating system (OS) or one of the OS system services.
4832 @value{GDBN} can catch some or all of the syscalls issued by the
4833 debuggee, and show the related information for each syscall. If no
4834 argument is specified, calls to and returns from all system calls
4837 @var{name} can be any system call name that is valid for the
4838 underlying OS. Just what syscalls are valid depends on the OS. On
4839 GNU and Unix systems, you can find the full list of valid syscall
4840 names on @file{/usr/include/asm/unistd.h}.
4842 @c For MS-Windows, the syscall names and the corresponding numbers
4843 @c can be found, e.g., on this URL:
4844 @c http://www.metasploit.com/users/opcode/syscalls.html
4845 @c but we don't support Windows syscalls yet.
4847 Normally, @value{GDBN} knows in advance which syscalls are valid for
4848 each OS, so you can use the @value{GDBN} command-line completion
4849 facilities (@pxref{Completion,, command completion}) to list the
4852 You may also specify the system call numerically. A syscall's
4853 number is the value passed to the OS's syscall dispatcher to
4854 identify the requested service. When you specify the syscall by its
4855 name, @value{GDBN} uses its database of syscalls to convert the name
4856 into the corresponding numeric code, but using the number directly
4857 may be useful if @value{GDBN}'s database does not have the complete
4858 list of syscalls on your system (e.g., because @value{GDBN} lags
4859 behind the OS upgrades).
4861 You may specify a group of related syscalls to be caught at once using
4862 the @code{group:} syntax (@code{g:} is a shorter equivalent). For
4863 instance, on some platforms @value{GDBN} allows you to catch all
4864 network related syscalls, by passing the argument @code{group:network}
4865 to @code{catch syscall}. Note that not all syscall groups are
4866 available in every system. You can use the command completion
4867 facilities (@pxref{Completion,, command completion}) to list the
4868 syscall groups available on your environment.
4870 The example below illustrates how this command works if you don't provide
4874 (@value{GDBP}) catch syscall
4875 Catchpoint 1 (syscall)
4877 Starting program: /tmp/catch-syscall
4879 Catchpoint 1 (call to syscall 'close'), \
4880 0xffffe424 in __kernel_vsyscall ()
4884 Catchpoint 1 (returned from syscall 'close'), \
4885 0xffffe424 in __kernel_vsyscall ()
4889 Here is an example of catching a system call by name:
4892 (@value{GDBP}) catch syscall chroot
4893 Catchpoint 1 (syscall 'chroot' [61])
4895 Starting program: /tmp/catch-syscall
4897 Catchpoint 1 (call to syscall 'chroot'), \
4898 0xffffe424 in __kernel_vsyscall ()
4902 Catchpoint 1 (returned from syscall 'chroot'), \
4903 0xffffe424 in __kernel_vsyscall ()
4907 An example of specifying a system call numerically. In the case
4908 below, the syscall number has a corresponding entry in the XML
4909 file, so @value{GDBN} finds its name and prints it:
4912 (@value{GDBP}) catch syscall 252
4913 Catchpoint 1 (syscall(s) 'exit_group')
4915 Starting program: /tmp/catch-syscall
4917 Catchpoint 1 (call to syscall 'exit_group'), \
4918 0xffffe424 in __kernel_vsyscall ()
4922 Program exited normally.
4926 Here is an example of catching a syscall group:
4929 (@value{GDBP}) catch syscall group:process
4930 Catchpoint 1 (syscalls 'exit' [1] 'fork' [2] 'waitpid' [7]
4931 'execve' [11] 'wait4' [114] 'clone' [120] 'vfork' [190]
4932 'exit_group' [252] 'waitid' [284] 'unshare' [310])
4934 Starting program: /tmp/catch-syscall
4936 Catchpoint 1 (call to syscall fork), 0x00007ffff7df4e27 in open64 ()
4937 from /lib64/ld-linux-x86-64.so.2
4943 However, there can be situations when there is no corresponding name
4944 in XML file for that syscall number. In this case, @value{GDBN} prints
4945 a warning message saying that it was not able to find the syscall name,
4946 but the catchpoint will be set anyway. See the example below:
4949 (@value{GDBP}) catch syscall 764
4950 warning: The number '764' does not represent a known syscall.
4951 Catchpoint 2 (syscall 764)
4955 If you configure @value{GDBN} using the @samp{--without-expat} option,
4956 it will not be able to display syscall names. Also, if your
4957 architecture does not have an XML file describing its system calls,
4958 you will not be able to see the syscall names. It is important to
4959 notice that these two features are used for accessing the syscall
4960 name database. In either case, you will see a warning like this:
4963 (@value{GDBP}) catch syscall
4964 warning: Could not open "syscalls/i386-linux.xml"
4965 warning: Could not load the syscall XML file 'syscalls/i386-linux.xml'.
4966 GDB will not be able to display syscall names.
4967 Catchpoint 1 (syscall)
4971 Of course, the file name will change depending on your architecture and system.
4973 Still using the example above, you can also try to catch a syscall by its
4974 number. In this case, you would see something like:
4977 (@value{GDBP}) catch syscall 252
4978 Catchpoint 1 (syscall(s) 252)
4981 Again, in this case @value{GDBN} would not be able to display syscall's names.
4985 A call to @code{fork}.
4989 A call to @code{vfork}.
4991 @item load @r{[}@var{regexp}@r{]}
4992 @itemx unload @r{[}@var{regexp}@r{]}
4994 @kindex catch unload
4995 The loading or unloading of a shared library. If @var{regexp} is
4996 given, then the catchpoint will stop only if the regular expression
4997 matches one of the affected libraries.
4999 @item signal @r{[}@var{signal}@dots{} @r{|} @samp{all}@r{]}
5000 @kindex catch signal
5001 The delivery of a signal.
5003 With no arguments, this catchpoint will catch any signal that is not
5004 used internally by @value{GDBN}, specifically, all signals except
5005 @samp{SIGTRAP} and @samp{SIGINT}.
5007 With the argument @samp{all}, all signals, including those used by
5008 @value{GDBN}, will be caught. This argument cannot be used with other
5011 Otherwise, the arguments are a list of signal names as given to
5012 @code{handle} (@pxref{Signals}). Only signals specified in this list
5015 One reason that @code{catch signal} can be more useful than
5016 @code{handle} is that you can attach commands and conditions to the
5019 When a signal is caught by a catchpoint, the signal's @code{stop} and
5020 @code{print} settings, as specified by @code{handle}, are ignored.
5021 However, whether the signal is still delivered to the inferior depends
5022 on the @code{pass} setting; this can be changed in the catchpoint's
5027 @item tcatch @var{event}
5029 Set a catchpoint that is enabled only for one stop. The catchpoint is
5030 automatically deleted after the first time the event is caught.
5034 Use the @code{info break} command to list the current catchpoints.
5038 @subsection Deleting Breakpoints
5040 @cindex clearing breakpoints, watchpoints, catchpoints
5041 @cindex deleting breakpoints, watchpoints, catchpoints
5042 It is often necessary to eliminate a breakpoint, watchpoint, or
5043 catchpoint once it has done its job and you no longer want your program
5044 to stop there. This is called @dfn{deleting} the breakpoint. A
5045 breakpoint that has been deleted no longer exists; it is forgotten.
5047 With the @code{clear} command you can delete breakpoints according to
5048 where they are in your program. With the @code{delete} command you can
5049 delete individual breakpoints, watchpoints, or catchpoints by specifying
5050 their breakpoint numbers.
5052 It is not necessary to delete a breakpoint to proceed past it. @value{GDBN}
5053 automatically ignores breakpoints on the first instruction to be executed
5054 when you continue execution without changing the execution address.
5059 Delete any breakpoints at the next instruction to be executed in the
5060 selected stack frame (@pxref{Selection, ,Selecting a Frame}). When
5061 the innermost frame is selected, this is a good way to delete a
5062 breakpoint where your program just stopped.
5064 @item clear @var{location}
5065 Delete any breakpoints set at the specified @var{location}.
5066 @xref{Specify Location}, for the various forms of @var{location}; the
5067 most useful ones are listed below:
5070 @item clear @var{function}
5071 @itemx clear @var{filename}:@var{function}
5072 Delete any breakpoints set at entry to the named @var{function}.
5074 @item clear @var{linenum}
5075 @itemx clear @var{filename}:@var{linenum}
5076 Delete any breakpoints set at or within the code of the specified
5077 @var{linenum} of the specified @var{filename}.
5080 @cindex delete breakpoints
5082 @kindex d @r{(@code{delete})}
5083 @item delete @r{[}breakpoints@r{]} @r{[}@var{list}@dots{}@r{]}
5084 Delete the breakpoints, watchpoints, or catchpoints of the breakpoint
5085 list specified as argument. If no argument is specified, delete all
5086 breakpoints (@value{GDBN} asks confirmation, unless you have @code{set
5087 confirm off}). You can abbreviate this command as @code{d}.
5091 @subsection Disabling Breakpoints
5093 @cindex enable/disable a breakpoint
5094 Rather than deleting a breakpoint, watchpoint, or catchpoint, you might
5095 prefer to @dfn{disable} it. This makes the breakpoint inoperative as if
5096 it had been deleted, but remembers the information on the breakpoint so
5097 that you can @dfn{enable} it again later.
5099 You disable and enable breakpoints, watchpoints, and catchpoints with
5100 the @code{enable} and @code{disable} commands, optionally specifying
5101 one or more breakpoint numbers as arguments. Use @code{info break} to
5102 print a list of all breakpoints, watchpoints, and catchpoints if you
5103 do not know which numbers to use.
5105 Disabling and enabling a breakpoint that has multiple locations
5106 affects all of its locations.
5108 A breakpoint, watchpoint, or catchpoint can have any of several
5109 different states of enablement:
5113 Enabled. The breakpoint stops your program. A breakpoint set
5114 with the @code{break} command starts out in this state.
5116 Disabled. The breakpoint has no effect on your program.
5118 Enabled once. The breakpoint stops your program, but then becomes
5121 Enabled for a count. The breakpoint stops your program for the next
5122 N times, then becomes disabled.
5124 Enabled for deletion. The breakpoint stops your program, but
5125 immediately after it does so it is deleted permanently. A breakpoint
5126 set with the @code{tbreak} command starts out in this state.
5129 You can use the following commands to enable or disable breakpoints,
5130 watchpoints, and catchpoints:
5134 @kindex dis @r{(@code{disable})}
5135 @item disable @r{[}breakpoints@r{]} @r{[}@var{list}@dots{}@r{]}
5136 Disable the specified breakpoints---or all breakpoints, if none are
5137 listed. A disabled breakpoint has no effect but is not forgotten. All
5138 options such as ignore-counts, conditions and commands are remembered in
5139 case the breakpoint is enabled again later. You may abbreviate
5140 @code{disable} as @code{dis}.
5143 @item enable @r{[}breakpoints@r{]} @r{[}@var{list}@dots{}@r{]}
5144 Enable the specified breakpoints (or all defined breakpoints). They
5145 become effective once again in stopping your program.
5147 @item enable @r{[}breakpoints@r{]} once @var{list}@dots{}
5148 Enable the specified breakpoints temporarily. @value{GDBN} disables any
5149 of these breakpoints immediately after stopping your program.
5151 @item enable @r{[}breakpoints@r{]} count @var{count} @var{list}@dots{}
5152 Enable the specified breakpoints temporarily. @value{GDBN} records
5153 @var{count} with each of the specified breakpoints, and decrements a
5154 breakpoint's count when it is hit. When any count reaches 0,
5155 @value{GDBN} disables that breakpoint. If a breakpoint has an ignore
5156 count (@pxref{Conditions, ,Break Conditions}), that will be
5157 decremented to 0 before @var{count} is affected.
5159 @item enable @r{[}breakpoints@r{]} delete @var{list}@dots{}
5160 Enable the specified breakpoints to work once, then die. @value{GDBN}
5161 deletes any of these breakpoints as soon as your program stops there.
5162 Breakpoints set by the @code{tbreak} command start out in this state.
5165 @c FIXME: I think the following ``Except for [...] @code{tbreak}'' is
5166 @c confusing: tbreak is also initially enabled.
5167 Except for a breakpoint set with @code{tbreak} (@pxref{Set Breaks,
5168 ,Setting Breakpoints}), breakpoints that you set are initially enabled;
5169 subsequently, they become disabled or enabled only when you use one of
5170 the commands above. (The command @code{until} can set and delete a
5171 breakpoint of its own, but it does not change the state of your other
5172 breakpoints; see @ref{Continuing and Stepping, ,Continuing and
5176 @subsection Break Conditions
5177 @cindex conditional breakpoints
5178 @cindex breakpoint conditions
5180 @c FIXME what is scope of break condition expr? Context where wanted?
5181 @c in particular for a watchpoint?
5182 The simplest sort of breakpoint breaks every time your program reaches a
5183 specified place. You can also specify a @dfn{condition} for a
5184 breakpoint. A condition is just a Boolean expression in your
5185 programming language (@pxref{Expressions, ,Expressions}). A breakpoint with
5186 a condition evaluates the expression each time your program reaches it,
5187 and your program stops only if the condition is @emph{true}.
5189 This is the converse of using assertions for program validation; in that
5190 situation, you want to stop when the assertion is violated---that is,
5191 when the condition is false. In C, if you want to test an assertion expressed
5192 by the condition @var{assert}, you should set the condition
5193 @samp{! @var{assert}} on the appropriate breakpoint.
5195 Conditions are also accepted for watchpoints; you may not need them,
5196 since a watchpoint is inspecting the value of an expression anyhow---but
5197 it might be simpler, say, to just set a watchpoint on a variable name,
5198 and specify a condition that tests whether the new value is an interesting
5201 Break conditions can have side effects, and may even call functions in
5202 your program. This can be useful, for example, to activate functions
5203 that log program progress, or to use your own print functions to
5204 format special data structures. The effects are completely predictable
5205 unless there is another enabled breakpoint at the same address. (In
5206 that case, @value{GDBN} might see the other breakpoint first and stop your
5207 program without checking the condition of this one.) Note that
5208 breakpoint commands are usually more convenient and flexible than break
5210 purpose of performing side effects when a breakpoint is reached
5211 (@pxref{Break Commands, ,Breakpoint Command Lists}).
5213 Breakpoint conditions can also be evaluated on the target's side if
5214 the target supports it. Instead of evaluating the conditions locally,
5215 @value{GDBN} encodes the expression into an agent expression
5216 (@pxref{Agent Expressions}) suitable for execution on the target,
5217 independently of @value{GDBN}. Global variables become raw memory
5218 locations, locals become stack accesses, and so forth.
5220 In this case, @value{GDBN} will only be notified of a breakpoint trigger
5221 when its condition evaluates to true. This mechanism may provide faster
5222 response times depending on the performance characteristics of the target
5223 since it does not need to keep @value{GDBN} informed about
5224 every breakpoint trigger, even those with false conditions.
5226 Break conditions can be specified when a breakpoint is set, by using
5227 @samp{if} in the arguments to the @code{break} command. @xref{Set
5228 Breaks, ,Setting Breakpoints}. They can also be changed at any time
5229 with the @code{condition} command.
5231 You can also use the @code{if} keyword with the @code{watch} command.
5232 The @code{catch} command does not recognize the @code{if} keyword;
5233 @code{condition} is the only way to impose a further condition on a
5238 @item condition @var{bnum} @var{expression}
5239 Specify @var{expression} as the break condition for breakpoint,
5240 watchpoint, or catchpoint number @var{bnum}. After you set a condition,
5241 breakpoint @var{bnum} stops your program only if the value of
5242 @var{expression} is true (nonzero, in C). When you use
5243 @code{condition}, @value{GDBN} checks @var{expression} immediately for
5244 syntactic correctness, and to determine whether symbols in it have
5245 referents in the context of your breakpoint. If @var{expression} uses
5246 symbols not referenced in the context of the breakpoint, @value{GDBN}
5247 prints an error message:
5250 No symbol "foo" in current context.
5255 not actually evaluate @var{expression} at the time the @code{condition}
5256 command (or a command that sets a breakpoint with a condition, like
5257 @code{break if @dots{}}) is given, however. @xref{Expressions, ,Expressions}.
5259 @item condition @var{bnum}
5260 Remove the condition from breakpoint number @var{bnum}. It becomes
5261 an ordinary unconditional breakpoint.
5264 @cindex ignore count (of breakpoint)
5265 A special case of a breakpoint condition is to stop only when the
5266 breakpoint has been reached a certain number of times. This is so
5267 useful that there is a special way to do it, using the @dfn{ignore
5268 count} of the breakpoint. Every breakpoint has an ignore count, which
5269 is an integer. Most of the time, the ignore count is zero, and
5270 therefore has no effect. But if your program reaches a breakpoint whose
5271 ignore count is positive, then instead of stopping, it just decrements
5272 the ignore count by one and continues. As a result, if the ignore count
5273 value is @var{n}, the breakpoint does not stop the next @var{n} times
5274 your program reaches it.
5278 @item ignore @var{bnum} @var{count}
5279 Set the ignore count of breakpoint number @var{bnum} to @var{count}.
5280 The next @var{count} times the breakpoint is reached, your program's
5281 execution does not stop; other than to decrement the ignore count, @value{GDBN}
5284 To make the breakpoint stop the next time it is reached, specify
5287 When you use @code{continue} to resume execution of your program from a
5288 breakpoint, you can specify an ignore count directly as an argument to
5289 @code{continue}, rather than using @code{ignore}. @xref{Continuing and
5290 Stepping,,Continuing and Stepping}.
5292 If a breakpoint has a positive ignore count and a condition, the
5293 condition is not checked. Once the ignore count reaches zero,
5294 @value{GDBN} resumes checking the condition.
5296 You could achieve the effect of the ignore count with a condition such
5297 as @w{@samp{$foo-- <= 0}} using a debugger convenience variable that
5298 is decremented each time. @xref{Convenience Vars, ,Convenience
5302 Ignore counts apply to breakpoints, watchpoints, and catchpoints.
5305 @node Break Commands
5306 @subsection Breakpoint Command Lists
5308 @cindex breakpoint commands
5309 You can give any breakpoint (or watchpoint or catchpoint) a series of
5310 commands to execute when your program stops due to that breakpoint. For
5311 example, you might want to print the values of certain expressions, or
5312 enable other breakpoints.
5316 @kindex end@r{ (breakpoint commands)}
5317 @item commands @r{[}@var{list}@dots{}@r{]}
5318 @itemx @dots{} @var{command-list} @dots{}
5320 Specify a list of commands for the given breakpoints. The commands
5321 themselves appear on the following lines. Type a line containing just
5322 @code{end} to terminate the commands.
5324 To remove all commands from a breakpoint, type @code{commands} and
5325 follow it immediately with @code{end}; that is, give no commands.
5327 With no argument, @code{commands} refers to the last breakpoint,
5328 watchpoint, or catchpoint set (not to the breakpoint most recently
5329 encountered). If the most recent breakpoints were set with a single
5330 command, then the @code{commands} will apply to all the breakpoints
5331 set by that command. This applies to breakpoints set by
5332 @code{rbreak}, and also applies when a single @code{break} command
5333 creates multiple breakpoints (@pxref{Ambiguous Expressions,,Ambiguous
5337 Pressing @key{RET} as a means of repeating the last @value{GDBN} command is
5338 disabled within a @var{command-list}.
5340 You can use breakpoint commands to start your program up again. Simply
5341 use the @code{continue} command, or @code{step}, or any other command
5342 that resumes execution.
5344 Any other commands in the command list, after a command that resumes
5345 execution, are ignored. This is because any time you resume execution
5346 (even with a simple @code{next} or @code{step}), you may encounter
5347 another breakpoint---which could have its own command list, leading to
5348 ambiguities about which list to execute.
5351 If the first command you specify in a command list is @code{silent}, the
5352 usual message about stopping at a breakpoint is not printed. This may
5353 be desirable for breakpoints that are to print a specific message and
5354 then continue. If none of the remaining commands print anything, you
5355 see no sign that the breakpoint was reached. @code{silent} is
5356 meaningful only at the beginning of a breakpoint command list.
5358 The commands @code{echo}, @code{output}, and @code{printf} allow you to
5359 print precisely controlled output, and are often useful in silent
5360 breakpoints. @xref{Output, ,Commands for Controlled Output}.
5362 For example, here is how you could use breakpoint commands to print the
5363 value of @code{x} at entry to @code{foo} whenever @code{x} is positive.
5369 printf "x is %d\n",x
5374 One application for breakpoint commands is to compensate for one bug so
5375 you can test for another. Put a breakpoint just after the erroneous line
5376 of code, give it a condition to detect the case in which something
5377 erroneous has been done, and give it commands to assign correct values
5378 to any variables that need them. End with the @code{continue} command
5379 so that your program does not stop, and start with the @code{silent}
5380 command so that no output is produced. Here is an example:
5391 @node Dynamic Printf
5392 @subsection Dynamic Printf
5394 @cindex dynamic printf
5396 The dynamic printf command @code{dprintf} combines a breakpoint with
5397 formatted printing of your program's data to give you the effect of
5398 inserting @code{printf} calls into your program on-the-fly, without
5399 having to recompile it.
5401 In its most basic form, the output goes to the GDB console. However,
5402 you can set the variable @code{dprintf-style} for alternate handling.
5403 For instance, you can ask to format the output by calling your
5404 program's @code{printf} function. This has the advantage that the
5405 characters go to the program's output device, so they can recorded in
5406 redirects to files and so forth.
5408 If you are doing remote debugging with a stub or agent, you can also
5409 ask to have the printf handled by the remote agent. In addition to
5410 ensuring that the output goes to the remote program's device along
5411 with any other output the program might produce, you can also ask that
5412 the dprintf remain active even after disconnecting from the remote
5413 target. Using the stub/agent is also more efficient, as it can do
5414 everything without needing to communicate with @value{GDBN}.
5418 @item dprintf @var{location},@var{template},@var{expression}[,@var{expression}@dots{}]
5419 Whenever execution reaches @var{location}, print the values of one or
5420 more @var{expressions} under the control of the string @var{template}.
5421 To print several values, separate them with commas.
5423 @item set dprintf-style @var{style}
5424 Set the dprintf output to be handled in one of several different
5425 styles enumerated below. A change of style affects all existing
5426 dynamic printfs immediately. (If you need individual control over the
5427 print commands, simply define normal breakpoints with
5428 explicitly-supplied command lists.)
5432 @kindex dprintf-style gdb
5433 Handle the output using the @value{GDBN} @code{printf} command.
5436 @kindex dprintf-style call
5437 Handle the output by calling a function in your program (normally
5441 @kindex dprintf-style agent
5442 Have the remote debugging agent (such as @code{gdbserver}) handle
5443 the output itself. This style is only available for agents that
5444 support running commands on the target.
5447 @item set dprintf-function @var{function}
5448 Set the function to call if the dprintf style is @code{call}. By
5449 default its value is @code{printf}. You may set it to any expression.
5450 that @value{GDBN} can evaluate to a function, as per the @code{call}
5453 @item set dprintf-channel @var{channel}
5454 Set a ``channel'' for dprintf. If set to a non-empty value,
5455 @value{GDBN} will evaluate it as an expression and pass the result as
5456 a first argument to the @code{dprintf-function}, in the manner of
5457 @code{fprintf} and similar functions. Otherwise, the dprintf format
5458 string will be the first argument, in the manner of @code{printf}.
5460 As an example, if you wanted @code{dprintf} output to go to a logfile
5461 that is a standard I/O stream assigned to the variable @code{mylog},
5462 you could do the following:
5465 (gdb) set dprintf-style call
5466 (gdb) set dprintf-function fprintf
5467 (gdb) set dprintf-channel mylog
5468 (gdb) dprintf 25,"at line 25, glob=%d\n",glob
5469 Dprintf 1 at 0x123456: file main.c, line 25.
5471 1 dprintf keep y 0x00123456 in main at main.c:25
5472 call (void) fprintf (mylog,"at line 25, glob=%d\n",glob)
5477 Note that the @code{info break} displays the dynamic printf commands
5478 as normal breakpoint commands; you can thus easily see the effect of
5479 the variable settings.
5481 @item set disconnected-dprintf on
5482 @itemx set disconnected-dprintf off
5483 @kindex set disconnected-dprintf
5484 Choose whether @code{dprintf} commands should continue to run if
5485 @value{GDBN} has disconnected from the target. This only applies
5486 if the @code{dprintf-style} is @code{agent}.
5488 @item show disconnected-dprintf off
5489 @kindex show disconnected-dprintf
5490 Show the current choice for disconnected @code{dprintf}.
5494 @value{GDBN} does not check the validity of function and channel,
5495 relying on you to supply values that are meaningful for the contexts
5496 in which they are being used. For instance, the function and channel
5497 may be the values of local variables, but if that is the case, then
5498 all enabled dynamic prints must be at locations within the scope of
5499 those locals. If evaluation fails, @value{GDBN} will report an error.
5501 @node Save Breakpoints
5502 @subsection How to save breakpoints to a file
5504 To save breakpoint definitions to a file use the @w{@code{save
5505 breakpoints}} command.
5508 @kindex save breakpoints
5509 @cindex save breakpoints to a file for future sessions
5510 @item save breakpoints [@var{filename}]
5511 This command saves all current breakpoint definitions together with
5512 their commands and ignore counts, into a file @file{@var{filename}}
5513 suitable for use in a later debugging session. This includes all
5514 types of breakpoints (breakpoints, watchpoints, catchpoints,
5515 tracepoints). To read the saved breakpoint definitions, use the
5516 @code{source} command (@pxref{Command Files}). Note that watchpoints
5517 with expressions involving local variables may fail to be recreated
5518 because it may not be possible to access the context where the
5519 watchpoint is valid anymore. Because the saved breakpoint definitions
5520 are simply a sequence of @value{GDBN} commands that recreate the
5521 breakpoints, you can edit the file in your favorite editing program,
5522 and remove the breakpoint definitions you're not interested in, or
5523 that can no longer be recreated.
5526 @node Static Probe Points
5527 @subsection Static Probe Points
5529 @cindex static probe point, SystemTap
5530 @cindex static probe point, DTrace
5531 @value{GDBN} supports @dfn{SDT} probes in the code. @acronym{SDT} stands
5532 for Statically Defined Tracing, and the probes are designed to have a tiny
5533 runtime code and data footprint, and no dynamic relocations.
5535 Currently, the following types of probes are supported on
5536 ELF-compatible systems:
5540 @item @code{SystemTap} (@uref{http://sourceware.org/systemtap/})
5541 @acronym{SDT} probes@footnote{See
5542 @uref{http://sourceware.org/systemtap/wiki/AddingUserSpaceProbingToApps}
5543 for more information on how to add @code{SystemTap} @acronym{SDT}
5544 probes in your applications.}. @code{SystemTap} probes are usable
5545 from assembly, C and C@t{++} languages@footnote{See
5546 @uref{http://sourceware.org/systemtap/wiki/UserSpaceProbeImplementation}
5547 for a good reference on how the @acronym{SDT} probes are implemented.}.
5549 @item @code{DTrace} (@uref{http://oss.oracle.com/projects/DTrace})
5550 @acronym{USDT} probes. @code{DTrace} probes are usable from C and
5554 @cindex semaphores on static probe points
5555 Some @code{SystemTap} probes have an associated semaphore variable;
5556 for instance, this happens automatically if you defined your probe
5557 using a DTrace-style @file{.d} file. If your probe has a semaphore,
5558 @value{GDBN} will automatically enable it when you specify a
5559 breakpoint using the @samp{-probe-stap} notation. But, if you put a
5560 breakpoint at a probe's location by some other method (e.g.,
5561 @code{break file:line}), then @value{GDBN} will not automatically set
5562 the semaphore. @code{DTrace} probes do not support semaphores.
5564 You can examine the available static static probes using @code{info
5565 probes}, with optional arguments:
5569 @item info probes @r{[}@var{type}@r{]} @r{[}@var{provider} @r{[}@var{name} @r{[}@var{objfile}@r{]}@r{]}@r{]}
5570 If given, @var{type} is either @code{stap} for listing
5571 @code{SystemTap} probes or @code{dtrace} for listing @code{DTrace}
5572 probes. If omitted all probes are listed regardless of their types.
5574 If given, @var{provider} is a regular expression used to match against provider
5575 names when selecting which probes to list. If omitted, probes by all
5576 probes from all providers are listed.
5578 If given, @var{name} is a regular expression to match against probe names
5579 when selecting which probes to list. If omitted, probe names are not
5580 considered when deciding whether to display them.
5582 If given, @var{objfile} is a regular expression used to select which
5583 object files (executable or shared libraries) to examine. If not
5584 given, all object files are considered.
5586 @item info probes all
5587 List the available static probes, from all types.
5590 @cindex enabling and disabling probes
5591 Some probe points can be enabled and/or disabled. The effect of
5592 enabling or disabling a probe depends on the type of probe being
5593 handled. Some @code{DTrace} probes can be enabled or
5594 disabled, but @code{SystemTap} probes cannot be disabled.
5596 You can enable (or disable) one or more probes using the following
5597 commands, with optional arguments:
5600 @kindex enable probes
5601 @item enable probes @r{[}@var{provider} @r{[}@var{name} @r{[}@var{objfile}@r{]}@r{]}@r{]}
5602 If given, @var{provider} is a regular expression used to match against
5603 provider names when selecting which probes to enable. If omitted,
5604 all probes from all providers are enabled.
5606 If given, @var{name} is a regular expression to match against probe
5607 names when selecting which probes to enable. If omitted, probe names
5608 are not considered when deciding whether to enable them.
5610 If given, @var{objfile} is a regular expression used to select which
5611 object files (executable or shared libraries) to examine. If not
5612 given, all object files are considered.
5614 @kindex disable probes
5615 @item disable probes @r{[}@var{provider} @r{[}@var{name} @r{[}@var{objfile}@r{]}@r{]}@r{]}
5616 See the @code{enable probes} command above for a description of the
5617 optional arguments accepted by this command.
5620 @vindex $_probe_arg@r{, convenience variable}
5621 A probe may specify up to twelve arguments. These are available at the
5622 point at which the probe is defined---that is, when the current PC is
5623 at the probe's location. The arguments are available using the
5624 convenience variables (@pxref{Convenience Vars})
5625 @code{$_probe_arg0}@dots{}@code{$_probe_arg11}. In @code{SystemTap}
5626 probes each probe argument is an integer of the appropriate size;
5627 types are not preserved. In @code{DTrace} probes types are preserved
5628 provided that they are recognized as such by @value{GDBN}; otherwise
5629 the value of the probe argument will be a long integer. The
5630 convenience variable @code{$_probe_argc} holds the number of arguments
5631 at the current probe point.
5633 These variables are always available, but attempts to access them at
5634 any location other than a probe point will cause @value{GDBN} to give
5638 @c @ifclear BARETARGET
5639 @node Error in Breakpoints
5640 @subsection ``Cannot insert breakpoints''
5642 If you request too many active hardware-assisted breakpoints and
5643 watchpoints, you will see this error message:
5645 @c FIXME: the precise wording of this message may change; the relevant
5646 @c source change is not committed yet (Sep 3, 1999).
5648 Stopped; cannot insert breakpoints.
5649 You may have requested too many hardware breakpoints and watchpoints.
5653 This message is printed when you attempt to resume the program, since
5654 only then @value{GDBN} knows exactly how many hardware breakpoints and
5655 watchpoints it needs to insert.
5657 When this message is printed, you need to disable or remove some of the
5658 hardware-assisted breakpoints and watchpoints, and then continue.
5660 @node Breakpoint-related Warnings
5661 @subsection ``Breakpoint address adjusted...''
5662 @cindex breakpoint address adjusted
5664 Some processor architectures place constraints on the addresses at
5665 which breakpoints may be placed. For architectures thus constrained,
5666 @value{GDBN} will attempt to adjust the breakpoint's address to comply
5667 with the constraints dictated by the architecture.
5669 One example of such an architecture is the Fujitsu FR-V. The FR-V is
5670 a VLIW architecture in which a number of RISC-like instructions may be
5671 bundled together for parallel execution. The FR-V architecture
5672 constrains the location of a breakpoint instruction within such a
5673 bundle to the instruction with the lowest address. @value{GDBN}
5674 honors this constraint by adjusting a breakpoint's address to the
5675 first in the bundle.
5677 It is not uncommon for optimized code to have bundles which contain
5678 instructions from different source statements, thus it may happen that
5679 a breakpoint's address will be adjusted from one source statement to
5680 another. Since this adjustment may significantly alter @value{GDBN}'s
5681 breakpoint related behavior from what the user expects, a warning is
5682 printed when the breakpoint is first set and also when the breakpoint
5685 A warning like the one below is printed when setting a breakpoint
5686 that's been subject to address adjustment:
5689 warning: Breakpoint address adjusted from 0x00010414 to 0x00010410.
5692 Such warnings are printed both for user settable and @value{GDBN}'s
5693 internal breakpoints. If you see one of these warnings, you should
5694 verify that a breakpoint set at the adjusted address will have the
5695 desired affect. If not, the breakpoint in question may be removed and
5696 other breakpoints may be set which will have the desired behavior.
5697 E.g., it may be sufficient to place the breakpoint at a later
5698 instruction. A conditional breakpoint may also be useful in some
5699 cases to prevent the breakpoint from triggering too often.
5701 @value{GDBN} will also issue a warning when stopping at one of these
5702 adjusted breakpoints:
5705 warning: Breakpoint 1 address previously adjusted from 0x00010414
5709 When this warning is encountered, it may be too late to take remedial
5710 action except in cases where the breakpoint is hit earlier or more
5711 frequently than expected.
5713 @node Continuing and Stepping
5714 @section Continuing and Stepping
5718 @cindex resuming execution
5719 @dfn{Continuing} means resuming program execution until your program
5720 completes normally. In contrast, @dfn{stepping} means executing just
5721 one more ``step'' of your program, where ``step'' may mean either one
5722 line of source code, or one machine instruction (depending on what
5723 particular command you use). Either when continuing or when stepping,
5724 your program may stop even sooner, due to a breakpoint or a signal. (If
5725 it stops due to a signal, you may want to use @code{handle}, or use
5726 @samp{signal 0} to resume execution (@pxref{Signals, ,Signals}),
5727 or you may step into the signal's handler (@pxref{stepping and signal
5732 @kindex c @r{(@code{continue})}
5733 @kindex fg @r{(resume foreground execution)}
5734 @item continue @r{[}@var{ignore-count}@r{]}
5735 @itemx c @r{[}@var{ignore-count}@r{]}
5736 @itemx fg @r{[}@var{ignore-count}@r{]}
5737 Resume program execution, at the address where your program last stopped;
5738 any breakpoints set at that address are bypassed. The optional argument
5739 @var{ignore-count} allows you to specify a further number of times to
5740 ignore a breakpoint at this location; its effect is like that of
5741 @code{ignore} (@pxref{Conditions, ,Break Conditions}).
5743 The argument @var{ignore-count} is meaningful only when your program
5744 stopped due to a breakpoint. At other times, the argument to
5745 @code{continue} is ignored.
5747 The synonyms @code{c} and @code{fg} (for @dfn{foreground}, as the
5748 debugged program is deemed to be the foreground program) are provided
5749 purely for convenience, and have exactly the same behavior as
5753 To resume execution at a different place, you can use @code{return}
5754 (@pxref{Returning, ,Returning from a Function}) to go back to the
5755 calling function; or @code{jump} (@pxref{Jumping, ,Continuing at a
5756 Different Address}) to go to an arbitrary location in your program.
5758 A typical technique for using stepping is to set a breakpoint
5759 (@pxref{Breakpoints, ,Breakpoints; Watchpoints; and Catchpoints}) at the
5760 beginning of the function or the section of your program where a problem
5761 is believed to lie, run your program until it stops at that breakpoint,
5762 and then step through the suspect area, examining the variables that are
5763 interesting, until you see the problem happen.
5767 @kindex s @r{(@code{step})}
5769 Continue running your program until control reaches a different source
5770 line, then stop it and return control to @value{GDBN}. This command is
5771 abbreviated @code{s}.
5774 @c "without debugging information" is imprecise; actually "without line
5775 @c numbers in the debugging information". (gcc -g1 has debugging info but
5776 @c not line numbers). But it seems complex to try to make that
5777 @c distinction here.
5778 @emph{Warning:} If you use the @code{step} command while control is
5779 within a function that was compiled without debugging information,
5780 execution proceeds until control reaches a function that does have
5781 debugging information. Likewise, it will not step into a function which
5782 is compiled without debugging information. To step through functions
5783 without debugging information, use the @code{stepi} command, described
5787 The @code{step} command only stops at the first instruction of a source
5788 line. This prevents the multiple stops that could otherwise occur in
5789 @code{switch} statements, @code{for} loops, etc. @code{step} continues
5790 to stop if a function that has debugging information is called within
5791 the line. In other words, @code{step} @emph{steps inside} any functions
5792 called within the line.
5794 Also, the @code{step} command only enters a function if there is line
5795 number information for the function. Otherwise it acts like the
5796 @code{next} command. This avoids problems when using @code{cc -gl}
5797 on @acronym{MIPS} machines. Previously, @code{step} entered subroutines if there
5798 was any debugging information about the routine.
5800 @item step @var{count}
5801 Continue running as in @code{step}, but do so @var{count} times. If a
5802 breakpoint is reached, or a signal not related to stepping occurs before
5803 @var{count} steps, stepping stops right away.
5806 @kindex n @r{(@code{next})}
5807 @item next @r{[}@var{count}@r{]}
5808 Continue to the next source line in the current (innermost) stack frame.
5809 This is similar to @code{step}, but function calls that appear within
5810 the line of code are executed without stopping. Execution stops when
5811 control reaches a different line of code at the original stack level
5812 that was executing when you gave the @code{next} command. This command
5813 is abbreviated @code{n}.
5815 An argument @var{count} is a repeat count, as for @code{step}.
5818 @c FIX ME!! Do we delete this, or is there a way it fits in with
5819 @c the following paragraph? --- Vctoria
5821 @c @code{next} within a function that lacks debugging information acts like
5822 @c @code{step}, but any function calls appearing within the code of the
5823 @c function are executed without stopping.
5825 The @code{next} command only stops at the first instruction of a
5826 source line. This prevents multiple stops that could otherwise occur in
5827 @code{switch} statements, @code{for} loops, etc.
5829 @kindex set step-mode
5831 @cindex functions without line info, and stepping
5832 @cindex stepping into functions with no line info
5833 @itemx set step-mode on
5834 The @code{set step-mode on} command causes the @code{step} command to
5835 stop at the first instruction of a function which contains no debug line
5836 information rather than stepping over it.
5838 This is useful in cases where you may be interested in inspecting the
5839 machine instructions of a function which has no symbolic info and do not
5840 want @value{GDBN} to automatically skip over this function.
5842 @item set step-mode off
5843 Causes the @code{step} command to step over any functions which contains no
5844 debug information. This is the default.
5846 @item show step-mode
5847 Show whether @value{GDBN} will stop in or step over functions without
5848 source line debug information.
5851 @kindex fin @r{(@code{finish})}
5853 Continue running until just after function in the selected stack frame
5854 returns. Print the returned value (if any). This command can be
5855 abbreviated as @code{fin}.
5857 Contrast this with the @code{return} command (@pxref{Returning,
5858 ,Returning from a Function}).
5860 @kindex set print finish
5861 @kindex show print finish
5862 @item set print finish @r{[}on|off@r{]}
5863 @itemx show print finish
5864 By default the @code{finish} command will show the value that is
5865 returned by the function. This can be disabled using @code{set print
5866 finish off}. When disabled, the value is still entered into the value
5867 history (@pxref{Value History}), but not displayed.
5870 @kindex u @r{(@code{until})}
5871 @cindex run until specified location
5874 Continue running until a source line past the current line, in the
5875 current stack frame, is reached. This command is used to avoid single
5876 stepping through a loop more than once. It is like the @code{next}
5877 command, except that when @code{until} encounters a jump, it
5878 automatically continues execution until the program counter is greater
5879 than the address of the jump.
5881 This means that when you reach the end of a loop after single stepping
5882 though it, @code{until} makes your program continue execution until it
5883 exits the loop. In contrast, a @code{next} command at the end of a loop
5884 simply steps back to the beginning of the loop, which forces you to step
5885 through the next iteration.
5887 @code{until} always stops your program if it attempts to exit the current
5890 @code{until} may produce somewhat counterintuitive results if the order
5891 of machine code does not match the order of the source lines. For
5892 example, in the following excerpt from a debugging session, the @code{f}
5893 (@code{frame}) command shows that execution is stopped at line
5894 @code{206}; yet when we use @code{until}, we get to line @code{195}:
5898 #0 main (argc=4, argv=0xf7fffae8) at m4.c:206
5900 (@value{GDBP}) until
5901 195 for ( ; argc > 0; NEXTARG) @{
5904 This happened because, for execution efficiency, the compiler had
5905 generated code for the loop closure test at the end, rather than the
5906 start, of the loop---even though the test in a C @code{for}-loop is
5907 written before the body of the loop. The @code{until} command appeared
5908 to step back to the beginning of the loop when it advanced to this
5909 expression; however, it has not really gone to an earlier
5910 statement---not in terms of the actual machine code.
5912 @code{until} with no argument works by means of single
5913 instruction stepping, and hence is slower than @code{until} with an
5916 @item until @var{location}
5917 @itemx u @var{location}
5918 Continue running your program until either the specified @var{location} is
5919 reached, or the current stack frame returns. The location is any of
5920 the forms described in @ref{Specify Location}.
5921 This form of the command uses temporary breakpoints, and
5922 hence is quicker than @code{until} without an argument. The specified
5923 location is actually reached only if it is in the current frame. This
5924 implies that @code{until} can be used to skip over recursive function
5925 invocations. For instance in the code below, if the current location is
5926 line @code{96}, issuing @code{until 99} will execute the program up to
5927 line @code{99} in the same invocation of factorial, i.e., after the inner
5928 invocations have returned.
5931 94 int factorial (int value)
5933 96 if (value > 1) @{
5934 97 value *= factorial (value - 1);
5941 @kindex advance @var{location}
5942 @item advance @var{location}
5943 Continue running the program up to the given @var{location}. An argument is
5944 required, which should be of one of the forms described in
5945 @ref{Specify Location}.
5946 Execution will also stop upon exit from the current stack
5947 frame. This command is similar to @code{until}, but @code{advance} will
5948 not skip over recursive function calls, and the target location doesn't
5949 have to be in the same frame as the current one.
5953 @kindex si @r{(@code{stepi})}
5955 @itemx stepi @var{arg}
5957 Execute one machine instruction, then stop and return to the debugger.
5959 It is often useful to do @samp{display/i $pc} when stepping by machine
5960 instructions. This makes @value{GDBN} automatically display the next
5961 instruction to be executed, each time your program stops. @xref{Auto
5962 Display,, Automatic Display}.
5964 An argument is a repeat count, as in @code{step}.
5968 @kindex ni @r{(@code{nexti})}
5970 @itemx nexti @var{arg}
5972 Execute one machine instruction, but if it is a function call,
5973 proceed until the function returns.
5975 An argument is a repeat count, as in @code{next}.
5979 @anchor{range stepping}
5980 @cindex range stepping
5981 @cindex target-assisted range stepping
5982 By default, and if available, @value{GDBN} makes use of
5983 target-assisted @dfn{range stepping}. In other words, whenever you
5984 use a stepping command (e.g., @code{step}, @code{next}), @value{GDBN}
5985 tells the target to step the corresponding range of instruction
5986 addresses instead of issuing multiple single-steps. This speeds up
5987 line stepping, particularly for remote targets. Ideally, there should
5988 be no reason you would want to turn range stepping off. However, it's
5989 possible that a bug in the debug info, a bug in the remote stub (for
5990 remote targets), or even a bug in @value{GDBN} could make line
5991 stepping behave incorrectly when target-assisted range stepping is
5992 enabled. You can use the following command to turn off range stepping
5996 @kindex set range-stepping
5997 @kindex show range-stepping
5998 @item set range-stepping
5999 @itemx show range-stepping
6000 Control whether range stepping is enabled.
6002 If @code{on}, and the target supports it, @value{GDBN} tells the
6003 target to step a range of addresses itself, instead of issuing
6004 multiple single-steps. If @code{off}, @value{GDBN} always issues
6005 single-steps, even if range stepping is supported by the target. The
6006 default is @code{on}.
6010 @node Skipping Over Functions and Files
6011 @section Skipping Over Functions and Files
6012 @cindex skipping over functions and files
6014 The program you are debugging may contain some functions which are
6015 uninteresting to debug. The @code{skip} command lets you tell @value{GDBN} to
6016 skip a function, all functions in a file or a particular function in
6017 a particular file when stepping.
6019 For example, consider the following C function:
6030 Suppose you wish to step into the functions @code{foo} and @code{bar}, but you
6031 are not interested in stepping through @code{boring}. If you run @code{step}
6032 at line 103, you'll enter @code{boring()}, but if you run @code{next}, you'll
6033 step over both @code{foo} and @code{boring}!
6035 One solution is to @code{step} into @code{boring} and use the @code{finish}
6036 command to immediately exit it. But this can become tedious if @code{boring}
6037 is called from many places.
6039 A more flexible solution is to execute @kbd{skip boring}. This instructs
6040 @value{GDBN} never to step into @code{boring}. Now when you execute
6041 @code{step} at line 103, you'll step over @code{boring} and directly into
6044 Functions may be skipped by providing either a function name, linespec
6045 (@pxref{Specify Location}), regular expression that matches the function's
6046 name, file name or a @code{glob}-style pattern that matches the file name.
6048 On Posix systems the form of the regular expression is
6049 ``Extended Regular Expressions''. See for example @samp{man 7 regex}
6050 on @sc{gnu}/Linux systems. On non-Posix systems the form of the regular
6051 expression is whatever is provided by the @code{regcomp} function of
6052 the underlying system.
6053 See for example @samp{man 7 glob} on @sc{gnu}/Linux systems for a
6054 description of @code{glob}-style patterns.
6058 @item skip @r{[}@var{options}@r{]}
6059 The basic form of the @code{skip} command takes zero or more options
6060 that specify what to skip.
6061 The @var{options} argument is any useful combination of the following:
6064 @item -file @var{file}
6065 @itemx -fi @var{file}
6066 Functions in @var{file} will be skipped over when stepping.
6068 @item -gfile @var{file-glob-pattern}
6069 @itemx -gfi @var{file-glob-pattern}
6070 @cindex skipping over files via glob-style patterns
6071 Functions in files matching @var{file-glob-pattern} will be skipped
6075 (gdb) skip -gfi utils/*.c
6078 @item -function @var{linespec}
6079 @itemx -fu @var{linespec}
6080 Functions named by @var{linespec} or the function containing the line
6081 named by @var{linespec} will be skipped over when stepping.
6082 @xref{Specify Location}.
6084 @item -rfunction @var{regexp}
6085 @itemx -rfu @var{regexp}
6086 @cindex skipping over functions via regular expressions
6087 Functions whose name matches @var{regexp} will be skipped over when stepping.
6089 This form is useful for complex function names.
6090 For example, there is generally no need to step into C@t{++} @code{std::string}
6091 constructors or destructors. Plus with C@t{++} templates it can be hard to
6092 write out the full name of the function, and often it doesn't matter what
6093 the template arguments are. Specifying the function to be skipped as a
6094 regular expression makes this easier.
6097 (gdb) skip -rfu ^std::(allocator|basic_string)<.*>::~?\1 *\(
6100 If you want to skip every templated C@t{++} constructor and destructor
6101 in the @code{std} namespace you can do:
6104 (gdb) skip -rfu ^std::([a-zA-z0-9_]+)<.*>::~?\1 *\(
6108 If no options are specified, the function you're currently debugging
6111 @kindex skip function
6112 @item skip function @r{[}@var{linespec}@r{]}
6113 After running this command, the function named by @var{linespec} or the
6114 function containing the line named by @var{linespec} will be skipped over when
6115 stepping. @xref{Specify Location}.
6117 If you do not specify @var{linespec}, the function you're currently debugging
6120 (If you have a function called @code{file} that you want to skip, use
6121 @kbd{skip function file}.)
6124 @item skip file @r{[}@var{filename}@r{]}
6125 After running this command, any function whose source lives in @var{filename}
6126 will be skipped over when stepping.
6129 (gdb) skip file boring.c
6130 File boring.c will be skipped when stepping.
6133 If you do not specify @var{filename}, functions whose source lives in the file
6134 you're currently debugging will be skipped.
6137 Skips can be listed, deleted, disabled, and enabled, much like breakpoints.
6138 These are the commands for managing your list of skips:
6142 @item info skip @r{[}@var{range}@r{]}
6143 Print details about the specified skip(s). If @var{range} is not specified,
6144 print a table with details about all functions and files marked for skipping.
6145 @code{info skip} prints the following information about each skip:
6149 A number identifying this skip.
6150 @item Enabled or Disabled
6151 Enabled skips are marked with @samp{y}.
6152 Disabled skips are marked with @samp{n}.
6154 If the file name is a @samp{glob} pattern this is @samp{y}.
6155 Otherwise it is @samp{n}.
6157 The name or @samp{glob} pattern of the file to be skipped.
6158 If no file is specified this is @samp{<none>}.
6160 If the function name is a @samp{regular expression} this is @samp{y}.
6161 Otherwise it is @samp{n}.
6163 The name or regular expression of the function to skip.
6164 If no function is specified this is @samp{<none>}.
6168 @item skip delete @r{[}@var{range}@r{]}
6169 Delete the specified skip(s). If @var{range} is not specified, delete all
6173 @item skip enable @r{[}@var{range}@r{]}
6174 Enable the specified skip(s). If @var{range} is not specified, enable all
6177 @kindex skip disable
6178 @item skip disable @r{[}@var{range}@r{]}
6179 Disable the specified skip(s). If @var{range} is not specified, disable all
6182 @kindex set debug skip
6183 @item set debug skip @r{[}on|off@r{]}
6184 Set whether to print the debug output about skipping files and functions.
6186 @kindex show debug skip
6187 @item show debug skip
6188 Show whether the debug output about skipping files and functions is printed.
6196 A signal is an asynchronous event that can happen in a program. The
6197 operating system defines the possible kinds of signals, and gives each
6198 kind a name and a number. For example, in Unix @code{SIGINT} is the
6199 signal a program gets when you type an interrupt character (often @kbd{Ctrl-c});
6200 @code{SIGSEGV} is the signal a program gets from referencing a place in
6201 memory far away from all the areas in use; @code{SIGALRM} occurs when
6202 the alarm clock timer goes off (which happens only if your program has
6203 requested an alarm).
6205 @cindex fatal signals
6206 Some signals, including @code{SIGALRM}, are a normal part of the
6207 functioning of your program. Others, such as @code{SIGSEGV}, indicate
6208 errors; these signals are @dfn{fatal} (they kill your program immediately) if the
6209 program has not specified in advance some other way to handle the signal.
6210 @code{SIGINT} does not indicate an error in your program, but it is normally
6211 fatal so it can carry out the purpose of the interrupt: to kill the program.
6213 @value{GDBN} has the ability to detect any occurrence of a signal in your
6214 program. You can tell @value{GDBN} in advance what to do for each kind of
6217 @cindex handling signals
6218 Normally, @value{GDBN} is set up to let the non-erroneous signals like
6219 @code{SIGALRM} be silently passed to your program
6220 (so as not to interfere with their role in the program's functioning)
6221 but to stop your program immediately whenever an error signal happens.
6222 You can change these settings with the @code{handle} command.
6225 @kindex info signals
6229 Print a table of all the kinds of signals and how @value{GDBN} has been told to
6230 handle each one. You can use this to see the signal numbers of all
6231 the defined types of signals.
6233 @item info signals @var{sig}
6234 Similar, but print information only about the specified signal number.
6236 @code{info handle} is an alias for @code{info signals}.
6238 @item catch signal @r{[}@var{signal}@dots{} @r{|} @samp{all}@r{]}
6239 Set a catchpoint for the indicated signals. @xref{Set Catchpoints},
6240 for details about this command.
6243 @item handle @var{signal} @r{[}@var{keywords}@dots{}@r{]}
6244 Change the way @value{GDBN} handles signal @var{signal}. The @var{signal}
6245 can be the number of a signal or its name (with or without the
6246 @samp{SIG} at the beginning); a list of signal numbers of the form
6247 @samp{@var{low}-@var{high}}; or the word @samp{all}, meaning all the
6248 known signals. Optional arguments @var{keywords}, described below,
6249 say what change to make.
6253 The keywords allowed by the @code{handle} command can be abbreviated.
6254 Their full names are:
6258 @value{GDBN} should not stop your program when this signal happens. It may
6259 still print a message telling you that the signal has come in.
6262 @value{GDBN} should stop your program when this signal happens. This implies
6263 the @code{print} keyword as well.
6266 @value{GDBN} should print a message when this signal happens.
6269 @value{GDBN} should not mention the occurrence of the signal at all. This
6270 implies the @code{nostop} keyword as well.
6274 @value{GDBN} should allow your program to see this signal; your program
6275 can handle the signal, or else it may terminate if the signal is fatal
6276 and not handled. @code{pass} and @code{noignore} are synonyms.
6280 @value{GDBN} should not allow your program to see this signal.
6281 @code{nopass} and @code{ignore} are synonyms.
6285 When a signal stops your program, the signal is not visible to the
6287 continue. Your program sees the signal then, if @code{pass} is in
6288 effect for the signal in question @emph{at that time}. In other words,
6289 after @value{GDBN} reports a signal, you can use the @code{handle}
6290 command with @code{pass} or @code{nopass} to control whether your
6291 program sees that signal when you continue.
6293 The default is set to @code{nostop}, @code{noprint}, @code{pass} for
6294 non-erroneous signals such as @code{SIGALRM}, @code{SIGWINCH} and
6295 @code{SIGCHLD}, and to @code{stop}, @code{print}, @code{pass} for the
6298 You can also use the @code{signal} command to prevent your program from
6299 seeing a signal, or cause it to see a signal it normally would not see,
6300 or to give it any signal at any time. For example, if your program stopped
6301 due to some sort of memory reference error, you might store correct
6302 values into the erroneous variables and continue, hoping to see more
6303 execution; but your program would probably terminate immediately as
6304 a result of the fatal signal once it saw the signal. To prevent this,
6305 you can continue with @samp{signal 0}. @xref{Signaling, ,Giving your
6308 @cindex stepping and signal handlers
6309 @anchor{stepping and signal handlers}
6311 @value{GDBN} optimizes for stepping the mainline code. If a signal
6312 that has @code{handle nostop} and @code{handle pass} set arrives while
6313 a stepping command (e.g., @code{stepi}, @code{step}, @code{next}) is
6314 in progress, @value{GDBN} lets the signal handler run and then resumes
6315 stepping the mainline code once the signal handler returns. In other
6316 words, @value{GDBN} steps over the signal handler. This prevents
6317 signals that you've specified as not interesting (with @code{handle
6318 nostop}) from changing the focus of debugging unexpectedly. Note that
6319 the signal handler itself may still hit a breakpoint, stop for another
6320 signal that has @code{handle stop} in effect, or for any other event
6321 that normally results in stopping the stepping command sooner. Also
6322 note that @value{GDBN} still informs you that the program received a
6323 signal if @code{handle print} is set.
6325 @anchor{stepping into signal handlers}
6327 If you set @code{handle pass} for a signal, and your program sets up a
6328 handler for it, then issuing a stepping command, such as @code{step}
6329 or @code{stepi}, when your program is stopped due to the signal will
6330 step @emph{into} the signal handler (if the target supports that).
6332 Likewise, if you use the @code{queue-signal} command to queue a signal
6333 to be delivered to the current thread when execution of the thread
6334 resumes (@pxref{Signaling, ,Giving your Program a Signal}), then a
6335 stepping command will step into the signal handler.
6337 Here's an example, using @code{stepi} to step to the first instruction
6338 of @code{SIGUSR1}'s handler:
6341 (@value{GDBP}) handle SIGUSR1
6342 Signal Stop Print Pass to program Description
6343 SIGUSR1 Yes Yes Yes User defined signal 1
6347 Program received signal SIGUSR1, User defined signal 1.
6348 main () sigusr1.c:28
6351 sigusr1_handler () at sigusr1.c:9
6355 The same, but using @code{queue-signal} instead of waiting for the
6356 program to receive the signal first:
6361 (@value{GDBP}) queue-signal SIGUSR1
6363 sigusr1_handler () at sigusr1.c:9
6368 @cindex extra signal information
6369 @anchor{extra signal information}
6371 On some targets, @value{GDBN} can inspect extra signal information
6372 associated with the intercepted signal, before it is actually
6373 delivered to the program being debugged. This information is exported
6374 by the convenience variable @code{$_siginfo}, and consists of data
6375 that is passed by the kernel to the signal handler at the time of the
6376 receipt of a signal. The data type of the information itself is
6377 target dependent. You can see the data type using the @code{ptype
6378 $_siginfo} command. On Unix systems, it typically corresponds to the
6379 standard @code{siginfo_t} type, as defined in the @file{signal.h}
6382 Here's an example, on a @sc{gnu}/Linux system, printing the stray
6383 referenced address that raised a segmentation fault.
6387 (@value{GDBP}) continue
6388 Program received signal SIGSEGV, Segmentation fault.
6389 0x0000000000400766 in main ()
6391 (@value{GDBP}) ptype $_siginfo
6398 struct @{...@} _kill;
6399 struct @{...@} _timer;
6401 struct @{...@} _sigchld;
6402 struct @{...@} _sigfault;
6403 struct @{...@} _sigpoll;
6406 (@value{GDBP}) ptype $_siginfo._sifields._sigfault
6410 (@value{GDBP}) p $_siginfo._sifields._sigfault.si_addr
6411 $1 = (void *) 0x7ffff7ff7000
6415 Depending on target support, @code{$_siginfo} may also be writable.
6417 @cindex Intel MPX boundary violations
6418 @cindex boundary violations, Intel MPX
6419 On some targets, a @code{SIGSEGV} can be caused by a boundary
6420 violation, i.e., accessing an address outside of the allowed range.
6421 In those cases @value{GDBN} may displays additional information,
6422 depending on how @value{GDBN} has been told to handle the signal.
6423 With @code{handle stop SIGSEGV}, @value{GDBN} displays the violation
6424 kind: "Upper" or "Lower", the memory address accessed and the
6425 bounds, while with @code{handle nostop SIGSEGV} no additional
6426 information is displayed.
6428 The usual output of a segfault is:
6430 Program received signal SIGSEGV, Segmentation fault
6431 0x0000000000400d7c in upper () at i386-mpx-sigsegv.c:68
6432 68 value = *(p + len);
6435 While a bound violation is presented as:
6437 Program received signal SIGSEGV, Segmentation fault
6438 Upper bound violation while accessing address 0x7fffffffc3b3
6439 Bounds: [lower = 0x7fffffffc390, upper = 0x7fffffffc3a3]
6440 0x0000000000400d7c in upper () at i386-mpx-sigsegv.c:68
6441 68 value = *(p + len);
6445 @section Stopping and Starting Multi-thread Programs
6447 @cindex stopped threads
6448 @cindex threads, stopped
6450 @cindex continuing threads
6451 @cindex threads, continuing
6453 @value{GDBN} supports debugging programs with multiple threads
6454 (@pxref{Threads,, Debugging Programs with Multiple Threads}). There
6455 are two modes of controlling execution of your program within the
6456 debugger. In the default mode, referred to as @dfn{all-stop mode},
6457 when any thread in your program stops (for example, at a breakpoint
6458 or while being stepped), all other threads in the program are also stopped by
6459 @value{GDBN}. On some targets, @value{GDBN} also supports
6460 @dfn{non-stop mode}, in which other threads can continue to run freely while
6461 you examine the stopped thread in the debugger.
6464 * All-Stop Mode:: All threads stop when GDB takes control
6465 * Non-Stop Mode:: Other threads continue to execute
6466 * Background Execution:: Running your program asynchronously
6467 * Thread-Specific Breakpoints:: Controlling breakpoints
6468 * Interrupted System Calls:: GDB may interfere with system calls
6469 * Observer Mode:: GDB does not alter program behavior
6473 @subsection All-Stop Mode
6475 @cindex all-stop mode
6477 In all-stop mode, whenever your program stops under @value{GDBN} for any reason,
6478 @emph{all} threads of execution stop, not just the current thread. This
6479 allows you to examine the overall state of the program, including
6480 switching between threads, without worrying that things may change
6483 Conversely, whenever you restart the program, @emph{all} threads start
6484 executing. @emph{This is true even when single-stepping} with commands
6485 like @code{step} or @code{next}.
6487 In particular, @value{GDBN} cannot single-step all threads in lockstep.
6488 Since thread scheduling is up to your debugging target's operating
6489 system (not controlled by @value{GDBN}), other threads may
6490 execute more than one statement while the current thread completes a
6491 single step. Moreover, in general other threads stop in the middle of a
6492 statement, rather than at a clean statement boundary, when the program
6495 You might even find your program stopped in another thread after
6496 continuing or even single-stepping. This happens whenever some other
6497 thread runs into a breakpoint, a signal, or an exception before the
6498 first thread completes whatever you requested.
6500 @cindex automatic thread selection
6501 @cindex switching threads automatically
6502 @cindex threads, automatic switching
6503 Whenever @value{GDBN} stops your program, due to a breakpoint or a
6504 signal, it automatically selects the thread where that breakpoint or
6505 signal happened. @value{GDBN} alerts you to the context switch with a
6506 message such as @samp{[Switching to Thread @var{n}]} to identify the
6509 On some OSes, you can modify @value{GDBN}'s default behavior by
6510 locking the OS scheduler to allow only a single thread to run.
6513 @item set scheduler-locking @var{mode}
6514 @cindex scheduler locking mode
6515 @cindex lock scheduler
6516 Set the scheduler locking mode. It applies to normal execution,
6517 record mode, and replay mode. If it is @code{off}, then there is no
6518 locking and any thread may run at any time. If @code{on}, then only
6519 the current thread may run when the inferior is resumed. The
6520 @code{step} mode optimizes for single-stepping; it prevents other
6521 threads from preempting the current thread while you are stepping, so
6522 that the focus of debugging does not change unexpectedly. Other
6523 threads never get a chance to run when you step, and they are
6524 completely free to run when you use commands like @samp{continue},
6525 @samp{until}, or @samp{finish}. However, unless another thread hits a
6526 breakpoint during its timeslice, @value{GDBN} does not change the
6527 current thread away from the thread that you are debugging. The
6528 @code{replay} mode behaves like @code{off} in record mode and like
6529 @code{on} in replay mode.
6531 @item show scheduler-locking
6532 Display the current scheduler locking mode.
6535 @cindex resume threads of multiple processes simultaneously
6536 By default, when you issue one of the execution commands such as
6537 @code{continue}, @code{next} or @code{step}, @value{GDBN} allows only
6538 threads of the current inferior to run. For example, if @value{GDBN}
6539 is attached to two inferiors, each with two threads, the
6540 @code{continue} command resumes only the two threads of the current
6541 inferior. This is useful, for example, when you debug a program that
6542 forks and you want to hold the parent stopped (so that, for instance,
6543 it doesn't run to exit), while you debug the child. In other
6544 situations, you may not be interested in inspecting the current state
6545 of any of the processes @value{GDBN} is attached to, and you may want
6546 to resume them all until some breakpoint is hit. In the latter case,
6547 you can instruct @value{GDBN} to allow all threads of all the
6548 inferiors to run with the @w{@code{set schedule-multiple}} command.
6551 @kindex set schedule-multiple
6552 @item set schedule-multiple
6553 Set the mode for allowing threads of multiple processes to be resumed
6554 when an execution command is issued. When @code{on}, all threads of
6555 all processes are allowed to run. When @code{off}, only the threads
6556 of the current process are resumed. The default is @code{off}. The
6557 @code{scheduler-locking} mode takes precedence when set to @code{on},
6558 or while you are stepping and set to @code{step}.
6560 @item show schedule-multiple
6561 Display the current mode for resuming the execution of threads of
6566 @subsection Non-Stop Mode
6568 @cindex non-stop mode
6570 @c This section is really only a place-holder, and needs to be expanded
6571 @c with more details.
6573 For some multi-threaded targets, @value{GDBN} supports an optional
6574 mode of operation in which you can examine stopped program threads in
6575 the debugger while other threads continue to execute freely. This
6576 minimizes intrusion when debugging live systems, such as programs
6577 where some threads have real-time constraints or must continue to
6578 respond to external events. This is referred to as @dfn{non-stop} mode.
6580 In non-stop mode, when a thread stops to report a debugging event,
6581 @emph{only} that thread is stopped; @value{GDBN} does not stop other
6582 threads as well, in contrast to the all-stop mode behavior. Additionally,
6583 execution commands such as @code{continue} and @code{step} apply by default
6584 only to the current thread in non-stop mode, rather than all threads as
6585 in all-stop mode. This allows you to control threads explicitly in
6586 ways that are not possible in all-stop mode --- for example, stepping
6587 one thread while allowing others to run freely, stepping
6588 one thread while holding all others stopped, or stepping several threads
6589 independently and simultaneously.
6591 To enter non-stop mode, use this sequence of commands before you run
6592 or attach to your program:
6595 # If using the CLI, pagination breaks non-stop.
6598 # Finally, turn it on!
6602 You can use these commands to manipulate the non-stop mode setting:
6605 @kindex set non-stop
6606 @item set non-stop on
6607 Enable selection of non-stop mode.
6608 @item set non-stop off
6609 Disable selection of non-stop mode.
6610 @kindex show non-stop
6612 Show the current non-stop enablement setting.
6615 Note these commands only reflect whether non-stop mode is enabled,
6616 not whether the currently-executing program is being run in non-stop mode.
6617 In particular, the @code{set non-stop} preference is only consulted when
6618 @value{GDBN} starts or connects to the target program, and it is generally
6619 not possible to switch modes once debugging has started. Furthermore,
6620 since not all targets support non-stop mode, even when you have enabled
6621 non-stop mode, @value{GDBN} may still fall back to all-stop operation by
6624 In non-stop mode, all execution commands apply only to the current thread
6625 by default. That is, @code{continue} only continues one thread.
6626 To continue all threads, issue @code{continue -a} or @code{c -a}.
6628 You can use @value{GDBN}'s background execution commands
6629 (@pxref{Background Execution}) to run some threads in the background
6630 while you continue to examine or step others from @value{GDBN}.
6631 The MI execution commands (@pxref{GDB/MI Program Execution}) are
6632 always executed asynchronously in non-stop mode.
6634 Suspending execution is done with the @code{interrupt} command when
6635 running in the background, or @kbd{Ctrl-c} during foreground execution.
6636 In all-stop mode, this stops the whole process;
6637 but in non-stop mode the interrupt applies only to the current thread.
6638 To stop the whole program, use @code{interrupt -a}.
6640 Other execution commands do not currently support the @code{-a} option.
6642 In non-stop mode, when a thread stops, @value{GDBN} doesn't automatically make
6643 that thread current, as it does in all-stop mode. This is because the
6644 thread stop notifications are asynchronous with respect to @value{GDBN}'s
6645 command interpreter, and it would be confusing if @value{GDBN} unexpectedly
6646 changed to a different thread just as you entered a command to operate on the
6647 previously current thread.
6649 @node Background Execution
6650 @subsection Background Execution
6652 @cindex foreground execution
6653 @cindex background execution
6654 @cindex asynchronous execution
6655 @cindex execution, foreground, background and asynchronous
6657 @value{GDBN}'s execution commands have two variants: the normal
6658 foreground (synchronous) behavior, and a background
6659 (asynchronous) behavior. In foreground execution, @value{GDBN} waits for
6660 the program to report that some thread has stopped before prompting for
6661 another command. In background execution, @value{GDBN} immediately gives
6662 a command prompt so that you can issue other commands while your program runs.
6664 If the target doesn't support async mode, @value{GDBN} issues an error
6665 message if you attempt to use the background execution commands.
6667 @cindex @code{&}, background execution of commands
6668 To specify background execution, add a @code{&} to the command. For example,
6669 the background form of the @code{continue} command is @code{continue&}, or
6670 just @code{c&}. The execution commands that accept background execution
6676 @xref{Starting, , Starting your Program}.
6680 @xref{Attach, , Debugging an Already-running Process}.
6684 @xref{Continuing and Stepping, step}.
6688 @xref{Continuing and Stepping, stepi}.
6692 @xref{Continuing and Stepping, next}.
6696 @xref{Continuing and Stepping, nexti}.
6700 @xref{Continuing and Stepping, continue}.
6704 @xref{Continuing and Stepping, finish}.
6708 @xref{Continuing and Stepping, until}.
6712 Background execution is especially useful in conjunction with non-stop
6713 mode for debugging programs with multiple threads; see @ref{Non-Stop Mode}.
6714 However, you can also use these commands in the normal all-stop mode with
6715 the restriction that you cannot issue another execution command until the
6716 previous one finishes. Examples of commands that are valid in all-stop
6717 mode while the program is running include @code{help} and @code{info break}.
6719 You can interrupt your program while it is running in the background by
6720 using the @code{interrupt} command.
6727 Suspend execution of the running program. In all-stop mode,
6728 @code{interrupt} stops the whole process, but in non-stop mode, it stops
6729 only the current thread. To stop the whole program in non-stop mode,
6730 use @code{interrupt -a}.
6733 @node Thread-Specific Breakpoints
6734 @subsection Thread-Specific Breakpoints
6736 When your program has multiple threads (@pxref{Threads,, Debugging
6737 Programs with Multiple Threads}), you can choose whether to set
6738 breakpoints on all threads, or on a particular thread.
6741 @cindex breakpoints and threads
6742 @cindex thread breakpoints
6743 @kindex break @dots{} thread @var{thread-id}
6744 @item break @var{location} thread @var{thread-id}
6745 @itemx break @var{location} thread @var{thread-id} if @dots{}
6746 @var{location} specifies source lines; there are several ways of
6747 writing them (@pxref{Specify Location}), but the effect is always to
6748 specify some source line.
6750 Use the qualifier @samp{thread @var{thread-id}} with a breakpoint command
6751 to specify that you only want @value{GDBN} to stop the program when a
6752 particular thread reaches this breakpoint. The @var{thread-id} specifier
6753 is one of the thread identifiers assigned by @value{GDBN}, shown
6754 in the first column of the @samp{info threads} display.
6756 If you do not specify @samp{thread @var{thread-id}} when you set a
6757 breakpoint, the breakpoint applies to @emph{all} threads of your
6760 You can use the @code{thread} qualifier on conditional breakpoints as
6761 well; in this case, place @samp{thread @var{thread-id}} before or
6762 after the breakpoint condition, like this:
6765 (@value{GDBP}) break frik.c:13 thread 28 if bartab > lim
6770 Thread-specific breakpoints are automatically deleted when
6771 @value{GDBN} detects the corresponding thread is no longer in the
6772 thread list. For example:
6776 Thread-specific breakpoint 3 deleted - thread 28 no longer in the thread list.
6779 There are several ways for a thread to disappear, such as a regular
6780 thread exit, but also when you detach from the process with the
6781 @code{detach} command (@pxref{Attach, ,Debugging an Already-running
6782 Process}), or if @value{GDBN} loses the remote connection
6783 (@pxref{Remote Debugging}), etc. Note that with some targets,
6784 @value{GDBN} is only able to detect a thread has exited when the user
6785 explictly asks for the thread list with the @code{info threads}
6788 @node Interrupted System Calls
6789 @subsection Interrupted System Calls
6791 @cindex thread breakpoints and system calls
6792 @cindex system calls and thread breakpoints
6793 @cindex premature return from system calls
6794 There is an unfortunate side effect when using @value{GDBN} to debug
6795 multi-threaded programs. If one thread stops for a
6796 breakpoint, or for some other reason, and another thread is blocked in a
6797 system call, then the system call may return prematurely. This is a
6798 consequence of the interaction between multiple threads and the signals
6799 that @value{GDBN} uses to implement breakpoints and other events that
6802 To handle this problem, your program should check the return value of
6803 each system call and react appropriately. This is good programming
6806 For example, do not write code like this:
6812 The call to @code{sleep} will return early if a different thread stops
6813 at a breakpoint or for some other reason.
6815 Instead, write this:
6820 unslept = sleep (unslept);
6823 A system call is allowed to return early, so the system is still
6824 conforming to its specification. But @value{GDBN} does cause your
6825 multi-threaded program to behave differently than it would without
6828 Also, @value{GDBN} uses internal breakpoints in the thread library to
6829 monitor certain events such as thread creation and thread destruction.
6830 When such an event happens, a system call in another thread may return
6831 prematurely, even though your program does not appear to stop.
6834 @subsection Observer Mode
6836 If you want to build on non-stop mode and observe program behavior
6837 without any chance of disruption by @value{GDBN}, you can set
6838 variables to disable all of the debugger's attempts to modify state,
6839 whether by writing memory, inserting breakpoints, etc. These operate
6840 at a low level, intercepting operations from all commands.
6842 When all of these are set to @code{off}, then @value{GDBN} is said to
6843 be @dfn{observer mode}. As a convenience, the variable
6844 @code{observer} can be set to disable these, plus enable non-stop
6847 Note that @value{GDBN} will not prevent you from making nonsensical
6848 combinations of these settings. For instance, if you have enabled
6849 @code{may-insert-breakpoints} but disabled @code{may-write-memory},
6850 then breakpoints that work by writing trap instructions into the code
6851 stream will still not be able to be placed.
6856 @item set observer on
6857 @itemx set observer off
6858 When set to @code{on}, this disables all the permission variables
6859 below (except for @code{insert-fast-tracepoints}), plus enables
6860 non-stop debugging. Setting this to @code{off} switches back to
6861 normal debugging, though remaining in non-stop mode.
6864 Show whether observer mode is on or off.
6866 @kindex may-write-registers
6867 @item set may-write-registers on
6868 @itemx set may-write-registers off
6869 This controls whether @value{GDBN} will attempt to alter the values of
6870 registers, such as with assignment expressions in @code{print}, or the
6871 @code{jump} command. It defaults to @code{on}.
6873 @item show may-write-registers
6874 Show the current permission to write registers.
6876 @kindex may-write-memory
6877 @item set may-write-memory on
6878 @itemx set may-write-memory off
6879 This controls whether @value{GDBN} will attempt to alter the contents
6880 of memory, such as with assignment expressions in @code{print}. It
6881 defaults to @code{on}.
6883 @item show may-write-memory
6884 Show the current permission to write memory.
6886 @kindex may-insert-breakpoints
6887 @item set may-insert-breakpoints on
6888 @itemx set may-insert-breakpoints off
6889 This controls whether @value{GDBN} will attempt to insert breakpoints.
6890 This affects all breakpoints, including internal breakpoints defined
6891 by @value{GDBN}. It defaults to @code{on}.
6893 @item show may-insert-breakpoints
6894 Show the current permission to insert breakpoints.
6896 @kindex may-insert-tracepoints
6897 @item set may-insert-tracepoints on
6898 @itemx set may-insert-tracepoints off
6899 This controls whether @value{GDBN} will attempt to insert (regular)
6900 tracepoints at the beginning of a tracing experiment. It affects only
6901 non-fast tracepoints, fast tracepoints being under the control of
6902 @code{may-insert-fast-tracepoints}. It defaults to @code{on}.
6904 @item show may-insert-tracepoints
6905 Show the current permission to insert tracepoints.
6907 @kindex may-insert-fast-tracepoints
6908 @item set may-insert-fast-tracepoints on
6909 @itemx set may-insert-fast-tracepoints off
6910 This controls whether @value{GDBN} will attempt to insert fast
6911 tracepoints at the beginning of a tracing experiment. It affects only
6912 fast tracepoints, regular (non-fast) tracepoints being under the
6913 control of @code{may-insert-tracepoints}. It defaults to @code{on}.
6915 @item show may-insert-fast-tracepoints
6916 Show the current permission to insert fast tracepoints.
6918 @kindex may-interrupt
6919 @item set may-interrupt on
6920 @itemx set may-interrupt off
6921 This controls whether @value{GDBN} will attempt to interrupt or stop
6922 program execution. When this variable is @code{off}, the
6923 @code{interrupt} command will have no effect, nor will
6924 @kbd{Ctrl-c}. It defaults to @code{on}.
6926 @item show may-interrupt
6927 Show the current permission to interrupt or stop the program.
6931 @node Reverse Execution
6932 @chapter Running programs backward
6933 @cindex reverse execution
6934 @cindex running programs backward
6936 When you are debugging a program, it is not unusual to realize that
6937 you have gone too far, and some event of interest has already happened.
6938 If the target environment supports it, @value{GDBN} can allow you to
6939 ``rewind'' the program by running it backward.
6941 A target environment that supports reverse execution should be able
6942 to ``undo'' the changes in machine state that have taken place as the
6943 program was executing normally. Variables, registers etc.@: should
6944 revert to their previous values. Obviously this requires a great
6945 deal of sophistication on the part of the target environment; not
6946 all target environments can support reverse execution.
6948 When a program is executed in reverse, the instructions that
6949 have most recently been executed are ``un-executed'', in reverse
6950 order. The program counter runs backward, following the previous
6951 thread of execution in reverse. As each instruction is ``un-executed'',
6952 the values of memory and/or registers that were changed by that
6953 instruction are reverted to their previous states. After executing
6954 a piece of source code in reverse, all side effects of that code
6955 should be ``undone'', and all variables should be returned to their
6956 prior values@footnote{
6957 Note that some side effects are easier to undo than others. For instance,
6958 memory and registers are relatively easy, but device I/O is hard. Some
6959 targets may be able undo things like device I/O, and some may not.
6961 The contract between @value{GDBN} and the reverse executing target
6962 requires only that the target do something reasonable when
6963 @value{GDBN} tells it to execute backwards, and then report the
6964 results back to @value{GDBN}. Whatever the target reports back to
6965 @value{GDBN}, @value{GDBN} will report back to the user. @value{GDBN}
6966 assumes that the memory and registers that the target reports are in a
6967 consistant state, but @value{GDBN} accepts whatever it is given.
6970 On some platforms, @value{GDBN} has built-in support for reverse
6971 execution, activated with the @code{record} or @code{record btrace}
6972 commands. @xref{Process Record and Replay}. Some remote targets,
6973 typically full system emulators, support reverse execution directly
6974 without requiring any special command.
6976 If you are debugging in a target environment that supports
6977 reverse execution, @value{GDBN} provides the following commands.
6980 @kindex reverse-continue
6981 @kindex rc @r{(@code{reverse-continue})}
6982 @item reverse-continue @r{[}@var{ignore-count}@r{]}
6983 @itemx rc @r{[}@var{ignore-count}@r{]}
6984 Beginning at the point where your program last stopped, start executing
6985 in reverse. Reverse execution will stop for breakpoints and synchronous
6986 exceptions (signals), just like normal execution. Behavior of
6987 asynchronous signals depends on the target environment.
6989 @kindex reverse-step
6990 @kindex rs @r{(@code{step})}
6991 @item reverse-step @r{[}@var{count}@r{]}
6992 Run the program backward until control reaches the start of a
6993 different source line; then stop it, and return control to @value{GDBN}.
6995 Like the @code{step} command, @code{reverse-step} will only stop
6996 at the beginning of a source line. It ``un-executes'' the previously
6997 executed source line. If the previous source line included calls to
6998 debuggable functions, @code{reverse-step} will step (backward) into
6999 the called function, stopping at the beginning of the @emph{last}
7000 statement in the called function (typically a return statement).
7002 Also, as with the @code{step} command, if non-debuggable functions are
7003 called, @code{reverse-step} will run thru them backward without stopping.
7005 @kindex reverse-stepi
7006 @kindex rsi @r{(@code{reverse-stepi})}
7007 @item reverse-stepi @r{[}@var{count}@r{]}
7008 Reverse-execute one machine instruction. Note that the instruction
7009 to be reverse-executed is @emph{not} the one pointed to by the program
7010 counter, but the instruction executed prior to that one. For instance,
7011 if the last instruction was a jump, @code{reverse-stepi} will take you
7012 back from the destination of the jump to the jump instruction itself.
7014 @kindex reverse-next
7015 @kindex rn @r{(@code{reverse-next})}
7016 @item reverse-next @r{[}@var{count}@r{]}
7017 Run backward to the beginning of the previous line executed in
7018 the current (innermost) stack frame. If the line contains function
7019 calls, they will be ``un-executed'' without stopping. Starting from
7020 the first line of a function, @code{reverse-next} will take you back
7021 to the caller of that function, @emph{before} the function was called,
7022 just as the normal @code{next} command would take you from the last
7023 line of a function back to its return to its caller
7024 @footnote{Unless the code is too heavily optimized.}.
7026 @kindex reverse-nexti
7027 @kindex rni @r{(@code{reverse-nexti})}
7028 @item reverse-nexti @r{[}@var{count}@r{]}
7029 Like @code{nexti}, @code{reverse-nexti} executes a single instruction
7030 in reverse, except that called functions are ``un-executed'' atomically.
7031 That is, if the previously executed instruction was a return from
7032 another function, @code{reverse-nexti} will continue to execute
7033 in reverse until the call to that function (from the current stack
7036 @kindex reverse-finish
7037 @item reverse-finish
7038 Just as the @code{finish} command takes you to the point where the
7039 current function returns, @code{reverse-finish} takes you to the point
7040 where it was called. Instead of ending up at the end of the current
7041 function invocation, you end up at the beginning.
7043 @kindex set exec-direction
7044 @item set exec-direction
7045 Set the direction of target execution.
7046 @item set exec-direction reverse
7047 @cindex execute forward or backward in time
7048 @value{GDBN} will perform all execution commands in reverse, until the
7049 exec-direction mode is changed to ``forward''. Affected commands include
7050 @code{step, stepi, next, nexti, continue, and finish}. The @code{return}
7051 command cannot be used in reverse mode.
7052 @item set exec-direction forward
7053 @value{GDBN} will perform all execution commands in the normal fashion.
7054 This is the default.
7058 @node Process Record and Replay
7059 @chapter Recording Inferior's Execution and Replaying It
7060 @cindex process record and replay
7061 @cindex recording inferior's execution and replaying it
7063 On some platforms, @value{GDBN} provides a special @dfn{process record
7064 and replay} target that can record a log of the process execution, and
7065 replay it later with both forward and reverse execution commands.
7068 When this target is in use, if the execution log includes the record
7069 for the next instruction, @value{GDBN} will debug in @dfn{replay
7070 mode}. In the replay mode, the inferior does not really execute code
7071 instructions. Instead, all the events that normally happen during
7072 code execution are taken from the execution log. While code is not
7073 really executed in replay mode, the values of registers (including the
7074 program counter register) and the memory of the inferior are still
7075 changed as they normally would. Their contents are taken from the
7079 If the record for the next instruction is not in the execution log,
7080 @value{GDBN} will debug in @dfn{record mode}. In this mode, the
7081 inferior executes normally, and @value{GDBN} records the execution log
7084 The process record and replay target supports reverse execution
7085 (@pxref{Reverse Execution}), even if the platform on which the
7086 inferior runs does not. However, the reverse execution is limited in
7087 this case by the range of the instructions recorded in the execution
7088 log. In other words, reverse execution on platforms that don't
7089 support it directly can only be done in the replay mode.
7091 When debugging in the reverse direction, @value{GDBN} will work in
7092 replay mode as long as the execution log includes the record for the
7093 previous instruction; otherwise, it will work in record mode, if the
7094 platform supports reverse execution, or stop if not.
7096 Currently, process record and replay is supported on ARM, Aarch64,
7097 Moxie, PowerPC, PowerPC64, S/390, and x86 (i386/amd64) running
7098 GNU/Linux. Process record and replay can be used both when native
7099 debugging, and when remote debugging via @code{gdbserver}.
7101 For architecture environments that support process record and replay,
7102 @value{GDBN} provides the following commands:
7105 @kindex target record
7106 @kindex target record-full
7107 @kindex target record-btrace
7110 @kindex record btrace
7111 @kindex record btrace bts
7112 @kindex record btrace pt
7118 @kindex rec btrace bts
7119 @kindex rec btrace pt
7122 @item record @var{method}
7123 This command starts the process record and replay target. The
7124 recording method can be specified as parameter. Without a parameter
7125 the command uses the @code{full} recording method. The following
7126 recording methods are available:
7130 Full record/replay recording using @value{GDBN}'s software record and
7131 replay implementation. This method allows replaying and reverse
7134 @item btrace @var{format}
7135 Hardware-supported instruction recording, supported on Intel
7136 processors. This method does not record data. Further, the data is
7137 collected in a ring buffer so old data will be overwritten when the
7138 buffer is full. It allows limited reverse execution. Variables and
7139 registers are not available during reverse execution. In remote
7140 debugging, recording continues on disconnect. Recorded data can be
7141 inspected after reconnecting. The recording may be stopped using
7144 The recording format can be specified as parameter. Without a parameter
7145 the command chooses the recording format. The following recording
7146 formats are available:
7150 @cindex branch trace store
7151 Use the @dfn{Branch Trace Store} (@acronym{BTS}) recording format. In
7152 this format, the processor stores a from/to record for each executed
7153 branch in the btrace ring buffer.
7156 @cindex Intel Processor Trace
7157 Use the @dfn{Intel Processor Trace} recording format. In this
7158 format, the processor stores the execution trace in a compressed form
7159 that is afterwards decoded by @value{GDBN}.
7161 The trace can be recorded with very low overhead. The compressed
7162 trace format also allows small trace buffers to already contain a big
7163 number of instructions compared to @acronym{BTS}.
7165 Decoding the recorded execution trace, on the other hand, is more
7166 expensive than decoding @acronym{BTS} trace. This is mostly due to the
7167 increased number of instructions to process. You should increase the
7168 buffer-size with care.
7171 Not all recording formats may be available on all processors.
7174 The process record and replay target can only debug a process that is
7175 already running. Therefore, you need first to start the process with
7176 the @kbd{run} or @kbd{start} commands, and then start the recording
7177 with the @kbd{record @var{method}} command.
7179 @cindex displaced stepping, and process record and replay
7180 Displaced stepping (@pxref{Maintenance Commands,, displaced stepping})
7181 will be automatically disabled when process record and replay target
7182 is started. That's because the process record and replay target
7183 doesn't support displaced stepping.
7185 @cindex non-stop mode, and process record and replay
7186 @cindex asynchronous execution, and process record and replay
7187 If the inferior is in the non-stop mode (@pxref{Non-Stop Mode}) or in
7188 the asynchronous execution mode (@pxref{Background Execution}), not
7189 all recording methods are available. The @code{full} recording method
7190 does not support these two modes.
7195 Stop the process record and replay target. When process record and
7196 replay target stops, the entire execution log will be deleted and the
7197 inferior will either be terminated, or will remain in its final state.
7199 When you stop the process record and replay target in record mode (at
7200 the end of the execution log), the inferior will be stopped at the
7201 next instruction that would have been recorded. In other words, if
7202 you record for a while and then stop recording, the inferior process
7203 will be left in the same state as if the recording never happened.
7205 On the other hand, if the process record and replay target is stopped
7206 while in replay mode (that is, not at the end of the execution log,
7207 but at some earlier point), the inferior process will become ``live''
7208 at that earlier state, and it will then be possible to continue the
7209 usual ``live'' debugging of the process from that state.
7211 When the inferior process exits, or @value{GDBN} detaches from it,
7212 process record and replay target will automatically stop itself.
7216 Go to a specific location in the execution log. There are several
7217 ways to specify the location to go to:
7220 @item record goto begin
7221 @itemx record goto start
7222 Go to the beginning of the execution log.
7224 @item record goto end
7225 Go to the end of the execution log.
7227 @item record goto @var{n}
7228 Go to instruction number @var{n} in the execution log.
7232 @item record save @var{filename}
7233 Save the execution log to a file @file{@var{filename}}.
7234 Default filename is @file{gdb_record.@var{process_id}}, where
7235 @var{process_id} is the process ID of the inferior.
7237 This command may not be available for all recording methods.
7239 @kindex record restore
7240 @item record restore @var{filename}
7241 Restore the execution log from a file @file{@var{filename}}.
7242 File must have been created with @code{record save}.
7244 @kindex set record full
7245 @item set record full insn-number-max @var{limit}
7246 @itemx set record full insn-number-max unlimited
7247 Set the limit of instructions to be recorded for the @code{full}
7248 recording method. Default value is 200000.
7250 If @var{limit} is a positive number, then @value{GDBN} will start
7251 deleting instructions from the log once the number of the record
7252 instructions becomes greater than @var{limit}. For every new recorded
7253 instruction, @value{GDBN} will delete the earliest recorded
7254 instruction to keep the number of recorded instructions at the limit.
7255 (Since deleting recorded instructions loses information, @value{GDBN}
7256 lets you control what happens when the limit is reached, by means of
7257 the @code{stop-at-limit} option, described below.)
7259 If @var{limit} is @code{unlimited} or zero, @value{GDBN} will never
7260 delete recorded instructions from the execution log. The number of
7261 recorded instructions is limited only by the available memory.
7263 @kindex show record full
7264 @item show record full insn-number-max
7265 Show the limit of instructions to be recorded with the @code{full}
7268 @item set record full stop-at-limit
7269 Control the behavior of the @code{full} recording method when the
7270 number of recorded instructions reaches the limit. If ON (the
7271 default), @value{GDBN} will stop when the limit is reached for the
7272 first time and ask you whether you want to stop the inferior or
7273 continue running it and recording the execution log. If you decide
7274 to continue recording, each new recorded instruction will cause the
7275 oldest one to be deleted.
7277 If this option is OFF, @value{GDBN} will automatically delete the
7278 oldest record to make room for each new one, without asking.
7280 @item show record full stop-at-limit
7281 Show the current setting of @code{stop-at-limit}.
7283 @item set record full memory-query
7284 Control the behavior when @value{GDBN} is unable to record memory
7285 changes caused by an instruction for the @code{full} recording method.
7286 If ON, @value{GDBN} will query whether to stop the inferior in that
7289 If this option is OFF (the default), @value{GDBN} will automatically
7290 ignore the effect of such instructions on memory. Later, when
7291 @value{GDBN} replays this execution log, it will mark the log of this
7292 instruction as not accessible, and it will not affect the replay
7295 @item show record full memory-query
7296 Show the current setting of @code{memory-query}.
7298 @kindex set record btrace
7299 The @code{btrace} record target does not trace data. As a
7300 convenience, when replaying, @value{GDBN} reads read-only memory off
7301 the live program directly, assuming that the addresses of the
7302 read-only areas don't change. This for example makes it possible to
7303 disassemble code while replaying, but not to print variables.
7304 In some cases, being able to inspect variables might be useful.
7305 You can use the following command for that:
7307 @item set record btrace replay-memory-access
7308 Control the behavior of the @code{btrace} recording method when
7309 accessing memory during replay. If @code{read-only} (the default),
7310 @value{GDBN} will only allow accesses to read-only memory.
7311 If @code{read-write}, @value{GDBN} will allow accesses to read-only
7312 and to read-write memory. Beware that the accessed memory corresponds
7313 to the live target and not necessarily to the current replay
7316 @item set record btrace cpu @var{identifier}
7317 Set the processor to be used for enabling workarounds for processor
7318 errata when decoding the trace.
7320 Processor errata are defects in processor operation, caused by its
7321 design or manufacture. They can cause a trace not to match the
7322 specification. This, in turn, may cause trace decode to fail.
7323 @value{GDBN} can detect erroneous trace packets and correct them, thus
7324 avoiding the decoding failures. These corrections are known as
7325 @dfn{errata workarounds}, and are enabled based on the processor on
7326 which the trace was recorded.
7328 By default, @value{GDBN} attempts to detect the processor
7329 automatically, and apply the necessary workarounds for it. However,
7330 you may need to specify the processor if @value{GDBN} does not yet
7331 support it. This command allows you to do that, and also allows to
7332 disable the workarounds.
7334 The argument @var{identifier} identifies the @sc{cpu} and is of the
7335 form: @code{@var{vendor}:@var{procesor identifier}}. In addition,
7336 there are two special identifiers, @code{none} and @code{auto}
7339 The following vendor identifiers and corresponding processor
7340 identifiers are currently supported:
7342 @multitable @columnfractions .1 .9
7345 @tab @var{family}/@var{model}[/@var{stepping}]
7349 On GNU/Linux systems, the processor @var{family}, @var{model}, and
7350 @var{stepping} can be obtained from @code{/proc/cpuinfo}.
7352 If @var{identifier} is @code{auto}, enable errata workarounds for the
7353 processor on which the trace was recorded. If @var{identifier} is
7354 @code{none}, errata workarounds are disabled.
7356 For example, when using an old @value{GDBN} on a new system, decode
7357 may fail because @value{GDBN} does not support the new processor. It
7358 often suffices to specify an older processor that @value{GDBN}
7363 Active record target: record-btrace
7364 Recording format: Intel Processor Trace.
7366 Failed to configure the Intel Processor Trace decoder: unknown cpu.
7367 (gdb) set record btrace cpu intel:6/158
7369 Active record target: record-btrace
7370 Recording format: Intel Processor Trace.
7372 Recorded 84872 instructions in 3189 functions (0 gaps) for thread 1 (...).
7375 @kindex show record btrace
7376 @item show record btrace replay-memory-access
7377 Show the current setting of @code{replay-memory-access}.
7379 @item show record btrace cpu
7380 Show the processor to be used for enabling trace decode errata
7383 @kindex set record btrace bts
7384 @item set record btrace bts buffer-size @var{size}
7385 @itemx set record btrace bts buffer-size unlimited
7386 Set the requested ring buffer size for branch tracing in @acronym{BTS}
7387 format. Default is 64KB.
7389 If @var{size} is a positive number, then @value{GDBN} will try to
7390 allocate a buffer of at least @var{size} bytes for each new thread
7391 that uses the btrace recording method and the @acronym{BTS} format.
7392 The actually obtained buffer size may differ from the requested
7393 @var{size}. Use the @code{info record} command to see the actual
7394 buffer size for each thread that uses the btrace recording method and
7395 the @acronym{BTS} format.
7397 If @var{limit} is @code{unlimited} or zero, @value{GDBN} will try to
7398 allocate a buffer of 4MB.
7400 Bigger buffers mean longer traces. On the other hand, @value{GDBN} will
7401 also need longer to process the branch trace data before it can be used.
7403 @item show record btrace bts buffer-size @var{size}
7404 Show the current setting of the requested ring buffer size for branch
7405 tracing in @acronym{BTS} format.
7407 @kindex set record btrace pt
7408 @item set record btrace pt buffer-size @var{size}
7409 @itemx set record btrace pt buffer-size unlimited
7410 Set the requested ring buffer size for branch tracing in Intel
7411 Processor Trace format. Default is 16KB.
7413 If @var{size} is a positive number, then @value{GDBN} will try to
7414 allocate a buffer of at least @var{size} bytes for each new thread
7415 that uses the btrace recording method and the Intel Processor Trace
7416 format. The actually obtained buffer size may differ from the
7417 requested @var{size}. Use the @code{info record} command to see the
7418 actual buffer size for each thread.
7420 If @var{limit} is @code{unlimited} or zero, @value{GDBN} will try to
7421 allocate a buffer of 4MB.
7423 Bigger buffers mean longer traces. On the other hand, @value{GDBN} will
7424 also need longer to process the branch trace data before it can be used.
7426 @item show record btrace pt buffer-size @var{size}
7427 Show the current setting of the requested ring buffer size for branch
7428 tracing in Intel Processor Trace format.
7432 Show various statistics about the recording depending on the recording
7437 For the @code{full} recording method, it shows the state of process
7438 record and its in-memory execution log buffer, including:
7442 Whether in record mode or replay mode.
7444 Lowest recorded instruction number (counting from when the current execution log started recording instructions).
7446 Highest recorded instruction number.
7448 Current instruction about to be replayed (if in replay mode).
7450 Number of instructions contained in the execution log.
7452 Maximum number of instructions that may be contained in the execution log.
7456 For the @code{btrace} recording method, it shows:
7462 Number of instructions that have been recorded.
7464 Number of blocks of sequential control-flow formed by the recorded
7467 Whether in record mode or replay mode.
7470 For the @code{bts} recording format, it also shows:
7473 Size of the perf ring buffer.
7476 For the @code{pt} recording format, it also shows:
7479 Size of the perf ring buffer.
7483 @kindex record delete
7486 When record target runs in replay mode (``in the past''), delete the
7487 subsequent execution log and begin to record a new execution log starting
7488 from the current address. This means you will abandon the previously
7489 recorded ``future'' and begin recording a new ``future''.
7491 @kindex record instruction-history
7492 @kindex rec instruction-history
7493 @item record instruction-history
7494 Disassembles instructions from the recorded execution log. By
7495 default, ten instructions are disassembled. This can be changed using
7496 the @code{set record instruction-history-size} command. Instructions
7497 are printed in execution order.
7499 It can also print mixed source+disassembly if you specify the the
7500 @code{/m} or @code{/s} modifier, and print the raw instructions in hex
7501 as well as in symbolic form by specifying the @code{/r} modifier.
7503 The current position marker is printed for the instruction at the
7504 current program counter value. This instruction can appear multiple
7505 times in the trace and the current position marker will be printed
7506 every time. To omit the current position marker, specify the
7509 To better align the printed instructions when the trace contains
7510 instructions from more than one function, the function name may be
7511 omitted by specifying the @code{/f} modifier.
7513 Speculatively executed instructions are prefixed with @samp{?}. This
7514 feature is not available for all recording formats.
7516 There are several ways to specify what part of the execution log to
7520 @item record instruction-history @var{insn}
7521 Disassembles ten instructions starting from instruction number
7524 @item record instruction-history @var{insn}, +/-@var{n}
7525 Disassembles @var{n} instructions around instruction number
7526 @var{insn}. If @var{n} is preceded with @code{+}, disassembles
7527 @var{n} instructions after instruction number @var{insn}. If
7528 @var{n} is preceded with @code{-}, disassembles @var{n}
7529 instructions before instruction number @var{insn}.
7531 @item record instruction-history
7532 Disassembles ten more instructions after the last disassembly.
7534 @item record instruction-history -
7535 Disassembles ten more instructions before the last disassembly.
7537 @item record instruction-history @var{begin}, @var{end}
7538 Disassembles instructions beginning with instruction number
7539 @var{begin} until instruction number @var{end}. The instruction
7540 number @var{end} is included.
7543 This command may not be available for all recording methods.
7546 @item set record instruction-history-size @var{size}
7547 @itemx set record instruction-history-size unlimited
7548 Define how many instructions to disassemble in the @code{record
7549 instruction-history} command. The default value is 10.
7550 A @var{size} of @code{unlimited} means unlimited instructions.
7553 @item show record instruction-history-size
7554 Show how many instructions to disassemble in the @code{record
7555 instruction-history} command.
7557 @kindex record function-call-history
7558 @kindex rec function-call-history
7559 @item record function-call-history
7560 Prints the execution history at function granularity. It prints one
7561 line for each sequence of instructions that belong to the same
7562 function giving the name of that function, the source lines
7563 for this instruction sequence (if the @code{/l} modifier is
7564 specified), and the instructions numbers that form the sequence (if
7565 the @code{/i} modifier is specified). The function names are indented
7566 to reflect the call stack depth if the @code{/c} modifier is
7567 specified. The @code{/l}, @code{/i}, and @code{/c} modifiers can be
7571 (@value{GDBP}) @b{list 1, 10}
7582 (@value{GDBP}) @b{record function-call-history /ilc}
7583 1 bar inst 1,4 at foo.c:6,8
7584 2 foo inst 5,10 at foo.c:2,3
7585 3 bar inst 11,13 at foo.c:9,10
7588 By default, ten lines are printed. This can be changed using the
7589 @code{set record function-call-history-size} command. Functions are
7590 printed in execution order. There are several ways to specify what
7594 @item record function-call-history @var{func}
7595 Prints ten functions starting from function number @var{func}.
7597 @item record function-call-history @var{func}, +/-@var{n}
7598 Prints @var{n} functions around function number @var{func}. If
7599 @var{n} is preceded with @code{+}, prints @var{n} functions after
7600 function number @var{func}. If @var{n} is preceded with @code{-},
7601 prints @var{n} functions before function number @var{func}.
7603 @item record function-call-history
7604 Prints ten more functions after the last ten-line print.
7606 @item record function-call-history -
7607 Prints ten more functions before the last ten-line print.
7609 @item record function-call-history @var{begin}, @var{end}
7610 Prints functions beginning with function number @var{begin} until
7611 function number @var{end}. The function number @var{end} is included.
7614 This command may not be available for all recording methods.
7616 @item set record function-call-history-size @var{size}
7617 @itemx set record function-call-history-size unlimited
7618 Define how many lines to print in the
7619 @code{record function-call-history} command. The default value is 10.
7620 A size of @code{unlimited} means unlimited lines.
7622 @item show record function-call-history-size
7623 Show how many lines to print in the
7624 @code{record function-call-history} command.
7629 @chapter Examining the Stack
7631 When your program has stopped, the first thing you need to know is where it
7632 stopped and how it got there.
7635 Each time your program performs a function call, information about the call
7637 That information includes the location of the call in your program,
7638 the arguments of the call,
7639 and the local variables of the function being called.
7640 The information is saved in a block of data called a @dfn{stack frame}.
7641 The stack frames are allocated in a region of memory called the @dfn{call
7644 When your program stops, the @value{GDBN} commands for examining the
7645 stack allow you to see all of this information.
7647 @cindex selected frame
7648 One of the stack frames is @dfn{selected} by @value{GDBN} and many
7649 @value{GDBN} commands refer implicitly to the selected frame. In
7650 particular, whenever you ask @value{GDBN} for the value of a variable in
7651 your program, the value is found in the selected frame. There are
7652 special @value{GDBN} commands to select whichever frame you are
7653 interested in. @xref{Selection, ,Selecting a Frame}.
7655 When your program stops, @value{GDBN} automatically selects the
7656 currently executing frame and describes it briefly, similar to the
7657 @code{frame} command (@pxref{Frame Info, ,Information about a Frame}).
7660 * Frames:: Stack frames
7661 * Backtrace:: Backtraces
7662 * Selection:: Selecting a frame
7663 * Frame Info:: Information on a frame
7664 * Frame Apply:: Applying a command to several frames
7665 * Frame Filter Management:: Managing frame filters
7670 @section Stack Frames
7672 @cindex frame, definition
7674 The call stack is divided up into contiguous pieces called @dfn{stack
7675 frames}, or @dfn{frames} for short; each frame is the data associated
7676 with one call to one function. The frame contains the arguments given
7677 to the function, the function's local variables, and the address at
7678 which the function is executing.
7680 @cindex initial frame
7681 @cindex outermost frame
7682 @cindex innermost frame
7683 When your program is started, the stack has only one frame, that of the
7684 function @code{main}. This is called the @dfn{initial} frame or the
7685 @dfn{outermost} frame. Each time a function is called, a new frame is
7686 made. Each time a function returns, the frame for that function invocation
7687 is eliminated. If a function is recursive, there can be many frames for
7688 the same function. The frame for the function in which execution is
7689 actually occurring is called the @dfn{innermost} frame. This is the most
7690 recently created of all the stack frames that still exist.
7692 @cindex frame pointer
7693 Inside your program, stack frames are identified by their addresses. A
7694 stack frame consists of many bytes, each of which has its own address; each
7695 kind of computer has a convention for choosing one byte whose
7696 address serves as the address of the frame. Usually this address is kept
7697 in a register called the @dfn{frame pointer register}
7698 (@pxref{Registers, $fp}) while execution is going on in that frame.
7701 @cindex frame number
7702 @value{GDBN} labels each existing stack frame with a @dfn{level}, a
7703 number that is zero for the innermost frame, one for the frame that
7704 called it, and so on upward. These level numbers give you a way of
7705 designating stack frames in @value{GDBN} commands. The terms
7706 @dfn{frame number} and @dfn{frame level} can be used interchangeably to
7707 describe this number.
7709 @c The -fomit-frame-pointer below perennially causes hbox overflow
7710 @c underflow problems.
7711 @cindex frameless execution
7712 Some compilers provide a way to compile functions so that they operate
7713 without stack frames. (For example, the @value{NGCC} option
7715 @samp{-fomit-frame-pointer}
7717 generates functions without a frame.)
7718 This is occasionally done with heavily used library functions to save
7719 the frame setup time. @value{GDBN} has limited facilities for dealing
7720 with these function invocations. If the innermost function invocation
7721 has no stack frame, @value{GDBN} nevertheless regards it as though
7722 it had a separate frame, which is numbered zero as usual, allowing
7723 correct tracing of the function call chain. However, @value{GDBN} has
7724 no provision for frameless functions elsewhere in the stack.
7730 @cindex call stack traces
7731 A backtrace is a summary of how your program got where it is. It shows one
7732 line per frame, for many frames, starting with the currently executing
7733 frame (frame zero), followed by its caller (frame one), and on up the
7736 @anchor{backtrace-command}
7738 @kindex bt @r{(@code{backtrace})}
7739 To print a backtrace of the entire stack, use the @code{backtrace}
7740 command, or its alias @code{bt}. This command will print one line per
7741 frame for frames in the stack. By default, all stack frames are
7742 printed. You can stop the backtrace at any time by typing the system
7743 interrupt character, normally @kbd{Ctrl-c}.
7746 @item backtrace [@var{option}]@dots{} [@var{qualifier}]@dots{} [@var{count}]
7747 @itemx bt [@var{option}]@dots{} [@var{qualifier}]@dots{} [@var{count}]
7748 Print the backtrace of the entire stack.
7750 The optional @var{count} can be one of the following:
7755 Print only the innermost @var{n} frames, where @var{n} is a positive
7760 Print only the outermost @var{n} frames, where @var{n} is a positive
7768 Print the values of the local variables also. This can be combined
7769 with the optional @var{count} to limit the number of frames shown.
7772 Do not run Python frame filters on this backtrace. @xref{Frame
7773 Filter API}, for more information. Additionally use @ref{disable
7774 frame-filter all} to turn off all frame filters. This is only
7775 relevant when @value{GDBN} has been configured with @code{Python}
7779 A Python frame filter might decide to ``elide'' some frames. Normally
7780 such elided frames are still printed, but they are indented relative
7781 to the filtered frames that cause them to be elided. The @code{-hide}
7782 option causes elided frames to not be printed at all.
7785 The @code{backtrace} command also supports a number of options that
7786 allow overriding relevant global print settings as set by @code{set
7787 backtrace} and @code{set print} subcommands:
7790 @item -past-main [@code{on}|@code{off}]
7791 Set whether backtraces should continue past @code{main}. Related setting:
7792 @ref{set backtrace past-main}.
7794 @item -past-entry [@code{on}|@code{off}]
7795 Set whether backtraces should continue past the entry point of a program.
7796 Related setting: @ref{set backtrace past-entry}.
7798 @item -entry-values @code{no}|@code{only}|@code{preferred}|@code{if-needed}|@code{both}|@code{compact}|@code{default}
7799 Set printing of function arguments at function entry.
7800 Related setting: @ref{set print entry-values}.
7802 @item -frame-arguments @code{all}|@code{scalars}|@code{none}
7803 Set printing of non-scalar frame arguments.
7804 Related setting: @ref{set print frame-arguments}.
7806 @item -raw-frame-arguments [@code{on}|@code{off}]
7807 Set whether to print frame arguments in raw form.
7808 Related setting: @ref{set print raw-frame-arguments}.
7810 @item -frame-info @code{auto}|@code{source-line}|@code{location}|@code{source-and-location}|@code{location-and-address}|@code{short-location}
7811 Set printing of frame information.
7812 Related setting: @ref{set print frame-info}.
7815 The optional @var{qualifier} is maintained for backward compatibility.
7816 It can be one of the following:
7820 Equivalent to the @code{-full} option.
7823 Equivalent to the @code{-no-filters} option.
7826 Equivalent to the @code{-hide} option.
7833 The names @code{where} and @code{info stack} (abbreviated @code{info s})
7834 are additional aliases for @code{backtrace}.
7836 @cindex multiple threads, backtrace
7837 In a multi-threaded program, @value{GDBN} by default shows the
7838 backtrace only for the current thread. To display the backtrace for
7839 several or all of the threads, use the command @code{thread apply}
7840 (@pxref{Threads, thread apply}). For example, if you type @kbd{thread
7841 apply all backtrace}, @value{GDBN} will display the backtrace for all
7842 the threads; this is handy when you debug a core dump of a
7843 multi-threaded program.
7845 Each line in the backtrace shows the frame number and the function name.
7846 The program counter value is also shown---unless you use @code{set
7847 print address off}. The backtrace also shows the source file name and
7848 line number, as well as the arguments to the function. The program
7849 counter value is omitted if it is at the beginning of the code for that
7852 Here is an example of a backtrace. It was made with the command
7853 @samp{bt 3}, so it shows the innermost three frames.
7857 #0 m4_traceon (obs=0x24eb0, argc=1, argv=0x2b8c8)
7859 #1 0x6e38 in expand_macro (sym=0x2b600, data=...) at macro.c:242
7860 #2 0x6840 in expand_token (obs=0x0, t=177664, td=0xf7fffb08)
7862 (More stack frames follow...)
7867 The display for frame zero does not begin with a program counter
7868 value, indicating that your program has stopped at the beginning of the
7869 code for line @code{993} of @code{builtin.c}.
7872 The value of parameter @code{data} in frame 1 has been replaced by
7873 @code{@dots{}}. By default, @value{GDBN} prints the value of a parameter
7874 only if it is a scalar (integer, pointer, enumeration, etc). See command
7875 @kbd{set print frame-arguments} in @ref{Print Settings} for more details
7876 on how to configure the way function parameter values are printed.
7877 The command @kbd{set print frame-info} (@pxref{Print Settings}) controls
7878 what frame information is printed.
7880 @cindex optimized out, in backtrace
7881 @cindex function call arguments, optimized out
7882 If your program was compiled with optimizations, some compilers will
7883 optimize away arguments passed to functions if those arguments are
7884 never used after the call. Such optimizations generate code that
7885 passes arguments through registers, but doesn't store those arguments
7886 in the stack frame. @value{GDBN} has no way of displaying such
7887 arguments in stack frames other than the innermost one. Here's what
7888 such a backtrace might look like:
7892 #0 m4_traceon (obs=0x24eb0, argc=1, argv=0x2b8c8)
7894 #1 0x6e38 in expand_macro (sym=<optimized out>) at macro.c:242
7895 #2 0x6840 in expand_token (obs=0x0, t=<optimized out>, td=0xf7fffb08)
7897 (More stack frames follow...)
7902 The values of arguments that were not saved in their stack frames are
7903 shown as @samp{<optimized out>}.
7905 If you need to display the values of such optimized-out arguments,
7906 either deduce that from other variables whose values depend on the one
7907 you are interested in, or recompile without optimizations.
7909 @cindex backtrace beyond @code{main} function
7910 @cindex program entry point
7911 @cindex startup code, and backtrace
7912 Most programs have a standard user entry point---a place where system
7913 libraries and startup code transition into user code. For C this is
7914 @code{main}@footnote{
7915 Note that embedded programs (the so-called ``free-standing''
7916 environment) are not required to have a @code{main} function as the
7917 entry point. They could even have multiple entry points.}.
7918 When @value{GDBN} finds the entry function in a backtrace
7919 it will terminate the backtrace, to avoid tracing into highly
7920 system-specific (and generally uninteresting) code.
7922 If you need to examine the startup code, or limit the number of levels
7923 in a backtrace, you can change this behavior:
7926 @item set backtrace past-main
7927 @itemx set backtrace past-main on
7928 @anchor{set backtrace past-main}
7929 @kindex set backtrace
7930 Backtraces will continue past the user entry point.
7932 @item set backtrace past-main off
7933 Backtraces will stop when they encounter the user entry point. This is the
7936 @item show backtrace past-main
7937 @kindex show backtrace
7938 Display the current user entry point backtrace policy.
7940 @item set backtrace past-entry
7941 @itemx set backtrace past-entry on
7942 @anchor{set backtrace past-entry}
7943 Backtraces will continue past the internal entry point of an application.
7944 This entry point is encoded by the linker when the application is built,
7945 and is likely before the user entry point @code{main} (or equivalent) is called.
7947 @item set backtrace past-entry off
7948 Backtraces will stop when they encounter the internal entry point of an
7949 application. This is the default.
7951 @item show backtrace past-entry
7952 Display the current internal entry point backtrace policy.
7954 @item set backtrace limit @var{n}
7955 @itemx set backtrace limit 0
7956 @itemx set backtrace limit unlimited
7957 @anchor{set backtrace limit}
7958 @cindex backtrace limit
7959 Limit the backtrace to @var{n} levels. A value of @code{unlimited}
7960 or zero means unlimited levels.
7962 @item show backtrace limit
7963 Display the current limit on backtrace levels.
7966 You can control how file names are displayed.
7969 @item set filename-display
7970 @itemx set filename-display relative
7971 @cindex filename-display
7972 Display file names relative to the compilation directory. This is the default.
7974 @item set filename-display basename
7975 Display only basename of a filename.
7977 @item set filename-display absolute
7978 Display an absolute filename.
7980 @item show filename-display
7981 Show the current way to display filenames.
7985 @section Selecting a Frame
7987 Most commands for examining the stack and other data in your program work on
7988 whichever stack frame is selected at the moment. Here are the commands for
7989 selecting a stack frame; all of them finish by printing a brief description
7990 of the stack frame just selected.
7993 @kindex frame@r{, selecting}
7994 @kindex f @r{(@code{frame})}
7995 @item frame @r{[} @var{frame-selection-spec} @r{]}
7996 @item f @r{[} @var{frame-selection-spec} @r{]}
7997 The @command{frame} command allows different stack frames to be
7998 selected. The @var{frame-selection-spec} can be any of the following:
8003 @item level @var{num}
8004 Select frame level @var{num}. Recall that frame zero is the innermost
8005 (currently executing) frame, frame one is the frame that called the
8006 innermost one, and so on. The highest level frame is usually the one
8009 As this is the most common method of navigating the frame stack, the
8010 string @command{level} can be omitted. For example, the following two
8011 commands are equivalent:
8014 (@value{GDBP}) frame 3
8015 (@value{GDBP}) frame level 3
8018 @kindex frame address
8019 @item address @var{stack-address}
8020 Select the frame with stack address @var{stack-address}. The
8021 @var{stack-address} for a frame can be seen in the output of
8022 @command{info frame}, for example:
8026 Stack level 1, frame at 0x7fffffffda30:
8027 rip = 0x40066d in b (amd64-entry-value.cc:59); saved rip 0x4004c5
8028 tail call frame, caller of frame at 0x7fffffffda30
8029 source language c++.
8030 Arglist at unknown address.
8031 Locals at unknown address, Previous frame's sp is 0x7fffffffda30
8034 The @var{stack-address} for this frame is @code{0x7fffffffda30} as
8035 indicated by the line:
8038 Stack level 1, frame at 0x7fffffffda30:
8041 @kindex frame function
8042 @item function @var{function-name}
8043 Select the stack frame for function @var{function-name}. If there are
8044 multiple stack frames for function @var{function-name} then the inner
8045 most stack frame is selected.
8048 @item view @var{stack-address} @r{[} @var{pc-addr} @r{]}
8049 View a frame that is not part of @value{GDBN}'s backtrace. The frame
8050 viewed has stack address @var{stack-addr}, and optionally, a program
8051 counter address of @var{pc-addr}.
8053 This is useful mainly if the chaining of stack frames has been
8054 damaged by a bug, making it impossible for @value{GDBN} to assign
8055 numbers properly to all frames. In addition, this can be useful
8056 when your program has multiple stacks and switches between them.
8058 When viewing a frame outside the current backtrace using
8059 @command{frame view} then you can always return to the original
8060 stack using one of the previous stack frame selection instructions,
8061 for example @command{frame level 0}.
8067 Move @var{n} frames up the stack; @var{n} defaults to 1. For positive
8068 numbers @var{n}, this advances toward the outermost frame, to higher
8069 frame numbers, to frames that have existed longer.
8072 @kindex do @r{(@code{down})}
8074 Move @var{n} frames down the stack; @var{n} defaults to 1. For
8075 positive numbers @var{n}, this advances toward the innermost frame, to
8076 lower frame numbers, to frames that were created more recently.
8077 You may abbreviate @code{down} as @code{do}.
8080 All of these commands end by printing two lines of output describing the
8081 frame. The first line shows the frame number, the function name, the
8082 arguments, and the source file and line number of execution in that
8083 frame. The second line shows the text of that source line.
8091 #1 0x22f0 in main (argc=1, argv=0xf7fffbf4, env=0xf7fffbfc)
8093 10 read_input_file (argv[i]);
8097 After such a printout, the @code{list} command with no arguments
8098 prints ten lines centered on the point of execution in the frame.
8099 You can also edit the program at the point of execution with your favorite
8100 editing program by typing @code{edit}.
8101 @xref{List, ,Printing Source Lines},
8105 @kindex select-frame
8106 @item select-frame @r{[} @var{frame-selection-spec} @r{]}
8107 The @code{select-frame} command is a variant of @code{frame} that does
8108 not display the new frame after selecting it. This command is
8109 intended primarily for use in @value{GDBN} command scripts, where the
8110 output might be unnecessary and distracting. The
8111 @var{frame-selection-spec} is as for the @command{frame} command
8112 described in @ref{Selection, ,Selecting a Frame}.
8114 @kindex down-silently
8116 @item up-silently @var{n}
8117 @itemx down-silently @var{n}
8118 These two commands are variants of @code{up} and @code{down},
8119 respectively; they differ in that they do their work silently, without
8120 causing display of the new frame. They are intended primarily for use
8121 in @value{GDBN} command scripts, where the output might be unnecessary and
8126 @section Information About a Frame
8128 There are several other commands to print information about the selected
8134 When used without any argument, this command does not change which
8135 frame is selected, but prints a brief description of the currently
8136 selected stack frame. It can be abbreviated @code{f}. With an
8137 argument, this command is used to select a stack frame.
8138 @xref{Selection, ,Selecting a Frame}.
8141 @kindex info f @r{(@code{info frame})}
8144 This command prints a verbose description of the selected stack frame,
8149 the address of the frame
8151 the address of the next frame down (called by this frame)
8153 the address of the next frame up (caller of this frame)
8155 the language in which the source code corresponding to this frame is written
8157 the address of the frame's arguments
8159 the address of the frame's local variables
8161 the program counter saved in it (the address of execution in the caller frame)
8163 which registers were saved in the frame
8166 @noindent The verbose description is useful when
8167 something has gone wrong that has made the stack format fail to fit
8168 the usual conventions.
8170 @item info frame @r{[} @var{frame-selection-spec} @r{]}
8171 @itemx info f @r{[} @var{frame-selection-spec} @r{]}
8172 Print a verbose description of the frame selected by
8173 @var{frame-selection-spec}. The @var{frame-selection-spec} is the
8174 same as for the @command{frame} command (@pxref{Selection, ,Selecting
8175 a Frame}). The selected frame remains unchanged by this command.
8178 @item info args [-q]
8179 Print the arguments of the selected frame, each on a separate line.
8181 The optional flag @samp{-q}, which stands for @samp{quiet}, disables
8182 printing header information and messages explaining why no argument
8185 @item info args [-q] [-t @var{type_regexp}] [@var{regexp}]
8186 Like @kbd{info args}, but only print the arguments selected
8187 with the provided regexp(s).
8189 If @var{regexp} is provided, print only the arguments whose names
8190 match the regular expression @var{regexp}.
8192 If @var{type_regexp} is provided, print only the arguments whose
8193 types, as printed by the @code{whatis} command, match
8194 the regular expression @var{type_regexp}.
8195 If @var{type_regexp} contains space(s), it should be enclosed in
8196 quote characters. If needed, use backslash to escape the meaning
8197 of special characters or quotes.
8199 If both @var{regexp} and @var{type_regexp} are provided, an argument
8200 is printed only if its name matches @var{regexp} and its type matches
8203 @item info locals [-q]
8205 Print the local variables of the selected frame, each on a separate
8206 line. These are all variables (declared either static or automatic)
8207 accessible at the point of execution of the selected frame.
8209 The optional flag @samp{-q}, which stands for @samp{quiet}, disables
8210 printing header information and messages explaining why no local variables
8213 @item info locals [-q] [-t @var{type_regexp}] [@var{regexp}]
8214 Like @kbd{info locals}, but only print the local variables selected
8215 with the provided regexp(s).
8217 If @var{regexp} is provided, print only the local variables whose names
8218 match the regular expression @var{regexp}.
8220 If @var{type_regexp} is provided, print only the local variables whose
8221 types, as printed by the @code{whatis} command, match
8222 the regular expression @var{type_regexp}.
8223 If @var{type_regexp} contains space(s), it should be enclosed in
8224 quote characters. If needed, use backslash to escape the meaning
8225 of special characters or quotes.
8227 If both @var{regexp} and @var{type_regexp} are provided, a local variable
8228 is printed only if its name matches @var{regexp} and its type matches
8231 The command @kbd{info locals -q -t @var{type_regexp}} can usefully be
8232 combined with the commands @kbd{frame apply} and @kbd{thread apply}.
8233 For example, your program might use Resource Acquisition Is
8234 Initialization types (RAII) such as @code{lock_something_t}: each
8235 local variable of type @code{lock_something_t} automatically places a
8236 lock that is destroyed when the variable goes out of scope. You can
8237 then list all acquired locks in your program by doing
8239 thread apply all -s frame apply all -s info locals -q -t lock_something_t
8242 or the equivalent shorter form
8244 tfaas i lo -q -t lock_something_t
8250 @section Applying a Command to Several Frames.
8251 @anchor{frame apply}
8253 @cindex apply command to several frames
8255 @item frame apply [all | @var{count} | @var{-count} | level @var{level}@dots{}] [@var{option}]@dots{} @var{command}
8256 The @code{frame apply} command allows you to apply the named
8257 @var{command} to one or more frames.
8261 Specify @code{all} to apply @var{command} to all frames.
8264 Use @var{count} to apply @var{command} to the innermost @var{count}
8265 frames, where @var{count} is a positive number.
8268 Use @var{-count} to apply @var{command} to the outermost @var{count}
8269 frames, where @var{count} is a positive number.
8272 Use @code{level} to apply @var{command} to the set of frames identified
8273 by the @var{level} list. @var{level} is a frame level or a range of frame
8274 levels as @var{level1}-@var{level2}. The frame level is the number shown
8275 in the first field of the @samp{backtrace} command output.
8276 E.g., @samp{2-4 6-8 3} indicates to apply @var{command} for the frames
8277 at levels 2, 3, 4, 6, 7, 8, and then again on frame at level 3.
8281 Note that the frames on which @code{frame apply} applies a command are
8282 also influenced by the @code{set backtrace} settings such as @code{set
8283 backtrace past-main} and @code{set backtrace limit N}.
8284 @xref{Backtrace,,Backtraces}.
8286 The @code{frame apply} command also supports a number of options that
8287 allow overriding relevant @code{set backtrace} settings:
8290 @item -past-main [@code{on}|@code{off}]
8291 Whether backtraces should continue past @code{main}.
8292 Related setting: @ref{set backtrace past-main}.
8294 @item -past-entry [@code{on}|@code{off}]
8295 Whether backtraces should continue past the entry point of a program.
8296 Related setting: @ref{set backtrace past-entry}.
8299 By default, @value{GDBN} displays some frame information before the
8300 output produced by @var{command}, and an error raised during the
8301 execution of a @var{command} will abort @code{frame apply}. The
8302 following options can be used to fine-tune these behaviors:
8306 The flag @code{-c}, which stands for @samp{continue}, causes any
8307 errors in @var{command} to be displayed, and the execution of
8308 @code{frame apply} then continues.
8310 The flag @code{-s}, which stands for @samp{silent}, causes any errors
8311 or empty output produced by a @var{command} to be silently ignored.
8312 That is, the execution continues, but the frame information and errors
8315 The flag @code{-q} (@samp{quiet}) disables printing the frame
8319 The following example shows how the flags @code{-c} and @code{-s} are
8320 working when applying the command @code{p j} to all frames, where
8321 variable @code{j} can only be successfully printed in the outermost
8322 @code{#1 main} frame.
8326 (gdb) frame apply all p j
8327 #0 some_function (i=5) at fun.c:4
8328 No symbol "j" in current context.
8329 (gdb) frame apply all -c p j
8330 #0 some_function (i=5) at fun.c:4
8331 No symbol "j" in current context.
8332 #1 0x565555fb in main (argc=1, argv=0xffffd2c4) at fun.c:11
8334 (gdb) frame apply all -s p j
8335 #1 0x565555fb in main (argc=1, argv=0xffffd2c4) at fun.c:11
8341 By default, @samp{frame apply}, prints the frame location
8342 information before the command output:
8346 (gdb) frame apply all p $sp
8347 #0 some_function (i=5) at fun.c:4
8348 $4 = (void *) 0xffffd1e0
8349 #1 0x565555fb in main (argc=1, argv=0xffffd2c4) at fun.c:11
8350 $5 = (void *) 0xffffd1f0
8355 If the flag @code{-q} is given, no frame information is printed:
8358 (gdb) frame apply all -q p $sp
8359 $12 = (void *) 0xffffd1e0
8360 $13 = (void *) 0xffffd1f0
8370 @cindex apply a command to all frames (ignoring errors and empty output)
8371 @item faas @var{command}
8372 Shortcut for @code{frame apply all -s @var{command}}.
8373 Applies @var{command} on all frames, ignoring errors and empty output.
8375 It can for example be used to print a local variable or a function
8376 argument without knowing the frame where this variable or argument
8379 (@value{GDBP}) faas p some_local_var_i_do_not_remember_where_it_is
8382 The @code{faas} command accepts the same options as the @code{frame
8383 apply} command. @xref{frame apply}.
8385 Note that the command @code{tfaas @var{command}} applies @var{command}
8386 on all frames of all threads. See @xref{Threads,,Threads}.
8390 @node Frame Filter Management
8391 @section Management of Frame Filters.
8392 @cindex managing frame filters
8394 Frame filters are Python based utilities to manage and decorate the
8395 output of frames. @xref{Frame Filter API}, for further information.
8397 Managing frame filters is performed by several commands available
8398 within @value{GDBN}, detailed here.
8401 @kindex info frame-filter
8402 @item info frame-filter
8403 Print a list of installed frame filters from all dictionaries, showing
8404 their name, priority and enabled status.
8406 @kindex disable frame-filter
8407 @anchor{disable frame-filter all}
8408 @item disable frame-filter @var{filter-dictionary} @var{filter-name}
8409 Disable a frame filter in the dictionary matching
8410 @var{filter-dictionary} and @var{filter-name}. The
8411 @var{filter-dictionary} may be @code{all}, @code{global},
8412 @code{progspace}, or the name of the object file where the frame filter
8413 dictionary resides. When @code{all} is specified, all frame filters
8414 across all dictionaries are disabled. The @var{filter-name} is the name
8415 of the frame filter and is used when @code{all} is not the option for
8416 @var{filter-dictionary}. A disabled frame-filter is not deleted, it
8417 may be enabled again later.
8419 @kindex enable frame-filter
8420 @item enable frame-filter @var{filter-dictionary} @var{filter-name}
8421 Enable a frame filter in the dictionary matching
8422 @var{filter-dictionary} and @var{filter-name}. The
8423 @var{filter-dictionary} may be @code{all}, @code{global},
8424 @code{progspace} or the name of the object file where the frame filter
8425 dictionary resides. When @code{all} is specified, all frame filters across
8426 all dictionaries are enabled. The @var{filter-name} is the name of the frame
8427 filter and is used when @code{all} is not the option for
8428 @var{filter-dictionary}.
8433 (gdb) info frame-filter
8435 global frame-filters:
8436 Priority Enabled Name
8437 1000 No PrimaryFunctionFilter
8440 progspace /build/test frame-filters:
8441 Priority Enabled Name
8442 100 Yes ProgspaceFilter
8444 objfile /build/test frame-filters:
8445 Priority Enabled Name
8446 999 Yes BuildProgra Filter
8448 (gdb) disable frame-filter /build/test BuildProgramFilter
8449 (gdb) info frame-filter
8451 global frame-filters:
8452 Priority Enabled Name
8453 1000 No PrimaryFunctionFilter
8456 progspace /build/test frame-filters:
8457 Priority Enabled Name
8458 100 Yes ProgspaceFilter
8460 objfile /build/test frame-filters:
8461 Priority Enabled Name
8462 999 No BuildProgramFilter
8464 (gdb) enable frame-filter global PrimaryFunctionFilter
8465 (gdb) info frame-filter
8467 global frame-filters:
8468 Priority Enabled Name
8469 1000 Yes PrimaryFunctionFilter
8472 progspace /build/test frame-filters:
8473 Priority Enabled Name
8474 100 Yes ProgspaceFilter
8476 objfile /build/test frame-filters:
8477 Priority Enabled Name
8478 999 No BuildProgramFilter
8481 @kindex set frame-filter priority
8482 @item set frame-filter priority @var{filter-dictionary} @var{filter-name} @var{priority}
8483 Set the @var{priority} of a frame filter in the dictionary matching
8484 @var{filter-dictionary}, and the frame filter name matching
8485 @var{filter-name}. The @var{filter-dictionary} may be @code{global},
8486 @code{progspace} or the name of the object file where the frame filter
8487 dictionary resides. The @var{priority} is an integer.
8489 @kindex show frame-filter priority
8490 @item show frame-filter priority @var{filter-dictionary} @var{filter-name}
8491 Show the @var{priority} of a frame filter in the dictionary matching
8492 @var{filter-dictionary}, and the frame filter name matching
8493 @var{filter-name}. The @var{filter-dictionary} may be @code{global},
8494 @code{progspace} or the name of the object file where the frame filter
8500 (gdb) info frame-filter
8502 global frame-filters:
8503 Priority Enabled Name
8504 1000 Yes PrimaryFunctionFilter
8507 progspace /build/test frame-filters:
8508 Priority Enabled Name
8509 100 Yes ProgspaceFilter
8511 objfile /build/test frame-filters:
8512 Priority Enabled Name
8513 999 No BuildProgramFilter
8515 (gdb) set frame-filter priority global Reverse 50
8516 (gdb) info frame-filter
8518 global frame-filters:
8519 Priority Enabled Name
8520 1000 Yes PrimaryFunctionFilter
8523 progspace /build/test frame-filters:
8524 Priority Enabled Name
8525 100 Yes ProgspaceFilter
8527 objfile /build/test frame-filters:
8528 Priority Enabled Name
8529 999 No BuildProgramFilter
8534 @chapter Examining Source Files
8536 @value{GDBN} can print parts of your program's source, since the debugging
8537 information recorded in the program tells @value{GDBN} what source files were
8538 used to build it. When your program stops, @value{GDBN} spontaneously prints
8539 the line where it stopped. Likewise, when you select a stack frame
8540 (@pxref{Selection, ,Selecting a Frame}), @value{GDBN} prints the line where
8541 execution in that frame has stopped. You can print other portions of
8542 source files by explicit command.
8544 If you use @value{GDBN} through its @sc{gnu} Emacs interface, you may
8545 prefer to use Emacs facilities to view source; see @ref{Emacs, ,Using
8546 @value{GDBN} under @sc{gnu} Emacs}.
8549 * List:: Printing source lines
8550 * Specify Location:: How to specify code locations
8551 * Edit:: Editing source files
8552 * Search:: Searching source files
8553 * Source Path:: Specifying source directories
8554 * Machine Code:: Source and machine code
8558 @section Printing Source Lines
8561 @kindex l @r{(@code{list})}
8562 To print lines from a source file, use the @code{list} command
8563 (abbreviated @code{l}). By default, ten lines are printed.
8564 There are several ways to specify what part of the file you want to
8565 print; see @ref{Specify Location}, for the full list.
8567 Here are the forms of the @code{list} command most commonly used:
8570 @item list @var{linenum}
8571 Print lines centered around line number @var{linenum} in the
8572 current source file.
8574 @item list @var{function}
8575 Print lines centered around the beginning of function
8579 Print more lines. If the last lines printed were printed with a
8580 @code{list} command, this prints lines following the last lines
8581 printed; however, if the last line printed was a solitary line printed
8582 as part of displaying a stack frame (@pxref{Stack, ,Examining the
8583 Stack}), this prints lines centered around that line.
8586 Print lines just before the lines last printed.
8589 @cindex @code{list}, how many lines to display
8590 By default, @value{GDBN} prints ten source lines with any of these forms of
8591 the @code{list} command. You can change this using @code{set listsize}:
8594 @kindex set listsize
8595 @item set listsize @var{count}
8596 @itemx set listsize unlimited
8597 Make the @code{list} command display @var{count} source lines (unless
8598 the @code{list} argument explicitly specifies some other number).
8599 Setting @var{count} to @code{unlimited} or 0 means there's no limit.
8601 @kindex show listsize
8603 Display the number of lines that @code{list} prints.
8606 Repeating a @code{list} command with @key{RET} discards the argument,
8607 so it is equivalent to typing just @code{list}. This is more useful
8608 than listing the same lines again. An exception is made for an
8609 argument of @samp{-}; that argument is preserved in repetition so that
8610 each repetition moves up in the source file.
8612 In general, the @code{list} command expects you to supply zero, one or two
8613 @dfn{locations}. Locations specify source lines; there are several ways
8614 of writing them (@pxref{Specify Location}), but the effect is always
8615 to specify some source line.
8617 Here is a complete description of the possible arguments for @code{list}:
8620 @item list @var{location}
8621 Print lines centered around the line specified by @var{location}.
8623 @item list @var{first},@var{last}
8624 Print lines from @var{first} to @var{last}. Both arguments are
8625 locations. When a @code{list} command has two locations, and the
8626 source file of the second location is omitted, this refers to
8627 the same source file as the first location.
8629 @item list ,@var{last}
8630 Print lines ending with @var{last}.
8632 @item list @var{first},
8633 Print lines starting with @var{first}.
8636 Print lines just after the lines last printed.
8639 Print lines just before the lines last printed.
8642 As described in the preceding table.
8645 @node Specify Location
8646 @section Specifying a Location
8647 @cindex specifying location
8649 @cindex source location
8652 * Linespec Locations:: Linespec locations
8653 * Explicit Locations:: Explicit locations
8654 * Address Locations:: Address locations
8657 Several @value{GDBN} commands accept arguments that specify a location
8658 of your program's code. Since @value{GDBN} is a source-level
8659 debugger, a location usually specifies some line in the source code.
8660 Locations may be specified using three different formats:
8661 linespec locations, explicit locations, or address locations.
8663 @node Linespec Locations
8664 @subsection Linespec Locations
8665 @cindex linespec locations
8667 A @dfn{linespec} is a colon-separated list of source location parameters such
8668 as file name, function name, etc. Here are all the different ways of
8669 specifying a linespec:
8673 Specifies the line number @var{linenum} of the current source file.
8676 @itemx +@var{offset}
8677 Specifies the line @var{offset} lines before or after the @dfn{current
8678 line}. For the @code{list} command, the current line is the last one
8679 printed; for the breakpoint commands, this is the line at which
8680 execution stopped in the currently selected @dfn{stack frame}
8681 (@pxref{Frames, ,Frames}, for a description of stack frames.) When
8682 used as the second of the two linespecs in a @code{list} command,
8683 this specifies the line @var{offset} lines up or down from the first
8686 @item @var{filename}:@var{linenum}
8687 Specifies the line @var{linenum} in the source file @var{filename}.
8688 If @var{filename} is a relative file name, then it will match any
8689 source file name with the same trailing components. For example, if
8690 @var{filename} is @samp{gcc/expr.c}, then it will match source file
8691 name of @file{/build/trunk/gcc/expr.c}, but not
8692 @file{/build/trunk/libcpp/expr.c} or @file{/build/trunk/gcc/x-expr.c}.
8694 @item @var{function}
8695 Specifies the line that begins the body of the function @var{function}.
8696 For example, in C, this is the line with the open brace.
8698 By default, in C@t{++} and Ada, @var{function} is interpreted as
8699 specifying all functions named @var{function} in all scopes. For
8700 C@t{++}, this means in all namespaces and classes. For Ada, this
8701 means in all packages.
8703 For example, assuming a program with C@t{++} symbols named
8704 @code{A::B::func} and @code{B::func}, both commands @w{@kbd{break
8705 func}} and @w{@kbd{break B::func}} set a breakpoint on both symbols.
8707 Commands that accept a linespec let you override this with the
8708 @code{-qualified} option. For example, @w{@kbd{break -qualified
8709 func}} sets a breakpoint on a free-function named @code{func} ignoring
8710 any C@t{++} class methods and namespace functions called @code{func}.
8712 @xref{Explicit Locations}.
8714 @item @var{function}:@var{label}
8715 Specifies the line where @var{label} appears in @var{function}.
8717 @item @var{filename}:@var{function}
8718 Specifies the line that begins the body of the function @var{function}
8719 in the file @var{filename}. You only need the file name with a
8720 function name to avoid ambiguity when there are identically named
8721 functions in different source files.
8724 Specifies the line at which the label named @var{label} appears
8725 in the function corresponding to the currently selected stack frame.
8726 If there is no current selected stack frame (for instance, if the inferior
8727 is not running), then @value{GDBN} will not search for a label.
8729 @cindex breakpoint at static probe point
8730 @item -pstap|-probe-stap @r{[}@var{objfile}:@r{[}@var{provider}:@r{]}@r{]}@var{name}
8731 The @sc{gnu}/Linux tool @code{SystemTap} provides a way for
8732 applications to embed static probes. @xref{Static Probe Points}, for more
8733 information on finding and using static probes. This form of linespec
8734 specifies the location of such a static probe.
8736 If @var{objfile} is given, only probes coming from that shared library
8737 or executable matching @var{objfile} as a regular expression are considered.
8738 If @var{provider} is given, then only probes from that provider are considered.
8739 If several probes match the spec, @value{GDBN} will insert a breakpoint at
8740 each one of those probes.
8743 @node Explicit Locations
8744 @subsection Explicit Locations
8745 @cindex explicit locations
8747 @dfn{Explicit locations} allow the user to directly specify the source
8748 location's parameters using option-value pairs.
8750 Explicit locations are useful when several functions, labels, or
8751 file names have the same name (base name for files) in the program's
8752 sources. In these cases, explicit locations point to the source
8753 line you meant more accurately and unambiguously. Also, using
8754 explicit locations might be faster in large programs.
8756 For example, the linespec @samp{foo:bar} may refer to a function @code{bar}
8757 defined in the file named @file{foo} or the label @code{bar} in a function
8758 named @code{foo}. @value{GDBN} must search either the file system or
8759 the symbol table to know.
8761 The list of valid explicit location options is summarized in the
8765 @item -source @var{filename}
8766 The value specifies the source file name. To differentiate between
8767 files with the same base name, prepend as many directories as is necessary
8768 to uniquely identify the desired file, e.g., @file{foo/bar/baz.c}. Otherwise
8769 @value{GDBN} will use the first file it finds with the given base
8770 name. This option requires the use of either @code{-function} or @code{-line}.
8772 @item -function @var{function}
8773 The value specifies the name of a function. Operations
8774 on function locations unmodified by other options (such as @code{-label}
8775 or @code{-line}) refer to the line that begins the body of the function.
8776 In C, for example, this is the line with the open brace.
8778 By default, in C@t{++} and Ada, @var{function} is interpreted as
8779 specifying all functions named @var{function} in all scopes. For
8780 C@t{++}, this means in all namespaces and classes. For Ada, this
8781 means in all packages.
8783 For example, assuming a program with C@t{++} symbols named
8784 @code{A::B::func} and @code{B::func}, both commands @w{@kbd{break
8785 -function func}} and @w{@kbd{break -function B::func}} set a
8786 breakpoint on both symbols.
8788 You can use the @kbd{-qualified} flag to override this (see below).
8792 This flag makes @value{GDBN} interpret a function name specified with
8793 @kbd{-function} as a complete fully-qualified name.
8795 For example, assuming a C@t{++} program with symbols named
8796 @code{A::B::func} and @code{B::func}, the @w{@kbd{break -qualified
8797 -function B::func}} command sets a breakpoint on @code{B::func}, only.
8799 (Note: the @kbd{-qualified} option can precede a linespec as well
8800 (@pxref{Linespec Locations}), so the particular example above could be
8801 simplified as @w{@kbd{break -qualified B::func}}.)
8803 @item -label @var{label}
8804 The value specifies the name of a label. When the function
8805 name is not specified, the label is searched in the function of the currently
8806 selected stack frame.
8808 @item -line @var{number}
8809 The value specifies a line offset for the location. The offset may either
8810 be absolute (@code{-line 3}) or relative (@code{-line +3}), depending on
8811 the command. When specified without any other options, the line offset is
8812 relative to the current line.
8815 Explicit location options may be abbreviated by omitting any non-unique
8816 trailing characters from the option name, e.g., @w{@kbd{break -s main.c -li 3}}.
8818 @node Address Locations
8819 @subsection Address Locations
8820 @cindex address locations
8822 @dfn{Address locations} indicate a specific program address. They have
8823 the generalized form *@var{address}.
8825 For line-oriented commands, such as @code{list} and @code{edit}, this
8826 specifies a source line that contains @var{address}. For @code{break} and
8827 other breakpoint-oriented commands, this can be used to set breakpoints in
8828 parts of your program which do not have debugging information or
8831 Here @var{address} may be any expression valid in the current working
8832 language (@pxref{Languages, working language}) that specifies a code
8833 address. In addition, as a convenience, @value{GDBN} extends the
8834 semantics of expressions used in locations to cover several situations
8835 that frequently occur during debugging. Here are the various forms
8839 @item @var{expression}
8840 Any expression valid in the current working language.
8842 @item @var{funcaddr}
8843 An address of a function or procedure derived from its name. In C,
8844 C@t{++}, Objective-C, Fortran, minimal, and assembly, this is
8845 simply the function's name @var{function} (and actually a special case
8846 of a valid expression). In Pascal and Modula-2, this is
8847 @code{&@var{function}}. In Ada, this is @code{@var{function}'Address}
8848 (although the Pascal form also works).
8850 This form specifies the address of the function's first instruction,
8851 before the stack frame and arguments have been set up.
8853 @item '@var{filename}':@var{funcaddr}
8854 Like @var{funcaddr} above, but also specifies the name of the source
8855 file explicitly. This is useful if the name of the function does not
8856 specify the function unambiguously, e.g., if there are several
8857 functions with identical names in different source files.
8861 @section Editing Source Files
8862 @cindex editing source files
8865 @kindex e @r{(@code{edit})}
8866 To edit the lines in a source file, use the @code{edit} command.
8867 The editing program of your choice
8868 is invoked with the current line set to
8869 the active line in the program.
8870 Alternatively, there are several ways to specify what part of the file you
8871 want to print if you want to see other parts of the program:
8874 @item edit @var{location}
8875 Edit the source file specified by @code{location}. Editing starts at
8876 that @var{location}, e.g., at the specified source line of the
8877 specified file. @xref{Specify Location}, for all the possible forms
8878 of the @var{location} argument; here are the forms of the @code{edit}
8879 command most commonly used:
8882 @item edit @var{number}
8883 Edit the current source file with @var{number} as the active line number.
8885 @item edit @var{function}
8886 Edit the file containing @var{function} at the beginning of its definition.
8891 @subsection Choosing your Editor
8892 You can customize @value{GDBN} to use any editor you want
8894 The only restriction is that your editor (say @code{ex}), recognizes the
8895 following command-line syntax:
8897 ex +@var{number} file
8899 The optional numeric value +@var{number} specifies the number of the line in
8900 the file where to start editing.}.
8901 By default, it is @file{@value{EDITOR}}, but you can change this
8902 by setting the environment variable @code{EDITOR} before using
8903 @value{GDBN}. For example, to configure @value{GDBN} to use the
8904 @code{vi} editor, you could use these commands with the @code{sh} shell:
8910 or in the @code{csh} shell,
8912 setenv EDITOR /usr/bin/vi
8917 @section Searching Source Files
8918 @cindex searching source files
8920 There are two commands for searching through the current source file for a
8925 @kindex forward-search
8926 @kindex fo @r{(@code{forward-search})}
8927 @item forward-search @var{regexp}
8928 @itemx search @var{regexp}
8929 The command @samp{forward-search @var{regexp}} checks each line,
8930 starting with the one following the last line listed, for a match for
8931 @var{regexp}. It lists the line that is found. You can use the
8932 synonym @samp{search @var{regexp}} or abbreviate the command name as
8935 @kindex reverse-search
8936 @item reverse-search @var{regexp}
8937 The command @samp{reverse-search @var{regexp}} checks each line, starting
8938 with the one before the last line listed and going backward, for a match
8939 for @var{regexp}. It lists the line that is found. You can abbreviate
8940 this command as @code{rev}.
8944 @section Specifying Source Directories
8947 @cindex directories for source files
8948 Executable programs sometimes do not record the directories of the source
8949 files from which they were compiled, just the names. Even when they do,
8950 the directories could be moved between the compilation and your debugging
8951 session. @value{GDBN} has a list of directories to search for source files;
8952 this is called the @dfn{source path}. Each time @value{GDBN} wants a source file,
8953 it tries all the directories in the list, in the order they are present
8954 in the list, until it finds a file with the desired name.
8956 For example, suppose an executable references the file
8957 @file{/usr/src/foo-1.0/lib/foo.c}, and our source path is
8958 @file{/mnt/cross}. The file is first looked up literally; if this
8959 fails, @file{/mnt/cross/usr/src/foo-1.0/lib/foo.c} is tried; if this
8960 fails, @file{/mnt/cross/foo.c} is opened; if this fails, an error
8961 message is printed. @value{GDBN} does not look up the parts of the
8962 source file name, such as @file{/mnt/cross/src/foo-1.0/lib/foo.c}.
8963 Likewise, the subdirectories of the source path are not searched: if
8964 the source path is @file{/mnt/cross}, and the binary refers to
8965 @file{foo.c}, @value{GDBN} would not find it under
8966 @file{/mnt/cross/usr/src/foo-1.0/lib}.
8968 Plain file names, relative file names with leading directories, file
8969 names containing dots, etc.@: are all treated as described above; for
8970 instance, if the source path is @file{/mnt/cross}, and the source file
8971 is recorded as @file{../lib/foo.c}, @value{GDBN} would first try
8972 @file{../lib/foo.c}, then @file{/mnt/cross/../lib/foo.c}, and after
8973 that---@file{/mnt/cross/foo.c}.
8975 Note that the executable search path is @emph{not} used to locate the
8978 Whenever you reset or rearrange the source path, @value{GDBN} clears out
8979 any information it has cached about where source files are found and where
8980 each line is in the file.
8984 When you start @value{GDBN}, its source path includes only @samp{cdir}
8985 and @samp{cwd}, in that order.
8986 To add other directories, use the @code{directory} command.
8988 The search path is used to find both program source files and @value{GDBN}
8989 script files (read using the @samp{-command} option and @samp{source} command).
8991 In addition to the source path, @value{GDBN} provides a set of commands
8992 that manage a list of source path substitution rules. A @dfn{substitution
8993 rule} specifies how to rewrite source directories stored in the program's
8994 debug information in case the sources were moved to a different
8995 directory between compilation and debugging. A rule is made of
8996 two strings, the first specifying what needs to be rewritten in
8997 the path, and the second specifying how it should be rewritten.
8998 In @ref{set substitute-path}, we name these two parts @var{from} and
8999 @var{to} respectively. @value{GDBN} does a simple string replacement
9000 of @var{from} with @var{to} at the start of the directory part of the
9001 source file name, and uses that result instead of the original file
9002 name to look up the sources.
9004 Using the previous example, suppose the @file{foo-1.0} tree has been
9005 moved from @file{/usr/src} to @file{/mnt/cross}, then you can tell
9006 @value{GDBN} to replace @file{/usr/src} in all source path names with
9007 @file{/mnt/cross}. The first lookup will then be
9008 @file{/mnt/cross/foo-1.0/lib/foo.c} in place of the original location
9009 of @file{/usr/src/foo-1.0/lib/foo.c}. To define a source path
9010 substitution rule, use the @code{set substitute-path} command
9011 (@pxref{set substitute-path}).
9013 To avoid unexpected substitution results, a rule is applied only if the
9014 @var{from} part of the directory name ends at a directory separator.
9015 For instance, a rule substituting @file{/usr/source} into
9016 @file{/mnt/cross} will be applied to @file{/usr/source/foo-1.0} but
9017 not to @file{/usr/sourceware/foo-2.0}. And because the substitution
9018 is applied only at the beginning of the directory name, this rule will
9019 not be applied to @file{/root/usr/source/baz.c} either.
9021 In many cases, you can achieve the same result using the @code{directory}
9022 command. However, @code{set substitute-path} can be more efficient in
9023 the case where the sources are organized in a complex tree with multiple
9024 subdirectories. With the @code{directory} command, you need to add each
9025 subdirectory of your project. If you moved the entire tree while
9026 preserving its internal organization, then @code{set substitute-path}
9027 allows you to direct the debugger to all the sources with one single
9030 @code{set substitute-path} is also more than just a shortcut command.
9031 The source path is only used if the file at the original location no
9032 longer exists. On the other hand, @code{set substitute-path} modifies
9033 the debugger behavior to look at the rewritten location instead. So, if
9034 for any reason a source file that is not relevant to your executable is
9035 located at the original location, a substitution rule is the only
9036 method available to point @value{GDBN} at the new location.
9038 @cindex @samp{--with-relocated-sources}
9039 @cindex default source path substitution
9040 You can configure a default source path substitution rule by
9041 configuring @value{GDBN} with the
9042 @samp{--with-relocated-sources=@var{dir}} option. The @var{dir}
9043 should be the name of a directory under @value{GDBN}'s configured
9044 prefix (set with @samp{--prefix} or @samp{--exec-prefix}), and
9045 directory names in debug information under @var{dir} will be adjusted
9046 automatically if the installed @value{GDBN} is moved to a new
9047 location. This is useful if @value{GDBN}, libraries or executables
9048 with debug information and corresponding source code are being moved
9052 @item directory @var{dirname} @dots{}
9053 @item dir @var{dirname} @dots{}
9054 Add directory @var{dirname} to the front of the source path. Several
9055 directory names may be given to this command, separated by @samp{:}
9056 (@samp{;} on MS-DOS and MS-Windows, where @samp{:} usually appears as
9057 part of absolute file names) or
9058 whitespace. You may specify a directory that is already in the source
9059 path; this moves it forward, so @value{GDBN} searches it sooner.
9063 @vindex $cdir@r{, convenience variable}
9064 @vindex $cwd@r{, convenience variable}
9065 @cindex compilation directory
9066 @cindex current directory
9067 @cindex working directory
9068 @cindex directory, current
9069 @cindex directory, compilation
9070 You can use the string @samp{$cdir} to refer to the compilation
9071 directory (if one is recorded), and @samp{$cwd} to refer to the current
9072 working directory. @samp{$cwd} is not the same as @samp{.}---the former
9073 tracks the current working directory as it changes during your @value{GDBN}
9074 session, while the latter is immediately expanded to the current
9075 directory at the time you add an entry to the source path.
9078 Reset the source path to its default value (@samp{$cdir:$cwd} on Unix systems). This requires confirmation.
9080 @c RET-repeat for @code{directory} is explicitly disabled, but since
9081 @c repeating it would be a no-op we do not say that. (thanks to RMS)
9083 @item set directories @var{path-list}
9084 @kindex set directories
9085 Set the source path to @var{path-list}.
9086 @samp{$cdir:$cwd} are added if missing.
9088 @item show directories
9089 @kindex show directories
9090 Print the source path: show which directories it contains.
9092 @anchor{set substitute-path}
9093 @item set substitute-path @var{from} @var{to}
9094 @kindex set substitute-path
9095 Define a source path substitution rule, and add it at the end of the
9096 current list of existing substitution rules. If a rule with the same
9097 @var{from} was already defined, then the old rule is also deleted.
9099 For example, if the file @file{/foo/bar/baz.c} was moved to
9100 @file{/mnt/cross/baz.c}, then the command
9103 (@value{GDBP}) set substitute-path /foo/bar /mnt/cross
9107 will tell @value{GDBN} to replace @samp{/foo/bar} with
9108 @samp{/mnt/cross}, which will allow @value{GDBN} to find the file
9109 @file{baz.c} even though it was moved.
9111 In the case when more than one substitution rule have been defined,
9112 the rules are evaluated one by one in the order where they have been
9113 defined. The first one matching, if any, is selected to perform
9116 For instance, if we had entered the following commands:
9119 (@value{GDBP}) set substitute-path /usr/src/include /mnt/include
9120 (@value{GDBP}) set substitute-path /usr/src /mnt/src
9124 @value{GDBN} would then rewrite @file{/usr/src/include/defs.h} into
9125 @file{/mnt/include/defs.h} by using the first rule. However, it would
9126 use the second rule to rewrite @file{/usr/src/lib/foo.c} into
9127 @file{/mnt/src/lib/foo.c}.
9130 @item unset substitute-path [path]
9131 @kindex unset substitute-path
9132 If a path is specified, search the current list of substitution rules
9133 for a rule that would rewrite that path. Delete that rule if found.
9134 A warning is emitted by the debugger if no rule could be found.
9136 If no path is specified, then all substitution rules are deleted.
9138 @item show substitute-path [path]
9139 @kindex show substitute-path
9140 If a path is specified, then print the source path substitution rule
9141 which would rewrite that path, if any.
9143 If no path is specified, then print all existing source path substitution
9148 If your source path is cluttered with directories that are no longer of
9149 interest, @value{GDBN} may sometimes cause confusion by finding the wrong
9150 versions of source. You can correct the situation as follows:
9154 Use @code{directory} with no argument to reset the source path to its default value.
9157 Use @code{directory} with suitable arguments to reinstall the
9158 directories you want in the source path. You can add all the
9159 directories in one command.
9163 @section Source and Machine Code
9164 @cindex source line and its code address
9166 You can use the command @code{info line} to map source lines to program
9167 addresses (and vice versa), and the command @code{disassemble} to display
9168 a range of addresses as machine instructions. You can use the command
9169 @code{set disassemble-next-line} to set whether to disassemble next
9170 source line when execution stops. When run under @sc{gnu} Emacs
9171 mode, the @code{info line} command causes the arrow to point to the
9172 line specified. Also, @code{info line} prints addresses in symbolic form as
9178 @itemx info line @var{location}
9179 Print the starting and ending addresses of the compiled code for
9180 source line @var{location}. You can specify source lines in any of
9181 the ways documented in @ref{Specify Location}. With no @var{location}
9182 information about the current source line is printed.
9185 For example, we can use @code{info line} to discover the location of
9186 the object code for the first line of function
9187 @code{m4_changequote}:
9190 (@value{GDBP}) info line m4_changequote
9191 Line 895 of "builtin.c" starts at pc 0x634c <m4_changequote> and \
9192 ends at 0x6350 <m4_changequote+4>.
9196 @cindex code address and its source line
9197 We can also inquire (using @code{*@var{addr}} as the form for
9198 @var{location}) what source line covers a particular address:
9200 (@value{GDBP}) info line *0x63ff
9201 Line 926 of "builtin.c" starts at pc 0x63e4 <m4_changequote+152> and \
9202 ends at 0x6404 <m4_changequote+184>.
9205 @cindex @code{$_} and @code{info line}
9206 @cindex @code{x} command, default address
9207 @kindex x@r{(examine), and} info line
9208 After @code{info line}, the default address for the @code{x} command
9209 is changed to the starting address of the line, so that @samp{x/i} is
9210 sufficient to begin examining the machine code (@pxref{Memory,
9211 ,Examining Memory}). Also, this address is saved as the value of the
9212 convenience variable @code{$_} (@pxref{Convenience Vars, ,Convenience
9215 @cindex info line, repeated calls
9216 After @code{info line}, using @code{info line} again without
9217 specifying a location will display information about the next source
9222 @cindex assembly instructions
9223 @cindex instructions, assembly
9224 @cindex machine instructions
9225 @cindex listing machine instructions
9227 @itemx disassemble /m
9228 @itemx disassemble /s
9229 @itemx disassemble /r
9230 This specialized command dumps a range of memory as machine
9231 instructions. It can also print mixed source+disassembly by specifying
9232 the @code{/m} or @code{/s} modifier and print the raw instructions in hex
9233 as well as in symbolic form by specifying the @code{/r} modifier.
9234 The default memory range is the function surrounding the
9235 program counter of the selected frame. A single argument to this
9236 command is a program counter value; @value{GDBN} dumps the function
9237 surrounding this value. When two arguments are given, they should
9238 be separated by a comma, possibly surrounded by whitespace. The
9239 arguments specify a range of addresses to dump, in one of two forms:
9242 @item @var{start},@var{end}
9243 the addresses from @var{start} (inclusive) to @var{end} (exclusive)
9244 @item @var{start},+@var{length}
9245 the addresses from @var{start} (inclusive) to
9246 @code{@var{start}+@var{length}} (exclusive).
9250 When 2 arguments are specified, the name of the function is also
9251 printed (since there could be several functions in the given range).
9253 The argument(s) can be any expression yielding a numeric value, such as
9254 @samp{0x32c4}, @samp{&main+10} or @samp{$pc - 8}.
9256 If the range of memory being disassembled contains current program counter,
9257 the instruction at that location is shown with a @code{=>} marker.
9260 The following example shows the disassembly of a range of addresses of
9261 HP PA-RISC 2.0 code:
9264 (@value{GDBP}) disas 0x32c4, 0x32e4
9265 Dump of assembler code from 0x32c4 to 0x32e4:
9266 0x32c4 <main+204>: addil 0,dp
9267 0x32c8 <main+208>: ldw 0x22c(sr0,r1),r26
9268 0x32cc <main+212>: ldil 0x3000,r31
9269 0x32d0 <main+216>: ble 0x3f8(sr4,r31)
9270 0x32d4 <main+220>: ldo 0(r31),rp
9271 0x32d8 <main+224>: addil -0x800,dp
9272 0x32dc <main+228>: ldo 0x588(r1),r26
9273 0x32e0 <main+232>: ldil 0x3000,r31
9274 End of assembler dump.
9277 Here is an example showing mixed source+assembly for Intel x86
9278 with @code{/m} or @code{/s}, when the program is stopped just after
9279 function prologue in a non-optimized function with no inline code.
9282 (@value{GDBP}) disas /m main
9283 Dump of assembler code for function main:
9285 0x08048330 <+0>: push %ebp
9286 0x08048331 <+1>: mov %esp,%ebp
9287 0x08048333 <+3>: sub $0x8,%esp
9288 0x08048336 <+6>: and $0xfffffff0,%esp
9289 0x08048339 <+9>: sub $0x10,%esp
9291 6 printf ("Hello.\n");
9292 => 0x0804833c <+12>: movl $0x8048440,(%esp)
9293 0x08048343 <+19>: call 0x8048284 <puts@@plt>
9297 0x08048348 <+24>: mov $0x0,%eax
9298 0x0804834d <+29>: leave
9299 0x0804834e <+30>: ret
9301 End of assembler dump.
9304 The @code{/m} option is deprecated as its output is not useful when
9305 there is either inlined code or re-ordered code.
9306 The @code{/s} option is the preferred choice.
9307 Here is an example for AMD x86-64 showing the difference between
9308 @code{/m} output and @code{/s} output.
9309 This example has one inline function defined in a header file,
9310 and the code is compiled with @samp{-O2} optimization.
9311 Note how the @code{/m} output is missing the disassembly of
9312 several instructions that are present in the @code{/s} output.
9342 (@value{GDBP}) disas /m main
9343 Dump of assembler code for function main:
9347 0x0000000000400400 <+0>: mov 0x200c2e(%rip),%eax # 0x601034 <y>
9348 0x0000000000400417 <+23>: mov %eax,0x200c13(%rip) # 0x601030 <x>
9352 0x000000000040041d <+29>: xor %eax,%eax
9353 0x000000000040041f <+31>: retq
9354 0x0000000000400420 <+32>: add %eax,%eax
9355 0x0000000000400422 <+34>: jmp 0x400417 <main+23>
9357 End of assembler dump.
9358 (@value{GDBP}) disas /s main
9359 Dump of assembler code for function main:
9363 0x0000000000400400 <+0>: mov 0x200c2e(%rip),%eax # 0x601034 <y>
9367 0x0000000000400406 <+6>: test %eax,%eax
9368 0x0000000000400408 <+8>: js 0x400420 <main+32>
9373 0x000000000040040a <+10>: lea 0xa(%rax),%edx
9374 0x000000000040040d <+13>: test %eax,%eax
9375 0x000000000040040f <+15>: mov $0x1,%eax
9376 0x0000000000400414 <+20>: cmovne %edx,%eax
9380 0x0000000000400417 <+23>: mov %eax,0x200c13(%rip) # 0x601030 <x>
9384 0x000000000040041d <+29>: xor %eax,%eax
9385 0x000000000040041f <+31>: retq
9389 0x0000000000400420 <+32>: add %eax,%eax
9390 0x0000000000400422 <+34>: jmp 0x400417 <main+23>
9391 End of assembler dump.
9394 Here is another example showing raw instructions in hex for AMD x86-64,
9397 (gdb) disas /r 0x400281,+10
9398 Dump of assembler code from 0x400281 to 0x40028b:
9399 0x0000000000400281: 38 36 cmp %dh,(%rsi)
9400 0x0000000000400283: 2d 36 34 2e 73 sub $0x732e3436,%eax
9401 0x0000000000400288: 6f outsl %ds:(%rsi),(%dx)
9402 0x0000000000400289: 2e 32 00 xor %cs:(%rax),%al
9403 End of assembler dump.
9406 Addresses cannot be specified as a location (@pxref{Specify Location}).
9407 So, for example, if you want to disassemble function @code{bar}
9408 in file @file{foo.c}, you must type @samp{disassemble 'foo.c'::bar}
9409 and not @samp{disassemble foo.c:bar}.
9411 Some architectures have more than one commonly-used set of instruction
9412 mnemonics or other syntax.
9414 For programs that were dynamically linked and use shared libraries,
9415 instructions that call functions or branch to locations in the shared
9416 libraries might show a seemingly bogus location---it's actually a
9417 location of the relocation table. On some architectures, @value{GDBN}
9418 might be able to resolve these to actual function names.
9421 @kindex set disassembler-options
9422 @cindex disassembler options
9423 @item set disassembler-options @var{option1}[,@var{option2}@dots{}]
9424 This command controls the passing of target specific information to
9425 the disassembler. For a list of valid options, please refer to the
9426 @code{-M}/@code{--disassembler-options} section of the @samp{objdump}
9427 manual and/or the output of @kbd{objdump --help}
9428 (@pxref{objdump,,objdump,binutils,The GNU Binary Utilities}).
9429 The default value is the empty string.
9431 If it is necessary to specify more than one disassembler option, then
9432 multiple options can be placed together into a comma separated list.
9433 Currently this command is only supported on targets ARM, MIPS, PowerPC
9436 @kindex show disassembler-options
9437 @item show disassembler-options
9438 Show the current setting of the disassembler options.
9442 @kindex set disassembly-flavor
9443 @cindex Intel disassembly flavor
9444 @cindex AT&T disassembly flavor
9445 @item set disassembly-flavor @var{instruction-set}
9446 Select the instruction set to use when disassembling the
9447 program via the @code{disassemble} or @code{x/i} commands.
9449 Currently this command is only defined for the Intel x86 family. You
9450 can set @var{instruction-set} to either @code{intel} or @code{att}.
9451 The default is @code{att}, the AT&T flavor used by default by Unix
9452 assemblers for x86-based targets.
9454 @kindex show disassembly-flavor
9455 @item show disassembly-flavor
9456 Show the current setting of the disassembly flavor.
9460 @kindex set disassemble-next-line
9461 @kindex show disassemble-next-line
9462 @item set disassemble-next-line
9463 @itemx show disassemble-next-line
9464 Control whether or not @value{GDBN} will disassemble the next source
9465 line or instruction when execution stops. If ON, @value{GDBN} will
9466 display disassembly of the next source line when execution of the
9467 program being debugged stops. This is @emph{in addition} to
9468 displaying the source line itself, which @value{GDBN} always does if
9469 possible. If the next source line cannot be displayed for some reason
9470 (e.g., if @value{GDBN} cannot find the source file, or there's no line
9471 info in the debug info), @value{GDBN} will display disassembly of the
9472 next @emph{instruction} instead of showing the next source line. If
9473 AUTO, @value{GDBN} will display disassembly of next instruction only
9474 if the source line cannot be displayed. This setting causes
9475 @value{GDBN} to display some feedback when you step through a function
9476 with no line info or whose source file is unavailable. The default is
9477 OFF, which means never display the disassembly of the next line or
9483 @chapter Examining Data
9485 @cindex printing data
9486 @cindex examining data
9489 The usual way to examine data in your program is with the @code{print}
9490 command (abbreviated @code{p}), or its synonym @code{inspect}. It
9491 evaluates and prints the value of an expression of the language your
9492 program is written in (@pxref{Languages, ,Using @value{GDBN} with
9493 Different Languages}). It may also print the expression using a
9494 Python-based pretty-printer (@pxref{Pretty Printing}).
9497 @item print [[@var{options}] --] @var{expr}
9498 @itemx print [[@var{options}] --] /@var{f} @var{expr}
9499 @var{expr} is an expression (in the source language). By default the
9500 value of @var{expr} is printed in a format appropriate to its data type;
9501 you can choose a different format by specifying @samp{/@var{f}}, where
9502 @var{f} is a letter specifying the format; see @ref{Output Formats,,Output
9505 @anchor{print options}
9506 The @code{print} command supports a number of options that allow
9507 overriding relevant global print settings as set by @code{set print}
9511 @item -address [@code{on}|@code{off}]
9512 Set printing of addresses.
9513 Related setting: @ref{set print address}.
9515 @item -array [@code{on}|@code{off}]
9516 Pretty formatting of arrays.
9517 Related setting: @ref{set print array}.
9519 @item -array-indexes [@code{on}|@code{off}]
9520 Set printing of array indexes.
9521 Related setting: @ref{set print array-indexes}.
9523 @item -elements @var{number-of-elements}|@code{unlimited}
9524 Set limit on string chars or array elements to print. The value
9525 @code{unlimited} causes there to be no limit. Related setting:
9526 @ref{set print elements}.
9528 @item -max-depth @var{depth}|@code{unlimited}
9529 Set the threshold after which nested structures are replaced with
9530 ellipsis. Related setting: @ref{set print max-depth}.
9532 @item -null-stop [@code{on}|@code{off}]
9533 Set printing of char arrays to stop at first null char. Related
9534 setting: @ref{set print null-stop}.
9536 @item -object [@code{on}|@code{off}]
9537 Set printing C@t{++} virtual function tables. Related setting:
9538 @ref{set print object}.
9540 @item -pretty [@code{on}|@code{off}]
9541 Set pretty formatting of structures. Related setting: @ref{set print
9544 @item -repeats @var{number-of-repeats}|@code{unlimited}
9545 Set threshold for repeated print elements. @code{unlimited} causes
9546 all elements to be individually printed. Related setting: @ref{set
9549 @item -static-members [@code{on}|@code{off}]
9550 Set printing C@t{++} static members. Related setting: @ref{set print
9553 @item -symbol [@code{on}|@code{off}]
9554 Set printing of symbol names when printing pointers. Related setting:
9555 @ref{set print symbol}.
9557 @item -union [@code{on}|@code{off}]
9558 Set printing of unions interior to structures. Related setting:
9559 @ref{set print union}.
9561 @item -vtbl [@code{on}|@code{off}]
9562 Set printing of C++ virtual function tables. Related setting:
9563 @ref{set print vtbl}.
9566 Because the @code{print} command accepts arbitrary expressions which
9567 may look like options (including abbreviations), if you specify any
9568 command option, then you must use a double dash (@code{--}) to mark
9569 the end of option processing.
9571 For example, this prints the value of the @code{-r} expression:
9574 (@value{GDBP}) print -r
9577 While this repeats the last value in the value history (see below)
9578 with the @code{-raw} option in effect:
9581 (@value{GDBP}) print -r --
9584 Here is an example including both on option and an expression:
9588 (@value{GDBP}) print -pretty -- *myptr
9600 @item print [@var{options}]
9601 @itemx print [@var{options}] /@var{f}
9602 @cindex reprint the last value
9603 If you omit @var{expr}, @value{GDBN} displays the last value again (from the
9604 @dfn{value history}; @pxref{Value History, ,Value History}). This allows you to
9605 conveniently inspect the same value in an alternative format.
9608 A more low-level way of examining data is with the @code{x} command.
9609 It examines data in memory at a specified address and prints it in a
9610 specified format. @xref{Memory, ,Examining Memory}.
9612 If you are interested in information about types, or about how the
9613 fields of a struct or a class are declared, use the @code{ptype @var{exp}}
9614 command rather than @code{print}. @xref{Symbols, ,Examining the Symbol
9617 @cindex exploring hierarchical data structures
9619 Another way of examining values of expressions and type information is
9620 through the Python extension command @code{explore} (available only if
9621 the @value{GDBN} build is configured with @code{--with-python}). It
9622 offers an interactive way to start at the highest level (or, the most
9623 abstract level) of the data type of an expression (or, the data type
9624 itself) and explore all the way down to leaf scalar values/fields
9625 embedded in the higher level data types.
9628 @item explore @var{arg}
9629 @var{arg} is either an expression (in the source language), or a type
9630 visible in the current context of the program being debugged.
9633 The working of the @code{explore} command can be illustrated with an
9634 example. If a data type @code{struct ComplexStruct} is defined in your
9644 struct ComplexStruct
9646 struct SimpleStruct *ss_p;
9652 followed by variable declarations as
9655 struct SimpleStruct ss = @{ 10, 1.11 @};
9656 struct ComplexStruct cs = @{ &ss, @{ 0, 1, 2, 3, 4, 5, 6, 7, 8, 9 @} @};
9660 then, the value of the variable @code{cs} can be explored using the
9661 @code{explore} command as follows.
9665 The value of `cs' is a struct/class of type `struct ComplexStruct' with
9666 the following fields:
9668 ss_p = <Enter 0 to explore this field of type `struct SimpleStruct *'>
9669 arr = <Enter 1 to explore this field of type `int [10]'>
9671 Enter the field number of choice:
9675 Since the fields of @code{cs} are not scalar values, you are being
9676 prompted to chose the field you want to explore. Let's say you choose
9677 the field @code{ss_p} by entering @code{0}. Then, since this field is a
9678 pointer, you will be asked if it is pointing to a single value. From
9679 the declaration of @code{cs} above, it is indeed pointing to a single
9680 value, hence you enter @code{y}. If you enter @code{n}, then you will
9681 be asked if it were pointing to an array of values, in which case this
9682 field will be explored as if it were an array.
9685 `cs.ss_p' is a pointer to a value of type `struct SimpleStruct'
9686 Continue exploring it as a pointer to a single value [y/n]: y
9687 The value of `*(cs.ss_p)' is a struct/class of type `struct
9688 SimpleStruct' with the following fields:
9690 i = 10 .. (Value of type `int')
9691 d = 1.1100000000000001 .. (Value of type `double')
9693 Press enter to return to parent value:
9697 If the field @code{arr} of @code{cs} was chosen for exploration by
9698 entering @code{1} earlier, then since it is as array, you will be
9699 prompted to enter the index of the element in the array that you want
9703 `cs.arr' is an array of `int'.
9704 Enter the index of the element you want to explore in `cs.arr': 5
9706 `(cs.arr)[5]' is a scalar value of type `int'.
9710 Press enter to return to parent value:
9713 In general, at any stage of exploration, you can go deeper towards the
9714 leaf values by responding to the prompts appropriately, or hit the
9715 return key to return to the enclosing data structure (the @i{higher}
9716 level data structure).
9718 Similar to exploring values, you can use the @code{explore} command to
9719 explore types. Instead of specifying a value (which is typically a
9720 variable name or an expression valid in the current context of the
9721 program being debugged), you specify a type name. If you consider the
9722 same example as above, your can explore the type
9723 @code{struct ComplexStruct} by passing the argument
9724 @code{struct ComplexStruct} to the @code{explore} command.
9727 (gdb) explore struct ComplexStruct
9731 By responding to the prompts appropriately in the subsequent interactive
9732 session, you can explore the type @code{struct ComplexStruct} in a
9733 manner similar to how the value @code{cs} was explored in the above
9736 The @code{explore} command also has two sub-commands,
9737 @code{explore value} and @code{explore type}. The former sub-command is
9738 a way to explicitly specify that value exploration of the argument is
9739 being invoked, while the latter is a way to explicitly specify that type
9740 exploration of the argument is being invoked.
9743 @item explore value @var{expr}
9744 @cindex explore value
9745 This sub-command of @code{explore} explores the value of the
9746 expression @var{expr} (if @var{expr} is an expression valid in the
9747 current context of the program being debugged). The behavior of this
9748 command is identical to that of the behavior of the @code{explore}
9749 command being passed the argument @var{expr}.
9751 @item explore type @var{arg}
9752 @cindex explore type
9753 This sub-command of @code{explore} explores the type of @var{arg} (if
9754 @var{arg} is a type visible in the current context of program being
9755 debugged), or the type of the value/expression @var{arg} (if @var{arg}
9756 is an expression valid in the current context of the program being
9757 debugged). If @var{arg} is a type, then the behavior of this command is
9758 identical to that of the @code{explore} command being passed the
9759 argument @var{arg}. If @var{arg} is an expression, then the behavior of
9760 this command will be identical to that of the @code{explore} command
9761 being passed the type of @var{arg} as the argument.
9765 * Expressions:: Expressions
9766 * Ambiguous Expressions:: Ambiguous Expressions
9767 * Variables:: Program variables
9768 * Arrays:: Artificial arrays
9769 * Output Formats:: Output formats
9770 * Memory:: Examining memory
9771 * Auto Display:: Automatic display
9772 * Print Settings:: Print settings
9773 * Pretty Printing:: Python pretty printing
9774 * Value History:: Value history
9775 * Convenience Vars:: Convenience variables
9776 * Convenience Funs:: Convenience functions
9777 * Registers:: Registers
9778 * Floating Point Hardware:: Floating point hardware
9779 * Vector Unit:: Vector Unit
9780 * OS Information:: Auxiliary data provided by operating system
9781 * Memory Region Attributes:: Memory region attributes
9782 * Dump/Restore Files:: Copy between memory and a file
9783 * Core File Generation:: Cause a program dump its core
9784 * Character Sets:: Debugging programs that use a different
9785 character set than GDB does
9786 * Caching Target Data:: Data caching for targets
9787 * Searching Memory:: Searching memory for a sequence of bytes
9788 * Value Sizes:: Managing memory allocated for values
9792 @section Expressions
9795 @code{print} and many other @value{GDBN} commands accept an expression and
9796 compute its value. Any kind of constant, variable or operator defined
9797 by the programming language you are using is valid in an expression in
9798 @value{GDBN}. This includes conditional expressions, function calls,
9799 casts, and string constants. It also includes preprocessor macros, if
9800 you compiled your program to include this information; see
9803 @cindex arrays in expressions
9804 @value{GDBN} supports array constants in expressions input by
9805 the user. The syntax is @{@var{element}, @var{element}@dots{}@}. For example,
9806 you can use the command @code{print @{1, 2, 3@}} to create an array
9807 of three integers. If you pass an array to a function or assign it
9808 to a program variable, @value{GDBN} copies the array to memory that
9809 is @code{malloc}ed in the target program.
9811 Because C is so widespread, most of the expressions shown in examples in
9812 this manual are in C. @xref{Languages, , Using @value{GDBN} with Different
9813 Languages}, for information on how to use expressions in other
9816 In this section, we discuss operators that you can use in @value{GDBN}
9817 expressions regardless of your programming language.
9819 @cindex casts, in expressions
9820 Casts are supported in all languages, not just in C, because it is so
9821 useful to cast a number into a pointer in order to examine a structure
9822 at that address in memory.
9823 @c FIXME: casts supported---Mod2 true?
9825 @value{GDBN} supports these operators, in addition to those common
9826 to programming languages:
9830 @samp{@@} is a binary operator for treating parts of memory as arrays.
9831 @xref{Arrays, ,Artificial Arrays}, for more information.
9834 @samp{::} allows you to specify a variable in terms of the file or
9835 function where it is defined. @xref{Variables, ,Program Variables}.
9837 @cindex @{@var{type}@}
9838 @cindex type casting memory
9839 @cindex memory, viewing as typed object
9840 @cindex casts, to view memory
9841 @item @{@var{type}@} @var{addr}
9842 Refers to an object of type @var{type} stored at address @var{addr} in
9843 memory. The address @var{addr} may be any expression whose value is
9844 an integer or pointer (but parentheses are required around binary
9845 operators, just as in a cast). This construct is allowed regardless
9846 of what kind of data is normally supposed to reside at @var{addr}.
9849 @node Ambiguous Expressions
9850 @section Ambiguous Expressions
9851 @cindex ambiguous expressions
9853 Expressions can sometimes contain some ambiguous elements. For instance,
9854 some programming languages (notably Ada, C@t{++} and Objective-C) permit
9855 a single function name to be defined several times, for application in
9856 different contexts. This is called @dfn{overloading}. Another example
9857 involving Ada is generics. A @dfn{generic package} is similar to C@t{++}
9858 templates and is typically instantiated several times, resulting in
9859 the same function name being defined in different contexts.
9861 In some cases and depending on the language, it is possible to adjust
9862 the expression to remove the ambiguity. For instance in C@t{++}, you
9863 can specify the signature of the function you want to break on, as in
9864 @kbd{break @var{function}(@var{types})}. In Ada, using the fully
9865 qualified name of your function often makes the expression unambiguous
9868 When an ambiguity that needs to be resolved is detected, the debugger
9869 has the capability to display a menu of numbered choices for each
9870 possibility, and then waits for the selection with the prompt @samp{>}.
9871 The first option is always @samp{[0] cancel}, and typing @kbd{0 @key{RET}}
9872 aborts the current command. If the command in which the expression was
9873 used allows more than one choice to be selected, the next option in the
9874 menu is @samp{[1] all}, and typing @kbd{1 @key{RET}} selects all possible
9877 For example, the following session excerpt shows an attempt to set a
9878 breakpoint at the overloaded symbol @code{String::after}.
9879 We choose three particular definitions of that function name:
9881 @c FIXME! This is likely to change to show arg type lists, at least
9884 (@value{GDBP}) b String::after
9887 [2] file:String.cc; line number:867
9888 [3] file:String.cc; line number:860
9889 [4] file:String.cc; line number:875
9890 [5] file:String.cc; line number:853
9891 [6] file:String.cc; line number:846
9892 [7] file:String.cc; line number:735
9894 Breakpoint 1 at 0xb26c: file String.cc, line 867.
9895 Breakpoint 2 at 0xb344: file String.cc, line 875.
9896 Breakpoint 3 at 0xafcc: file String.cc, line 846.
9897 Multiple breakpoints were set.
9898 Use the "delete" command to delete unwanted
9905 @kindex set multiple-symbols
9906 @item set multiple-symbols @var{mode}
9907 @cindex multiple-symbols menu
9909 This option allows you to adjust the debugger behavior when an expression
9912 By default, @var{mode} is set to @code{all}. If the command with which
9913 the expression is used allows more than one choice, then @value{GDBN}
9914 automatically selects all possible choices. For instance, inserting
9915 a breakpoint on a function using an ambiguous name results in a breakpoint
9916 inserted on each possible match. However, if a unique choice must be made,
9917 then @value{GDBN} uses the menu to help you disambiguate the expression.
9918 For instance, printing the address of an overloaded function will result
9919 in the use of the menu.
9921 When @var{mode} is set to @code{ask}, the debugger always uses the menu
9922 when an ambiguity is detected.
9924 Finally, when @var{mode} is set to @code{cancel}, the debugger reports
9925 an error due to the ambiguity and the command is aborted.
9927 @kindex show multiple-symbols
9928 @item show multiple-symbols
9929 Show the current value of the @code{multiple-symbols} setting.
9933 @section Program Variables
9935 The most common kind of expression to use is the name of a variable
9938 Variables in expressions are understood in the selected stack frame
9939 (@pxref{Selection, ,Selecting a Frame}); they must be either:
9943 global (or file-static)
9950 visible according to the scope rules of the
9951 programming language from the point of execution in that frame
9954 @noindent This means that in the function
9969 you can examine and use the variable @code{a} whenever your program is
9970 executing within the function @code{foo}, but you can only use or
9971 examine the variable @code{b} while your program is executing inside
9972 the block where @code{b} is declared.
9974 @cindex variable name conflict
9975 There is an exception: you can refer to a variable or function whose
9976 scope is a single source file even if the current execution point is not
9977 in this file. But it is possible to have more than one such variable or
9978 function with the same name (in different source files). If that
9979 happens, referring to that name has unpredictable effects. If you wish,
9980 you can specify a static variable in a particular function or file by
9981 using the colon-colon (@code{::}) notation:
9983 @cindex colon-colon, context for variables/functions
9985 @c info cannot cope with a :: index entry, but why deprive hard copy readers?
9986 @cindex @code{::}, context for variables/functions
9989 @var{file}::@var{variable}
9990 @var{function}::@var{variable}
9994 Here @var{file} or @var{function} is the name of the context for the
9995 static @var{variable}. In the case of file names, you can use quotes to
9996 make sure @value{GDBN} parses the file name as a single word---for example,
9997 to print a global value of @code{x} defined in @file{f2.c}:
10000 (@value{GDBP}) p 'f2.c'::x
10003 The @code{::} notation is normally used for referring to
10004 static variables, since you typically disambiguate uses of local variables
10005 in functions by selecting the appropriate frame and using the
10006 simple name of the variable. However, you may also use this notation
10007 to refer to local variables in frames enclosing the selected frame:
10016 process (a); /* Stop here */
10027 For example, if there is a breakpoint at the commented line,
10028 here is what you might see
10029 when the program stops after executing the call @code{bar(0)}:
10034 (@value{GDBP}) p bar::a
10036 (@value{GDBP}) up 2
10037 #2 0x080483d0 in foo (a=5) at foobar.c:12
10040 (@value{GDBP}) p bar::a
10044 @cindex C@t{++} scope resolution
10045 These uses of @samp{::} are very rarely in conflict with the very
10046 similar use of the same notation in C@t{++}. When they are in
10047 conflict, the C@t{++} meaning takes precedence; however, this can be
10048 overridden by quoting the file or function name with single quotes.
10050 For example, suppose the program is stopped in a method of a class
10051 that has a field named @code{includefile}, and there is also an
10052 include file named @file{includefile} that defines a variable,
10053 @code{some_global}.
10056 (@value{GDBP}) p includefile
10058 (@value{GDBP}) p includefile::some_global
10059 A syntax error in expression, near `'.
10060 (@value{GDBP}) p 'includefile'::some_global
10064 @cindex wrong values
10065 @cindex variable values, wrong
10066 @cindex function entry/exit, wrong values of variables
10067 @cindex optimized code, wrong values of variables
10069 @emph{Warning:} Occasionally, a local variable may appear to have the
10070 wrong value at certain points in a function---just after entry to a new
10071 scope, and just before exit.
10073 You may see this problem when you are stepping by machine instructions.
10074 This is because, on most machines, it takes more than one instruction to
10075 set up a stack frame (including local variable definitions); if you are
10076 stepping by machine instructions, variables may appear to have the wrong
10077 values until the stack frame is completely built. On exit, it usually
10078 also takes more than one machine instruction to destroy a stack frame;
10079 after you begin stepping through that group of instructions, local
10080 variable definitions may be gone.
10082 This may also happen when the compiler does significant optimizations.
10083 To be sure of always seeing accurate values, turn off all optimization
10086 @cindex ``No symbol "foo" in current context''
10087 Another possible effect of compiler optimizations is to optimize
10088 unused variables out of existence, or assign variables to registers (as
10089 opposed to memory addresses). Depending on the support for such cases
10090 offered by the debug info format used by the compiler, @value{GDBN}
10091 might not be able to display values for such local variables. If that
10092 happens, @value{GDBN} will print a message like this:
10095 No symbol "foo" in current context.
10098 To solve such problems, either recompile without optimizations, or use a
10099 different debug info format, if the compiler supports several such
10100 formats. @xref{Compilation}, for more information on choosing compiler
10101 options. @xref{C, ,C and C@t{++}}, for more information about debug
10102 info formats that are best suited to C@t{++} programs.
10104 If you ask to print an object whose contents are unknown to
10105 @value{GDBN}, e.g., because its data type is not completely specified
10106 by the debug information, @value{GDBN} will say @samp{<incomplete
10107 type>}. @xref{Symbols, incomplete type}, for more about this.
10109 @cindex no debug info variables
10110 If you try to examine or use the value of a (global) variable for
10111 which @value{GDBN} has no type information, e.g., because the program
10112 includes no debug information, @value{GDBN} displays an error message.
10113 @xref{Symbols, unknown type}, for more about unknown types. If you
10114 cast the variable to its declared type, @value{GDBN} gets the
10115 variable's value using the cast-to type as the variable's type. For
10116 example, in a C program:
10119 (@value{GDBP}) p var
10120 'var' has unknown type; cast it to its declared type
10121 (@value{GDBP}) p (float) var
10125 If you append @kbd{@@entry} string to a function parameter name you get its
10126 value at the time the function got called. If the value is not available an
10127 error message is printed. Entry values are available only with some compilers.
10128 Entry values are normally also printed at the function parameter list according
10129 to @ref{set print entry-values}.
10132 Breakpoint 1, d (i=30) at gdb.base/entry-value.c:29
10138 (gdb) print i@@entry
10142 Strings are identified as arrays of @code{char} values without specified
10143 signedness. Arrays of either @code{signed char} or @code{unsigned char} get
10144 printed as arrays of 1 byte sized integers. @code{-fsigned-char} or
10145 @code{-funsigned-char} @value{NGCC} options have no effect as @value{GDBN}
10146 defines literal string type @code{"char"} as @code{char} without a sign.
10151 signed char var1[] = "A";
10154 You get during debugging
10159 $2 = @{65 'A', 0 '\0'@}
10163 @section Artificial Arrays
10165 @cindex artificial array
10167 @kindex @@@r{, referencing memory as an array}
10168 It is often useful to print out several successive objects of the
10169 same type in memory; a section of an array, or an array of
10170 dynamically determined size for which only a pointer exists in the
10173 You can do this by referring to a contiguous span of memory as an
10174 @dfn{artificial array}, using the binary operator @samp{@@}. The left
10175 operand of @samp{@@} should be the first element of the desired array
10176 and be an individual object. The right operand should be the desired length
10177 of the array. The result is an array value whose elements are all of
10178 the type of the left argument. The first element is actually the left
10179 argument; the second element comes from bytes of memory immediately
10180 following those that hold the first element, and so on. Here is an
10181 example. If a program says
10184 int *array = (int *) malloc (len * sizeof (int));
10188 you can print the contents of @code{array} with
10194 The left operand of @samp{@@} must reside in memory. Array values made
10195 with @samp{@@} in this way behave just like other arrays in terms of
10196 subscripting, and are coerced to pointers when used in expressions.
10197 Artificial arrays most often appear in expressions via the value history
10198 (@pxref{Value History, ,Value History}), after printing one out.
10200 Another way to create an artificial array is to use a cast.
10201 This re-interprets a value as if it were an array.
10202 The value need not be in memory:
10204 (@value{GDBP}) p/x (short[2])0x12345678
10205 $1 = @{0x1234, 0x5678@}
10208 As a convenience, if you leave the array length out (as in
10209 @samp{(@var{type}[])@var{value}}) @value{GDBN} calculates the size to fill
10210 the value (as @samp{sizeof(@var{value})/sizeof(@var{type})}:
10212 (@value{GDBP}) p/x (short[])0x12345678
10213 $2 = @{0x1234, 0x5678@}
10216 Sometimes the artificial array mechanism is not quite enough; in
10217 moderately complex data structures, the elements of interest may not
10218 actually be adjacent---for example, if you are interested in the values
10219 of pointers in an array. One useful work-around in this situation is
10220 to use a convenience variable (@pxref{Convenience Vars, ,Convenience
10221 Variables}) as a counter in an expression that prints the first
10222 interesting value, and then repeat that expression via @key{RET}. For
10223 instance, suppose you have an array @code{dtab} of pointers to
10224 structures, and you are interested in the values of a field @code{fv}
10225 in each structure. Here is an example of what you might type:
10235 @node Output Formats
10236 @section Output Formats
10238 @cindex formatted output
10239 @cindex output formats
10240 By default, @value{GDBN} prints a value according to its data type. Sometimes
10241 this is not what you want. For example, you might want to print a number
10242 in hex, or a pointer in decimal. Or you might want to view data in memory
10243 at a certain address as a character string or as an instruction. To do
10244 these things, specify an @dfn{output format} when you print a value.
10246 The simplest use of output formats is to say how to print a value
10247 already computed. This is done by starting the arguments of the
10248 @code{print} command with a slash and a format letter. The format
10249 letters supported are:
10253 Regard the bits of the value as an integer, and print the integer in
10257 Print as integer in signed decimal.
10260 Print as integer in unsigned decimal.
10263 Print as integer in octal.
10266 Print as integer in binary. The letter @samp{t} stands for ``two''.
10267 @footnote{@samp{b} cannot be used because these format letters are also
10268 used with the @code{x} command, where @samp{b} stands for ``byte'';
10269 see @ref{Memory,,Examining Memory}.}
10272 @cindex unknown address, locating
10273 @cindex locate address
10274 Print as an address, both absolute in hexadecimal and as an offset from
10275 the nearest preceding symbol. You can use this format used to discover
10276 where (in what function) an unknown address is located:
10279 (@value{GDBP}) p/a 0x54320
10280 $3 = 0x54320 <_initialize_vx+396>
10284 The command @code{info symbol 0x54320} yields similar results.
10285 @xref{Symbols, info symbol}.
10288 Regard as an integer and print it as a character constant. This
10289 prints both the numerical value and its character representation. The
10290 character representation is replaced with the octal escape @samp{\nnn}
10291 for characters outside the 7-bit @sc{ascii} range.
10293 Without this format, @value{GDBN} displays @code{char},
10294 @w{@code{unsigned char}}, and @w{@code{signed char}} data as character
10295 constants. Single-byte members of vectors are displayed as integer
10299 Regard the bits of the value as a floating point number and print
10300 using typical floating point syntax.
10303 @cindex printing strings
10304 @cindex printing byte arrays
10305 Regard as a string, if possible. With this format, pointers to single-byte
10306 data are displayed as null-terminated strings and arrays of single-byte data
10307 are displayed as fixed-length strings. Other values are displayed in their
10310 Without this format, @value{GDBN} displays pointers to and arrays of
10311 @code{char}, @w{@code{unsigned char}}, and @w{@code{signed char}} as
10312 strings. Single-byte members of a vector are displayed as an integer
10316 Like @samp{x} formatting, the value is treated as an integer and
10317 printed as hexadecimal, but leading zeros are printed to pad the value
10318 to the size of the integer type.
10321 @cindex raw printing
10322 Print using the @samp{raw} formatting. By default, @value{GDBN} will
10323 use a Python-based pretty-printer, if one is available (@pxref{Pretty
10324 Printing}). This typically results in a higher-level display of the
10325 value's contents. The @samp{r} format bypasses any Python
10326 pretty-printer which might exist.
10329 For example, to print the program counter in hex (@pxref{Registers}), type
10336 Note that no space is required before the slash; this is because command
10337 names in @value{GDBN} cannot contain a slash.
10339 To reprint the last value in the value history with a different format,
10340 you can use the @code{print} command with just a format and no
10341 expression. For example, @samp{p/x} reprints the last value in hex.
10344 @section Examining Memory
10346 You can use the command @code{x} (for ``examine'') to examine memory in
10347 any of several formats, independently of your program's data types.
10349 @cindex examining memory
10351 @kindex x @r{(examine memory)}
10352 @item x/@var{nfu} @var{addr}
10353 @itemx x @var{addr}
10355 Use the @code{x} command to examine memory.
10358 @var{n}, @var{f}, and @var{u} are all optional parameters that specify how
10359 much memory to display and how to format it; @var{addr} is an
10360 expression giving the address where you want to start displaying memory.
10361 If you use defaults for @var{nfu}, you need not type the slash @samp{/}.
10362 Several commands set convenient defaults for @var{addr}.
10365 @item @var{n}, the repeat count
10366 The repeat count is a decimal integer; the default is 1. It specifies
10367 how much memory (counting by units @var{u}) to display. If a negative
10368 number is specified, memory is examined backward from @var{addr}.
10369 @c This really is **decimal**; unaffected by 'set radix' as of GDB
10372 @item @var{f}, the display format
10373 The display format is one of the formats used by @code{print}
10374 (@samp{x}, @samp{d}, @samp{u}, @samp{o}, @samp{t}, @samp{a}, @samp{c},
10375 @samp{f}, @samp{s}), and in addition @samp{i} (for machine instructions).
10376 The default is @samp{x} (hexadecimal) initially. The default changes
10377 each time you use either @code{x} or @code{print}.
10379 @item @var{u}, the unit size
10380 The unit size is any of
10386 Halfwords (two bytes).
10388 Words (four bytes). This is the initial default.
10390 Giant words (eight bytes).
10393 Each time you specify a unit size with @code{x}, that size becomes the
10394 default unit the next time you use @code{x}. For the @samp{i} format,
10395 the unit size is ignored and is normally not written. For the @samp{s} format,
10396 the unit size defaults to @samp{b}, unless it is explicitly given.
10397 Use @kbd{x /hs} to display 16-bit char strings and @kbd{x /ws} to display
10398 32-bit strings. The next use of @kbd{x /s} will again display 8-bit strings.
10399 Note that the results depend on the programming language of the
10400 current compilation unit. If the language is C, the @samp{s}
10401 modifier will use the UTF-16 encoding while @samp{w} will use
10402 UTF-32. The encoding is set by the programming language and cannot
10405 @item @var{addr}, starting display address
10406 @var{addr} is the address where you want @value{GDBN} to begin displaying
10407 memory. The expression need not have a pointer value (though it may);
10408 it is always interpreted as an integer address of a byte of memory.
10409 @xref{Expressions, ,Expressions}, for more information on expressions. The default for
10410 @var{addr} is usually just after the last address examined---but several
10411 other commands also set the default address: @code{info breakpoints} (to
10412 the address of the last breakpoint listed), @code{info line} (to the
10413 starting address of a line), and @code{print} (if you use it to display
10414 a value from memory).
10417 For example, @samp{x/3uh 0x54320} is a request to display three halfwords
10418 (@code{h}) of memory, formatted as unsigned decimal integers (@samp{u}),
10419 starting at address @code{0x54320}. @samp{x/4xw $sp} prints the four
10420 words (@samp{w}) of memory above the stack pointer (here, @samp{$sp};
10421 @pxref{Registers, ,Registers}) in hexadecimal (@samp{x}).
10423 You can also specify a negative repeat count to examine memory backward
10424 from the given address. For example, @samp{x/-3uh 0x54320} prints three
10425 halfwords (@code{h}) at @code{0x54314}, @code{0x54328}, and @code{0x5431c}.
10427 Since the letters indicating unit sizes are all distinct from the
10428 letters specifying output formats, you do not have to remember whether
10429 unit size or format comes first; either order works. The output
10430 specifications @samp{4xw} and @samp{4wx} mean exactly the same thing.
10431 (However, the count @var{n} must come first; @samp{wx4} does not work.)
10433 Even though the unit size @var{u} is ignored for the formats @samp{s}
10434 and @samp{i}, you might still want to use a count @var{n}; for example,
10435 @samp{3i} specifies that you want to see three machine instructions,
10436 including any operands. For convenience, especially when used with
10437 the @code{display} command, the @samp{i} format also prints branch delay
10438 slot instructions, if any, beyond the count specified, which immediately
10439 follow the last instruction that is within the count. The command
10440 @code{disassemble} gives an alternative way of inspecting machine
10441 instructions; see @ref{Machine Code,,Source and Machine Code}.
10443 If a negative repeat count is specified for the formats @samp{s} or @samp{i},
10444 the command displays null-terminated strings or instructions before the given
10445 address as many as the absolute value of the given number. For the @samp{i}
10446 format, we use line number information in the debug info to accurately locate
10447 instruction boundaries while disassembling backward. If line info is not
10448 available, the command stops examining memory with an error message.
10450 All the defaults for the arguments to @code{x} are designed to make it
10451 easy to continue scanning memory with minimal specifications each time
10452 you use @code{x}. For example, after you have inspected three machine
10453 instructions with @samp{x/3i @var{addr}}, you can inspect the next seven
10454 with just @samp{x/7}. If you use @key{RET} to repeat the @code{x} command,
10455 the repeat count @var{n} is used again; the other arguments default as
10456 for successive uses of @code{x}.
10458 When examining machine instructions, the instruction at current program
10459 counter is shown with a @code{=>} marker. For example:
10462 (@value{GDBP}) x/5i $pc-6
10463 0x804837f <main+11>: mov %esp,%ebp
10464 0x8048381 <main+13>: push %ecx
10465 0x8048382 <main+14>: sub $0x4,%esp
10466 => 0x8048385 <main+17>: movl $0x8048460,(%esp)
10467 0x804838c <main+24>: call 0x80482d4 <puts@@plt>
10470 @cindex @code{$_}, @code{$__}, and value history
10471 The addresses and contents printed by the @code{x} command are not saved
10472 in the value history because there is often too much of them and they
10473 would get in the way. Instead, @value{GDBN} makes these values available for
10474 subsequent use in expressions as values of the convenience variables
10475 @code{$_} and @code{$__}. After an @code{x} command, the last address
10476 examined is available for use in expressions in the convenience variable
10477 @code{$_}. The contents of that address, as examined, are available in
10478 the convenience variable @code{$__}.
10480 If the @code{x} command has a repeat count, the address and contents saved
10481 are from the last memory unit printed; this is not the same as the last
10482 address printed if several units were printed on the last line of output.
10484 @anchor{addressable memory unit}
10485 @cindex addressable memory unit
10486 Most targets have an addressable memory unit size of 8 bits. This means
10487 that to each memory address are associated 8 bits of data. Some
10488 targets, however, have other addressable memory unit sizes.
10489 Within @value{GDBN} and this document, the term
10490 @dfn{addressable memory unit} (or @dfn{memory unit} for short) is used
10491 when explicitly referring to a chunk of data of that size. The word
10492 @dfn{byte} is used to refer to a chunk of data of 8 bits, regardless of
10493 the addressable memory unit size of the target. For most systems,
10494 addressable memory unit is a synonym of byte.
10496 @cindex remote memory comparison
10497 @cindex target memory comparison
10498 @cindex verify remote memory image
10499 @cindex verify target memory image
10500 When you are debugging a program running on a remote target machine
10501 (@pxref{Remote Debugging}), you may wish to verify the program's image
10502 in the remote machine's memory against the executable file you
10503 downloaded to the target. Or, on any target, you may want to check
10504 whether the program has corrupted its own read-only sections. The
10505 @code{compare-sections} command is provided for such situations.
10508 @kindex compare-sections
10509 @item compare-sections @r{[}@var{section-name}@r{|}@code{-r}@r{]}
10510 Compare the data of a loadable section @var{section-name} in the
10511 executable file of the program being debugged with the same section in
10512 the target machine's memory, and report any mismatches. With no
10513 arguments, compares all loadable sections. With an argument of
10514 @code{-r}, compares all loadable read-only sections.
10516 Note: for remote targets, this command can be accelerated if the
10517 target supports computing the CRC checksum of a block of memory
10518 (@pxref{qCRC packet}).
10522 @section Automatic Display
10523 @cindex automatic display
10524 @cindex display of expressions
10526 If you find that you want to print the value of an expression frequently
10527 (to see how it changes), you might want to add it to the @dfn{automatic
10528 display list} so that @value{GDBN} prints its value each time your program stops.
10529 Each expression added to the list is given a number to identify it;
10530 to remove an expression from the list, you specify that number.
10531 The automatic display looks like this:
10535 3: bar[5] = (struct hack *) 0x3804
10539 This display shows item numbers, expressions and their current values. As with
10540 displays you request manually using @code{x} or @code{print}, you can
10541 specify the output format you prefer; in fact, @code{display} decides
10542 whether to use @code{print} or @code{x} depending your format
10543 specification---it uses @code{x} if you specify either the @samp{i}
10544 or @samp{s} format, or a unit size; otherwise it uses @code{print}.
10548 @item display @var{expr}
10549 Add the expression @var{expr} to the list of expressions to display
10550 each time your program stops. @xref{Expressions, ,Expressions}.
10552 @code{display} does not repeat if you press @key{RET} again after using it.
10554 @item display/@var{fmt} @var{expr}
10555 For @var{fmt} specifying only a display format and not a size or
10556 count, add the expression @var{expr} to the auto-display list but
10557 arrange to display it each time in the specified format @var{fmt}.
10558 @xref{Output Formats,,Output Formats}.
10560 @item display/@var{fmt} @var{addr}
10561 For @var{fmt} @samp{i} or @samp{s}, or including a unit-size or a
10562 number of units, add the expression @var{addr} as a memory address to
10563 be examined each time your program stops. Examining means in effect
10564 doing @samp{x/@var{fmt} @var{addr}}. @xref{Memory, ,Examining Memory}.
10567 For example, @samp{display/i $pc} can be helpful, to see the machine
10568 instruction about to be executed each time execution stops (@samp{$pc}
10569 is a common name for the program counter; @pxref{Registers, ,Registers}).
10572 @kindex delete display
10574 @item undisplay @var{dnums}@dots{}
10575 @itemx delete display @var{dnums}@dots{}
10576 Remove items from the list of expressions to display. Specify the
10577 numbers of the displays that you want affected with the command
10578 argument @var{dnums}. It can be a single display number, one of the
10579 numbers shown in the first field of the @samp{info display} display;
10580 or it could be a range of display numbers, as in @code{2-4}.
10582 @code{undisplay} does not repeat if you press @key{RET} after using it.
10583 (Otherwise you would just get the error @samp{No display number @dots{}}.)
10585 @kindex disable display
10586 @item disable display @var{dnums}@dots{}
10587 Disable the display of item numbers @var{dnums}. A disabled display
10588 item is not printed automatically, but is not forgotten. It may be
10589 enabled again later. Specify the numbers of the displays that you
10590 want affected with the command argument @var{dnums}. It can be a
10591 single display number, one of the numbers shown in the first field of
10592 the @samp{info display} display; or it could be a range of display
10593 numbers, as in @code{2-4}.
10595 @kindex enable display
10596 @item enable display @var{dnums}@dots{}
10597 Enable display of item numbers @var{dnums}. It becomes effective once
10598 again in auto display of its expression, until you specify otherwise.
10599 Specify the numbers of the displays that you want affected with the
10600 command argument @var{dnums}. It can be a single display number, one
10601 of the numbers shown in the first field of the @samp{info display}
10602 display; or it could be a range of display numbers, as in @code{2-4}.
10605 Display the current values of the expressions on the list, just as is
10606 done when your program stops.
10608 @kindex info display
10610 Print the list of expressions previously set up to display
10611 automatically, each one with its item number, but without showing the
10612 values. This includes disabled expressions, which are marked as such.
10613 It also includes expressions which would not be displayed right now
10614 because they refer to automatic variables not currently available.
10617 @cindex display disabled out of scope
10618 If a display expression refers to local variables, then it does not make
10619 sense outside the lexical context for which it was set up. Such an
10620 expression is disabled when execution enters a context where one of its
10621 variables is not defined. For example, if you give the command
10622 @code{display last_char} while inside a function with an argument
10623 @code{last_char}, @value{GDBN} displays this argument while your program
10624 continues to stop inside that function. When it stops elsewhere---where
10625 there is no variable @code{last_char}---the display is disabled
10626 automatically. The next time your program stops where @code{last_char}
10627 is meaningful, you can enable the display expression once again.
10629 @node Print Settings
10630 @section Print Settings
10632 @cindex format options
10633 @cindex print settings
10634 @value{GDBN} provides the following ways to control how arrays, structures,
10635 and symbols are printed.
10638 These settings are useful for debugging programs in any language:
10642 @anchor{set print address}
10643 @item set print address
10644 @itemx set print address on
10645 @cindex print/don't print memory addresses
10646 @value{GDBN} prints memory addresses showing the location of stack
10647 traces, structure values, pointer values, breakpoints, and so forth,
10648 even when it also displays the contents of those addresses. The default
10649 is @code{on}. For example, this is what a stack frame display looks like with
10650 @code{set print address on}:
10655 #0 set_quotes (lq=0x34c78 "<<", rq=0x34c88 ">>")
10657 530 if (lquote != def_lquote)
10661 @item set print address off
10662 Do not print addresses when displaying their contents. For example,
10663 this is the same stack frame displayed with @code{set print address off}:
10667 (@value{GDBP}) set print addr off
10669 #0 set_quotes (lq="<<", rq=">>") at input.c:530
10670 530 if (lquote != def_lquote)
10674 You can use @samp{set print address off} to eliminate all machine
10675 dependent displays from the @value{GDBN} interface. For example, with
10676 @code{print address off}, you should get the same text for backtraces on
10677 all machines---whether or not they involve pointer arguments.
10680 @item show print address
10681 Show whether or not addresses are to be printed.
10684 When @value{GDBN} prints a symbolic address, it normally prints the
10685 closest earlier symbol plus an offset. If that symbol does not uniquely
10686 identify the address (for example, it is a name whose scope is a single
10687 source file), you may need to clarify. One way to do this is with
10688 @code{info line}, for example @samp{info line *0x4537}. Alternately,
10689 you can set @value{GDBN} to print the source file and line number when
10690 it prints a symbolic address:
10693 @item set print symbol-filename on
10694 @cindex source file and line of a symbol
10695 @cindex symbol, source file and line
10696 Tell @value{GDBN} to print the source file name and line number of a
10697 symbol in the symbolic form of an address.
10699 @item set print symbol-filename off
10700 Do not print source file name and line number of a symbol. This is the
10703 @item show print symbol-filename
10704 Show whether or not @value{GDBN} will print the source file name and
10705 line number of a symbol in the symbolic form of an address.
10708 Another situation where it is helpful to show symbol filenames and line
10709 numbers is when disassembling code; @value{GDBN} shows you the line
10710 number and source file that corresponds to each instruction.
10712 Also, you may wish to see the symbolic form only if the address being
10713 printed is reasonably close to the closest earlier symbol:
10716 @item set print max-symbolic-offset @var{max-offset}
10717 @itemx set print max-symbolic-offset unlimited
10718 @cindex maximum value for offset of closest symbol
10719 Tell @value{GDBN} to only display the symbolic form of an address if the
10720 offset between the closest earlier symbol and the address is less than
10721 @var{max-offset}. The default is @code{unlimited}, which tells @value{GDBN}
10722 to always print the symbolic form of an address if any symbol precedes
10723 it. Zero is equivalent to @code{unlimited}.
10725 @item show print max-symbolic-offset
10726 Ask how large the maximum offset is that @value{GDBN} prints in a
10730 @cindex wild pointer, interpreting
10731 @cindex pointer, finding referent
10732 If you have a pointer and you are not sure where it points, try
10733 @samp{set print symbol-filename on}. Then you can determine the name
10734 and source file location of the variable where it points, using
10735 @samp{p/a @var{pointer}}. This interprets the address in symbolic form.
10736 For example, here @value{GDBN} shows that a variable @code{ptt} points
10737 at another variable @code{t}, defined in @file{hi2.c}:
10740 (@value{GDBP}) set print symbol-filename on
10741 (@value{GDBP}) p/a ptt
10742 $4 = 0xe008 <t in hi2.c>
10746 @emph{Warning:} For pointers that point to a local variable, @samp{p/a}
10747 does not show the symbol name and filename of the referent, even with
10748 the appropriate @code{set print} options turned on.
10751 You can also enable @samp{/a}-like formatting all the time using
10752 @samp{set print symbol on}:
10754 @anchor{set print symbol}
10756 @item set print symbol on
10757 Tell @value{GDBN} to print the symbol corresponding to an address, if
10760 @item set print symbol off
10761 Tell @value{GDBN} not to print the symbol corresponding to an
10762 address. In this mode, @value{GDBN} will still print the symbol
10763 corresponding to pointers to functions. This is the default.
10765 @item show print symbol
10766 Show whether @value{GDBN} will display the symbol corresponding to an
10770 Other settings control how different kinds of objects are printed:
10773 @anchor{set print array}
10774 @item set print array
10775 @itemx set print array on
10776 @cindex pretty print arrays
10777 Pretty print arrays. This format is more convenient to read,
10778 but uses more space. The default is off.
10780 @item set print array off
10781 Return to compressed format for arrays.
10783 @item show print array
10784 Show whether compressed or pretty format is selected for displaying
10787 @cindex print array indexes
10788 @anchor{set print array-indexes}
10789 @item set print array-indexes
10790 @itemx set print array-indexes on
10791 Print the index of each element when displaying arrays. May be more
10792 convenient to locate a given element in the array or quickly find the
10793 index of a given element in that printed array. The default is off.
10795 @item set print array-indexes off
10796 Stop printing element indexes when displaying arrays.
10798 @item show print array-indexes
10799 Show whether the index of each element is printed when displaying
10802 @anchor{set print elements}
10803 @item set print elements @var{number-of-elements}
10804 @itemx set print elements unlimited
10805 @cindex number of array elements to print
10806 @cindex limit on number of printed array elements
10807 Set a limit on how many elements of an array @value{GDBN} will print.
10808 If @value{GDBN} is printing a large array, it stops printing after it has
10809 printed the number of elements set by the @code{set print elements} command.
10810 This limit also applies to the display of strings.
10811 When @value{GDBN} starts, this limit is set to 200.
10812 Setting @var{number-of-elements} to @code{unlimited} or zero means
10813 that the number of elements to print is unlimited.
10815 @item show print elements
10816 Display the number of elements of a large array that @value{GDBN} will print.
10817 If the number is 0, then the printing is unlimited.
10819 @anchor{set print frame-arguments}
10820 @item set print frame-arguments @var{value}
10821 @kindex set print frame-arguments
10822 @cindex printing frame argument values
10823 @cindex print all frame argument values
10824 @cindex print frame argument values for scalars only
10825 @cindex do not print frame arguments
10826 This command allows to control how the values of arguments are printed
10827 when the debugger prints a frame (@pxref{Frames}). The possible
10832 The values of all arguments are printed.
10835 Print the value of an argument only if it is a scalar. The value of more
10836 complex arguments such as arrays, structures, unions, etc, is replaced
10837 by @code{@dots{}}. This is the default. Here is an example where
10838 only scalar arguments are shown:
10841 #1 0x08048361 in call_me (i=3, s=@dots{}, ss=0xbf8d508c, u=@dots{}, e=green)
10846 None of the argument values are printed. Instead, the value of each argument
10847 is replaced by @code{@dots{}}. In this case, the example above now becomes:
10850 #1 0x08048361 in call_me (i=@dots{}, s=@dots{}, ss=@dots{}, u=@dots{}, e=@dots{})
10855 Only the presence of arguments is indicated by @code{@dots{}}.
10856 The @code{@dots{}} are not printed for function without any arguments.
10857 None of the argument names and values are printed.
10858 In this case, the example above now becomes:
10861 #1 0x08048361 in call_me (@dots{}) at frame-args.c:23
10866 By default, only scalar arguments are printed. This command can be used
10867 to configure the debugger to print the value of all arguments, regardless
10868 of their type. However, it is often advantageous to not print the value
10869 of more complex parameters. For instance, it reduces the amount of
10870 information printed in each frame, making the backtrace more readable.
10871 Also, it improves performance when displaying Ada frames, because
10872 the computation of large arguments can sometimes be CPU-intensive,
10873 especially in large applications. Setting @code{print frame-arguments}
10874 to @code{scalars} (the default), @code{none} or @code{presence} avoids
10875 this computation, thus speeding up the display of each Ada frame.
10877 @item show print frame-arguments
10878 Show how the value of arguments should be displayed when printing a frame.
10880 @anchor{set print raw-frame-arguments}
10881 @item set print raw-frame-arguments on
10882 Print frame arguments in raw, non pretty-printed, form.
10884 @item set print raw-frame-arguments off
10885 Print frame arguments in pretty-printed form, if there is a pretty-printer
10886 for the value (@pxref{Pretty Printing}),
10887 otherwise print the value in raw form.
10888 This is the default.
10890 @item show print raw-frame-arguments
10891 Show whether to print frame arguments in raw form.
10893 @anchor{set print entry-values}
10894 @item set print entry-values @var{value}
10895 @kindex set print entry-values
10896 Set printing of frame argument values at function entry. In some cases
10897 @value{GDBN} can determine the value of function argument which was passed by
10898 the function caller, even if the value was modified inside the called function
10899 and therefore is different. With optimized code, the current value could be
10900 unavailable, but the entry value may still be known.
10902 The default value is @code{default} (see below for its description). Older
10903 @value{GDBN} behaved as with the setting @code{no}. Compilers not supporting
10904 this feature will behave in the @code{default} setting the same way as with the
10907 This functionality is currently supported only by DWARF 2 debugging format and
10908 the compiler has to produce @samp{DW_TAG_call_site} tags. With
10909 @value{NGCC}, you need to specify @option{-O -g} during compilation, to get
10912 The @var{value} parameter can be one of the following:
10916 Print only actual parameter values, never print values from function entry
10920 #0 different (val=6)
10921 #0 lost (val=<optimized out>)
10923 #0 invalid (val=<optimized out>)
10927 Print only parameter values from function entry point. The actual parameter
10928 values are never printed.
10930 #0 equal (val@@entry=5)
10931 #0 different (val@@entry=5)
10932 #0 lost (val@@entry=5)
10933 #0 born (val@@entry=<optimized out>)
10934 #0 invalid (val@@entry=<optimized out>)
10938 Print only parameter values from function entry point. If value from function
10939 entry point is not known while the actual value is known, print the actual
10940 value for such parameter.
10942 #0 equal (val@@entry=5)
10943 #0 different (val@@entry=5)
10944 #0 lost (val@@entry=5)
10946 #0 invalid (val@@entry=<optimized out>)
10950 Print actual parameter values. If actual parameter value is not known while
10951 value from function entry point is known, print the entry point value for such
10955 #0 different (val=6)
10956 #0 lost (val@@entry=5)
10958 #0 invalid (val=<optimized out>)
10962 Always print both the actual parameter value and its value from function entry
10963 point, even if values of one or both are not available due to compiler
10966 #0 equal (val=5, val@@entry=5)
10967 #0 different (val=6, val@@entry=5)
10968 #0 lost (val=<optimized out>, val@@entry=5)
10969 #0 born (val=10, val@@entry=<optimized out>)
10970 #0 invalid (val=<optimized out>, val@@entry=<optimized out>)
10974 Print the actual parameter value if it is known and also its value from
10975 function entry point if it is known. If neither is known, print for the actual
10976 value @code{<optimized out>}. If not in MI mode (@pxref{GDB/MI}) and if both
10977 values are known and identical, print the shortened
10978 @code{param=param@@entry=VALUE} notation.
10980 #0 equal (val=val@@entry=5)
10981 #0 different (val=6, val@@entry=5)
10982 #0 lost (val@@entry=5)
10984 #0 invalid (val=<optimized out>)
10988 Always print the actual parameter value. Print also its value from function
10989 entry point, but only if it is known. If not in MI mode (@pxref{GDB/MI}) and
10990 if both values are known and identical, print the shortened
10991 @code{param=param@@entry=VALUE} notation.
10993 #0 equal (val=val@@entry=5)
10994 #0 different (val=6, val@@entry=5)
10995 #0 lost (val=<optimized out>, val@@entry=5)
10997 #0 invalid (val=<optimized out>)
11001 For analysis messages on possible failures of frame argument values at function
11002 entry resolution see @ref{set debug entry-values}.
11004 @item show print entry-values
11005 Show the method being used for printing of frame argument values at function
11008 @anchor{set print frame-info}
11009 @item set print frame-info @var{value}
11010 @kindex set print frame-info
11011 @cindex printing frame information
11012 @cindex frame information, printing
11013 This command allows to control the information printed when
11014 the debugger prints a frame. See @ref{Frames}, @ref{Backtrace},
11015 for a general explanation about frames and frame information.
11016 Note that some other settings (such as @code{set print frame-arguments}
11017 and @code{set print address}) are also influencing if and how some frame
11018 information is displayed. In particular, the frame program counter is never
11019 printed if @code{set print address} is off.
11021 The possible values for @code{set print frame-info} are:
11023 @item short-location
11024 Print the frame level, the program counter (if not at the
11025 beginning of the location source line), the function, the function
11028 Same as @code{short-location} but also print the source file and source line
11030 @item location-and-address
11031 Same as @code{location} but print the program counter even if located at the
11032 beginning of the location source line.
11034 Print the program counter (if not at the beginning of the location
11035 source line), the line number and the source line.
11036 @item source-and-location
11037 Print what @code{location} and @code{source-line} are printing.
11039 The information printed for a frame is decided automatically
11040 by the @value{GDBN} command that prints a frame.
11041 For example, @code{frame} prints the information printed by
11042 @code{source-and-location} while @code{stepi} will switch between
11043 @code{source-line} and @code{source-and-location} depending on the program
11045 The default value is @code{auto}.
11048 @anchor{set print repeats}
11049 @item set print repeats @var{number-of-repeats}
11050 @itemx set print repeats unlimited
11051 @cindex repeated array elements
11052 Set the threshold for suppressing display of repeated array
11053 elements. When the number of consecutive identical elements of an
11054 array exceeds the threshold, @value{GDBN} prints the string
11055 @code{"<repeats @var{n} times>"}, where @var{n} is the number of
11056 identical repetitions, instead of displaying the identical elements
11057 themselves. Setting the threshold to @code{unlimited} or zero will
11058 cause all elements to be individually printed. The default threshold
11061 @item show print repeats
11062 Display the current threshold for printing repeated identical
11065 @anchor{set print max-depth}
11066 @item set print max-depth @var{depth}
11067 @item set print max-depth unlimited
11068 @cindex printing nested structures
11069 Set the threshold after which nested structures are replaced with
11070 ellipsis, this can make visualising deeply nested structures easier.
11072 For example, given this C code
11075 typedef struct s1 @{ int a; @} s1;
11076 typedef struct s2 @{ s1 b; @} s2;
11077 typedef struct s3 @{ s2 c; @} s3;
11078 typedef struct s4 @{ s3 d; @} s4;
11080 s4 var = @{ @{ @{ @{ 3 @} @} @} @};
11083 The following table shows how different values of @var{depth} will
11084 effect how @code{var} is printed by @value{GDBN}:
11086 @multitable @columnfractions .3 .7
11087 @headitem @var{depth} setting @tab Result of @samp{p var}
11089 @tab @code{$1 = @{d = @{c = @{b = @{a = 3@}@}@}@}}
11091 @tab @code{$1 = @{...@}}
11093 @tab @code{$1 = @{d = @{...@}@}}
11095 @tab @code{$1 = @{d = @{c = @{...@}@}@}}
11097 @tab @code{$1 = @{d = @{c = @{b = @{...@}@}@}@}}
11099 @tab @code{$1 = @{d = @{c = @{b = @{a = 3@}@}@}@}}
11102 To see the contents of structures that have been hidden the user can
11103 either increase the print max-depth, or they can print the elements of
11104 the structure that are visible, for example
11107 (gdb) set print max-depth 2
11109 $1 = @{d = @{c = @{...@}@}@}
11111 $2 = @{c = @{b = @{...@}@}@}
11113 $3 = @{b = @{a = 3@}@}
11116 The pattern used to replace nested structures varies based on
11117 language, for most languages @code{@{...@}} is used, but Fortran uses
11120 @item show print max-depth
11121 Display the current threshold after which nested structures are
11122 replaces with ellipsis.
11124 @anchor{set print null-stop}
11125 @item set print null-stop
11126 @cindex @sc{null} elements in arrays
11127 Cause @value{GDBN} to stop printing the characters of an array when the first
11128 @sc{null} is encountered. This is useful when large arrays actually
11129 contain only short strings.
11130 The default is off.
11132 @item show print null-stop
11133 Show whether @value{GDBN} stops printing an array on the first
11134 @sc{null} character.
11136 @anchor{set print pretty}
11137 @item set print pretty on
11138 @cindex print structures in indented form
11139 @cindex indentation in structure display
11140 Cause @value{GDBN} to print structures in an indented format with one member
11141 per line, like this:
11156 @item set print pretty off
11157 Cause @value{GDBN} to print structures in a compact format, like this:
11161 $1 = @{next = 0x0, flags = @{sweet = 1, sour = 1@}, \
11162 meat = 0x54 "Pork"@}
11167 This is the default format.
11169 @item show print pretty
11170 Show which format @value{GDBN} is using to print structures.
11172 @item set print sevenbit-strings on
11173 @cindex eight-bit characters in strings
11174 @cindex octal escapes in strings
11175 Print using only seven-bit characters; if this option is set,
11176 @value{GDBN} displays any eight-bit characters (in strings or
11177 character values) using the notation @code{\}@var{nnn}. This setting is
11178 best if you are working in English (@sc{ascii}) and you use the
11179 high-order bit of characters as a marker or ``meta'' bit.
11181 @item set print sevenbit-strings off
11182 Print full eight-bit characters. This allows the use of more
11183 international character sets, and is the default.
11185 @item show print sevenbit-strings
11186 Show whether or not @value{GDBN} is printing only seven-bit characters.
11188 @anchor{set print union}
11189 @item set print union on
11190 @cindex unions in structures, printing
11191 Tell @value{GDBN} to print unions which are contained in structures
11192 and other unions. This is the default setting.
11194 @item set print union off
11195 Tell @value{GDBN} not to print unions which are contained in
11196 structures and other unions. @value{GDBN} will print @code{"@{...@}"}
11199 @item show print union
11200 Ask @value{GDBN} whether or not it will print unions which are contained in
11201 structures and other unions.
11203 For example, given the declarations
11206 typedef enum @{Tree, Bug@} Species;
11207 typedef enum @{Big_tree, Acorn, Seedling@} Tree_forms;
11208 typedef enum @{Caterpillar, Cocoon, Butterfly@}
11219 struct thing foo = @{Tree, @{Acorn@}@};
11223 with @code{set print union on} in effect @samp{p foo} would print
11226 $1 = @{it = Tree, form = @{tree = Acorn, bug = Cocoon@}@}
11230 and with @code{set print union off} in effect it would print
11233 $1 = @{it = Tree, form = @{...@}@}
11237 @code{set print union} affects programs written in C-like languages
11243 These settings are of interest when debugging C@t{++} programs:
11246 @cindex demangling C@t{++} names
11247 @item set print demangle
11248 @itemx set print demangle on
11249 Print C@t{++} names in their source form rather than in the encoded
11250 (``mangled'') form passed to the assembler and linker for type-safe
11251 linkage. The default is on.
11253 @item show print demangle
11254 Show whether C@t{++} names are printed in mangled or demangled form.
11256 @item set print asm-demangle
11257 @itemx set print asm-demangle on
11258 Print C@t{++} names in their source form rather than their mangled form, even
11259 in assembler code printouts such as instruction disassemblies.
11260 The default is off.
11262 @item show print asm-demangle
11263 Show whether C@t{++} names in assembly listings are printed in mangled
11266 @cindex C@t{++} symbol decoding style
11267 @cindex symbol decoding style, C@t{++}
11268 @kindex set demangle-style
11269 @item set demangle-style @var{style}
11270 Choose among several encoding schemes used by different compilers to represent
11271 C@t{++} names. If you omit @var{style}, you will see a list of possible
11272 formats. The default value is @var{auto}, which lets @value{GDBN} choose a
11273 decoding style by inspecting your program.
11275 @item show demangle-style
11276 Display the encoding style currently in use for decoding C@t{++} symbols.
11278 @anchor{set print object}
11279 @item set print object
11280 @itemx set print object on
11281 @cindex derived type of an object, printing
11282 @cindex display derived types
11283 When displaying a pointer to an object, identify the @emph{actual}
11284 (derived) type of the object rather than the @emph{declared} type, using
11285 the virtual function table. Note that the virtual function table is
11286 required---this feature can only work for objects that have run-time
11287 type identification; a single virtual method in the object's declared
11288 type is sufficient. Note that this setting is also taken into account when
11289 working with variable objects via MI (@pxref{GDB/MI}).
11291 @item set print object off
11292 Display only the declared type of objects, without reference to the
11293 virtual function table. This is the default setting.
11295 @item show print object
11296 Show whether actual, or declared, object types are displayed.
11298 @anchor{set print static-members}
11299 @item set print static-members
11300 @itemx set print static-members on
11301 @cindex static members of C@t{++} objects
11302 Print static members when displaying a C@t{++} object. The default is on.
11304 @item set print static-members off
11305 Do not print static members when displaying a C@t{++} object.
11307 @item show print static-members
11308 Show whether C@t{++} static members are printed or not.
11310 @item set print pascal_static-members
11311 @itemx set print pascal_static-members on
11312 @cindex static members of Pascal objects
11313 @cindex Pascal objects, static members display
11314 Print static members when displaying a Pascal object. The default is on.
11316 @item set print pascal_static-members off
11317 Do not print static members when displaying a Pascal object.
11319 @item show print pascal_static-members
11320 Show whether Pascal static members are printed or not.
11322 @c These don't work with HP ANSI C++ yet.
11323 @anchor{set print vtbl}
11324 @item set print vtbl
11325 @itemx set print vtbl on
11326 @cindex pretty print C@t{++} virtual function tables
11327 @cindex virtual functions (C@t{++}) display
11328 @cindex VTBL display
11329 Pretty print C@t{++} virtual function tables. The default is off.
11330 (The @code{vtbl} commands do not work on programs compiled with the HP
11331 ANSI C@t{++} compiler (@code{aCC}).)
11333 @item set print vtbl off
11334 Do not pretty print C@t{++} virtual function tables.
11336 @item show print vtbl
11337 Show whether C@t{++} virtual function tables are pretty printed, or not.
11340 @node Pretty Printing
11341 @section Pretty Printing
11343 @value{GDBN} provides a mechanism to allow pretty-printing of values using
11344 Python code. It greatly simplifies the display of complex objects. This
11345 mechanism works for both MI and the CLI.
11348 * Pretty-Printer Introduction:: Introduction to pretty-printers
11349 * Pretty-Printer Example:: An example pretty-printer
11350 * Pretty-Printer Commands:: Pretty-printer commands
11353 @node Pretty-Printer Introduction
11354 @subsection Pretty-Printer Introduction
11356 When @value{GDBN} prints a value, it first sees if there is a pretty-printer
11357 registered for the value. If there is then @value{GDBN} invokes the
11358 pretty-printer to print the value. Otherwise the value is printed normally.
11360 Pretty-printers are normally named. This makes them easy to manage.
11361 The @samp{info pretty-printer} command will list all the installed
11362 pretty-printers with their names.
11363 If a pretty-printer can handle multiple data types, then its
11364 @dfn{subprinters} are the printers for the individual data types.
11365 Each such subprinter has its own name.
11366 The format of the name is @var{printer-name};@var{subprinter-name}.
11368 Pretty-printers are installed by @dfn{registering} them with @value{GDBN}.
11369 Typically they are automatically loaded and registered when the corresponding
11370 debug information is loaded, thus making them available without having to
11371 do anything special.
11373 There are three places where a pretty-printer can be registered.
11377 Pretty-printers registered globally are available when debugging
11381 Pretty-printers registered with a program space are available only
11382 when debugging that program.
11383 @xref{Progspaces In Python}, for more details on program spaces in Python.
11386 Pretty-printers registered with an objfile are loaded and unloaded
11387 with the corresponding objfile (e.g., shared library).
11388 @xref{Objfiles In Python}, for more details on objfiles in Python.
11391 @xref{Selecting Pretty-Printers}, for further information on how
11392 pretty-printers are selected,
11394 @xref{Writing a Pretty-Printer}, for implementing pretty printers
11397 @node Pretty-Printer Example
11398 @subsection Pretty-Printer Example
11400 Here is how a C@t{++} @code{std::string} looks without a pretty-printer:
11403 (@value{GDBP}) print s
11405 static npos = 4294967295,
11407 <std::allocator<char>> = @{
11408 <__gnu_cxx::new_allocator<char>> = @{
11409 <No data fields>@}, <No data fields>
11411 members of std::basic_string<char, std::char_traits<char>,
11412 std::allocator<char> >::_Alloc_hider:
11413 _M_p = 0x804a014 "abcd"
11418 With a pretty-printer for @code{std::string} only the contents are printed:
11421 (@value{GDBP}) print s
11425 @node Pretty-Printer Commands
11426 @subsection Pretty-Printer Commands
11427 @cindex pretty-printer commands
11430 @kindex info pretty-printer
11431 @item info pretty-printer [@var{object-regexp} [@var{name-regexp}]]
11432 Print the list of installed pretty-printers.
11433 This includes disabled pretty-printers, which are marked as such.
11435 @var{object-regexp} is a regular expression matching the objects
11436 whose pretty-printers to list.
11437 Objects can be @code{global}, the program space's file
11438 (@pxref{Progspaces In Python}),
11439 and the object files within that program space (@pxref{Objfiles In Python}).
11440 @xref{Selecting Pretty-Printers}, for details on how @value{GDBN}
11441 looks up a printer from these three objects.
11443 @var{name-regexp} is a regular expression matching the name of the printers
11446 @kindex disable pretty-printer
11447 @item disable pretty-printer [@var{object-regexp} [@var{name-regexp}]]
11448 Disable pretty-printers matching @var{object-regexp} and @var{name-regexp}.
11449 A disabled pretty-printer is not forgotten, it may be enabled again later.
11451 @kindex enable pretty-printer
11452 @item enable pretty-printer [@var{object-regexp} [@var{name-regexp}]]
11453 Enable pretty-printers matching @var{object-regexp} and @var{name-regexp}.
11458 Suppose we have three pretty-printers installed: one from library1.so
11459 named @code{foo} that prints objects of type @code{foo}, and
11460 another from library2.so named @code{bar} that prints two types of objects,
11461 @code{bar1} and @code{bar2}.
11464 (gdb) info pretty-printer
11471 (gdb) info pretty-printer library2
11476 (gdb) disable pretty-printer library1
11478 2 of 3 printers enabled
11479 (gdb) info pretty-printer
11486 (gdb) disable pretty-printer library2 bar;bar1
11488 1 of 3 printers enabled
11489 (gdb) info pretty-printer library2
11496 (gdb) disable pretty-printer library2 bar
11498 0 of 3 printers enabled
11499 (gdb) info pretty-printer library2
11508 Note that for @code{bar} the entire printer can be disabled,
11509 as can each individual subprinter.
11511 @node Value History
11512 @section Value History
11514 @cindex value history
11515 @cindex history of values printed by @value{GDBN}
11516 Values printed by the @code{print} command are saved in the @value{GDBN}
11517 @dfn{value history}. This allows you to refer to them in other expressions.
11518 Values are kept until the symbol table is re-read or discarded
11519 (for example with the @code{file} or @code{symbol-file} commands).
11520 When the symbol table changes, the value history is discarded,
11521 since the values may contain pointers back to the types defined in the
11526 @cindex history number
11527 The values printed are given @dfn{history numbers} by which you can
11528 refer to them. These are successive integers starting with one.
11529 @code{print} shows you the history number assigned to a value by
11530 printing @samp{$@var{num} = } before the value; here @var{num} is the
11533 To refer to any previous value, use @samp{$} followed by the value's
11534 history number. The way @code{print} labels its output is designed to
11535 remind you of this. Just @code{$} refers to the most recent value in
11536 the history, and @code{$$} refers to the value before that.
11537 @code{$$@var{n}} refers to the @var{n}th value from the end; @code{$$2}
11538 is the value just prior to @code{$$}, @code{$$1} is equivalent to
11539 @code{$$}, and @code{$$0} is equivalent to @code{$}.
11541 For example, suppose you have just printed a pointer to a structure and
11542 want to see the contents of the structure. It suffices to type
11548 If you have a chain of structures where the component @code{next} points
11549 to the next one, you can print the contents of the next one with this:
11556 You can print successive links in the chain by repeating this
11557 command---which you can do by just typing @key{RET}.
11559 Note that the history records values, not expressions. If the value of
11560 @code{x} is 4 and you type these commands:
11568 then the value recorded in the value history by the @code{print} command
11569 remains 4 even though the value of @code{x} has changed.
11572 @kindex show values
11574 Print the last ten values in the value history, with their item numbers.
11575 This is like @samp{p@ $$9} repeated ten times, except that @code{show
11576 values} does not change the history.
11578 @item show values @var{n}
11579 Print ten history values centered on history item number @var{n}.
11581 @item show values +
11582 Print ten history values just after the values last printed. If no more
11583 values are available, @code{show values +} produces no display.
11586 Pressing @key{RET} to repeat @code{show values @var{n}} has exactly the
11587 same effect as @samp{show values +}.
11589 @node Convenience Vars
11590 @section Convenience Variables
11592 @cindex convenience variables
11593 @cindex user-defined variables
11594 @value{GDBN} provides @dfn{convenience variables} that you can use within
11595 @value{GDBN} to hold on to a value and refer to it later. These variables
11596 exist entirely within @value{GDBN}; they are not part of your program, and
11597 setting a convenience variable has no direct effect on further execution
11598 of your program. That is why you can use them freely.
11600 Convenience variables are prefixed with @samp{$}. Any name preceded by
11601 @samp{$} can be used for a convenience variable, unless it is one of
11602 the predefined machine-specific register names (@pxref{Registers, ,Registers}).
11603 (Value history references, in contrast, are @emph{numbers} preceded
11604 by @samp{$}. @xref{Value History, ,Value History}.)
11606 You can save a value in a convenience variable with an assignment
11607 expression, just as you would set a variable in your program.
11611 set $foo = *object_ptr
11615 would save in @code{$foo} the value contained in the object pointed to by
11618 Using a convenience variable for the first time creates it, but its
11619 value is @code{void} until you assign a new value. You can alter the
11620 value with another assignment at any time.
11622 Convenience variables have no fixed types. You can assign a convenience
11623 variable any type of value, including structures and arrays, even if
11624 that variable already has a value of a different type. The convenience
11625 variable, when used as an expression, has the type of its current value.
11628 @kindex show convenience
11629 @cindex show all user variables and functions
11630 @item show convenience
11631 Print a list of convenience variables used so far, and their values,
11632 as well as a list of the convenience functions.
11633 Abbreviated @code{show conv}.
11635 @kindex init-if-undefined
11636 @cindex convenience variables, initializing
11637 @item init-if-undefined $@var{variable} = @var{expression}
11638 Set a convenience variable if it has not already been set. This is useful
11639 for user-defined commands that keep some state. It is similar, in concept,
11640 to using local static variables with initializers in C (except that
11641 convenience variables are global). It can also be used to allow users to
11642 override default values used in a command script.
11644 If the variable is already defined then the expression is not evaluated so
11645 any side-effects do not occur.
11648 One of the ways to use a convenience variable is as a counter to be
11649 incremented or a pointer to be advanced. For example, to print
11650 a field from successive elements of an array of structures:
11654 print bar[$i++]->contents
11658 Repeat that command by typing @key{RET}.
11660 Some convenience variables are created automatically by @value{GDBN} and given
11661 values likely to be useful.
11664 @vindex $_@r{, convenience variable}
11666 The variable @code{$_} is automatically set by the @code{x} command to
11667 the last address examined (@pxref{Memory, ,Examining Memory}). Other
11668 commands which provide a default address for @code{x} to examine also
11669 set @code{$_} to that address; these commands include @code{info line}
11670 and @code{info breakpoint}. The type of @code{$_} is @code{void *}
11671 except when set by the @code{x} command, in which case it is a pointer
11672 to the type of @code{$__}.
11674 @vindex $__@r{, convenience variable}
11676 The variable @code{$__} is automatically set by the @code{x} command
11677 to the value found in the last address examined. Its type is chosen
11678 to match the format in which the data was printed.
11681 @vindex $_exitcode@r{, convenience variable}
11682 When the program being debugged terminates normally, @value{GDBN}
11683 automatically sets this variable to the exit code of the program, and
11684 resets @code{$_exitsignal} to @code{void}.
11687 @vindex $_exitsignal@r{, convenience variable}
11688 When the program being debugged dies due to an uncaught signal,
11689 @value{GDBN} automatically sets this variable to that signal's number,
11690 and resets @code{$_exitcode} to @code{void}.
11692 To distinguish between whether the program being debugged has exited
11693 (i.e., @code{$_exitcode} is not @code{void}) or signalled (i.e.,
11694 @code{$_exitsignal} is not @code{void}), the convenience function
11695 @code{$_isvoid} can be used (@pxref{Convenience Funs,, Convenience
11696 Functions}). For example, considering the following source code:
11699 #include <signal.h>
11702 main (int argc, char *argv[])
11709 A valid way of telling whether the program being debugged has exited
11710 or signalled would be:
11713 (@value{GDBP}) define has_exited_or_signalled
11714 Type commands for definition of ``has_exited_or_signalled''.
11715 End with a line saying just ``end''.
11716 >if $_isvoid ($_exitsignal)
11717 >echo The program has exited\n
11719 >echo The program has signalled\n
11725 Program terminated with signal SIGALRM, Alarm clock.
11726 The program no longer exists.
11727 (@value{GDBP}) has_exited_or_signalled
11728 The program has signalled
11731 As can be seen, @value{GDBN} correctly informs that the program being
11732 debugged has signalled, since it calls @code{raise} and raises a
11733 @code{SIGALRM} signal. If the program being debugged had not called
11734 @code{raise}, then @value{GDBN} would report a normal exit:
11737 (@value{GDBP}) has_exited_or_signalled
11738 The program has exited
11742 The variable @code{$_exception} is set to the exception object being
11743 thrown at an exception-related catchpoint. @xref{Set Catchpoints}.
11746 @itemx $_probe_arg0@dots{}$_probe_arg11
11747 Arguments to a static probe. @xref{Static Probe Points}.
11750 @vindex $_sdata@r{, inspect, convenience variable}
11751 The variable @code{$_sdata} contains extra collected static tracepoint
11752 data. @xref{Tracepoint Actions,,Tracepoint Action Lists}. Note that
11753 @code{$_sdata} could be empty, if not inspecting a trace buffer, or
11754 if extra static tracepoint data has not been collected.
11757 @vindex $_siginfo@r{, convenience variable}
11758 The variable @code{$_siginfo} contains extra signal information
11759 (@pxref{extra signal information}). Note that @code{$_siginfo}
11760 could be empty, if the application has not yet received any signals.
11761 For example, it will be empty before you execute the @code{run} command.
11764 @vindex $_tlb@r{, convenience variable}
11765 The variable @code{$_tlb} is automatically set when debugging
11766 applications running on MS-Windows in native mode or connected to
11767 gdbserver that supports the @code{qGetTIBAddr} request.
11768 @xref{General Query Packets}.
11769 This variable contains the address of the thread information block.
11772 The number of the current inferior. @xref{Inferiors and
11773 Programs, ,Debugging Multiple Inferiors and Programs}.
11776 The thread number of the current thread. @xref{thread numbers}.
11779 The global number of the current thread. @xref{global thread numbers}.
11783 @vindex $_gdb_major@r{, convenience variable}
11784 @vindex $_gdb_minor@r{, convenience variable}
11785 The major and minor version numbers of the running @value{GDBN}.
11786 Development snapshots and pretest versions have their minor version
11787 incremented by one; thus, @value{GDBN} pretest 9.11.90 will produce
11788 the value 12 for @code{$_gdb_minor}. These variables allow you to
11789 write scripts that work with different versions of @value{GDBN}
11790 without errors caused by features unavailable in some of those
11793 @item $_shell_exitcode
11794 @itemx $_shell_exitsignal
11795 @vindex $_shell_exitcode@r{, convenience variable}
11796 @vindex $_shell_exitsignal@r{, convenience variable}
11797 @cindex shell command, exit code
11798 @cindex shell command, exit signal
11799 @cindex exit status of shell commands
11800 @value{GDBN} commands such as @code{shell} and @code{|} are launching
11801 shell commands. When a launched command terminates, @value{GDBN}
11802 automatically maintains the variables @code{$_shell_exitcode}
11803 and @code{$_shell_exitsignal} according to the exit status of the last
11804 launched command. These variables are set and used similarly to
11805 the variables @code{$_exitcode} and @code{$_exitsignal}.
11809 @node Convenience Funs
11810 @section Convenience Functions
11812 @cindex convenience functions
11813 @value{GDBN} also supplies some @dfn{convenience functions}. These
11814 have a syntax similar to convenience variables. A convenience
11815 function can be used in an expression just like an ordinary function;
11816 however, a convenience function is implemented internally to
11819 These functions do not require @value{GDBN} to be configured with
11820 @code{Python} support, which means that they are always available.
11824 @item $_isvoid (@var{expr})
11825 @findex $_isvoid@r{, convenience function}
11826 Return one if the expression @var{expr} is @code{void}. Otherwise it
11829 A @code{void} expression is an expression where the type of the result
11830 is @code{void}. For example, you can examine a convenience variable
11831 (see @ref{Convenience Vars,, Convenience Variables}) to check whether
11835 (@value{GDBP}) print $_exitcode
11837 (@value{GDBP}) print $_isvoid ($_exitcode)
11840 Starting program: ./a.out
11841 [Inferior 1 (process 29572) exited normally]
11842 (@value{GDBP}) print $_exitcode
11844 (@value{GDBP}) print $_isvoid ($_exitcode)
11848 In the example above, we used @code{$_isvoid} to check whether
11849 @code{$_exitcode} is @code{void} before and after the execution of the
11850 program being debugged. Before the execution there is no exit code to
11851 be examined, therefore @code{$_exitcode} is @code{void}. After the
11852 execution the program being debugged returned zero, therefore
11853 @code{$_exitcode} is zero, which means that it is not @code{void}
11856 The @code{void} expression can also be a call of a function from the
11857 program being debugged. For example, given the following function:
11866 The result of calling it inside @value{GDBN} is @code{void}:
11869 (@value{GDBP}) print foo ()
11871 (@value{GDBP}) print $_isvoid (foo ())
11873 (@value{GDBP}) set $v = foo ()
11874 (@value{GDBP}) print $v
11876 (@value{GDBP}) print $_isvoid ($v)
11882 These functions require @value{GDBN} to be configured with
11883 @code{Python} support.
11887 @item $_memeq(@var{buf1}, @var{buf2}, @var{length})
11888 @findex $_memeq@r{, convenience function}
11889 Returns one if the @var{length} bytes at the addresses given by
11890 @var{buf1} and @var{buf2} are equal.
11891 Otherwise it returns zero.
11893 @item $_regex(@var{str}, @var{regex})
11894 @findex $_regex@r{, convenience function}
11895 Returns one if the string @var{str} matches the regular expression
11896 @var{regex}. Otherwise it returns zero.
11897 The syntax of the regular expression is that specified by @code{Python}'s
11898 regular expression support.
11900 @item $_streq(@var{str1}, @var{str2})
11901 @findex $_streq@r{, convenience function}
11902 Returns one if the strings @var{str1} and @var{str2} are equal.
11903 Otherwise it returns zero.
11905 @item $_strlen(@var{str})
11906 @findex $_strlen@r{, convenience function}
11907 Returns the length of string @var{str}.
11909 @item $_caller_is(@var{name}@r{[}, @var{number_of_frames}@r{]})
11910 @findex $_caller_is@r{, convenience function}
11911 Returns one if the calling function's name is equal to @var{name}.
11912 Otherwise it returns zero.
11914 If the optional argument @var{number_of_frames} is provided,
11915 it is the number of frames up in the stack to look.
11923 at testsuite/gdb.python/py-caller-is.c:21
11924 #1 0x00000000004005a0 in middle_func ()
11925 at testsuite/gdb.python/py-caller-is.c:27
11926 #2 0x00000000004005ab in top_func ()
11927 at testsuite/gdb.python/py-caller-is.c:33
11928 #3 0x00000000004005b6 in main ()
11929 at testsuite/gdb.python/py-caller-is.c:39
11930 (gdb) print $_caller_is ("middle_func")
11932 (gdb) print $_caller_is ("top_func", 2)
11936 @item $_caller_matches(@var{regexp}@r{[}, @var{number_of_frames}@r{]})
11937 @findex $_caller_matches@r{, convenience function}
11938 Returns one if the calling function's name matches the regular expression
11939 @var{regexp}. Otherwise it returns zero.
11941 If the optional argument @var{number_of_frames} is provided,
11942 it is the number of frames up in the stack to look.
11945 @item $_any_caller_is(@var{name}@r{[}, @var{number_of_frames}@r{]})
11946 @findex $_any_caller_is@r{, convenience function}
11947 Returns one if any calling function's name is equal to @var{name}.
11948 Otherwise it returns zero.
11950 If the optional argument @var{number_of_frames} is provided,
11951 it is the number of frames up in the stack to look.
11954 This function differs from @code{$_caller_is} in that this function
11955 checks all stack frames from the immediate caller to the frame specified
11956 by @var{number_of_frames}, whereas @code{$_caller_is} only checks the
11957 frame specified by @var{number_of_frames}.
11959 @item $_any_caller_matches(@var{regexp}@r{[}, @var{number_of_frames}@r{]})
11960 @findex $_any_caller_matches@r{, convenience function}
11961 Returns one if any calling function's name matches the regular expression
11962 @var{regexp}. Otherwise it returns zero.
11964 If the optional argument @var{number_of_frames} is provided,
11965 it is the number of frames up in the stack to look.
11968 This function differs from @code{$_caller_matches} in that this function
11969 checks all stack frames from the immediate caller to the frame specified
11970 by @var{number_of_frames}, whereas @code{$_caller_matches} only checks the
11971 frame specified by @var{number_of_frames}.
11973 @item $_as_string(@var{value})
11974 @findex $_as_string@r{, convenience function}
11975 Return the string representation of @var{value}.
11977 This function is useful to obtain the textual label (enumerator) of an
11978 enumeration value. For example, assuming the variable @var{node} is of
11979 an enumerated type:
11982 (gdb) printf "Visiting node of type %s\n", $_as_string(node)
11983 Visiting node of type NODE_INTEGER
11986 @item $_cimag(@var{value})
11987 @itemx $_creal(@var{value})
11988 @findex $_cimag@r{, convenience function}
11989 @findex $_creal@r{, convenience function}
11990 Return the imaginary (@code{$_cimag}) or real (@code{$_creal}) part of
11991 the complex number @var{value}.
11993 The type of the imaginary or real part depends on the type of the
11994 complex number, e.g., using @code{$_cimag} on a @code{float complex}
11995 will return an imaginary part of type @code{float}.
11999 @value{GDBN} provides the ability to list and get help on
12000 convenience functions.
12003 @item help function
12004 @kindex help function
12005 @cindex show all convenience functions
12006 Print a list of all convenience functions.
12013 You can refer to machine register contents, in expressions, as variables
12014 with names starting with @samp{$}. The names of registers are different
12015 for each machine; use @code{info registers} to see the names used on
12019 @kindex info registers
12020 @item info registers
12021 Print the names and values of all registers except floating-point
12022 and vector registers (in the selected stack frame).
12024 @kindex info all-registers
12025 @cindex floating point registers
12026 @item info all-registers
12027 Print the names and values of all registers, including floating-point
12028 and vector registers (in the selected stack frame).
12030 @item info registers @var{reggroup} @dots{}
12031 Print the name and value of the registers in each of the specified
12032 @var{reggroup}s. The @var{reggoup} can be any of those returned by
12033 @code{maint print reggroups} (@pxref{Maintenance Commands}).
12035 @item info registers @var{regname} @dots{}
12036 Print the @dfn{relativized} value of each specified register @var{regname}.
12037 As discussed in detail below, register values are normally relative to
12038 the selected stack frame. The @var{regname} may be any register name valid on
12039 the machine you are using, with or without the initial @samp{$}.
12042 @anchor{standard registers}
12043 @cindex stack pointer register
12044 @cindex program counter register
12045 @cindex process status register
12046 @cindex frame pointer register
12047 @cindex standard registers
12048 @value{GDBN} has four ``standard'' register names that are available (in
12049 expressions) on most machines---whenever they do not conflict with an
12050 architecture's canonical mnemonics for registers. The register names
12051 @code{$pc} and @code{$sp} are used for the program counter register and
12052 the stack pointer. @code{$fp} is used for a register that contains a
12053 pointer to the current stack frame, and @code{$ps} is used for a
12054 register that contains the processor status. For example,
12055 you could print the program counter in hex with
12062 or print the instruction to be executed next with
12069 or add four to the stack pointer@footnote{This is a way of removing
12070 one word from the stack, on machines where stacks grow downward in
12071 memory (most machines, nowadays). This assumes that the innermost
12072 stack frame is selected; setting @code{$sp} is not allowed when other
12073 stack frames are selected. To pop entire frames off the stack,
12074 regardless of machine architecture, use @code{return};
12075 see @ref{Returning, ,Returning from a Function}.} with
12081 Whenever possible, these four standard register names are available on
12082 your machine even though the machine has different canonical mnemonics,
12083 so long as there is no conflict. The @code{info registers} command
12084 shows the canonical names. For example, on the SPARC, @code{info
12085 registers} displays the processor status register as @code{$psr} but you
12086 can also refer to it as @code{$ps}; and on x86-based machines @code{$ps}
12087 is an alias for the @sc{eflags} register.
12089 @value{GDBN} always considers the contents of an ordinary register as an
12090 integer when the register is examined in this way. Some machines have
12091 special registers which can hold nothing but floating point; these
12092 registers are considered to have floating point values. There is no way
12093 to refer to the contents of an ordinary register as floating point value
12094 (although you can @emph{print} it as a floating point value with
12095 @samp{print/f $@var{regname}}).
12097 Some registers have distinct ``raw'' and ``virtual'' data formats. This
12098 means that the data format in which the register contents are saved by
12099 the operating system is not the same one that your program normally
12100 sees. For example, the registers of the 68881 floating point
12101 coprocessor are always saved in ``extended'' (raw) format, but all C
12102 programs expect to work with ``double'' (virtual) format. In such
12103 cases, @value{GDBN} normally works with the virtual format only (the format
12104 that makes sense for your program), but the @code{info registers} command
12105 prints the data in both formats.
12107 @cindex SSE registers (x86)
12108 @cindex MMX registers (x86)
12109 Some machines have special registers whose contents can be interpreted
12110 in several different ways. For example, modern x86-based machines
12111 have SSE and MMX registers that can hold several values packed
12112 together in several different formats. @value{GDBN} refers to such
12113 registers in @code{struct} notation:
12116 (@value{GDBP}) print $xmm1
12118 v4_float = @{0, 3.43859137e-038, 1.54142831e-044, 1.821688e-044@},
12119 v2_double = @{9.92129282474342e-303, 2.7585945287983262e-313@},
12120 v16_int8 = "\000\000\000\000\3706;\001\v\000\000\000\r\000\000",
12121 v8_int16 = @{0, 0, 14072, 315, 11, 0, 13, 0@},
12122 v4_int32 = @{0, 20657912, 11, 13@},
12123 v2_int64 = @{88725056443645952, 55834574859@},
12124 uint128 = 0x0000000d0000000b013b36f800000000
12129 To set values of such registers, you need to tell @value{GDBN} which
12130 view of the register you wish to change, as if you were assigning
12131 value to a @code{struct} member:
12134 (@value{GDBP}) set $xmm1.uint128 = 0x000000000000000000000000FFFFFFFF
12137 Normally, register values are relative to the selected stack frame
12138 (@pxref{Selection, ,Selecting a Frame}). This means that you get the
12139 value that the register would contain if all stack frames farther in
12140 were exited and their saved registers restored. In order to see the
12141 true contents of hardware registers, you must select the innermost
12142 frame (with @samp{frame 0}).
12144 @cindex caller-saved registers
12145 @cindex call-clobbered registers
12146 @cindex volatile registers
12147 @cindex <not saved> values
12148 Usually ABIs reserve some registers as not needed to be saved by the
12149 callee (a.k.a.: ``caller-saved'', ``call-clobbered'' or ``volatile''
12150 registers). It may therefore not be possible for @value{GDBN} to know
12151 the value a register had before the call (in other words, in the outer
12152 frame), if the register value has since been changed by the callee.
12153 @value{GDBN} tries to deduce where the inner frame saved
12154 (``callee-saved'') registers, from the debug info, unwind info, or the
12155 machine code generated by your compiler. If some register is not
12156 saved, and @value{GDBN} knows the register is ``caller-saved'' (via
12157 its own knowledge of the ABI, or because the debug/unwind info
12158 explicitly says the register's value is undefined), @value{GDBN}
12159 displays @w{@samp{<not saved>}} as the register's value. With targets
12160 that @value{GDBN} has no knowledge of the register saving convention,
12161 if a register was not saved by the callee, then its value and location
12162 in the outer frame are assumed to be the same of the inner frame.
12163 This is usually harmless, because if the register is call-clobbered,
12164 the caller either does not care what is in the register after the
12165 call, or has code to restore the value that it does care about. Note,
12166 however, that if you change such a register in the outer frame, you
12167 may also be affecting the inner frame. Also, the more ``outer'' the
12168 frame is you're looking at, the more likely a call-clobbered
12169 register's value is to be wrong, in the sense that it doesn't actually
12170 represent the value the register had just before the call.
12172 @node Floating Point Hardware
12173 @section Floating Point Hardware
12174 @cindex floating point
12176 Depending on the configuration, @value{GDBN} may be able to give
12177 you more information about the status of the floating point hardware.
12182 Display hardware-dependent information about the floating
12183 point unit. The exact contents and layout vary depending on the
12184 floating point chip. Currently, @samp{info float} is supported on
12185 the ARM and x86 machines.
12189 @section Vector Unit
12190 @cindex vector unit
12192 Depending on the configuration, @value{GDBN} may be able to give you
12193 more information about the status of the vector unit.
12196 @kindex info vector
12198 Display information about the vector unit. The exact contents and
12199 layout vary depending on the hardware.
12202 @node OS Information
12203 @section Operating System Auxiliary Information
12204 @cindex OS information
12206 @value{GDBN} provides interfaces to useful OS facilities that can help
12207 you debug your program.
12209 @cindex auxiliary vector
12210 @cindex vector, auxiliary
12211 Some operating systems supply an @dfn{auxiliary vector} to programs at
12212 startup. This is akin to the arguments and environment that you
12213 specify for a program, but contains a system-dependent variety of
12214 binary values that tell system libraries important details about the
12215 hardware, operating system, and process. Each value's purpose is
12216 identified by an integer tag; the meanings are well-known but system-specific.
12217 Depending on the configuration and operating system facilities,
12218 @value{GDBN} may be able to show you this information. For remote
12219 targets, this functionality may further depend on the remote stub's
12220 support of the @samp{qXfer:auxv:read} packet, see
12221 @ref{qXfer auxiliary vector read}.
12226 Display the auxiliary vector of the inferior, which can be either a
12227 live process or a core dump file. @value{GDBN} prints each tag value
12228 numerically, and also shows names and text descriptions for recognized
12229 tags. Some values in the vector are numbers, some bit masks, and some
12230 pointers to strings or other data. @value{GDBN} displays each value in the
12231 most appropriate form for a recognized tag, and in hexadecimal for
12232 an unrecognized tag.
12235 On some targets, @value{GDBN} can access operating system-specific
12236 information and show it to you. The types of information available
12237 will differ depending on the type of operating system running on the
12238 target. The mechanism used to fetch the data is described in
12239 @ref{Operating System Information}. For remote targets, this
12240 functionality depends on the remote stub's support of the
12241 @samp{qXfer:osdata:read} packet, see @ref{qXfer osdata read}.
12245 @item info os @var{infotype}
12247 Display OS information of the requested type.
12249 On @sc{gnu}/Linux, the following values of @var{infotype} are valid:
12251 @anchor{linux info os infotypes}
12253 @kindex info os cpus
12255 Display the list of all CPUs/cores. For each CPU/core, @value{GDBN} prints
12256 the available fields from /proc/cpuinfo. For each supported architecture
12257 different fields are available. Two common entries are processor which gives
12258 CPU number and bogomips; a system constant that is calculated during
12259 kernel initialization.
12261 @kindex info os files
12263 Display the list of open file descriptors on the target. For each
12264 file descriptor, @value{GDBN} prints the identifier of the process
12265 owning the descriptor, the command of the owning process, the value
12266 of the descriptor, and the target of the descriptor.
12268 @kindex info os modules
12270 Display the list of all loaded kernel modules on the target. For each
12271 module, @value{GDBN} prints the module name, the size of the module in
12272 bytes, the number of times the module is used, the dependencies of the
12273 module, the status of the module, and the address of the loaded module
12276 @kindex info os msg
12278 Display the list of all System V message queues on the target. For each
12279 message queue, @value{GDBN} prints the message queue key, the message
12280 queue identifier, the access permissions, the current number of bytes
12281 on the queue, the current number of messages on the queue, the processes
12282 that last sent and received a message on the queue, the user and group
12283 of the owner and creator of the message queue, the times at which a
12284 message was last sent and received on the queue, and the time at which
12285 the message queue was last changed.
12287 @kindex info os processes
12289 Display the list of processes on the target. For each process,
12290 @value{GDBN} prints the process identifier, the name of the user, the
12291 command corresponding to the process, and the list of processor cores
12292 that the process is currently running on. (To understand what these
12293 properties mean, for this and the following info types, please consult
12294 the general @sc{gnu}/Linux documentation.)
12296 @kindex info os procgroups
12298 Display the list of process groups on the target. For each process,
12299 @value{GDBN} prints the identifier of the process group that it belongs
12300 to, the command corresponding to the process group leader, the process
12301 identifier, and the command line of the process. The list is sorted
12302 first by the process group identifier, then by the process identifier,
12303 so that processes belonging to the same process group are grouped together
12304 and the process group leader is listed first.
12306 @kindex info os semaphores
12308 Display the list of all System V semaphore sets on the target. For each
12309 semaphore set, @value{GDBN} prints the semaphore set key, the semaphore
12310 set identifier, the access permissions, the number of semaphores in the
12311 set, the user and group of the owner and creator of the semaphore set,
12312 and the times at which the semaphore set was operated upon and changed.
12314 @kindex info os shm
12316 Display the list of all System V shared-memory regions on the target.
12317 For each shared-memory region, @value{GDBN} prints the region key,
12318 the shared-memory identifier, the access permissions, the size of the
12319 region, the process that created the region, the process that last
12320 attached to or detached from the region, the current number of live
12321 attaches to the region, and the times at which the region was last
12322 attached to, detach from, and changed.
12324 @kindex info os sockets
12326 Display the list of Internet-domain sockets on the target. For each
12327 socket, @value{GDBN} prints the address and port of the local and
12328 remote endpoints, the current state of the connection, the creator of
12329 the socket, the IP address family of the socket, and the type of the
12332 @kindex info os threads
12334 Display the list of threads running on the target. For each thread,
12335 @value{GDBN} prints the identifier of the process that the thread
12336 belongs to, the command of the process, the thread identifier, and the
12337 processor core that it is currently running on. The main thread of a
12338 process is not listed.
12342 If @var{infotype} is omitted, then list the possible values for
12343 @var{infotype} and the kind of OS information available for each
12344 @var{infotype}. If the target does not return a list of possible
12345 types, this command will report an error.
12348 @node Memory Region Attributes
12349 @section Memory Region Attributes
12350 @cindex memory region attributes
12352 @dfn{Memory region attributes} allow you to describe special handling
12353 required by regions of your target's memory. @value{GDBN} uses
12354 attributes to determine whether to allow certain types of memory
12355 accesses; whether to use specific width accesses; and whether to cache
12356 target memory. By default the description of memory regions is
12357 fetched from the target (if the current target supports this), but the
12358 user can override the fetched regions.
12360 Defined memory regions can be individually enabled and disabled. When a
12361 memory region is disabled, @value{GDBN} uses the default attributes when
12362 accessing memory in that region. Similarly, if no memory regions have
12363 been defined, @value{GDBN} uses the default attributes when accessing
12366 When a memory region is defined, it is given a number to identify it;
12367 to enable, disable, or remove a memory region, you specify that number.
12371 @item mem @var{lower} @var{upper} @var{attributes}@dots{}
12372 Define a memory region bounded by @var{lower} and @var{upper} with
12373 attributes @var{attributes}@dots{}, and add it to the list of regions
12374 monitored by @value{GDBN}. Note that @var{upper} == 0 is a special
12375 case: it is treated as the target's maximum memory address.
12376 (0xffff on 16 bit targets, 0xffffffff on 32 bit targets, etc.)
12379 Discard any user changes to the memory regions and use target-supplied
12380 regions, if available, or no regions if the target does not support.
12383 @item delete mem @var{nums}@dots{}
12384 Remove memory regions @var{nums}@dots{} from the list of regions
12385 monitored by @value{GDBN}.
12387 @kindex disable mem
12388 @item disable mem @var{nums}@dots{}
12389 Disable monitoring of memory regions @var{nums}@dots{}.
12390 A disabled memory region is not forgotten.
12391 It may be enabled again later.
12394 @item enable mem @var{nums}@dots{}
12395 Enable monitoring of memory regions @var{nums}@dots{}.
12399 Print a table of all defined memory regions, with the following columns
12403 @item Memory Region Number
12404 @item Enabled or Disabled.
12405 Enabled memory regions are marked with @samp{y}.
12406 Disabled memory regions are marked with @samp{n}.
12409 The address defining the inclusive lower bound of the memory region.
12412 The address defining the exclusive upper bound of the memory region.
12415 The list of attributes set for this memory region.
12420 @subsection Attributes
12422 @subsubsection Memory Access Mode
12423 The access mode attributes set whether @value{GDBN} may make read or
12424 write accesses to a memory region.
12426 While these attributes prevent @value{GDBN} from performing invalid
12427 memory accesses, they do nothing to prevent the target system, I/O DMA,
12428 etc.@: from accessing memory.
12432 Memory is read only.
12434 Memory is write only.
12436 Memory is read/write. This is the default.
12439 @subsubsection Memory Access Size
12440 The access size attribute tells @value{GDBN} to use specific sized
12441 accesses in the memory region. Often memory mapped device registers
12442 require specific sized accesses. If no access size attribute is
12443 specified, @value{GDBN} may use accesses of any size.
12447 Use 8 bit memory accesses.
12449 Use 16 bit memory accesses.
12451 Use 32 bit memory accesses.
12453 Use 64 bit memory accesses.
12456 @c @subsubsection Hardware/Software Breakpoints
12457 @c The hardware/software breakpoint attributes set whether @value{GDBN}
12458 @c will use hardware or software breakpoints for the internal breakpoints
12459 @c used by the step, next, finish, until, etc. commands.
12463 @c Always use hardware breakpoints
12464 @c @item swbreak (default)
12467 @subsubsection Data Cache
12468 The data cache attributes set whether @value{GDBN} will cache target
12469 memory. While this generally improves performance by reducing debug
12470 protocol overhead, it can lead to incorrect results because @value{GDBN}
12471 does not know about volatile variables or memory mapped device
12476 Enable @value{GDBN} to cache target memory.
12478 Disable @value{GDBN} from caching target memory. This is the default.
12481 @subsection Memory Access Checking
12482 @value{GDBN} can be instructed to refuse accesses to memory that is
12483 not explicitly described. This can be useful if accessing such
12484 regions has undesired effects for a specific target, or to provide
12485 better error checking. The following commands control this behaviour.
12488 @kindex set mem inaccessible-by-default
12489 @item set mem inaccessible-by-default [on|off]
12490 If @code{on} is specified, make @value{GDBN} treat memory not
12491 explicitly described by the memory ranges as non-existent and refuse accesses
12492 to such memory. The checks are only performed if there's at least one
12493 memory range defined. If @code{off} is specified, make @value{GDBN}
12494 treat the memory not explicitly described by the memory ranges as RAM.
12495 The default value is @code{on}.
12496 @kindex show mem inaccessible-by-default
12497 @item show mem inaccessible-by-default
12498 Show the current handling of accesses to unknown memory.
12502 @c @subsubsection Memory Write Verification
12503 @c The memory write verification attributes set whether @value{GDBN}
12504 @c will re-reads data after each write to verify the write was successful.
12508 @c @item noverify (default)
12511 @node Dump/Restore Files
12512 @section Copy Between Memory and a File
12513 @cindex dump/restore files
12514 @cindex append data to a file
12515 @cindex dump data to a file
12516 @cindex restore data from a file
12518 You can use the commands @code{dump}, @code{append}, and
12519 @code{restore} to copy data between target memory and a file. The
12520 @code{dump} and @code{append} commands write data to a file, and the
12521 @code{restore} command reads data from a file back into the inferior's
12522 memory. Files may be in binary, Motorola S-record, Intel hex,
12523 Tektronix Hex, or Verilog Hex format; however, @value{GDBN} can only
12524 append to binary files, and cannot read from Verilog Hex files.
12529 @item dump @r{[}@var{format}@r{]} memory @var{filename} @var{start_addr} @var{end_addr}
12530 @itemx dump @r{[}@var{format}@r{]} value @var{filename} @var{expr}
12531 Dump the contents of memory from @var{start_addr} to @var{end_addr},
12532 or the value of @var{expr}, to @var{filename} in the given format.
12534 The @var{format} parameter may be any one of:
12541 Motorola S-record format.
12543 Tektronix Hex format.
12545 Verilog Hex format.
12548 @value{GDBN} uses the same definitions of these formats as the
12549 @sc{gnu} binary utilities, like @samp{objdump} and @samp{objcopy}. If
12550 @var{format} is omitted, @value{GDBN} dumps the data in raw binary
12554 @item append @r{[}binary@r{]} memory @var{filename} @var{start_addr} @var{end_addr}
12555 @itemx append @r{[}binary@r{]} value @var{filename} @var{expr}
12556 Append the contents of memory from @var{start_addr} to @var{end_addr},
12557 or the value of @var{expr}, to the file @var{filename}, in raw binary form.
12558 (@value{GDBN} can only append data to files in raw binary form.)
12561 @item restore @var{filename} @r{[}binary@r{]} @var{bias} @var{start} @var{end}
12562 Restore the contents of file @var{filename} into memory. The
12563 @code{restore} command can automatically recognize any known @sc{bfd}
12564 file format, except for raw binary. To restore a raw binary file you
12565 must specify the optional keyword @code{binary} after the filename.
12567 If @var{bias} is non-zero, its value will be added to the addresses
12568 contained in the file. Binary files always start at address zero, so
12569 they will be restored at address @var{bias}. Other bfd files have
12570 a built-in location; they will be restored at offset @var{bias}
12571 from that location.
12573 If @var{start} and/or @var{end} are non-zero, then only data between
12574 file offset @var{start} and file offset @var{end} will be restored.
12575 These offsets are relative to the addresses in the file, before
12576 the @var{bias} argument is applied.
12580 @node Core File Generation
12581 @section How to Produce a Core File from Your Program
12582 @cindex dump core from inferior
12584 A @dfn{core file} or @dfn{core dump} is a file that records the memory
12585 image of a running process and its process status (register values
12586 etc.). Its primary use is post-mortem debugging of a program that
12587 crashed while it ran outside a debugger. A program that crashes
12588 automatically produces a core file, unless this feature is disabled by
12589 the user. @xref{Files}, for information on invoking @value{GDBN} in
12590 the post-mortem debugging mode.
12592 Occasionally, you may wish to produce a core file of the program you
12593 are debugging in order to preserve a snapshot of its state.
12594 @value{GDBN} has a special command for that.
12598 @kindex generate-core-file
12599 @item generate-core-file [@var{file}]
12600 @itemx gcore [@var{file}]
12601 Produce a core dump of the inferior process. The optional argument
12602 @var{file} specifies the file name where to put the core dump. If not
12603 specified, the file name defaults to @file{core.@var{pid}}, where
12604 @var{pid} is the inferior process ID.
12606 Note that this command is implemented only for some systems (as of
12607 this writing, @sc{gnu}/Linux, FreeBSD, Solaris, and S390).
12609 On @sc{gnu}/Linux, this command can take into account the value of the
12610 file @file{/proc/@var{pid}/coredump_filter} when generating the core
12611 dump (@pxref{set use-coredump-filter}), and by default honors the
12612 @code{VM_DONTDUMP} flag for mappings where it is present in the file
12613 @file{/proc/@var{pid}/smaps} (@pxref{set dump-excluded-mappings}).
12615 @kindex set use-coredump-filter
12616 @anchor{set use-coredump-filter}
12617 @item set use-coredump-filter on
12618 @itemx set use-coredump-filter off
12619 Enable or disable the use of the file
12620 @file{/proc/@var{pid}/coredump_filter} when generating core dump
12621 files. This file is used by the Linux kernel to decide what types of
12622 memory mappings will be dumped or ignored when generating a core dump
12623 file. @var{pid} is the process ID of a currently running process.
12625 To make use of this feature, you have to write in the
12626 @file{/proc/@var{pid}/coredump_filter} file a value, in hexadecimal,
12627 which is a bit mask representing the memory mapping types. If a bit
12628 is set in the bit mask, then the memory mappings of the corresponding
12629 types will be dumped; otherwise, they will be ignored. This
12630 configuration is inherited by child processes. For more information
12631 about the bits that can be set in the
12632 @file{/proc/@var{pid}/coredump_filter} file, please refer to the
12633 manpage of @code{core(5)}.
12635 By default, this option is @code{on}. If this option is turned
12636 @code{off}, @value{GDBN} does not read the @file{coredump_filter} file
12637 and instead uses the same default value as the Linux kernel in order
12638 to decide which pages will be dumped in the core dump file. This
12639 value is currently @code{0x33}, which means that bits @code{0}
12640 (anonymous private mappings), @code{1} (anonymous shared mappings),
12641 @code{4} (ELF headers) and @code{5} (private huge pages) are active.
12642 This will cause these memory mappings to be dumped automatically.
12644 @kindex set dump-excluded-mappings
12645 @anchor{set dump-excluded-mappings}
12646 @item set dump-excluded-mappings on
12647 @itemx set dump-excluded-mappings off
12648 If @code{on} is specified, @value{GDBN} will dump memory mappings
12649 marked with the @code{VM_DONTDUMP} flag. This flag is represented in
12650 the file @file{/proc/@var{pid}/smaps} with the acronym @code{dd}.
12652 The default value is @code{off}.
12655 @node Character Sets
12656 @section Character Sets
12657 @cindex character sets
12659 @cindex translating between character sets
12660 @cindex host character set
12661 @cindex target character set
12663 If the program you are debugging uses a different character set to
12664 represent characters and strings than the one @value{GDBN} uses itself,
12665 @value{GDBN} can automatically translate between the character sets for
12666 you. The character set @value{GDBN} uses we call the @dfn{host
12667 character set}; the one the inferior program uses we call the
12668 @dfn{target character set}.
12670 For example, if you are running @value{GDBN} on a @sc{gnu}/Linux system, which
12671 uses the ISO Latin 1 character set, but you are using @value{GDBN}'s
12672 remote protocol (@pxref{Remote Debugging}) to debug a program
12673 running on an IBM mainframe, which uses the @sc{ebcdic} character set,
12674 then the host character set is Latin-1, and the target character set is
12675 @sc{ebcdic}. If you give @value{GDBN} the command @code{set
12676 target-charset EBCDIC-US}, then @value{GDBN} translates between
12677 @sc{ebcdic} and Latin 1 as you print character or string values, or use
12678 character and string literals in expressions.
12680 @value{GDBN} has no way to automatically recognize which character set
12681 the inferior program uses; you must tell it, using the @code{set
12682 target-charset} command, described below.
12684 Here are the commands for controlling @value{GDBN}'s character set
12688 @item set target-charset @var{charset}
12689 @kindex set target-charset
12690 Set the current target character set to @var{charset}. To display the
12691 list of supported target character sets, type
12692 @kbd{@w{set target-charset @key{TAB}@key{TAB}}}.
12694 @item set host-charset @var{charset}
12695 @kindex set host-charset
12696 Set the current host character set to @var{charset}.
12698 By default, @value{GDBN} uses a host character set appropriate to the
12699 system it is running on; you can override that default using the
12700 @code{set host-charset} command. On some systems, @value{GDBN} cannot
12701 automatically determine the appropriate host character set. In this
12702 case, @value{GDBN} uses @samp{UTF-8}.
12704 @value{GDBN} can only use certain character sets as its host character
12705 set. If you type @kbd{@w{set host-charset @key{TAB}@key{TAB}}},
12706 @value{GDBN} will list the host character sets it supports.
12708 @item set charset @var{charset}
12709 @kindex set charset
12710 Set the current host and target character sets to @var{charset}. As
12711 above, if you type @kbd{@w{set charset @key{TAB}@key{TAB}}},
12712 @value{GDBN} will list the names of the character sets that can be used
12713 for both host and target.
12716 @kindex show charset
12717 Show the names of the current host and target character sets.
12719 @item show host-charset
12720 @kindex show host-charset
12721 Show the name of the current host character set.
12723 @item show target-charset
12724 @kindex show target-charset
12725 Show the name of the current target character set.
12727 @item set target-wide-charset @var{charset}
12728 @kindex set target-wide-charset
12729 Set the current target's wide character set to @var{charset}. This is
12730 the character set used by the target's @code{wchar_t} type. To
12731 display the list of supported wide character sets, type
12732 @kbd{@w{set target-wide-charset @key{TAB}@key{TAB}}}.
12734 @item show target-wide-charset
12735 @kindex show target-wide-charset
12736 Show the name of the current target's wide character set.
12739 Here is an example of @value{GDBN}'s character set support in action.
12740 Assume that the following source code has been placed in the file
12741 @file{charset-test.c}:
12747 = @{72, 101, 108, 108, 111, 44, 32, 119,
12748 111, 114, 108, 100, 33, 10, 0@};
12749 char ibm1047_hello[]
12750 = @{200, 133, 147, 147, 150, 107, 64, 166,
12751 150, 153, 147, 132, 90, 37, 0@};
12755 printf ("Hello, world!\n");
12759 In this program, @code{ascii_hello} and @code{ibm1047_hello} are arrays
12760 containing the string @samp{Hello, world!} followed by a newline,
12761 encoded in the @sc{ascii} and @sc{ibm1047} character sets.
12763 We compile the program, and invoke the debugger on it:
12766 $ gcc -g charset-test.c -o charset-test
12767 $ gdb -nw charset-test
12768 GNU gdb 2001-12-19-cvs
12769 Copyright 2001 Free Software Foundation, Inc.
12774 We can use the @code{show charset} command to see what character sets
12775 @value{GDBN} is currently using to interpret and display characters and
12779 (@value{GDBP}) show charset
12780 The current host and target character set is `ISO-8859-1'.
12784 For the sake of printing this manual, let's use @sc{ascii} as our
12785 initial character set:
12787 (@value{GDBP}) set charset ASCII
12788 (@value{GDBP}) show charset
12789 The current host and target character set is `ASCII'.
12793 Let's assume that @sc{ascii} is indeed the correct character set for our
12794 host system --- in other words, let's assume that if @value{GDBN} prints
12795 characters using the @sc{ascii} character set, our terminal will display
12796 them properly. Since our current target character set is also
12797 @sc{ascii}, the contents of @code{ascii_hello} print legibly:
12800 (@value{GDBP}) print ascii_hello
12801 $1 = 0x401698 "Hello, world!\n"
12802 (@value{GDBP}) print ascii_hello[0]
12807 @value{GDBN} uses the target character set for character and string
12808 literals you use in expressions:
12811 (@value{GDBP}) print '+'
12816 The @sc{ascii} character set uses the number 43 to encode the @samp{+}
12819 @value{GDBN} relies on the user to tell it which character set the
12820 target program uses. If we print @code{ibm1047_hello} while our target
12821 character set is still @sc{ascii}, we get jibberish:
12824 (@value{GDBP}) print ibm1047_hello
12825 $4 = 0x4016a8 "\310\205\223\223\226k@@\246\226\231\223\204Z%"
12826 (@value{GDBP}) print ibm1047_hello[0]
12831 If we invoke the @code{set target-charset} followed by @key{TAB}@key{TAB},
12832 @value{GDBN} tells us the character sets it supports:
12835 (@value{GDBP}) set target-charset
12836 ASCII EBCDIC-US IBM1047 ISO-8859-1
12837 (@value{GDBP}) set target-charset
12840 We can select @sc{ibm1047} as our target character set, and examine the
12841 program's strings again. Now the @sc{ascii} string is wrong, but
12842 @value{GDBN} translates the contents of @code{ibm1047_hello} from the
12843 target character set, @sc{ibm1047}, to the host character set,
12844 @sc{ascii}, and they display correctly:
12847 (@value{GDBP}) set target-charset IBM1047
12848 (@value{GDBP}) show charset
12849 The current host character set is `ASCII'.
12850 The current target character set is `IBM1047'.
12851 (@value{GDBP}) print ascii_hello
12852 $6 = 0x401698 "\110\145%%?\054\040\167?\162%\144\041\012"
12853 (@value{GDBP}) print ascii_hello[0]
12855 (@value{GDBP}) print ibm1047_hello
12856 $8 = 0x4016a8 "Hello, world!\n"
12857 (@value{GDBP}) print ibm1047_hello[0]
12862 As above, @value{GDBN} uses the target character set for character and
12863 string literals you use in expressions:
12866 (@value{GDBP}) print '+'
12871 The @sc{ibm1047} character set uses the number 78 to encode the @samp{+}
12874 @node Caching Target Data
12875 @section Caching Data of Targets
12876 @cindex caching data of targets
12878 @value{GDBN} caches data exchanged between the debugger and a target.
12879 Each cache is associated with the address space of the inferior.
12880 @xref{Inferiors and Programs}, about inferior and address space.
12881 Such caching generally improves performance in remote debugging
12882 (@pxref{Remote Debugging}), because it reduces the overhead of the
12883 remote protocol by bundling memory reads and writes into large chunks.
12884 Unfortunately, simply caching everything would lead to incorrect results,
12885 since @value{GDBN} does not necessarily know anything about volatile
12886 values, memory-mapped I/O addresses, etc. Furthermore, in non-stop mode
12887 (@pxref{Non-Stop Mode}) memory can be changed @emph{while} a gdb command
12889 Therefore, by default, @value{GDBN} only caches data
12890 known to be on the stack@footnote{In non-stop mode, it is moderately
12891 rare for a running thread to modify the stack of a stopped thread
12892 in a way that would interfere with a backtrace, and caching of
12893 stack reads provides a significant speed up of remote backtraces.} or
12894 in the code segment.
12895 Other regions of memory can be explicitly marked as
12896 cacheable; @pxref{Memory Region Attributes}.
12899 @kindex set remotecache
12900 @item set remotecache on
12901 @itemx set remotecache off
12902 This option no longer does anything; it exists for compatibility
12905 @kindex show remotecache
12906 @item show remotecache
12907 Show the current state of the obsolete remotecache flag.
12909 @kindex set stack-cache
12910 @item set stack-cache on
12911 @itemx set stack-cache off
12912 Enable or disable caching of stack accesses. When @code{on}, use
12913 caching. By default, this option is @code{on}.
12915 @kindex show stack-cache
12916 @item show stack-cache
12917 Show the current state of data caching for memory accesses.
12919 @kindex set code-cache
12920 @item set code-cache on
12921 @itemx set code-cache off
12922 Enable or disable caching of code segment accesses. When @code{on},
12923 use caching. By default, this option is @code{on}. This improves
12924 performance of disassembly in remote debugging.
12926 @kindex show code-cache
12927 @item show code-cache
12928 Show the current state of target memory cache for code segment
12931 @kindex info dcache
12932 @item info dcache @r{[}line@r{]}
12933 Print the information about the performance of data cache of the
12934 current inferior's address space. The information displayed
12935 includes the dcache width and depth, and for each cache line, its
12936 number, address, and how many times it was referenced. This
12937 command is useful for debugging the data cache operation.
12939 If a line number is specified, the contents of that line will be
12942 @item set dcache size @var{size}
12943 @cindex dcache size
12944 @kindex set dcache size
12945 Set maximum number of entries in dcache (dcache depth above).
12947 @item set dcache line-size @var{line-size}
12948 @cindex dcache line-size
12949 @kindex set dcache line-size
12950 Set number of bytes each dcache entry caches (dcache width above).
12951 Must be a power of 2.
12953 @item show dcache size
12954 @kindex show dcache size
12955 Show maximum number of dcache entries. @xref{Caching Target Data, info dcache}.
12957 @item show dcache line-size
12958 @kindex show dcache line-size
12959 Show default size of dcache lines.
12963 @node Searching Memory
12964 @section Search Memory
12965 @cindex searching memory
12967 Memory can be searched for a particular sequence of bytes with the
12968 @code{find} command.
12972 @item find @r{[}/@var{sn}@r{]} @var{start_addr}, +@var{len}, @var{val1} @r{[}, @var{val2}, @dots{}@r{]}
12973 @itemx find @r{[}/@var{sn}@r{]} @var{start_addr}, @var{end_addr}, @var{val1} @r{[}, @var{val2}, @dots{}@r{]}
12974 Search memory for the sequence of bytes specified by @var{val1}, @var{val2},
12975 etc. The search begins at address @var{start_addr} and continues for either
12976 @var{len} bytes or through to @var{end_addr} inclusive.
12979 @var{s} and @var{n} are optional parameters.
12980 They may be specified in either order, apart or together.
12983 @item @var{s}, search query size
12984 The size of each search query value.
12990 halfwords (two bytes)
12994 giant words (eight bytes)
12997 All values are interpreted in the current language.
12998 This means, for example, that if the current source language is C/C@t{++}
12999 then searching for the string ``hello'' includes the trailing '\0'.
13000 The null terminator can be removed from searching by using casts,
13001 e.g.: @samp{@{char[5]@}"hello"}.
13003 If the value size is not specified, it is taken from the
13004 value's type in the current language.
13005 This is useful when one wants to specify the search
13006 pattern as a mixture of types.
13007 Note that this means, for example, that in the case of C-like languages
13008 a search for an untyped 0x42 will search for @samp{(int) 0x42}
13009 which is typically four bytes.
13011 @item @var{n}, maximum number of finds
13012 The maximum number of matches to print. The default is to print all finds.
13015 You can use strings as search values. Quote them with double-quotes
13017 The string value is copied into the search pattern byte by byte,
13018 regardless of the endianness of the target and the size specification.
13020 The address of each match found is printed as well as a count of the
13021 number of matches found.
13023 The address of the last value found is stored in convenience variable
13025 A count of the number of matches is stored in @samp{$numfound}.
13027 For example, if stopped at the @code{printf} in this function:
13033 static char hello[] = "hello-hello";
13034 static struct @{ char c; short s; int i; @}
13035 __attribute__ ((packed)) mixed
13036 = @{ 'c', 0x1234, 0x87654321 @};
13037 printf ("%s\n", hello);
13042 you get during debugging:
13045 (gdb) find &hello[0], +sizeof(hello), "hello"
13046 0x804956d <hello.1620+6>
13048 (gdb) find &hello[0], +sizeof(hello), 'h', 'e', 'l', 'l', 'o'
13049 0x8049567 <hello.1620>
13050 0x804956d <hello.1620+6>
13052 (gdb) find &hello[0], +sizeof(hello), @{char[5]@}"hello"
13053 0x8049567 <hello.1620>
13054 0x804956d <hello.1620+6>
13056 (gdb) find /b1 &hello[0], +sizeof(hello), 'h', 0x65, 'l'
13057 0x8049567 <hello.1620>
13059 (gdb) find &mixed, +sizeof(mixed), (char) 'c', (short) 0x1234, (int) 0x87654321
13060 0x8049560 <mixed.1625>
13062 (gdb) print $numfound
13065 $2 = (void *) 0x8049560
13069 @section Value Sizes
13071 Whenever @value{GDBN} prints a value memory will be allocated within
13072 @value{GDBN} to hold the contents of the value. It is possible in
13073 some languages with dynamic typing systems, that an invalid program
13074 may indicate a value that is incorrectly large, this in turn may cause
13075 @value{GDBN} to try and allocate an overly large ammount of memory.
13078 @kindex set max-value-size
13079 @item set max-value-size @var{bytes}
13080 @itemx set max-value-size unlimited
13081 Set the maximum size of memory that @value{GDBN} will allocate for the
13082 contents of a value to @var{bytes}, trying to display a value that
13083 requires more memory than that will result in an error.
13085 Setting this variable does not effect values that have already been
13086 allocated within @value{GDBN}, only future allocations.
13088 There's a minimum size that @code{max-value-size} can be set to in
13089 order that @value{GDBN} can still operate correctly, this minimum is
13090 currently 16 bytes.
13092 The limit applies to the results of some subexpressions as well as to
13093 complete expressions. For example, an expression denoting a simple
13094 integer component, such as @code{x.y.z}, may fail if the size of
13095 @var{x.y} is dynamic and exceeds @var{bytes}. On the other hand,
13096 @value{GDBN} is sometimes clever; the expression @code{A[i]}, where
13097 @var{A} is an array variable with non-constant size, will generally
13098 succeed regardless of the bounds on @var{A}, as long as the component
13099 size is less than @var{bytes}.
13101 The default value of @code{max-value-size} is currently 64k.
13103 @kindex show max-value-size
13104 @item show max-value-size
13105 Show the maximum size of memory, in bytes, that @value{GDBN} will
13106 allocate for the contents of a value.
13109 @node Optimized Code
13110 @chapter Debugging Optimized Code
13111 @cindex optimized code, debugging
13112 @cindex debugging optimized code
13114 Almost all compilers support optimization. With optimization
13115 disabled, the compiler generates assembly code that corresponds
13116 directly to your source code, in a simplistic way. As the compiler
13117 applies more powerful optimizations, the generated assembly code
13118 diverges from your original source code. With help from debugging
13119 information generated by the compiler, @value{GDBN} can map from
13120 the running program back to constructs from your original source.
13122 @value{GDBN} is more accurate with optimization disabled. If you
13123 can recompile without optimization, it is easier to follow the
13124 progress of your program during debugging. But, there are many cases
13125 where you may need to debug an optimized version.
13127 When you debug a program compiled with @samp{-g -O}, remember that the
13128 optimizer has rearranged your code; the debugger shows you what is
13129 really there. Do not be too surprised when the execution path does not
13130 exactly match your source file! An extreme example: if you define a
13131 variable, but never use it, @value{GDBN} never sees that
13132 variable---because the compiler optimizes it out of existence.
13134 Some things do not work as well with @samp{-g -O} as with just
13135 @samp{-g}, particularly on machines with instruction scheduling. If in
13136 doubt, recompile with @samp{-g} alone, and if this fixes the problem,
13137 please report it to us as a bug (including a test case!).
13138 @xref{Variables}, for more information about debugging optimized code.
13141 * Inline Functions:: How @value{GDBN} presents inlining
13142 * Tail Call Frames:: @value{GDBN} analysis of jumps to functions
13145 @node Inline Functions
13146 @section Inline Functions
13147 @cindex inline functions, debugging
13149 @dfn{Inlining} is an optimization that inserts a copy of the function
13150 body directly at each call site, instead of jumping to a shared
13151 routine. @value{GDBN} displays inlined functions just like
13152 non-inlined functions. They appear in backtraces. You can view their
13153 arguments and local variables, step into them with @code{step}, skip
13154 them with @code{next}, and escape from them with @code{finish}.
13155 You can check whether a function was inlined by using the
13156 @code{info frame} command.
13158 For @value{GDBN} to support inlined functions, the compiler must
13159 record information about inlining in the debug information ---
13160 @value{NGCC} using the @sc{dwarf 2} format does this, and several
13161 other compilers do also. @value{GDBN} only supports inlined functions
13162 when using @sc{dwarf 2}. Versions of @value{NGCC} before 4.1
13163 do not emit two required attributes (@samp{DW_AT_call_file} and
13164 @samp{DW_AT_call_line}); @value{GDBN} does not display inlined
13165 function calls with earlier versions of @value{NGCC}. It instead
13166 displays the arguments and local variables of inlined functions as
13167 local variables in the caller.
13169 The body of an inlined function is directly included at its call site;
13170 unlike a non-inlined function, there are no instructions devoted to
13171 the call. @value{GDBN} still pretends that the call site and the
13172 start of the inlined function are different instructions. Stepping to
13173 the call site shows the call site, and then stepping again shows
13174 the first line of the inlined function, even though no additional
13175 instructions are executed.
13177 This makes source-level debugging much clearer; you can see both the
13178 context of the call and then the effect of the call. Only stepping by
13179 a single instruction using @code{stepi} or @code{nexti} does not do
13180 this; single instruction steps always show the inlined body.
13182 There are some ways that @value{GDBN} does not pretend that inlined
13183 function calls are the same as normal calls:
13187 Setting breakpoints at the call site of an inlined function may not
13188 work, because the call site does not contain any code. @value{GDBN}
13189 may incorrectly move the breakpoint to the next line of the enclosing
13190 function, after the call. This limitation will be removed in a future
13191 version of @value{GDBN}; until then, set a breakpoint on an earlier line
13192 or inside the inlined function instead.
13195 @value{GDBN} cannot locate the return value of inlined calls after
13196 using the @code{finish} command. This is a limitation of compiler-generated
13197 debugging information; after @code{finish}, you can step to the next line
13198 and print a variable where your program stored the return value.
13202 @node Tail Call Frames
13203 @section Tail Call Frames
13204 @cindex tail call frames, debugging
13206 Function @code{B} can call function @code{C} in its very last statement. In
13207 unoptimized compilation the call of @code{C} is immediately followed by return
13208 instruction at the end of @code{B} code. Optimizing compiler may replace the
13209 call and return in function @code{B} into one jump to function @code{C}
13210 instead. Such use of a jump instruction is called @dfn{tail call}.
13212 During execution of function @code{C}, there will be no indication in the
13213 function call stack frames that it was tail-called from @code{B}. If function
13214 @code{A} regularly calls function @code{B} which tail-calls function @code{C},
13215 then @value{GDBN} will see @code{A} as the caller of @code{C}. However, in
13216 some cases @value{GDBN} can determine that @code{C} was tail-called from
13217 @code{B}, and it will then create fictitious call frame for that, with the
13218 return address set up as if @code{B} called @code{C} normally.
13220 This functionality is currently supported only by DWARF 2 debugging format and
13221 the compiler has to produce @samp{DW_TAG_call_site} tags. With
13222 @value{NGCC}, you need to specify @option{-O -g} during compilation, to get
13225 @kbd{info frame} command (@pxref{Frame Info}) will indicate the tail call frame
13226 kind by text @code{tail call frame} such as in this sample @value{GDBN} output:
13230 0x40066b <b(int, double)+11>: jmp 0x400640 <c(int, double)>
13232 Stack level 1, frame at 0x7fffffffda30:
13233 rip = 0x40066d in b (amd64-entry-value.cc:59); saved rip 0x4004c5
13234 tail call frame, caller of frame at 0x7fffffffda30
13235 source language c++.
13236 Arglist at unknown address.
13237 Locals at unknown address, Previous frame's sp is 0x7fffffffda30
13240 The detection of all the possible code path executions can find them ambiguous.
13241 There is no execution history stored (possible @ref{Reverse Execution} is never
13242 used for this purpose) and the last known caller could have reached the known
13243 callee by multiple different jump sequences. In such case @value{GDBN} still
13244 tries to show at least all the unambiguous top tail callers and all the
13245 unambiguous bottom tail calees, if any.
13248 @anchor{set debug entry-values}
13249 @item set debug entry-values
13250 @kindex set debug entry-values
13251 When set to on, enables printing of analysis messages for both frame argument
13252 values at function entry and tail calls. It will show all the possible valid
13253 tail calls code paths it has considered. It will also print the intersection
13254 of them with the final unambiguous (possibly partial or even empty) code path
13257 @item show debug entry-values
13258 @kindex show debug entry-values
13259 Show the current state of analysis messages printing for both frame argument
13260 values at function entry and tail calls.
13263 The analysis messages for tail calls can for example show why the virtual tail
13264 call frame for function @code{c} has not been recognized (due to the indirect
13265 reference by variable @code{x}):
13268 static void __attribute__((noinline, noclone)) c (void);
13269 void (*x) (void) = c;
13270 static void __attribute__((noinline, noclone)) a (void) @{ x++; @}
13271 static void __attribute__((noinline, noclone)) c (void) @{ a (); @}
13272 int main (void) @{ x (); return 0; @}
13274 Breakpoint 1, DW_OP_entry_value resolving cannot find
13275 DW_TAG_call_site 0x40039a in main
13277 3 static void __attribute__((noinline, noclone)) a (void) @{ x++; @}
13280 #1 0x000000000040039a in main () at t.c:5
13283 Another possibility is an ambiguous virtual tail call frames resolution:
13287 static void __attribute__((noinline, noclone)) f (void) @{ i++; @}
13288 static void __attribute__((noinline, noclone)) e (void) @{ f (); @}
13289 static void __attribute__((noinline, noclone)) d (void) @{ f (); @}
13290 static void __attribute__((noinline, noclone)) c (void) @{ d (); @}
13291 static void __attribute__((noinline, noclone)) b (void)
13292 @{ if (i) c (); else e (); @}
13293 static void __attribute__((noinline, noclone)) a (void) @{ b (); @}
13294 int main (void) @{ a (); return 0; @}
13296 tailcall: initial: 0x4004d2(a) 0x4004ce(b) 0x4004b2(c) 0x4004a2(d)
13297 tailcall: compare: 0x4004d2(a) 0x4004cc(b) 0x400492(e)
13298 tailcall: reduced: 0x4004d2(a) |
13301 #1 0x00000000004004d2 in a () at t.c:8
13302 #2 0x0000000000400395 in main () at t.c:9
13305 @set CALLSEQ1A @code{main@value{ARROW}a@value{ARROW}b@value{ARROW}c@value{ARROW}d@value{ARROW}f}
13306 @set CALLSEQ2A @code{main@value{ARROW}a@value{ARROW}b@value{ARROW}e@value{ARROW}f}
13308 @c Convert CALLSEQ#A to CALLSEQ#B depending on HAVE_MAKEINFO_CLICK.
13309 @ifset HAVE_MAKEINFO_CLICK
13310 @set ARROW @click{}
13311 @set CALLSEQ1B @clicksequence{@value{CALLSEQ1A}}
13312 @set CALLSEQ2B @clicksequence{@value{CALLSEQ2A}}
13314 @ifclear HAVE_MAKEINFO_CLICK
13316 @set CALLSEQ1B @value{CALLSEQ1A}
13317 @set CALLSEQ2B @value{CALLSEQ2A}
13320 Frames #0 and #2 are real, #1 is a virtual tail call frame.
13321 The code can have possible execution paths @value{CALLSEQ1B} or
13322 @value{CALLSEQ2B}, @value{GDBN} cannot find which one from the inferior state.
13324 @code{initial:} state shows some random possible calling sequence @value{GDBN}
13325 has found. It then finds another possible calling sequcen - that one is
13326 prefixed by @code{compare:}. The non-ambiguous intersection of these two is
13327 printed as the @code{reduced:} calling sequence. That one could have many
13328 futher @code{compare:} and @code{reduced:} statements as long as there remain
13329 any non-ambiguous sequence entries.
13331 For the frame of function @code{b} in both cases there are different possible
13332 @code{$pc} values (@code{0x4004cc} or @code{0x4004ce}), therefore this frame is
13333 also ambigous. The only non-ambiguous frame is the one for function @code{a},
13334 therefore this one is displayed to the user while the ambiguous frames are
13337 There can be also reasons why printing of frame argument values at function
13342 static void __attribute__((noinline, noclone)) c (int i) @{ v++; @}
13343 static void __attribute__((noinline, noclone)) a (int i);
13344 static void __attribute__((noinline, noclone)) b (int i) @{ a (i); @}
13345 static void __attribute__((noinline, noclone)) a (int i)
13346 @{ if (i) b (i - 1); else c (0); @}
13347 int main (void) @{ a (5); return 0; @}
13350 #0 c (i=i@@entry=0) at t.c:2
13351 #1 0x0000000000400428 in a (DW_OP_entry_value resolving has found
13352 function "a" at 0x400420 can call itself via tail calls
13353 i=<optimized out>) at t.c:6
13354 #2 0x000000000040036e in main () at t.c:7
13357 @value{GDBN} cannot find out from the inferior state if and how many times did
13358 function @code{a} call itself (via function @code{b}) as these calls would be
13359 tail calls. Such tail calls would modify thue @code{i} variable, therefore
13360 @value{GDBN} cannot be sure the value it knows would be right - @value{GDBN}
13361 prints @code{<optimized out>} instead.
13364 @chapter C Preprocessor Macros
13366 Some languages, such as C and C@t{++}, provide a way to define and invoke
13367 ``preprocessor macros'' which expand into strings of tokens.
13368 @value{GDBN} can evaluate expressions containing macro invocations, show
13369 the result of macro expansion, and show a macro's definition, including
13370 where it was defined.
13372 You may need to compile your program specially to provide @value{GDBN}
13373 with information about preprocessor macros. Most compilers do not
13374 include macros in their debugging information, even when you compile
13375 with the @option{-g} flag. @xref{Compilation}.
13377 A program may define a macro at one point, remove that definition later,
13378 and then provide a different definition after that. Thus, at different
13379 points in the program, a macro may have different definitions, or have
13380 no definition at all. If there is a current stack frame, @value{GDBN}
13381 uses the macros in scope at that frame's source code line. Otherwise,
13382 @value{GDBN} uses the macros in scope at the current listing location;
13385 Whenever @value{GDBN} evaluates an expression, it always expands any
13386 macro invocations present in the expression. @value{GDBN} also provides
13387 the following commands for working with macros explicitly.
13391 @kindex macro expand
13392 @cindex macro expansion, showing the results of preprocessor
13393 @cindex preprocessor macro expansion, showing the results of
13394 @cindex expanding preprocessor macros
13395 @item macro expand @var{expression}
13396 @itemx macro exp @var{expression}
13397 Show the results of expanding all preprocessor macro invocations in
13398 @var{expression}. Since @value{GDBN} simply expands macros, but does
13399 not parse the result, @var{expression} need not be a valid expression;
13400 it can be any string of tokens.
13403 @item macro expand-once @var{expression}
13404 @itemx macro exp1 @var{expression}
13405 @cindex expand macro once
13406 @i{(This command is not yet implemented.)} Show the results of
13407 expanding those preprocessor macro invocations that appear explicitly in
13408 @var{expression}. Macro invocations appearing in that expansion are
13409 left unchanged. This command allows you to see the effect of a
13410 particular macro more clearly, without being confused by further
13411 expansions. Since @value{GDBN} simply expands macros, but does not
13412 parse the result, @var{expression} need not be a valid expression; it
13413 can be any string of tokens.
13416 @cindex macro definition, showing
13417 @cindex definition of a macro, showing
13418 @cindex macros, from debug info
13419 @item info macro [-a|-all] [--] @var{macro}
13420 Show the current definition or all definitions of the named @var{macro},
13421 and describe the source location or compiler command-line where that
13422 definition was established. The optional double dash is to signify the end of
13423 argument processing and the beginning of @var{macro} for non C-like macros where
13424 the macro may begin with a hyphen.
13426 @kindex info macros
13427 @item info macros @var{location}
13428 Show all macro definitions that are in effect at the location specified
13429 by @var{location}, and describe the source location or compiler
13430 command-line where those definitions were established.
13432 @kindex macro define
13433 @cindex user-defined macros
13434 @cindex defining macros interactively
13435 @cindex macros, user-defined
13436 @item macro define @var{macro} @var{replacement-list}
13437 @itemx macro define @var{macro}(@var{arglist}) @var{replacement-list}
13438 Introduce a definition for a preprocessor macro named @var{macro},
13439 invocations of which are replaced by the tokens given in
13440 @var{replacement-list}. The first form of this command defines an
13441 ``object-like'' macro, which takes no arguments; the second form
13442 defines a ``function-like'' macro, which takes the arguments given in
13445 A definition introduced by this command is in scope in every
13446 expression evaluated in @value{GDBN}, until it is removed with the
13447 @code{macro undef} command, described below. The definition overrides
13448 all definitions for @var{macro} present in the program being debugged,
13449 as well as any previous user-supplied definition.
13451 @kindex macro undef
13452 @item macro undef @var{macro}
13453 Remove any user-supplied definition for the macro named @var{macro}.
13454 This command only affects definitions provided with the @code{macro
13455 define} command, described above; it cannot remove definitions present
13456 in the program being debugged.
13460 List all the macros defined using the @code{macro define} command.
13463 @cindex macros, example of debugging with
13464 Here is a transcript showing the above commands in action. First, we
13465 show our source files:
13470 #include "sample.h"
13473 #define ADD(x) (M + x)
13478 printf ("Hello, world!\n");
13480 printf ("We're so creative.\n");
13482 printf ("Goodbye, world!\n");
13489 Now, we compile the program using the @sc{gnu} C compiler,
13490 @value{NGCC}. We pass the @option{-gdwarf-2}@footnote{This is the
13491 minimum. Recent versions of @value{NGCC} support @option{-gdwarf-3}
13492 and @option{-gdwarf-4}; we recommend always choosing the most recent
13493 version of DWARF.} @emph{and} @option{-g3} flags to ensure the compiler
13494 includes information about preprocessor macros in the debugging
13498 $ gcc -gdwarf-2 -g3 sample.c -o sample
13502 Now, we start @value{GDBN} on our sample program:
13506 GNU gdb 2002-05-06-cvs
13507 Copyright 2002 Free Software Foundation, Inc.
13508 GDB is free software, @dots{}
13512 We can expand macros and examine their definitions, even when the
13513 program is not running. @value{GDBN} uses the current listing position
13514 to decide which macro definitions are in scope:
13517 (@value{GDBP}) list main
13520 5 #define ADD(x) (M + x)
13525 10 printf ("Hello, world!\n");
13527 12 printf ("We're so creative.\n");
13528 (@value{GDBP}) info macro ADD
13529 Defined at /home/jimb/gdb/macros/play/sample.c:5
13530 #define ADD(x) (M + x)
13531 (@value{GDBP}) info macro Q
13532 Defined at /home/jimb/gdb/macros/play/sample.h:1
13533 included at /home/jimb/gdb/macros/play/sample.c:2
13535 (@value{GDBP}) macro expand ADD(1)
13536 expands to: (42 + 1)
13537 (@value{GDBP}) macro expand-once ADD(1)
13538 expands to: once (M + 1)
13542 In the example above, note that @code{macro expand-once} expands only
13543 the macro invocation explicit in the original text --- the invocation of
13544 @code{ADD} --- but does not expand the invocation of the macro @code{M},
13545 which was introduced by @code{ADD}.
13547 Once the program is running, @value{GDBN} uses the macro definitions in
13548 force at the source line of the current stack frame:
13551 (@value{GDBP}) break main
13552 Breakpoint 1 at 0x8048370: file sample.c, line 10.
13554 Starting program: /home/jimb/gdb/macros/play/sample
13556 Breakpoint 1, main () at sample.c:10
13557 10 printf ("Hello, world!\n");
13561 At line 10, the definition of the macro @code{N} at line 9 is in force:
13564 (@value{GDBP}) info macro N
13565 Defined at /home/jimb/gdb/macros/play/sample.c:9
13567 (@value{GDBP}) macro expand N Q M
13568 expands to: 28 < 42
13569 (@value{GDBP}) print N Q M
13574 As we step over directives that remove @code{N}'s definition, and then
13575 give it a new definition, @value{GDBN} finds the definition (or lack
13576 thereof) in force at each point:
13579 (@value{GDBP}) next
13581 12 printf ("We're so creative.\n");
13582 (@value{GDBP}) info macro N
13583 The symbol `N' has no definition as a C/C++ preprocessor macro
13584 at /home/jimb/gdb/macros/play/sample.c:12
13585 (@value{GDBP}) next
13587 14 printf ("Goodbye, world!\n");
13588 (@value{GDBP}) info macro N
13589 Defined at /home/jimb/gdb/macros/play/sample.c:13
13591 (@value{GDBP}) macro expand N Q M
13592 expands to: 1729 < 42
13593 (@value{GDBP}) print N Q M
13598 In addition to source files, macros can be defined on the compilation command
13599 line using the @option{-D@var{name}=@var{value}} syntax. For macros defined in
13600 such a way, @value{GDBN} displays the location of their definition as line zero
13601 of the source file submitted to the compiler.
13604 (@value{GDBP}) info macro __STDC__
13605 Defined at /home/jimb/gdb/macros/play/sample.c:0
13612 @chapter Tracepoints
13613 @c This chapter is based on the documentation written by Michael
13614 @c Snyder, David Taylor, Jim Blandy, and Elena Zannoni.
13616 @cindex tracepoints
13617 In some applications, it is not feasible for the debugger to interrupt
13618 the program's execution long enough for the developer to learn
13619 anything helpful about its behavior. If the program's correctness
13620 depends on its real-time behavior, delays introduced by a debugger
13621 might cause the program to change its behavior drastically, or perhaps
13622 fail, even when the code itself is correct. It is useful to be able
13623 to observe the program's behavior without interrupting it.
13625 Using @value{GDBN}'s @code{trace} and @code{collect} commands, you can
13626 specify locations in the program, called @dfn{tracepoints}, and
13627 arbitrary expressions to evaluate when those tracepoints are reached.
13628 Later, using the @code{tfind} command, you can examine the values
13629 those expressions had when the program hit the tracepoints. The
13630 expressions may also denote objects in memory---structures or arrays,
13631 for example---whose values @value{GDBN} should record; while visiting
13632 a particular tracepoint, you may inspect those objects as if they were
13633 in memory at that moment. However, because @value{GDBN} records these
13634 values without interacting with you, it can do so quickly and
13635 unobtrusively, hopefully not disturbing the program's behavior.
13637 The tracepoint facility is currently available only for remote
13638 targets. @xref{Targets}. In addition, your remote target must know
13639 how to collect trace data. This functionality is implemented in the
13640 remote stub; however, none of the stubs distributed with @value{GDBN}
13641 support tracepoints as of this writing. The format of the remote
13642 packets used to implement tracepoints are described in @ref{Tracepoint
13645 It is also possible to get trace data from a file, in a manner reminiscent
13646 of corefiles; you specify the filename, and use @code{tfind} to search
13647 through the file. @xref{Trace Files}, for more details.
13649 This chapter describes the tracepoint commands and features.
13652 * Set Tracepoints::
13653 * Analyze Collected Data::
13654 * Tracepoint Variables::
13658 @node Set Tracepoints
13659 @section Commands to Set Tracepoints
13661 Before running such a @dfn{trace experiment}, an arbitrary number of
13662 tracepoints can be set. A tracepoint is actually a special type of
13663 breakpoint (@pxref{Set Breaks}), so you can manipulate it using
13664 standard breakpoint commands. For instance, as with breakpoints,
13665 tracepoint numbers are successive integers starting from one, and many
13666 of the commands associated with tracepoints take the tracepoint number
13667 as their argument, to identify which tracepoint to work on.
13669 For each tracepoint, you can specify, in advance, some arbitrary set
13670 of data that you want the target to collect in the trace buffer when
13671 it hits that tracepoint. The collected data can include registers,
13672 local variables, or global data. Later, you can use @value{GDBN}
13673 commands to examine the values these data had at the time the
13674 tracepoint was hit.
13676 Tracepoints do not support every breakpoint feature. Ignore counts on
13677 tracepoints have no effect, and tracepoints cannot run @value{GDBN}
13678 commands when they are hit. Tracepoints may not be thread-specific
13681 @cindex fast tracepoints
13682 Some targets may support @dfn{fast tracepoints}, which are inserted in
13683 a different way (such as with a jump instead of a trap), that is
13684 faster but possibly restricted in where they may be installed.
13686 @cindex static tracepoints
13687 @cindex markers, static tracepoints
13688 @cindex probing markers, static tracepoints
13689 Regular and fast tracepoints are dynamic tracing facilities, meaning
13690 that they can be used to insert tracepoints at (almost) any location
13691 in the target. Some targets may also support controlling @dfn{static
13692 tracepoints} from @value{GDBN}. With static tracing, a set of
13693 instrumentation points, also known as @dfn{markers}, are embedded in
13694 the target program, and can be activated or deactivated by name or
13695 address. These are usually placed at locations which facilitate
13696 investigating what the target is actually doing. @value{GDBN}'s
13697 support for static tracing includes being able to list instrumentation
13698 points, and attach them with @value{GDBN} defined high level
13699 tracepoints that expose the whole range of convenience of
13700 @value{GDBN}'s tracepoints support. Namely, support for collecting
13701 registers values and values of global or local (to the instrumentation
13702 point) variables; tracepoint conditions and trace state variables.
13703 The act of installing a @value{GDBN} static tracepoint on an
13704 instrumentation point, or marker, is referred to as @dfn{probing} a
13705 static tracepoint marker.
13707 @code{gdbserver} supports tracepoints on some target systems.
13708 @xref{Server,,Tracepoints support in @code{gdbserver}}.
13710 This section describes commands to set tracepoints and associated
13711 conditions and actions.
13714 * Create and Delete Tracepoints::
13715 * Enable and Disable Tracepoints::
13716 * Tracepoint Passcounts::
13717 * Tracepoint Conditions::
13718 * Trace State Variables::
13719 * Tracepoint Actions::
13720 * Listing Tracepoints::
13721 * Listing Static Tracepoint Markers::
13722 * Starting and Stopping Trace Experiments::
13723 * Tracepoint Restrictions::
13726 @node Create and Delete Tracepoints
13727 @subsection Create and Delete Tracepoints
13730 @cindex set tracepoint
13732 @item trace @var{location}
13733 The @code{trace} command is very similar to the @code{break} command.
13734 Its argument @var{location} can be any valid location.
13735 @xref{Specify Location}. The @code{trace} command defines a tracepoint,
13736 which is a point in the target program where the debugger will briefly stop,
13737 collect some data, and then allow the program to continue. Setting a tracepoint
13738 or changing its actions takes effect immediately if the remote stub
13739 supports the @samp{InstallInTrace} feature (@pxref{install tracepoint
13741 If remote stub doesn't support the @samp{InstallInTrace} feature, all
13742 these changes don't take effect until the next @code{tstart}
13743 command, and once a trace experiment is running, further changes will
13744 not have any effect until the next trace experiment starts. In addition,
13745 @value{GDBN} supports @dfn{pending tracepoints}---tracepoints whose
13746 address is not yet resolved. (This is similar to pending breakpoints.)
13747 Pending tracepoints are not downloaded to the target and not installed
13748 until they are resolved. The resolution of pending tracepoints requires
13749 @value{GDBN} support---when debugging with the remote target, and
13750 @value{GDBN} disconnects from the remote stub (@pxref{disconnected
13751 tracing}), pending tracepoints can not be resolved (and downloaded to
13752 the remote stub) while @value{GDBN} is disconnected.
13754 Here are some examples of using the @code{trace} command:
13757 (@value{GDBP}) @b{trace foo.c:121} // a source file and line number
13759 (@value{GDBP}) @b{trace +2} // 2 lines forward
13761 (@value{GDBP}) @b{trace my_function} // first source line of function
13763 (@value{GDBP}) @b{trace *my_function} // EXACT start address of function
13765 (@value{GDBP}) @b{trace *0x2117c4} // an address
13769 You can abbreviate @code{trace} as @code{tr}.
13771 @item trace @var{location} if @var{cond}
13772 Set a tracepoint with condition @var{cond}; evaluate the expression
13773 @var{cond} each time the tracepoint is reached, and collect data only
13774 if the value is nonzero---that is, if @var{cond} evaluates as true.
13775 @xref{Tracepoint Conditions, ,Tracepoint Conditions}, for more
13776 information on tracepoint conditions.
13778 @item ftrace @var{location} [ if @var{cond} ]
13779 @cindex set fast tracepoint
13780 @cindex fast tracepoints, setting
13782 The @code{ftrace} command sets a fast tracepoint. For targets that
13783 support them, fast tracepoints will use a more efficient but possibly
13784 less general technique to trigger data collection, such as a jump
13785 instruction instead of a trap, or some sort of hardware support. It
13786 may not be possible to create a fast tracepoint at the desired
13787 location, in which case the command will exit with an explanatory
13790 @value{GDBN} handles arguments to @code{ftrace} exactly as for
13793 On 32-bit x86-architecture systems, fast tracepoints normally need to
13794 be placed at an instruction that is 5 bytes or longer, but can be
13795 placed at 4-byte instructions if the low 64K of memory of the target
13796 program is available to install trampolines. Some Unix-type systems,
13797 such as @sc{gnu}/Linux, exclude low addresses from the program's
13798 address space; but for instance with the Linux kernel it is possible
13799 to let @value{GDBN} use this area by doing a @command{sysctl} command
13800 to set the @code{mmap_min_addr} kernel parameter, as in
13803 sudo sysctl -w vm.mmap_min_addr=32768
13807 which sets the low address to 32K, which leaves plenty of room for
13808 trampolines. The minimum address should be set to a page boundary.
13810 @item strace @var{location} [ if @var{cond} ]
13811 @cindex set static tracepoint
13812 @cindex static tracepoints, setting
13813 @cindex probe static tracepoint marker
13815 The @code{strace} command sets a static tracepoint. For targets that
13816 support it, setting a static tracepoint probes a static
13817 instrumentation point, or marker, found at @var{location}. It may not
13818 be possible to set a static tracepoint at the desired location, in
13819 which case the command will exit with an explanatory message.
13821 @value{GDBN} handles arguments to @code{strace} exactly as for
13822 @code{trace}, with the addition that the user can also specify
13823 @code{-m @var{marker}} as @var{location}. This probes the marker
13824 identified by the @var{marker} string identifier. This identifier
13825 depends on the static tracepoint backend library your program is
13826 using. You can find all the marker identifiers in the @samp{ID} field
13827 of the @code{info static-tracepoint-markers} command output.
13828 @xref{Listing Static Tracepoint Markers,,Listing Static Tracepoint
13829 Markers}. For example, in the following small program using the UST
13835 trace_mark(ust, bar33, "str %s", "FOOBAZ");
13840 the marker id is composed of joining the first two arguments to the
13841 @code{trace_mark} call with a slash, which translates to:
13844 (@value{GDBP}) info static-tracepoint-markers
13845 Cnt Enb ID Address What
13846 1 n ust/bar33 0x0000000000400ddc in main at stexample.c:22
13852 so you may probe the marker above with:
13855 (@value{GDBP}) strace -m ust/bar33
13858 Static tracepoints accept an extra collect action --- @code{collect
13859 $_sdata}. This collects arbitrary user data passed in the probe point
13860 call to the tracing library. In the UST example above, you'll see
13861 that the third argument to @code{trace_mark} is a printf-like format
13862 string. The user data is then the result of running that formating
13863 string against the following arguments. Note that @code{info
13864 static-tracepoint-markers} command output lists that format string in
13865 the @samp{Data:} field.
13867 You can inspect this data when analyzing the trace buffer, by printing
13868 the $_sdata variable like any other variable available to
13869 @value{GDBN}. @xref{Tracepoint Actions,,Tracepoint Action Lists}.
13872 @cindex last tracepoint number
13873 @cindex recent tracepoint number
13874 @cindex tracepoint number
13875 The convenience variable @code{$tpnum} records the tracepoint number
13876 of the most recently set tracepoint.
13878 @kindex delete tracepoint
13879 @cindex tracepoint deletion
13880 @item delete tracepoint @r{[}@var{num}@r{]}
13881 Permanently delete one or more tracepoints. With no argument, the
13882 default is to delete all tracepoints. Note that the regular
13883 @code{delete} command can remove tracepoints also.
13888 (@value{GDBP}) @b{delete trace 1 2 3} // remove three tracepoints
13890 (@value{GDBP}) @b{delete trace} // remove all tracepoints
13894 You can abbreviate this command as @code{del tr}.
13897 @node Enable and Disable Tracepoints
13898 @subsection Enable and Disable Tracepoints
13900 These commands are deprecated; they are equivalent to plain @code{disable} and @code{enable}.
13903 @kindex disable tracepoint
13904 @item disable tracepoint @r{[}@var{num}@r{]}
13905 Disable tracepoint @var{num}, or all tracepoints if no argument
13906 @var{num} is given. A disabled tracepoint will have no effect during
13907 a trace experiment, but it is not forgotten. You can re-enable
13908 a disabled tracepoint using the @code{enable tracepoint} command.
13909 If the command is issued during a trace experiment and the debug target
13910 has support for disabling tracepoints during a trace experiment, then the
13911 change will be effective immediately. Otherwise, it will be applied to the
13912 next trace experiment.
13914 @kindex enable tracepoint
13915 @item enable tracepoint @r{[}@var{num}@r{]}
13916 Enable tracepoint @var{num}, or all tracepoints. If this command is
13917 issued during a trace experiment and the debug target supports enabling
13918 tracepoints during a trace experiment, then the enabled tracepoints will
13919 become effective immediately. Otherwise, they will become effective the
13920 next time a trace experiment is run.
13923 @node Tracepoint Passcounts
13924 @subsection Tracepoint Passcounts
13928 @cindex tracepoint pass count
13929 @item passcount @r{[}@var{n} @r{[}@var{num}@r{]]}
13930 Set the @dfn{passcount} of a tracepoint. The passcount is a way to
13931 automatically stop a trace experiment. If a tracepoint's passcount is
13932 @var{n}, then the trace experiment will be automatically stopped on
13933 the @var{n}'th time that tracepoint is hit. If the tracepoint number
13934 @var{num} is not specified, the @code{passcount} command sets the
13935 passcount of the most recently defined tracepoint. If no passcount is
13936 given, the trace experiment will run until stopped explicitly by the
13942 (@value{GDBP}) @b{passcount 5 2} // Stop on the 5th execution of
13943 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// tracepoint 2}
13945 (@value{GDBP}) @b{passcount 12} // Stop on the 12th execution of the
13946 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// most recently defined tracepoint.}
13947 (@value{GDBP}) @b{trace foo}
13948 (@value{GDBP}) @b{pass 3}
13949 (@value{GDBP}) @b{trace bar}
13950 (@value{GDBP}) @b{pass 2}
13951 (@value{GDBP}) @b{trace baz}
13952 (@value{GDBP}) @b{pass 1} // Stop tracing when foo has been
13953 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// executed 3 times OR when bar has}
13954 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// been executed 2 times}
13955 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// OR when baz has been executed 1 time.}
13959 @node Tracepoint Conditions
13960 @subsection Tracepoint Conditions
13961 @cindex conditional tracepoints
13962 @cindex tracepoint conditions
13964 The simplest sort of tracepoint collects data every time your program
13965 reaches a specified place. You can also specify a @dfn{condition} for
13966 a tracepoint. A condition is just a Boolean expression in your
13967 programming language (@pxref{Expressions, ,Expressions}). A
13968 tracepoint with a condition evaluates the expression each time your
13969 program reaches it, and data collection happens only if the condition
13972 Tracepoint conditions can be specified when a tracepoint is set, by
13973 using @samp{if} in the arguments to the @code{trace} command.
13974 @xref{Create and Delete Tracepoints, ,Setting Tracepoints}. They can
13975 also be set or changed at any time with the @code{condition} command,
13976 just as with breakpoints.
13978 Unlike breakpoint conditions, @value{GDBN} does not actually evaluate
13979 the conditional expression itself. Instead, @value{GDBN} encodes the
13980 expression into an agent expression (@pxref{Agent Expressions})
13981 suitable for execution on the target, independently of @value{GDBN}.
13982 Global variables become raw memory locations, locals become stack
13983 accesses, and so forth.
13985 For instance, suppose you have a function that is usually called
13986 frequently, but should not be called after an error has occurred. You
13987 could use the following tracepoint command to collect data about calls
13988 of that function that happen while the error code is propagating
13989 through the program; an unconditional tracepoint could end up
13990 collecting thousands of useless trace frames that you would have to
13994 (@value{GDBP}) @kbd{trace normal_operation if errcode > 0}
13997 @node Trace State Variables
13998 @subsection Trace State Variables
13999 @cindex trace state variables
14001 A @dfn{trace state variable} is a special type of variable that is
14002 created and managed by target-side code. The syntax is the same as
14003 that for GDB's convenience variables (a string prefixed with ``$''),
14004 but they are stored on the target. They must be created explicitly,
14005 using a @code{tvariable} command. They are always 64-bit signed
14008 Trace state variables are remembered by @value{GDBN}, and downloaded
14009 to the target along with tracepoint information when the trace
14010 experiment starts. There are no intrinsic limits on the number of
14011 trace state variables, beyond memory limitations of the target.
14013 @cindex convenience variables, and trace state variables
14014 Although trace state variables are managed by the target, you can use
14015 them in print commands and expressions as if they were convenience
14016 variables; @value{GDBN} will get the current value from the target
14017 while the trace experiment is running. Trace state variables share
14018 the same namespace as other ``$'' variables, which means that you
14019 cannot have trace state variables with names like @code{$23} or
14020 @code{$pc}, nor can you have a trace state variable and a convenience
14021 variable with the same name.
14025 @item tvariable $@var{name} [ = @var{expression} ]
14027 The @code{tvariable} command creates a new trace state variable named
14028 @code{$@var{name}}, and optionally gives it an initial value of
14029 @var{expression}. The @var{expression} is evaluated when this command is
14030 entered; the result will be converted to an integer if possible,
14031 otherwise @value{GDBN} will report an error. A subsequent
14032 @code{tvariable} command specifying the same name does not create a
14033 variable, but instead assigns the supplied initial value to the
14034 existing variable of that name, overwriting any previous initial
14035 value. The default initial value is 0.
14037 @item info tvariables
14038 @kindex info tvariables
14039 List all the trace state variables along with their initial values.
14040 Their current values may also be displayed, if the trace experiment is
14043 @item delete tvariable @r{[} $@var{name} @dots{} @r{]}
14044 @kindex delete tvariable
14045 Delete the given trace state variables, or all of them if no arguments
14050 @node Tracepoint Actions
14051 @subsection Tracepoint Action Lists
14055 @cindex tracepoint actions
14056 @item actions @r{[}@var{num}@r{]}
14057 This command will prompt for a list of actions to be taken when the
14058 tracepoint is hit. If the tracepoint number @var{num} is not
14059 specified, this command sets the actions for the one that was most
14060 recently defined (so that you can define a tracepoint and then say
14061 @code{actions} without bothering about its number). You specify the
14062 actions themselves on the following lines, one action at a time, and
14063 terminate the actions list with a line containing just @code{end}. So
14064 far, the only defined actions are @code{collect}, @code{teval}, and
14065 @code{while-stepping}.
14067 @code{actions} is actually equivalent to @code{commands} (@pxref{Break
14068 Commands, ,Breakpoint Command Lists}), except that only the defined
14069 actions are allowed; any other @value{GDBN} command is rejected.
14071 @cindex remove actions from a tracepoint
14072 To remove all actions from a tracepoint, type @samp{actions @var{num}}
14073 and follow it immediately with @samp{end}.
14076 (@value{GDBP}) @b{collect @var{data}} // collect some data
14078 (@value{GDBP}) @b{while-stepping 5} // single-step 5 times, collect data
14080 (@value{GDBP}) @b{end} // signals the end of actions.
14083 In the following example, the action list begins with @code{collect}
14084 commands indicating the things to be collected when the tracepoint is
14085 hit. Then, in order to single-step and collect additional data
14086 following the tracepoint, a @code{while-stepping} command is used,
14087 followed by the list of things to be collected after each step in a
14088 sequence of single steps. The @code{while-stepping} command is
14089 terminated by its own separate @code{end} command. Lastly, the action
14090 list is terminated by an @code{end} command.
14093 (@value{GDBP}) @b{trace foo}
14094 (@value{GDBP}) @b{actions}
14095 Enter actions for tracepoint 1, one per line:
14098 > while-stepping 12
14099 > collect $pc, arr[i]
14104 @kindex collect @r{(tracepoints)}
14105 @item collect@r{[}/@var{mods}@r{]} @var{expr1}, @var{expr2}, @dots{}
14106 Collect values of the given expressions when the tracepoint is hit.
14107 This command accepts a comma-separated list of any valid expressions.
14108 In addition to global, static, or local variables, the following
14109 special arguments are supported:
14113 Collect all registers.
14116 Collect all function arguments.
14119 Collect all local variables.
14122 Collect the return address. This is helpful if you want to see more
14125 @emph{Note:} The return address location can not always be reliably
14126 determined up front, and the wrong address / registers may end up
14127 collected instead. On some architectures the reliability is higher
14128 for tracepoints at function entry, while on others it's the opposite.
14129 When this happens, backtracing will stop because the return address is
14130 found unavailable (unless another collect rule happened to match it).
14133 Collects the number of arguments from the static probe at which the
14134 tracepoint is located.
14135 @xref{Static Probe Points}.
14137 @item $_probe_arg@var{n}
14138 @var{n} is an integer between 0 and 11. Collects the @var{n}th argument
14139 from the static probe at which the tracepoint is located.
14140 @xref{Static Probe Points}.
14143 @vindex $_sdata@r{, collect}
14144 Collect static tracepoint marker specific data. Only available for
14145 static tracepoints. @xref{Tracepoint Actions,,Tracepoint Action
14146 Lists}. On the UST static tracepoints library backend, an
14147 instrumentation point resembles a @code{printf} function call. The
14148 tracing library is able to collect user specified data formatted to a
14149 character string using the format provided by the programmer that
14150 instrumented the program. Other backends have similar mechanisms.
14151 Here's an example of a UST marker call:
14154 const char master_name[] = "$your_name";
14155 trace_mark(channel1, marker1, "hello %s", master_name)
14158 In this case, collecting @code{$_sdata} collects the string
14159 @samp{hello $yourname}. When analyzing the trace buffer, you can
14160 inspect @samp{$_sdata} like any other variable available to
14164 You can give several consecutive @code{collect} commands, each one
14165 with a single argument, or one @code{collect} command with several
14166 arguments separated by commas; the effect is the same.
14168 The optional @var{mods} changes the usual handling of the arguments.
14169 @code{s} requests that pointers to chars be handled as strings, in
14170 particular collecting the contents of the memory being pointed at, up
14171 to the first zero. The upper bound is by default the value of the
14172 @code{print elements} variable; if @code{s} is followed by a decimal
14173 number, that is the upper bound instead. So for instance
14174 @samp{collect/s25 mystr} collects as many as 25 characters at
14177 The command @code{info scope} (@pxref{Symbols, info scope}) is
14178 particularly useful for figuring out what data to collect.
14180 @kindex teval @r{(tracepoints)}
14181 @item teval @var{expr1}, @var{expr2}, @dots{}
14182 Evaluate the given expressions when the tracepoint is hit. This
14183 command accepts a comma-separated list of expressions. The results
14184 are discarded, so this is mainly useful for assigning values to trace
14185 state variables (@pxref{Trace State Variables}) without adding those
14186 values to the trace buffer, as would be the case if the @code{collect}
14189 @kindex while-stepping @r{(tracepoints)}
14190 @item while-stepping @var{n}
14191 Perform @var{n} single-step instruction traces after the tracepoint,
14192 collecting new data after each step. The @code{while-stepping}
14193 command is followed by the list of what to collect while stepping
14194 (followed by its own @code{end} command):
14197 > while-stepping 12
14198 > collect $regs, myglobal
14204 Note that @code{$pc} is not automatically collected by
14205 @code{while-stepping}; you need to explicitly collect that register if
14206 you need it. You may abbreviate @code{while-stepping} as @code{ws} or
14209 @item set default-collect @var{expr1}, @var{expr2}, @dots{}
14210 @kindex set default-collect
14211 @cindex default collection action
14212 This variable is a list of expressions to collect at each tracepoint
14213 hit. It is effectively an additional @code{collect} action prepended
14214 to every tracepoint action list. The expressions are parsed
14215 individually for each tracepoint, so for instance a variable named
14216 @code{xyz} may be interpreted as a global for one tracepoint, and a
14217 local for another, as appropriate to the tracepoint's location.
14219 @item show default-collect
14220 @kindex show default-collect
14221 Show the list of expressions that are collected by default at each
14226 @node Listing Tracepoints
14227 @subsection Listing Tracepoints
14230 @kindex info tracepoints @r{[}@var{n}@dots{}@r{]}
14231 @kindex info tp @r{[}@var{n}@dots{}@r{]}
14232 @cindex information about tracepoints
14233 @item info tracepoints @r{[}@var{num}@dots{}@r{]}
14234 Display information about the tracepoint @var{num}. If you don't
14235 specify a tracepoint number, displays information about all the
14236 tracepoints defined so far. The format is similar to that used for
14237 @code{info breakpoints}; in fact, @code{info tracepoints} is the same
14238 command, simply restricting itself to tracepoints.
14240 A tracepoint's listing may include additional information specific to
14245 its passcount as given by the @code{passcount @var{n}} command
14248 the state about installed on target of each location
14252 (@value{GDBP}) @b{info trace}
14253 Num Type Disp Enb Address What
14254 1 tracepoint keep y 0x0804ab57 in foo() at main.cxx:7
14256 collect globfoo, $regs
14261 2 tracepoint keep y <MULTIPLE>
14263 2.1 y 0x0804859c in func4 at change-loc.h:35
14264 installed on target
14265 2.2 y 0xb7ffc480 in func4 at change-loc.h:35
14266 installed on target
14267 2.3 y <PENDING> set_tracepoint
14268 3 tracepoint keep y 0x080485b1 in foo at change-loc.c:29
14269 not installed on target
14274 This command can be abbreviated @code{info tp}.
14277 @node Listing Static Tracepoint Markers
14278 @subsection Listing Static Tracepoint Markers
14281 @kindex info static-tracepoint-markers
14282 @cindex information about static tracepoint markers
14283 @item info static-tracepoint-markers
14284 Display information about all static tracepoint markers defined in the
14287 For each marker, the following columns are printed:
14291 An incrementing counter, output to help readability. This is not a
14294 The marker ID, as reported by the target.
14295 @item Enabled or Disabled
14296 Probed markers are tagged with @samp{y}. @samp{n} identifies marks
14297 that are not enabled.
14299 Where the marker is in your program, as a memory address.
14301 Where the marker is in the source for your program, as a file and line
14302 number. If the debug information included in the program does not
14303 allow @value{GDBN} to locate the source of the marker, this column
14304 will be left blank.
14308 In addition, the following information may be printed for each marker:
14312 User data passed to the tracing library by the marker call. In the
14313 UST backend, this is the format string passed as argument to the
14315 @item Static tracepoints probing the marker
14316 The list of static tracepoints attached to the marker.
14320 (@value{GDBP}) info static-tracepoint-markers
14321 Cnt ID Enb Address What
14322 1 ust/bar2 y 0x0000000000400e1a in main at stexample.c:25
14323 Data: number1 %d number2 %d
14324 Probed by static tracepoints: #2
14325 2 ust/bar33 n 0x0000000000400c87 in main at stexample.c:24
14331 @node Starting and Stopping Trace Experiments
14332 @subsection Starting and Stopping Trace Experiments
14335 @kindex tstart [ @var{notes} ]
14336 @cindex start a new trace experiment
14337 @cindex collected data discarded
14339 This command starts the trace experiment, and begins collecting data.
14340 It has the side effect of discarding all the data collected in the
14341 trace buffer during the previous trace experiment. If any arguments
14342 are supplied, they are taken as a note and stored with the trace
14343 experiment's state. The notes may be arbitrary text, and are
14344 especially useful with disconnected tracing in a multi-user context;
14345 the notes can explain what the trace is doing, supply user contact
14346 information, and so forth.
14348 @kindex tstop [ @var{notes} ]
14349 @cindex stop a running trace experiment
14351 This command stops the trace experiment. If any arguments are
14352 supplied, they are recorded with the experiment as a note. This is
14353 useful if you are stopping a trace started by someone else, for
14354 instance if the trace is interfering with the system's behavior and
14355 needs to be stopped quickly.
14357 @strong{Note}: a trace experiment and data collection may stop
14358 automatically if any tracepoint's passcount is reached
14359 (@pxref{Tracepoint Passcounts}), or if the trace buffer becomes full.
14362 @cindex status of trace data collection
14363 @cindex trace experiment, status of
14365 This command displays the status of the current trace data
14369 Here is an example of the commands we described so far:
14372 (@value{GDBP}) @b{trace gdb_c_test}
14373 (@value{GDBP}) @b{actions}
14374 Enter actions for tracepoint #1, one per line.
14375 > collect $regs,$locals,$args
14376 > while-stepping 11
14380 (@value{GDBP}) @b{tstart}
14381 [time passes @dots{}]
14382 (@value{GDBP}) @b{tstop}
14385 @anchor{disconnected tracing}
14386 @cindex disconnected tracing
14387 You can choose to continue running the trace experiment even if
14388 @value{GDBN} disconnects from the target, voluntarily or
14389 involuntarily. For commands such as @code{detach}, the debugger will
14390 ask what you want to do with the trace. But for unexpected
14391 terminations (@value{GDBN} crash, network outage), it would be
14392 unfortunate to lose hard-won trace data, so the variable
14393 @code{disconnected-tracing} lets you decide whether the trace should
14394 continue running without @value{GDBN}.
14397 @item set disconnected-tracing on
14398 @itemx set disconnected-tracing off
14399 @kindex set disconnected-tracing
14400 Choose whether a tracing run should continue to run if @value{GDBN}
14401 has disconnected from the target. Note that @code{detach} or
14402 @code{quit} will ask you directly what to do about a running trace no
14403 matter what this variable's setting, so the variable is mainly useful
14404 for handling unexpected situations, such as loss of the network.
14406 @item show disconnected-tracing
14407 @kindex show disconnected-tracing
14408 Show the current choice for disconnected tracing.
14412 When you reconnect to the target, the trace experiment may or may not
14413 still be running; it might have filled the trace buffer in the
14414 meantime, or stopped for one of the other reasons. If it is running,
14415 it will continue after reconnection.
14417 Upon reconnection, the target will upload information about the
14418 tracepoints in effect. @value{GDBN} will then compare that
14419 information to the set of tracepoints currently defined, and attempt
14420 to match them up, allowing for the possibility that the numbers may
14421 have changed due to creation and deletion in the meantime. If one of
14422 the target's tracepoints does not match any in @value{GDBN}, the
14423 debugger will create a new tracepoint, so that you have a number with
14424 which to specify that tracepoint. This matching-up process is
14425 necessarily heuristic, and it may result in useless tracepoints being
14426 created; you may simply delete them if they are of no use.
14428 @cindex circular trace buffer
14429 If your target agent supports a @dfn{circular trace buffer}, then you
14430 can run a trace experiment indefinitely without filling the trace
14431 buffer; when space runs out, the agent deletes already-collected trace
14432 frames, oldest first, until there is enough room to continue
14433 collecting. This is especially useful if your tracepoints are being
14434 hit too often, and your trace gets terminated prematurely because the
14435 buffer is full. To ask for a circular trace buffer, simply set
14436 @samp{circular-trace-buffer} to on. You can set this at any time,
14437 including during tracing; if the agent can do it, it will change
14438 buffer handling on the fly, otherwise it will not take effect until
14442 @item set circular-trace-buffer on
14443 @itemx set circular-trace-buffer off
14444 @kindex set circular-trace-buffer
14445 Choose whether a tracing run should use a linear or circular buffer
14446 for trace data. A linear buffer will not lose any trace data, but may
14447 fill up prematurely, while a circular buffer will discard old trace
14448 data, but it will have always room for the latest tracepoint hits.
14450 @item show circular-trace-buffer
14451 @kindex show circular-trace-buffer
14452 Show the current choice for the trace buffer. Note that this may not
14453 match the agent's current buffer handling, nor is it guaranteed to
14454 match the setting that might have been in effect during a past run,
14455 for instance if you are looking at frames from a trace file.
14460 @item set trace-buffer-size @var{n}
14461 @itemx set trace-buffer-size unlimited
14462 @kindex set trace-buffer-size
14463 Request that the target use a trace buffer of @var{n} bytes. Not all
14464 targets will honor the request; they may have a compiled-in size for
14465 the trace buffer, or some other limitation. Set to a value of
14466 @code{unlimited} or @code{-1} to let the target use whatever size it
14467 likes. This is also the default.
14469 @item show trace-buffer-size
14470 @kindex show trace-buffer-size
14471 Show the current requested size for the trace buffer. Note that this
14472 will only match the actual size if the target supports size-setting,
14473 and was able to handle the requested size. For instance, if the
14474 target can only change buffer size between runs, this variable will
14475 not reflect the change until the next run starts. Use @code{tstatus}
14476 to get a report of the actual buffer size.
14480 @item set trace-user @var{text}
14481 @kindex set trace-user
14483 @item show trace-user
14484 @kindex show trace-user
14486 @item set trace-notes @var{text}
14487 @kindex set trace-notes
14488 Set the trace run's notes.
14490 @item show trace-notes
14491 @kindex show trace-notes
14492 Show the trace run's notes.
14494 @item set trace-stop-notes @var{text}
14495 @kindex set trace-stop-notes
14496 Set the trace run's stop notes. The handling of the note is as for
14497 @code{tstop} arguments; the set command is convenient way to fix a
14498 stop note that is mistaken or incomplete.
14500 @item show trace-stop-notes
14501 @kindex show trace-stop-notes
14502 Show the trace run's stop notes.
14506 @node Tracepoint Restrictions
14507 @subsection Tracepoint Restrictions
14509 @cindex tracepoint restrictions
14510 There are a number of restrictions on the use of tracepoints. As
14511 described above, tracepoint data gathering occurs on the target
14512 without interaction from @value{GDBN}. Thus the full capabilities of
14513 the debugger are not available during data gathering, and then at data
14514 examination time, you will be limited by only having what was
14515 collected. The following items describe some common problems, but it
14516 is not exhaustive, and you may run into additional difficulties not
14522 Tracepoint expressions are intended to gather objects (lvalues). Thus
14523 the full flexibility of GDB's expression evaluator is not available.
14524 You cannot call functions, cast objects to aggregate types, access
14525 convenience variables or modify values (except by assignment to trace
14526 state variables). Some language features may implicitly call
14527 functions (for instance Objective-C fields with accessors), and therefore
14528 cannot be collected either.
14531 Collection of local variables, either individually or in bulk with
14532 @code{$locals} or @code{$args}, during @code{while-stepping} may
14533 behave erratically. The stepping action may enter a new scope (for
14534 instance by stepping into a function), or the location of the variable
14535 may change (for instance it is loaded into a register). The
14536 tracepoint data recorded uses the location information for the
14537 variables that is correct for the tracepoint location. When the
14538 tracepoint is created, it is not possible, in general, to determine
14539 where the steps of a @code{while-stepping} sequence will advance the
14540 program---particularly if a conditional branch is stepped.
14543 Collection of an incompletely-initialized or partially-destroyed object
14544 may result in something that @value{GDBN} cannot display, or displays
14545 in a misleading way.
14548 When @value{GDBN} displays a pointer to character it automatically
14549 dereferences the pointer to also display characters of the string
14550 being pointed to. However, collecting the pointer during tracing does
14551 not automatically collect the string. You need to explicitly
14552 dereference the pointer and provide size information if you want to
14553 collect not only the pointer, but the memory pointed to. For example,
14554 @code{*ptr@@50} can be used to collect the 50 element array pointed to
14558 It is not possible to collect a complete stack backtrace at a
14559 tracepoint. Instead, you may collect the registers and a few hundred
14560 bytes from the stack pointer with something like @code{*(unsigned char *)$esp@@300}
14561 (adjust to use the name of the actual stack pointer register on your
14562 target architecture, and the amount of stack you wish to capture).
14563 Then the @code{backtrace} command will show a partial backtrace when
14564 using a trace frame. The number of stack frames that can be examined
14565 depends on the sizes of the frames in the collected stack. Note that
14566 if you ask for a block so large that it goes past the bottom of the
14567 stack, the target agent may report an error trying to read from an
14571 If you do not collect registers at a tracepoint, @value{GDBN} can
14572 infer that the value of @code{$pc} must be the same as the address of
14573 the tracepoint and use that when you are looking at a trace frame
14574 for that tracepoint. However, this cannot work if the tracepoint has
14575 multiple locations (for instance if it was set in a function that was
14576 inlined), or if it has a @code{while-stepping} loop. In those cases
14577 @value{GDBN} will warn you that it can't infer @code{$pc}, and default
14582 @node Analyze Collected Data
14583 @section Using the Collected Data
14585 After the tracepoint experiment ends, you use @value{GDBN} commands
14586 for examining the trace data. The basic idea is that each tracepoint
14587 collects a trace @dfn{snapshot} every time it is hit and another
14588 snapshot every time it single-steps. All these snapshots are
14589 consecutively numbered from zero and go into a buffer, and you can
14590 examine them later. The way you examine them is to @dfn{focus} on a
14591 specific trace snapshot. When the remote stub is focused on a trace
14592 snapshot, it will respond to all @value{GDBN} requests for memory and
14593 registers by reading from the buffer which belongs to that snapshot,
14594 rather than from @emph{real} memory or registers of the program being
14595 debugged. This means that @strong{all} @value{GDBN} commands
14596 (@code{print}, @code{info registers}, @code{backtrace}, etc.) will
14597 behave as if we were currently debugging the program state as it was
14598 when the tracepoint occurred. Any requests for data that are not in
14599 the buffer will fail.
14602 * tfind:: How to select a trace snapshot
14603 * tdump:: How to display all data for a snapshot
14604 * save tracepoints:: How to save tracepoints for a future run
14608 @subsection @code{tfind @var{n}}
14611 @cindex select trace snapshot
14612 @cindex find trace snapshot
14613 The basic command for selecting a trace snapshot from the buffer is
14614 @code{tfind @var{n}}, which finds trace snapshot number @var{n},
14615 counting from zero. If no argument @var{n} is given, the next
14616 snapshot is selected.
14618 Here are the various forms of using the @code{tfind} command.
14622 Find the first snapshot in the buffer. This is a synonym for
14623 @code{tfind 0} (since 0 is the number of the first snapshot).
14626 Stop debugging trace snapshots, resume @emph{live} debugging.
14629 Same as @samp{tfind none}.
14632 No argument means find the next trace snapshot or find the first
14633 one if no trace snapshot is selected.
14636 Find the previous trace snapshot before the current one. This permits
14637 retracing earlier steps.
14639 @item tfind tracepoint @var{num}
14640 Find the next snapshot associated with tracepoint @var{num}. Search
14641 proceeds forward from the last examined trace snapshot. If no
14642 argument @var{num} is given, it means find the next snapshot collected
14643 for the same tracepoint as the current snapshot.
14645 @item tfind pc @var{addr}
14646 Find the next snapshot associated with the value @var{addr} of the
14647 program counter. Search proceeds forward from the last examined trace
14648 snapshot. If no argument @var{addr} is given, it means find the next
14649 snapshot with the same value of PC as the current snapshot.
14651 @item tfind outside @var{addr1}, @var{addr2}
14652 Find the next snapshot whose PC is outside the given range of
14653 addresses (exclusive).
14655 @item tfind range @var{addr1}, @var{addr2}
14656 Find the next snapshot whose PC is between @var{addr1} and
14657 @var{addr2} (inclusive).
14659 @item tfind line @r{[}@var{file}:@r{]}@var{n}
14660 Find the next snapshot associated with the source line @var{n}. If
14661 the optional argument @var{file} is given, refer to line @var{n} in
14662 that source file. Search proceeds forward from the last examined
14663 trace snapshot. If no argument @var{n} is given, it means find the
14664 next line other than the one currently being examined; thus saying
14665 @code{tfind line} repeatedly can appear to have the same effect as
14666 stepping from line to line in a @emph{live} debugging session.
14669 The default arguments for the @code{tfind} commands are specifically
14670 designed to make it easy to scan through the trace buffer. For
14671 instance, @code{tfind} with no argument selects the next trace
14672 snapshot, and @code{tfind -} with no argument selects the previous
14673 trace snapshot. So, by giving one @code{tfind} command, and then
14674 simply hitting @key{RET} repeatedly you can examine all the trace
14675 snapshots in order. Or, by saying @code{tfind -} and then hitting
14676 @key{RET} repeatedly you can examine the snapshots in reverse order.
14677 The @code{tfind line} command with no argument selects the snapshot
14678 for the next source line executed. The @code{tfind pc} command with
14679 no argument selects the next snapshot with the same program counter
14680 (PC) as the current frame. The @code{tfind tracepoint} command with
14681 no argument selects the next trace snapshot collected by the same
14682 tracepoint as the current one.
14684 In addition to letting you scan through the trace buffer manually,
14685 these commands make it easy to construct @value{GDBN} scripts that
14686 scan through the trace buffer and print out whatever collected data
14687 you are interested in. Thus, if we want to examine the PC, FP, and SP
14688 registers from each trace frame in the buffer, we can say this:
14691 (@value{GDBP}) @b{tfind start}
14692 (@value{GDBP}) @b{while ($trace_frame != -1)}
14693 > printf "Frame %d, PC = %08X, SP = %08X, FP = %08X\n", \
14694 $trace_frame, $pc, $sp, $fp
14698 Frame 0, PC = 0020DC64, SP = 0030BF3C, FP = 0030BF44
14699 Frame 1, PC = 0020DC6C, SP = 0030BF38, FP = 0030BF44
14700 Frame 2, PC = 0020DC70, SP = 0030BF34, FP = 0030BF44
14701 Frame 3, PC = 0020DC74, SP = 0030BF30, FP = 0030BF44
14702 Frame 4, PC = 0020DC78, SP = 0030BF2C, FP = 0030BF44
14703 Frame 5, PC = 0020DC7C, SP = 0030BF28, FP = 0030BF44
14704 Frame 6, PC = 0020DC80, SP = 0030BF24, FP = 0030BF44
14705 Frame 7, PC = 0020DC84, SP = 0030BF20, FP = 0030BF44
14706 Frame 8, PC = 0020DC88, SP = 0030BF1C, FP = 0030BF44
14707 Frame 9, PC = 0020DC8E, SP = 0030BF18, FP = 0030BF44
14708 Frame 10, PC = 00203F6C, SP = 0030BE3C, FP = 0030BF14
14711 Or, if we want to examine the variable @code{X} at each source line in
14715 (@value{GDBP}) @b{tfind start}
14716 (@value{GDBP}) @b{while ($trace_frame != -1)}
14717 > printf "Frame %d, X == %d\n", $trace_frame, X
14727 @subsection @code{tdump}
14729 @cindex dump all data collected at tracepoint
14730 @cindex tracepoint data, display
14732 This command takes no arguments. It prints all the data collected at
14733 the current trace snapshot.
14736 (@value{GDBP}) @b{trace 444}
14737 (@value{GDBP}) @b{actions}
14738 Enter actions for tracepoint #2, one per line:
14739 > collect $regs, $locals, $args, gdb_long_test
14742 (@value{GDBP}) @b{tstart}
14744 (@value{GDBP}) @b{tfind line 444}
14745 #0 gdb_test (p1=0x11, p2=0x22, p3=0x33, p4=0x44, p5=0x55, p6=0x66)
14747 444 printp( "%s: arguments = 0x%X 0x%X 0x%X 0x%X 0x%X 0x%X\n", )
14749 (@value{GDBP}) @b{tdump}
14750 Data collected at tracepoint 2, trace frame 1:
14751 d0 0xc4aa0085 -995491707
14755 d4 0x71aea3d 119204413
14758 d7 0x380035 3670069
14759 a0 0x19e24a 1696330
14760 a1 0x3000668 50333288
14762 a3 0x322000 3284992
14763 a4 0x3000698 50333336
14764 a5 0x1ad3cc 1758156
14765 fp 0x30bf3c 0x30bf3c
14766 sp 0x30bf34 0x30bf34
14768 pc 0x20b2c8 0x20b2c8
14772 p = 0x20e5b4 "gdb-test"
14779 gdb_long_test = 17 '\021'
14784 @code{tdump} works by scanning the tracepoint's current collection
14785 actions and printing the value of each expression listed. So
14786 @code{tdump} can fail, if after a run, you change the tracepoint's
14787 actions to mention variables that were not collected during the run.
14789 Also, for tracepoints with @code{while-stepping} loops, @code{tdump}
14790 uses the collected value of @code{$pc} to distinguish between trace
14791 frames that were collected at the tracepoint hit, and frames that were
14792 collected while stepping. This allows it to correctly choose whether
14793 to display the basic list of collections, or the collections from the
14794 body of the while-stepping loop. However, if @code{$pc} was not collected,
14795 then @code{tdump} will always attempt to dump using the basic collection
14796 list, and may fail if a while-stepping frame does not include all the
14797 same data that is collected at the tracepoint hit.
14798 @c This is getting pretty arcane, example would be good.
14800 @node save tracepoints
14801 @subsection @code{save tracepoints @var{filename}}
14802 @kindex save tracepoints
14803 @kindex save-tracepoints
14804 @cindex save tracepoints for future sessions
14806 This command saves all current tracepoint definitions together with
14807 their actions and passcounts, into a file @file{@var{filename}}
14808 suitable for use in a later debugging session. To read the saved
14809 tracepoint definitions, use the @code{source} command (@pxref{Command
14810 Files}). The @w{@code{save-tracepoints}} command is a deprecated
14811 alias for @w{@code{save tracepoints}}
14813 @node Tracepoint Variables
14814 @section Convenience Variables for Tracepoints
14815 @cindex tracepoint variables
14816 @cindex convenience variables for tracepoints
14819 @vindex $trace_frame
14820 @item (int) $trace_frame
14821 The current trace snapshot (a.k.a.@: @dfn{frame}) number, or -1 if no
14822 snapshot is selected.
14824 @vindex $tracepoint
14825 @item (int) $tracepoint
14826 The tracepoint for the current trace snapshot.
14828 @vindex $trace_line
14829 @item (int) $trace_line
14830 The line number for the current trace snapshot.
14832 @vindex $trace_file
14833 @item (char []) $trace_file
14834 The source file for the current trace snapshot.
14836 @vindex $trace_func
14837 @item (char []) $trace_func
14838 The name of the function containing @code{$tracepoint}.
14841 Note: @code{$trace_file} is not suitable for use in @code{printf},
14842 use @code{output} instead.
14844 Here's a simple example of using these convenience variables for
14845 stepping through all the trace snapshots and printing some of their
14846 data. Note that these are not the same as trace state variables,
14847 which are managed by the target.
14850 (@value{GDBP}) @b{tfind start}
14852 (@value{GDBP}) @b{while $trace_frame != -1}
14853 > output $trace_file
14854 > printf ", line %d (tracepoint #%d)\n", $trace_line, $tracepoint
14860 @section Using Trace Files
14861 @cindex trace files
14863 In some situations, the target running a trace experiment may no
14864 longer be available; perhaps it crashed, or the hardware was needed
14865 for a different activity. To handle these cases, you can arrange to
14866 dump the trace data into a file, and later use that file as a source
14867 of trace data, via the @code{target tfile} command.
14872 @item tsave [ -r ] @var{filename}
14873 @itemx tsave [-ctf] @var{dirname}
14874 Save the trace data to @var{filename}. By default, this command
14875 assumes that @var{filename} refers to the host filesystem, so if
14876 necessary @value{GDBN} will copy raw trace data up from the target and
14877 then save it. If the target supports it, you can also supply the
14878 optional argument @code{-r} (``remote'') to direct the target to save
14879 the data directly into @var{filename} in its own filesystem, which may be
14880 more efficient if the trace buffer is very large. (Note, however, that
14881 @code{target tfile} can only read from files accessible to the host.)
14882 By default, this command will save trace frame in tfile format.
14883 You can supply the optional argument @code{-ctf} to save data in CTF
14884 format. The @dfn{Common Trace Format} (CTF) is proposed as a trace format
14885 that can be shared by multiple debugging and tracing tools. Please go to
14886 @indicateurl{http://www.efficios.com/ctf} to get more information.
14888 @kindex target tfile
14892 @item target tfile @var{filename}
14893 @itemx target ctf @var{dirname}
14894 Use the file named @var{filename} or directory named @var{dirname} as
14895 a source of trace data. Commands that examine data work as they do with
14896 a live target, but it is not possible to run any new trace experiments.
14897 @code{tstatus} will report the state of the trace run at the moment
14898 the data was saved, as well as the current trace frame you are examining.
14899 Both @var{filename} and @var{dirname} must be on a filesystem accessible to
14903 (@value{GDBP}) target ctf ctf.ctf
14904 (@value{GDBP}) tfind
14905 Found trace frame 0, tracepoint 2
14906 39 ++a; /* set tracepoint 1 here */
14907 (@value{GDBP}) tdump
14908 Data collected at tracepoint 2, trace frame 0:
14912 c = @{"123", "456", "789", "123", "456", "789"@}
14913 d = @{@{@{a = 1, b = 2@}, @{a = 3, b = 4@}@}, @{@{a = 5, b = 6@}, @{a = 7, b = 8@}@}@}
14921 @chapter Debugging Programs That Use Overlays
14924 If your program is too large to fit completely in your target system's
14925 memory, you can sometimes use @dfn{overlays} to work around this
14926 problem. @value{GDBN} provides some support for debugging programs that
14930 * How Overlays Work:: A general explanation of overlays.
14931 * Overlay Commands:: Managing overlays in @value{GDBN}.
14932 * Automatic Overlay Debugging:: @value{GDBN} can find out which overlays are
14933 mapped by asking the inferior.
14934 * Overlay Sample Program:: A sample program using overlays.
14937 @node How Overlays Work
14938 @section How Overlays Work
14939 @cindex mapped overlays
14940 @cindex unmapped overlays
14941 @cindex load address, overlay's
14942 @cindex mapped address
14943 @cindex overlay area
14945 Suppose you have a computer whose instruction address space is only 64
14946 kilobytes long, but which has much more memory which can be accessed by
14947 other means: special instructions, segment registers, or memory
14948 management hardware, for example. Suppose further that you want to
14949 adapt a program which is larger than 64 kilobytes to run on this system.
14951 One solution is to identify modules of your program which are relatively
14952 independent, and need not call each other directly; call these modules
14953 @dfn{overlays}. Separate the overlays from the main program, and place
14954 their machine code in the larger memory. Place your main program in
14955 instruction memory, but leave at least enough space there to hold the
14956 largest overlay as well.
14958 Now, to call a function located in an overlay, you must first copy that
14959 overlay's machine code from the large memory into the space set aside
14960 for it in the instruction memory, and then jump to its entry point
14963 @c NB: In the below the mapped area's size is greater or equal to the
14964 @c size of all overlays. This is intentional to remind the developer
14965 @c that overlays don't necessarily need to be the same size.
14969 Data Instruction Larger
14970 Address Space Address Space Address Space
14971 +-----------+ +-----------+ +-----------+
14973 +-----------+ +-----------+ +-----------+<-- overlay 1
14974 | program | | main | .----| overlay 1 | load address
14975 | variables | | program | | +-----------+
14976 | and heap | | | | | |
14977 +-----------+ | | | +-----------+<-- overlay 2
14978 | | +-----------+ | | | load address
14979 +-----------+ | | | .-| overlay 2 |
14981 mapped --->+-----------+ | | +-----------+
14982 address | | | | | |
14983 | overlay | <-' | | |
14984 | area | <---' +-----------+<-- overlay 3
14985 | | <---. | | load address
14986 +-----------+ `--| overlay 3 |
14993 @anchor{A code overlay}A code overlay
14997 The diagram (@pxref{A code overlay}) shows a system with separate data
14998 and instruction address spaces. To map an overlay, the program copies
14999 its code from the larger address space to the instruction address space.
15000 Since the overlays shown here all use the same mapped address, only one
15001 may be mapped at a time. For a system with a single address space for
15002 data and instructions, the diagram would be similar, except that the
15003 program variables and heap would share an address space with the main
15004 program and the overlay area.
15006 An overlay loaded into instruction memory and ready for use is called a
15007 @dfn{mapped} overlay; its @dfn{mapped address} is its address in the
15008 instruction memory. An overlay not present (or only partially present)
15009 in instruction memory is called @dfn{unmapped}; its @dfn{load address}
15010 is its address in the larger memory. The mapped address is also called
15011 the @dfn{virtual memory address}, or @dfn{VMA}; the load address is also
15012 called the @dfn{load memory address}, or @dfn{LMA}.
15014 Unfortunately, overlays are not a completely transparent way to adapt a
15015 program to limited instruction memory. They introduce a new set of
15016 global constraints you must keep in mind as you design your program:
15021 Before calling or returning to a function in an overlay, your program
15022 must make sure that overlay is actually mapped. Otherwise, the call or
15023 return will transfer control to the right address, but in the wrong
15024 overlay, and your program will probably crash.
15027 If the process of mapping an overlay is expensive on your system, you
15028 will need to choose your overlays carefully to minimize their effect on
15029 your program's performance.
15032 The executable file you load onto your system must contain each
15033 overlay's instructions, appearing at the overlay's load address, not its
15034 mapped address. However, each overlay's instructions must be relocated
15035 and its symbols defined as if the overlay were at its mapped address.
15036 You can use GNU linker scripts to specify different load and relocation
15037 addresses for pieces of your program; see @ref{Overlay Description,,,
15038 ld.info, Using ld: the GNU linker}.
15041 The procedure for loading executable files onto your system must be able
15042 to load their contents into the larger address space as well as the
15043 instruction and data spaces.
15047 The overlay system described above is rather simple, and could be
15048 improved in many ways:
15053 If your system has suitable bank switch registers or memory management
15054 hardware, you could use those facilities to make an overlay's load area
15055 contents simply appear at their mapped address in instruction space.
15056 This would probably be faster than copying the overlay to its mapped
15057 area in the usual way.
15060 If your overlays are small enough, you could set aside more than one
15061 overlay area, and have more than one overlay mapped at a time.
15064 You can use overlays to manage data, as well as instructions. In
15065 general, data overlays are even less transparent to your design than
15066 code overlays: whereas code overlays only require care when you call or
15067 return to functions, data overlays require care every time you access
15068 the data. Also, if you change the contents of a data overlay, you
15069 must copy its contents back out to its load address before you can copy a
15070 different data overlay into the same mapped area.
15075 @node Overlay Commands
15076 @section Overlay Commands
15078 To use @value{GDBN}'s overlay support, each overlay in your program must
15079 correspond to a separate section of the executable file. The section's
15080 virtual memory address and load memory address must be the overlay's
15081 mapped and load addresses. Identifying overlays with sections allows
15082 @value{GDBN} to determine the appropriate address of a function or
15083 variable, depending on whether the overlay is mapped or not.
15085 @value{GDBN}'s overlay commands all start with the word @code{overlay};
15086 you can abbreviate this as @code{ov} or @code{ovly}. The commands are:
15091 Disable @value{GDBN}'s overlay support. When overlay support is
15092 disabled, @value{GDBN} assumes that all functions and variables are
15093 always present at their mapped addresses. By default, @value{GDBN}'s
15094 overlay support is disabled.
15096 @item overlay manual
15097 @cindex manual overlay debugging
15098 Enable @dfn{manual} overlay debugging. In this mode, @value{GDBN}
15099 relies on you to tell it which overlays are mapped, and which are not,
15100 using the @code{overlay map-overlay} and @code{overlay unmap-overlay}
15101 commands described below.
15103 @item overlay map-overlay @var{overlay}
15104 @itemx overlay map @var{overlay}
15105 @cindex map an overlay
15106 Tell @value{GDBN} that @var{overlay} is now mapped; @var{overlay} must
15107 be the name of the object file section containing the overlay. When an
15108 overlay is mapped, @value{GDBN} assumes it can find the overlay's
15109 functions and variables at their mapped addresses. @value{GDBN} assumes
15110 that any other overlays whose mapped ranges overlap that of
15111 @var{overlay} are now unmapped.
15113 @item overlay unmap-overlay @var{overlay}
15114 @itemx overlay unmap @var{overlay}
15115 @cindex unmap an overlay
15116 Tell @value{GDBN} that @var{overlay} is no longer mapped; @var{overlay}
15117 must be the name of the object file section containing the overlay.
15118 When an overlay is unmapped, @value{GDBN} assumes it can find the
15119 overlay's functions and variables at their load addresses.
15122 Enable @dfn{automatic} overlay debugging. In this mode, @value{GDBN}
15123 consults a data structure the overlay manager maintains in the inferior
15124 to see which overlays are mapped. For details, see @ref{Automatic
15125 Overlay Debugging}.
15127 @item overlay load-target
15128 @itemx overlay load
15129 @cindex reloading the overlay table
15130 Re-read the overlay table from the inferior. Normally, @value{GDBN}
15131 re-reads the table @value{GDBN} automatically each time the inferior
15132 stops, so this command should only be necessary if you have changed the
15133 overlay mapping yourself using @value{GDBN}. This command is only
15134 useful when using automatic overlay debugging.
15136 @item overlay list-overlays
15137 @itemx overlay list
15138 @cindex listing mapped overlays
15139 Display a list of the overlays currently mapped, along with their mapped
15140 addresses, load addresses, and sizes.
15144 Normally, when @value{GDBN} prints a code address, it includes the name
15145 of the function the address falls in:
15148 (@value{GDBP}) print main
15149 $3 = @{int ()@} 0x11a0 <main>
15152 When overlay debugging is enabled, @value{GDBN} recognizes code in
15153 unmapped overlays, and prints the names of unmapped functions with
15154 asterisks around them. For example, if @code{foo} is a function in an
15155 unmapped overlay, @value{GDBN} prints it this way:
15158 (@value{GDBP}) overlay list
15159 No sections are mapped.
15160 (@value{GDBP}) print foo
15161 $5 = @{int (int)@} 0x100000 <*foo*>
15164 When @code{foo}'s overlay is mapped, @value{GDBN} prints the function's
15168 (@value{GDBP}) overlay list
15169 Section .ov.foo.text, loaded at 0x100000 - 0x100034,
15170 mapped at 0x1016 - 0x104a
15171 (@value{GDBP}) print foo
15172 $6 = @{int (int)@} 0x1016 <foo>
15175 When overlay debugging is enabled, @value{GDBN} can find the correct
15176 address for functions and variables in an overlay, whether or not the
15177 overlay is mapped. This allows most @value{GDBN} commands, like
15178 @code{break} and @code{disassemble}, to work normally, even on unmapped
15179 code. However, @value{GDBN}'s breakpoint support has some limitations:
15183 @cindex breakpoints in overlays
15184 @cindex overlays, setting breakpoints in
15185 You can set breakpoints in functions in unmapped overlays, as long as
15186 @value{GDBN} can write to the overlay at its load address.
15188 @value{GDBN} can not set hardware or simulator-based breakpoints in
15189 unmapped overlays. However, if you set a breakpoint at the end of your
15190 overlay manager (and tell @value{GDBN} which overlays are now mapped, if
15191 you are using manual overlay management), @value{GDBN} will re-set its
15192 breakpoints properly.
15196 @node Automatic Overlay Debugging
15197 @section Automatic Overlay Debugging
15198 @cindex automatic overlay debugging
15200 @value{GDBN} can automatically track which overlays are mapped and which
15201 are not, given some simple co-operation from the overlay manager in the
15202 inferior. If you enable automatic overlay debugging with the
15203 @code{overlay auto} command (@pxref{Overlay Commands}), @value{GDBN}
15204 looks in the inferior's memory for certain variables describing the
15205 current state of the overlays.
15207 Here are the variables your overlay manager must define to support
15208 @value{GDBN}'s automatic overlay debugging:
15212 @item @code{_ovly_table}:
15213 This variable must be an array of the following structures:
15218 /* The overlay's mapped address. */
15221 /* The size of the overlay, in bytes. */
15222 unsigned long size;
15224 /* The overlay's load address. */
15227 /* Non-zero if the overlay is currently mapped;
15229 unsigned long mapped;
15233 @item @code{_novlys}:
15234 This variable must be a four-byte signed integer, holding the total
15235 number of elements in @code{_ovly_table}.
15239 To decide whether a particular overlay is mapped or not, @value{GDBN}
15240 looks for an entry in @w{@code{_ovly_table}} whose @code{vma} and
15241 @code{lma} members equal the VMA and LMA of the overlay's section in the
15242 executable file. When @value{GDBN} finds a matching entry, it consults
15243 the entry's @code{mapped} member to determine whether the overlay is
15246 In addition, your overlay manager may define a function called
15247 @code{_ovly_debug_event}. If this function is defined, @value{GDBN}
15248 will silently set a breakpoint there. If the overlay manager then
15249 calls this function whenever it has changed the overlay table, this
15250 will enable @value{GDBN} to accurately keep track of which overlays
15251 are in program memory, and update any breakpoints that may be set
15252 in overlays. This will allow breakpoints to work even if the
15253 overlays are kept in ROM or other non-writable memory while they
15254 are not being executed.
15256 @node Overlay Sample Program
15257 @section Overlay Sample Program
15258 @cindex overlay example program
15260 When linking a program which uses overlays, you must place the overlays
15261 at their load addresses, while relocating them to run at their mapped
15262 addresses. To do this, you must write a linker script (@pxref{Overlay
15263 Description,,, ld.info, Using ld: the GNU linker}). Unfortunately,
15264 since linker scripts are specific to a particular host system, target
15265 architecture, and target memory layout, this manual cannot provide
15266 portable sample code demonstrating @value{GDBN}'s overlay support.
15268 However, the @value{GDBN} source distribution does contain an overlaid
15269 program, with linker scripts for a few systems, as part of its test
15270 suite. The program consists of the following files from
15271 @file{gdb/testsuite/gdb.base}:
15275 The main program file.
15277 A simple overlay manager, used by @file{overlays.c}.
15282 Overlay modules, loaded and used by @file{overlays.c}.
15285 Linker scripts for linking the test program on the @code{d10v-elf}
15286 and @code{m32r-elf} targets.
15289 You can build the test program using the @code{d10v-elf} GCC
15290 cross-compiler like this:
15293 $ d10v-elf-gcc -g -c overlays.c
15294 $ d10v-elf-gcc -g -c ovlymgr.c
15295 $ d10v-elf-gcc -g -c foo.c
15296 $ d10v-elf-gcc -g -c bar.c
15297 $ d10v-elf-gcc -g -c baz.c
15298 $ d10v-elf-gcc -g -c grbx.c
15299 $ d10v-elf-gcc -g overlays.o ovlymgr.o foo.o bar.o \
15300 baz.o grbx.o -Wl,-Td10v.ld -o overlays
15303 The build process is identical for any other architecture, except that
15304 you must substitute the appropriate compiler and linker script for the
15305 target system for @code{d10v-elf-gcc} and @code{d10v.ld}.
15309 @chapter Using @value{GDBN} with Different Languages
15312 Although programming languages generally have common aspects, they are
15313 rarely expressed in the same manner. For instance, in ANSI C,
15314 dereferencing a pointer @code{p} is accomplished by @code{*p}, but in
15315 Modula-2, it is accomplished by @code{p^}. Values can also be
15316 represented (and displayed) differently. Hex numbers in C appear as
15317 @samp{0x1ae}, while in Modula-2 they appear as @samp{1AEH}.
15319 @cindex working language
15320 Language-specific information is built into @value{GDBN} for some languages,
15321 allowing you to express operations like the above in your program's
15322 native language, and allowing @value{GDBN} to output values in a manner
15323 consistent with the syntax of your program's native language. The
15324 language you use to build expressions is called the @dfn{working
15328 * Setting:: Switching between source languages
15329 * Show:: Displaying the language
15330 * Checks:: Type and range checks
15331 * Supported Languages:: Supported languages
15332 * Unsupported Languages:: Unsupported languages
15336 @section Switching Between Source Languages
15338 There are two ways to control the working language---either have @value{GDBN}
15339 set it automatically, or select it manually yourself. You can use the
15340 @code{set language} command for either purpose. On startup, @value{GDBN}
15341 defaults to setting the language automatically. The working language is
15342 used to determine how expressions you type are interpreted, how values
15345 In addition to the working language, every source file that
15346 @value{GDBN} knows about has its own working language. For some object
15347 file formats, the compiler might indicate which language a particular
15348 source file is in. However, most of the time @value{GDBN} infers the
15349 language from the name of the file. The language of a source file
15350 controls whether C@t{++} names are demangled---this way @code{backtrace} can
15351 show each frame appropriately for its own language. There is no way to
15352 set the language of a source file from within @value{GDBN}, but you can
15353 set the language associated with a filename extension. @xref{Show, ,
15354 Displaying the Language}.
15356 This is most commonly a problem when you use a program, such
15357 as @code{cfront} or @code{f2c}, that generates C but is written in
15358 another language. In that case, make the
15359 program use @code{#line} directives in its C output; that way
15360 @value{GDBN} will know the correct language of the source code of the original
15361 program, and will display that source code, not the generated C code.
15364 * Filenames:: Filename extensions and languages.
15365 * Manually:: Setting the working language manually
15366 * Automatically:: Having @value{GDBN} infer the source language
15370 @subsection List of Filename Extensions and Languages
15372 If a source file name ends in one of the following extensions, then
15373 @value{GDBN} infers that its language is the one indicated.
15391 C@t{++} source file
15397 Objective-C source file
15401 Fortran source file
15404 Modula-2 source file
15408 Assembler source file. This actually behaves almost like C, but
15409 @value{GDBN} does not skip over function prologues when stepping.
15412 In addition, you may set the language associated with a filename
15413 extension. @xref{Show, , Displaying the Language}.
15416 @subsection Setting the Working Language
15418 If you allow @value{GDBN} to set the language automatically,
15419 expressions are interpreted the same way in your debugging session and
15422 @kindex set language
15423 If you wish, you may set the language manually. To do this, issue the
15424 command @samp{set language @var{lang}}, where @var{lang} is the name of
15425 a language, such as
15426 @code{c} or @code{modula-2}.
15427 For a list of the supported languages, type @samp{set language}.
15429 Setting the language manually prevents @value{GDBN} from updating the working
15430 language automatically. This can lead to confusion if you try
15431 to debug a program when the working language is not the same as the
15432 source language, when an expression is acceptable to both
15433 languages---but means different things. For instance, if the current
15434 source file were written in C, and @value{GDBN} was parsing Modula-2, a
15442 might not have the effect you intended. In C, this means to add
15443 @code{b} and @code{c} and place the result in @code{a}. The result
15444 printed would be the value of @code{a}. In Modula-2, this means to compare
15445 @code{a} to the result of @code{b+c}, yielding a @code{BOOLEAN} value.
15447 @node Automatically
15448 @subsection Having @value{GDBN} Infer the Source Language
15450 To have @value{GDBN} set the working language automatically, use
15451 @samp{set language local} or @samp{set language auto}. @value{GDBN}
15452 then infers the working language. That is, when your program stops in a
15453 frame (usually by encountering a breakpoint), @value{GDBN} sets the
15454 working language to the language recorded for the function in that
15455 frame. If the language for a frame is unknown (that is, if the function
15456 or block corresponding to the frame was defined in a source file that
15457 does not have a recognized extension), the current working language is
15458 not changed, and @value{GDBN} issues a warning.
15460 This may not seem necessary for most programs, which are written
15461 entirely in one source language. However, program modules and libraries
15462 written in one source language can be used by a main program written in
15463 a different source language. Using @samp{set language auto} in this
15464 case frees you from having to set the working language manually.
15467 @section Displaying the Language
15469 The following commands help you find out which language is the
15470 working language, and also what language source files were written in.
15473 @item show language
15474 @anchor{show language}
15475 @kindex show language
15476 Display the current working language. This is the
15477 language you can use with commands such as @code{print} to
15478 build and compute expressions that may involve variables in your program.
15481 @kindex info frame@r{, show the source language}
15482 Display the source language for this frame. This language becomes the
15483 working language if you use an identifier from this frame.
15484 @xref{Frame Info, ,Information about a Frame}, to identify the other
15485 information listed here.
15488 @kindex info source@r{, show the source language}
15489 Display the source language of this source file.
15490 @xref{Symbols, ,Examining the Symbol Table}, to identify the other
15491 information listed here.
15494 In unusual circumstances, you may have source files with extensions
15495 not in the standard list. You can then set the extension associated
15496 with a language explicitly:
15499 @item set extension-language @var{ext} @var{language}
15500 @kindex set extension-language
15501 Tell @value{GDBN} that source files with extension @var{ext} are to be
15502 assumed as written in the source language @var{language}.
15504 @item info extensions
15505 @kindex info extensions
15506 List all the filename extensions and the associated languages.
15510 @section Type and Range Checking
15512 Some languages are designed to guard you against making seemingly common
15513 errors through a series of compile- and run-time checks. These include
15514 checking the type of arguments to functions and operators and making
15515 sure mathematical overflows are caught at run time. Checks such as
15516 these help to ensure a program's correctness once it has been compiled
15517 by eliminating type mismatches and providing active checks for range
15518 errors when your program is running.
15520 By default @value{GDBN} checks for these errors according to the
15521 rules of the current source language. Although @value{GDBN} does not check
15522 the statements in your program, it can check expressions entered directly
15523 into @value{GDBN} for evaluation via the @code{print} command, for example.
15526 * Type Checking:: An overview of type checking
15527 * Range Checking:: An overview of range checking
15530 @cindex type checking
15531 @cindex checks, type
15532 @node Type Checking
15533 @subsection An Overview of Type Checking
15535 Some languages, such as C and C@t{++}, are strongly typed, meaning that the
15536 arguments to operators and functions have to be of the correct type,
15537 otherwise an error occurs. These checks prevent type mismatch
15538 errors from ever causing any run-time problems. For example,
15541 int klass::my_method(char *b) @{ return b ? 1 : 2; @}
15543 (@value{GDBP}) print obj.my_method (0)
15546 (@value{GDBP}) print obj.my_method (0x1234)
15547 Cannot resolve method klass::my_method to any overloaded instance
15550 The second example fails because in C@t{++} the integer constant
15551 @samp{0x1234} is not type-compatible with the pointer parameter type.
15553 For the expressions you use in @value{GDBN} commands, you can tell
15554 @value{GDBN} to not enforce strict type checking or
15555 to treat any mismatches as errors and abandon the expression;
15556 When type checking is disabled, @value{GDBN} successfully evaluates
15557 expressions like the second example above.
15559 Even if type checking is off, there may be other reasons
15560 related to type that prevent @value{GDBN} from evaluating an expression.
15561 For instance, @value{GDBN} does not know how to add an @code{int} and
15562 a @code{struct foo}. These particular type errors have nothing to do
15563 with the language in use and usually arise from expressions which make
15564 little sense to evaluate anyway.
15566 @value{GDBN} provides some additional commands for controlling type checking:
15568 @kindex set check type
15569 @kindex show check type
15571 @item set check type on
15572 @itemx set check type off
15573 Set strict type checking on or off. If any type mismatches occur in
15574 evaluating an expression while type checking is on, @value{GDBN} prints a
15575 message and aborts evaluation of the expression.
15577 @item show check type
15578 Show the current setting of type checking and whether @value{GDBN}
15579 is enforcing strict type checking rules.
15582 @cindex range checking
15583 @cindex checks, range
15584 @node Range Checking
15585 @subsection An Overview of Range Checking
15587 In some languages (such as Modula-2), it is an error to exceed the
15588 bounds of a type; this is enforced with run-time checks. Such range
15589 checking is meant to ensure program correctness by making sure
15590 computations do not overflow, or indices on an array element access do
15591 not exceed the bounds of the array.
15593 For expressions you use in @value{GDBN} commands, you can tell
15594 @value{GDBN} to treat range errors in one of three ways: ignore them,
15595 always treat them as errors and abandon the expression, or issue
15596 warnings but evaluate the expression anyway.
15598 A range error can result from numerical overflow, from exceeding an
15599 array index bound, or when you type a constant that is not a member
15600 of any type. Some languages, however, do not treat overflows as an
15601 error. In many implementations of C, mathematical overflow causes the
15602 result to ``wrap around'' to lower values---for example, if @var{m} is
15603 the largest integer value, and @var{s} is the smallest, then
15606 @var{m} + 1 @result{} @var{s}
15609 This, too, is specific to individual languages, and in some cases
15610 specific to individual compilers or machines. @xref{Supported Languages, ,
15611 Supported Languages}, for further details on specific languages.
15613 @value{GDBN} provides some additional commands for controlling the range checker:
15615 @kindex set check range
15616 @kindex show check range
15618 @item set check range auto
15619 Set range checking on or off based on the current working language.
15620 @xref{Supported Languages, ,Supported Languages}, for the default settings for
15623 @item set check range on
15624 @itemx set check range off
15625 Set range checking on or off, overriding the default setting for the
15626 current working language. A warning is issued if the setting does not
15627 match the language default. If a range error occurs and range checking is on,
15628 then a message is printed and evaluation of the expression is aborted.
15630 @item set check range warn
15631 Output messages when the @value{GDBN} range checker detects a range error,
15632 but attempt to evaluate the expression anyway. Evaluating the
15633 expression may still be impossible for other reasons, such as accessing
15634 memory that the process does not own (a typical example from many Unix
15638 Show the current setting of the range checker, and whether or not it is
15639 being set automatically by @value{GDBN}.
15642 @node Supported Languages
15643 @section Supported Languages
15645 @value{GDBN} supports C, C@t{++}, D, Go, Objective-C, Fortran,
15646 OpenCL C, Pascal, Rust, assembly, Modula-2, and Ada.
15647 @c This is false ...
15648 Some @value{GDBN} features may be used in expressions regardless of the
15649 language you use: the @value{GDBN} @code{@@} and @code{::} operators,
15650 and the @samp{@{type@}addr} construct (@pxref{Expressions,
15651 ,Expressions}) can be used with the constructs of any supported
15654 The following sections detail to what degree each source language is
15655 supported by @value{GDBN}. These sections are not meant to be language
15656 tutorials or references, but serve only as a reference guide to what the
15657 @value{GDBN} expression parser accepts, and what input and output
15658 formats should look like for different languages. There are many good
15659 books written on each of these languages; please look to these for a
15660 language reference or tutorial.
15663 * C:: C and C@t{++}
15666 * Objective-C:: Objective-C
15667 * OpenCL C:: OpenCL C
15668 * Fortran:: Fortran
15671 * Modula-2:: Modula-2
15676 @subsection C and C@t{++}
15678 @cindex C and C@t{++}
15679 @cindex expressions in C or C@t{++}
15681 Since C and C@t{++} are so closely related, many features of @value{GDBN} apply
15682 to both languages. Whenever this is the case, we discuss those languages
15686 @cindex @code{g++}, @sc{gnu} C@t{++} compiler
15687 @cindex @sc{gnu} C@t{++}
15688 The C@t{++} debugging facilities are jointly implemented by the C@t{++}
15689 compiler and @value{GDBN}. Therefore, to debug your C@t{++} code
15690 effectively, you must compile your C@t{++} programs with a supported
15691 C@t{++} compiler, such as @sc{gnu} @code{g++}, or the HP ANSI C@t{++}
15692 compiler (@code{aCC}).
15695 * C Operators:: C and C@t{++} operators
15696 * C Constants:: C and C@t{++} constants
15697 * C Plus Plus Expressions:: C@t{++} expressions
15698 * C Defaults:: Default settings for C and C@t{++}
15699 * C Checks:: C and C@t{++} type and range checks
15700 * Debugging C:: @value{GDBN} and C
15701 * Debugging C Plus Plus:: @value{GDBN} features for C@t{++}
15702 * Decimal Floating Point:: Numbers in Decimal Floating Point format
15706 @subsubsection C and C@t{++} Operators
15708 @cindex C and C@t{++} operators
15710 Operators must be defined on values of specific types. For instance,
15711 @code{+} is defined on numbers, but not on structures. Operators are
15712 often defined on groups of types.
15714 For the purposes of C and C@t{++}, the following definitions hold:
15719 @emph{Integral types} include @code{int} with any of its storage-class
15720 specifiers; @code{char}; @code{enum}; and, for C@t{++}, @code{bool}.
15723 @emph{Floating-point types} include @code{float}, @code{double}, and
15724 @code{long double} (if supported by the target platform).
15727 @emph{Pointer types} include all types defined as @code{(@var{type} *)}.
15730 @emph{Scalar types} include all of the above.
15735 The following operators are supported. They are listed here
15736 in order of increasing precedence:
15740 The comma or sequencing operator. Expressions in a comma-separated list
15741 are evaluated from left to right, with the result of the entire
15742 expression being the last expression evaluated.
15745 Assignment. The value of an assignment expression is the value
15746 assigned. Defined on scalar types.
15749 Used in an expression of the form @w{@code{@var{a} @var{op}= @var{b}}},
15750 and translated to @w{@code{@var{a} = @var{a op b}}}.
15751 @w{@code{@var{op}=}} and @code{=} have the same precedence. The operator
15752 @var{op} is any one of the operators @code{|}, @code{^}, @code{&},
15753 @code{<<}, @code{>>}, @code{+}, @code{-}, @code{*}, @code{/}, @code{%}.
15756 The ternary operator. @code{@var{a} ? @var{b} : @var{c}} can be thought
15757 of as: if @var{a} then @var{b} else @var{c}. The argument @var{a}
15758 should be of an integral type.
15761 Logical @sc{or}. Defined on integral types.
15764 Logical @sc{and}. Defined on integral types.
15767 Bitwise @sc{or}. Defined on integral types.
15770 Bitwise exclusive-@sc{or}. Defined on integral types.
15773 Bitwise @sc{and}. Defined on integral types.
15776 Equality and inequality. Defined on scalar types. The value of these
15777 expressions is 0 for false and non-zero for true.
15779 @item <@r{, }>@r{, }<=@r{, }>=
15780 Less than, greater than, less than or equal, greater than or equal.
15781 Defined on scalar types. The value of these expressions is 0 for false
15782 and non-zero for true.
15785 left shift, and right shift. Defined on integral types.
15788 The @value{GDBN} ``artificial array'' operator (@pxref{Expressions, ,Expressions}).
15791 Addition and subtraction. Defined on integral types, floating-point types and
15794 @item *@r{, }/@r{, }%
15795 Multiplication, division, and modulus. Multiplication and division are
15796 defined on integral and floating-point types. Modulus is defined on
15800 Increment and decrement. When appearing before a variable, the
15801 operation is performed before the variable is used in an expression;
15802 when appearing after it, the variable's value is used before the
15803 operation takes place.
15806 Pointer dereferencing. Defined on pointer types. Same precedence as
15810 Address operator. Defined on variables. Same precedence as @code{++}.
15812 For debugging C@t{++}, @value{GDBN} implements a use of @samp{&} beyond what is
15813 allowed in the C@t{++} language itself: you can use @samp{&(&@var{ref})}
15814 to examine the address
15815 where a C@t{++} reference variable (declared with @samp{&@var{ref}}) is
15819 Negative. Defined on integral and floating-point types. Same
15820 precedence as @code{++}.
15823 Logical negation. Defined on integral types. Same precedence as
15827 Bitwise complement operator. Defined on integral types. Same precedence as
15832 Structure member, and pointer-to-structure member. For convenience,
15833 @value{GDBN} regards the two as equivalent, choosing whether to dereference a
15834 pointer based on the stored type information.
15835 Defined on @code{struct} and @code{union} data.
15838 Dereferences of pointers to members.
15841 Array indexing. @code{@var{a}[@var{i}]} is defined as
15842 @code{*(@var{a}+@var{i})}. Same precedence as @code{->}.
15845 Function parameter list. Same precedence as @code{->}.
15848 C@t{++} scope resolution operator. Defined on @code{struct}, @code{union},
15849 and @code{class} types.
15852 Doubled colons also represent the @value{GDBN} scope operator
15853 (@pxref{Expressions, ,Expressions}). Same precedence as @code{::},
15857 If an operator is redefined in the user code, @value{GDBN} usually
15858 attempts to invoke the redefined version instead of using the operator's
15859 predefined meaning.
15862 @subsubsection C and C@t{++} Constants
15864 @cindex C and C@t{++} constants
15866 @value{GDBN} allows you to express the constants of C and C@t{++} in the
15871 Integer constants are a sequence of digits. Octal constants are
15872 specified by a leading @samp{0} (i.e.@: zero), and hexadecimal constants
15873 by a leading @samp{0x} or @samp{0X}. Constants may also end with a letter
15874 @samp{l}, specifying that the constant should be treated as a
15878 Floating point constants are a sequence of digits, followed by a decimal
15879 point, followed by a sequence of digits, and optionally followed by an
15880 exponent. An exponent is of the form:
15881 @samp{@w{e@r{[[}+@r{]|}-@r{]}@var{nnn}}}, where @var{nnn} is another
15882 sequence of digits. The @samp{+} is optional for positive exponents.
15883 A floating-point constant may also end with a letter @samp{f} or
15884 @samp{F}, specifying that the constant should be treated as being of
15885 the @code{float} (as opposed to the default @code{double}) type; or with
15886 a letter @samp{l} or @samp{L}, which specifies a @code{long double}
15890 Enumerated constants consist of enumerated identifiers, or their
15891 integral equivalents.
15894 Character constants are a single character surrounded by single quotes
15895 (@code{'}), or a number---the ordinal value of the corresponding character
15896 (usually its @sc{ascii} value). Within quotes, the single character may
15897 be represented by a letter or by @dfn{escape sequences}, which are of
15898 the form @samp{\@var{nnn}}, where @var{nnn} is the octal representation
15899 of the character's ordinal value; or of the form @samp{\@var{x}}, where
15900 @samp{@var{x}} is a predefined special character---for example,
15901 @samp{\n} for newline.
15903 Wide character constants can be written by prefixing a character
15904 constant with @samp{L}, as in C. For example, @samp{L'x'} is the wide
15905 form of @samp{x}. The target wide character set is used when
15906 computing the value of this constant (@pxref{Character Sets}).
15909 String constants are a sequence of character constants surrounded by
15910 double quotes (@code{"}). Any valid character constant (as described
15911 above) may appear. Double quotes within the string must be preceded by
15912 a backslash, so for instance @samp{"a\"b'c"} is a string of five
15915 Wide string constants can be written by prefixing a string constant
15916 with @samp{L}, as in C. The target wide character set is used when
15917 computing the value of this constant (@pxref{Character Sets}).
15920 Pointer constants are an integral value. You can also write pointers
15921 to constants using the C operator @samp{&}.
15924 Array constants are comma-separated lists surrounded by braces @samp{@{}
15925 and @samp{@}}; for example, @samp{@{1,2,3@}} is a three-element array of
15926 integers, @samp{@{@{1,2@}, @{3,4@}, @{5,6@}@}} is a three-by-two array,
15927 and @samp{@{&"hi", &"there", &"fred"@}} is a three-element array of pointers.
15930 @node C Plus Plus Expressions
15931 @subsubsection C@t{++} Expressions
15933 @cindex expressions in C@t{++}
15934 @value{GDBN} expression handling can interpret most C@t{++} expressions.
15936 @cindex debugging C@t{++} programs
15937 @cindex C@t{++} compilers
15938 @cindex debug formats and C@t{++}
15939 @cindex @value{NGCC} and C@t{++}
15941 @emph{Warning:} @value{GDBN} can only debug C@t{++} code if you use
15942 the proper compiler and the proper debug format. Currently,
15943 @value{GDBN} works best when debugging C@t{++} code that is compiled
15944 with the most recent version of @value{NGCC} possible. The DWARF
15945 debugging format is preferred; @value{NGCC} defaults to this on most
15946 popular platforms. Other compilers and/or debug formats are likely to
15947 work badly or not at all when using @value{GDBN} to debug C@t{++}
15948 code. @xref{Compilation}.
15953 @cindex member functions
15955 Member function calls are allowed; you can use expressions like
15958 count = aml->GetOriginal(x, y)
15961 @vindex this@r{, inside C@t{++} member functions}
15962 @cindex namespace in C@t{++}
15964 While a member function is active (in the selected stack frame), your
15965 expressions have the same namespace available as the member function;
15966 that is, @value{GDBN} allows implicit references to the class instance
15967 pointer @code{this} following the same rules as C@t{++}. @code{using}
15968 declarations in the current scope are also respected by @value{GDBN}.
15970 @cindex call overloaded functions
15971 @cindex overloaded functions, calling
15972 @cindex type conversions in C@t{++}
15974 You can call overloaded functions; @value{GDBN} resolves the function
15975 call to the right definition, with some restrictions. @value{GDBN} does not
15976 perform overload resolution involving user-defined type conversions,
15977 calls to constructors, or instantiations of templates that do not exist
15978 in the program. It also cannot handle ellipsis argument lists or
15981 It does perform integral conversions and promotions, floating-point
15982 promotions, arithmetic conversions, pointer conversions, conversions of
15983 class objects to base classes, and standard conversions such as those of
15984 functions or arrays to pointers; it requires an exact match on the
15985 number of function arguments.
15987 Overload resolution is always performed, unless you have specified
15988 @code{set overload-resolution off}. @xref{Debugging C Plus Plus,
15989 ,@value{GDBN} Features for C@t{++}}.
15991 You must specify @code{set overload-resolution off} in order to use an
15992 explicit function signature to call an overloaded function, as in
15994 p 'foo(char,int)'('x', 13)
15997 The @value{GDBN} command-completion facility can simplify this;
15998 see @ref{Completion, ,Command Completion}.
16000 @cindex reference declarations
16002 @value{GDBN} understands variables declared as C@t{++} lvalue or rvalue
16003 references; you can use them in expressions just as you do in C@t{++}
16004 source---they are automatically dereferenced.
16006 In the parameter list shown when @value{GDBN} displays a frame, the values of
16007 reference variables are not displayed (unlike other variables); this
16008 avoids clutter, since references are often used for large structures.
16009 The @emph{address} of a reference variable is always shown, unless
16010 you have specified @samp{set print address off}.
16013 @value{GDBN} supports the C@t{++} name resolution operator @code{::}---your
16014 expressions can use it just as expressions in your program do. Since
16015 one scope may be defined in another, you can use @code{::} repeatedly if
16016 necessary, for example in an expression like
16017 @samp{@var{scope1}::@var{scope2}::@var{name}}. @value{GDBN} also allows
16018 resolving name scope by reference to source files, in both C and C@t{++}
16019 debugging (@pxref{Variables, ,Program Variables}).
16022 @value{GDBN} performs argument-dependent lookup, following the C@t{++}
16027 @subsubsection C and C@t{++} Defaults
16029 @cindex C and C@t{++} defaults
16031 If you allow @value{GDBN} to set range checking automatically, it
16032 defaults to @code{off} whenever the working language changes to
16033 C or C@t{++}. This happens regardless of whether you or @value{GDBN}
16034 selects the working language.
16036 If you allow @value{GDBN} to set the language automatically, it
16037 recognizes source files whose names end with @file{.c}, @file{.C}, or
16038 @file{.cc}, etc, and when @value{GDBN} enters code compiled from one of
16039 these files, it sets the working language to C or C@t{++}.
16040 @xref{Automatically, ,Having @value{GDBN} Infer the Source Language},
16041 for further details.
16044 @subsubsection C and C@t{++} Type and Range Checks
16046 @cindex C and C@t{++} checks
16048 By default, when @value{GDBN} parses C or C@t{++} expressions, strict type
16049 checking is used. However, if you turn type checking off, @value{GDBN}
16050 will allow certain non-standard conversions, such as promoting integer
16051 constants to pointers.
16053 Range checking, if turned on, is done on mathematical operations. Array
16054 indices are not checked, since they are often used to index a pointer
16055 that is not itself an array.
16058 @subsubsection @value{GDBN} and C
16060 The @code{set print union} and @code{show print union} commands apply to
16061 the @code{union} type. When set to @samp{on}, any @code{union} that is
16062 inside a @code{struct} or @code{class} is also printed. Otherwise, it
16063 appears as @samp{@{...@}}.
16065 The @code{@@} operator aids in the debugging of dynamic arrays, formed
16066 with pointers and a memory allocation function. @xref{Expressions,
16069 @node Debugging C Plus Plus
16070 @subsubsection @value{GDBN} Features for C@t{++}
16072 @cindex commands for C@t{++}
16074 Some @value{GDBN} commands are particularly useful with C@t{++}, and some are
16075 designed specifically for use with C@t{++}. Here is a summary:
16078 @cindex break in overloaded functions
16079 @item @r{breakpoint menus}
16080 When you want a breakpoint in a function whose name is overloaded,
16081 @value{GDBN} has the capability to display a menu of possible breakpoint
16082 locations to help you specify which function definition you want.
16083 @xref{Ambiguous Expressions,,Ambiguous Expressions}.
16085 @cindex overloading in C@t{++}
16086 @item rbreak @var{regex}
16087 Setting breakpoints using regular expressions is helpful for setting
16088 breakpoints on overloaded functions that are not members of any special
16090 @xref{Set Breaks, ,Setting Breakpoints}.
16092 @cindex C@t{++} exception handling
16094 @itemx catch rethrow
16096 Debug C@t{++} exception handling using these commands. @xref{Set
16097 Catchpoints, , Setting Catchpoints}.
16099 @cindex inheritance
16100 @item ptype @var{typename}
16101 Print inheritance relationships as well as other information for type
16103 @xref{Symbols, ,Examining the Symbol Table}.
16105 @item info vtbl @var{expression}.
16106 The @code{info vtbl} command can be used to display the virtual
16107 method tables of the object computed by @var{expression}. This shows
16108 one entry per virtual table; there may be multiple virtual tables when
16109 multiple inheritance is in use.
16111 @cindex C@t{++} demangling
16112 @item demangle @var{name}
16113 Demangle @var{name}.
16114 @xref{Symbols}, for a more complete description of the @code{demangle} command.
16116 @cindex C@t{++} symbol display
16117 @item set print demangle
16118 @itemx show print demangle
16119 @itemx set print asm-demangle
16120 @itemx show print asm-demangle
16121 Control whether C@t{++} symbols display in their source form, both when
16122 displaying code as C@t{++} source and when displaying disassemblies.
16123 @xref{Print Settings, ,Print Settings}.
16125 @item set print object
16126 @itemx show print object
16127 Choose whether to print derived (actual) or declared types of objects.
16128 @xref{Print Settings, ,Print Settings}.
16130 @item set print vtbl
16131 @itemx show print vtbl
16132 Control the format for printing virtual function tables.
16133 @xref{Print Settings, ,Print Settings}.
16134 (The @code{vtbl} commands do not work on programs compiled with the HP
16135 ANSI C@t{++} compiler (@code{aCC}).)
16137 @kindex set overload-resolution
16138 @cindex overloaded functions, overload resolution
16139 @item set overload-resolution on
16140 Enable overload resolution for C@t{++} expression evaluation. The default
16141 is on. For overloaded functions, @value{GDBN} evaluates the arguments
16142 and searches for a function whose signature matches the argument types,
16143 using the standard C@t{++} conversion rules (see @ref{C Plus Plus
16144 Expressions, ,C@t{++} Expressions}, for details).
16145 If it cannot find a match, it emits a message.
16147 @item set overload-resolution off
16148 Disable overload resolution for C@t{++} expression evaluation. For
16149 overloaded functions that are not class member functions, @value{GDBN}
16150 chooses the first function of the specified name that it finds in the
16151 symbol table, whether or not its arguments are of the correct type. For
16152 overloaded functions that are class member functions, @value{GDBN}
16153 searches for a function whose signature @emph{exactly} matches the
16156 @kindex show overload-resolution
16157 @item show overload-resolution
16158 Show the current setting of overload resolution.
16160 @item @r{Overloaded symbol names}
16161 You can specify a particular definition of an overloaded symbol, using
16162 the same notation that is used to declare such symbols in C@t{++}: type
16163 @code{@var{symbol}(@var{types})} rather than just @var{symbol}. You can
16164 also use the @value{GDBN} command-line word completion facilities to list the
16165 available choices, or to finish the type list for you.
16166 @xref{Completion,, Command Completion}, for details on how to do this.
16168 @item @r{Breakpoints in functions with ABI tags}
16170 The GNU C@t{++} compiler introduced the notion of ABI ``tags'', which
16171 correspond to changes in the ABI of a type, function, or variable that
16172 would not otherwise be reflected in a mangled name. See
16173 @url{https://developers.redhat.com/blog/2015/02/05/gcc5-and-the-c11-abi/}
16176 The ABI tags are visible in C@t{++} demangled names. For example, a
16177 function that returns a std::string:
16180 std::string function(int);
16184 when compiled for the C++11 ABI is marked with the @code{cxx11} ABI
16185 tag, and @value{GDBN} displays the symbol like this:
16188 function[abi:cxx11](int)
16191 You can set a breakpoint on such functions simply as if they had no
16195 (gdb) b function(int)
16196 Breakpoint 2 at 0x40060d: file main.cc, line 10.
16197 (gdb) info breakpoints
16198 Num Type Disp Enb Address What
16199 1 breakpoint keep y 0x0040060d in function[abi:cxx11](int)
16203 On the rare occasion you need to disambiguate between different ABI
16204 tags, you can do so by simply including the ABI tag in the function
16208 (@value{GDBP}) b ambiguous[abi:other_tag](int)
16212 @node Decimal Floating Point
16213 @subsubsection Decimal Floating Point format
16214 @cindex decimal floating point format
16216 @value{GDBN} can examine, set and perform computations with numbers in
16217 decimal floating point format, which in the C language correspond to the
16218 @code{_Decimal32}, @code{_Decimal64} and @code{_Decimal128} types as
16219 specified by the extension to support decimal floating-point arithmetic.
16221 There are two encodings in use, depending on the architecture: BID (Binary
16222 Integer Decimal) for x86 and x86-64, and DPD (Densely Packed Decimal) for
16223 PowerPC and S/390. @value{GDBN} will use the appropriate encoding for the
16226 Because of a limitation in @file{libdecnumber}, the library used by @value{GDBN}
16227 to manipulate decimal floating point numbers, it is not possible to convert
16228 (using a cast, for example) integers wider than 32-bit to decimal float.
16230 In addition, in order to imitate @value{GDBN}'s behaviour with binary floating
16231 point computations, error checking in decimal float operations ignores
16232 underflow, overflow and divide by zero exceptions.
16234 In the PowerPC architecture, @value{GDBN} provides a set of pseudo-registers
16235 to inspect @code{_Decimal128} values stored in floating point registers.
16236 See @ref{PowerPC,,PowerPC} for more details.
16242 @value{GDBN} can be used to debug programs written in D and compiled with
16243 GDC, LDC or DMD compilers. Currently @value{GDBN} supports only one D
16244 specific feature --- dynamic arrays.
16249 @cindex Go (programming language)
16250 @value{GDBN} can be used to debug programs written in Go and compiled with
16251 @file{gccgo} or @file{6g} compilers.
16253 Here is a summary of the Go-specific features and restrictions:
16256 @cindex current Go package
16257 @item The current Go package
16258 The name of the current package does not need to be specified when
16259 specifying global variables and functions.
16261 For example, given the program:
16265 var myglob = "Shall we?"
16271 When stopped inside @code{main} either of these work:
16275 (gdb) p main.myglob
16278 @cindex builtin Go types
16279 @item Builtin Go types
16280 The @code{string} type is recognized by @value{GDBN} and is printed
16283 @cindex builtin Go functions
16284 @item Builtin Go functions
16285 The @value{GDBN} expression parser recognizes the @code{unsafe.Sizeof}
16286 function and handles it internally.
16288 @cindex restrictions on Go expressions
16289 @item Restrictions on Go expressions
16290 All Go operators are supported except @code{&^}.
16291 The Go @code{_} ``blank identifier'' is not supported.
16292 Automatic dereferencing of pointers is not supported.
16296 @subsection Objective-C
16298 @cindex Objective-C
16299 This section provides information about some commands and command
16300 options that are useful for debugging Objective-C code. See also
16301 @ref{Symbols, info classes}, and @ref{Symbols, info selectors}, for a
16302 few more commands specific to Objective-C support.
16305 * Method Names in Commands::
16306 * The Print Command with Objective-C::
16309 @node Method Names in Commands
16310 @subsubsection Method Names in Commands
16312 The following commands have been extended to accept Objective-C method
16313 names as line specifications:
16315 @kindex clear@r{, and Objective-C}
16316 @kindex break@r{, and Objective-C}
16317 @kindex info line@r{, and Objective-C}
16318 @kindex jump@r{, and Objective-C}
16319 @kindex list@r{, and Objective-C}
16323 @item @code{info line}
16328 A fully qualified Objective-C method name is specified as
16331 -[@var{Class} @var{methodName}]
16334 where the minus sign is used to indicate an instance method and a
16335 plus sign (not shown) is used to indicate a class method. The class
16336 name @var{Class} and method name @var{methodName} are enclosed in
16337 brackets, similar to the way messages are specified in Objective-C
16338 source code. For example, to set a breakpoint at the @code{create}
16339 instance method of class @code{Fruit} in the program currently being
16343 break -[Fruit create]
16346 To list ten program lines around the @code{initialize} class method,
16350 list +[NSText initialize]
16353 In the current version of @value{GDBN}, the plus or minus sign is
16354 required. In future versions of @value{GDBN}, the plus or minus
16355 sign will be optional, but you can use it to narrow the search. It
16356 is also possible to specify just a method name:
16362 You must specify the complete method name, including any colons. If
16363 your program's source files contain more than one @code{create} method,
16364 you'll be presented with a numbered list of classes that implement that
16365 method. Indicate your choice by number, or type @samp{0} to exit if
16368 As another example, to clear a breakpoint established at the
16369 @code{makeKeyAndOrderFront:} method of the @code{NSWindow} class, enter:
16372 clear -[NSWindow makeKeyAndOrderFront:]
16375 @node The Print Command with Objective-C
16376 @subsubsection The Print Command With Objective-C
16377 @cindex Objective-C, print objects
16378 @kindex print-object
16379 @kindex po @r{(@code{print-object})}
16381 The print command has also been extended to accept methods. For example:
16384 print -[@var{object} hash]
16387 @cindex print an Objective-C object description
16388 @cindex @code{_NSPrintForDebugger}, and printing Objective-C objects
16390 will tell @value{GDBN} to send the @code{hash} message to @var{object}
16391 and print the result. Also, an additional command has been added,
16392 @code{print-object} or @code{po} for short, which is meant to print
16393 the description of an object. However, this command may only work
16394 with certain Objective-C libraries that have a particular hook
16395 function, @code{_NSPrintForDebugger}, defined.
16398 @subsection OpenCL C
16401 This section provides information about @value{GDBN}s OpenCL C support.
16404 * OpenCL C Datatypes::
16405 * OpenCL C Expressions::
16406 * OpenCL C Operators::
16409 @node OpenCL C Datatypes
16410 @subsubsection OpenCL C Datatypes
16412 @cindex OpenCL C Datatypes
16413 @value{GDBN} supports the builtin scalar and vector datatypes specified
16414 by OpenCL 1.1. In addition the half- and double-precision floating point
16415 data types of the @code{cl_khr_fp16} and @code{cl_khr_fp64} OpenCL
16416 extensions are also known to @value{GDBN}.
16418 @node OpenCL C Expressions
16419 @subsubsection OpenCL C Expressions
16421 @cindex OpenCL C Expressions
16422 @value{GDBN} supports accesses to vector components including the access as
16423 lvalue where possible. Since OpenCL C is based on C99 most C expressions
16424 supported by @value{GDBN} can be used as well.
16426 @node OpenCL C Operators
16427 @subsubsection OpenCL C Operators
16429 @cindex OpenCL C Operators
16430 @value{GDBN} supports the operators specified by OpenCL 1.1 for scalar and
16434 @subsection Fortran
16435 @cindex Fortran-specific support in @value{GDBN}
16437 @value{GDBN} can be used to debug programs written in Fortran, but it
16438 currently supports only the features of Fortran 77 language.
16440 @cindex trailing underscore, in Fortran symbols
16441 Some Fortran compilers (@sc{gnu} Fortran 77 and Fortran 95 compilers
16442 among them) append an underscore to the names of variables and
16443 functions. When you debug programs compiled by those compilers, you
16444 will need to refer to variables and functions with a trailing
16448 * Fortran Operators:: Fortran operators and expressions
16449 * Fortran Defaults:: Default settings for Fortran
16450 * Special Fortran Commands:: Special @value{GDBN} commands for Fortran
16453 @node Fortran Operators
16454 @subsubsection Fortran Operators and Expressions
16456 @cindex Fortran operators and expressions
16458 Operators must be defined on values of specific types. For instance,
16459 @code{+} is defined on numbers, but not on characters or other non-
16460 arithmetic types. Operators are often defined on groups of types.
16464 The exponentiation operator. It raises the first operand to the power
16468 The range operator. Normally used in the form of array(low:high) to
16469 represent a section of array.
16472 The access component operator. Normally used to access elements in derived
16473 types. Also suitable for unions. As unions aren't part of regular Fortran,
16474 this can only happen when accessing a register that uses a gdbarch-defined
16478 @node Fortran Defaults
16479 @subsubsection Fortran Defaults
16481 @cindex Fortran Defaults
16483 Fortran symbols are usually case-insensitive, so @value{GDBN} by
16484 default uses case-insensitive matches for Fortran symbols. You can
16485 change that with the @samp{set case-insensitive} command, see
16486 @ref{Symbols}, for the details.
16488 @node Special Fortran Commands
16489 @subsubsection Special Fortran Commands
16491 @cindex Special Fortran commands
16493 @value{GDBN} has some commands to support Fortran-specific features,
16494 such as displaying common blocks.
16497 @cindex @code{COMMON} blocks, Fortran
16498 @kindex info common
16499 @item info common @r{[}@var{common-name}@r{]}
16500 This command prints the values contained in the Fortran @code{COMMON}
16501 block whose name is @var{common-name}. With no argument, the names of
16502 all @code{COMMON} blocks visible at the current program location are
16509 @cindex Pascal support in @value{GDBN}, limitations
16510 Debugging Pascal programs which use sets, subranges, file variables, or
16511 nested functions does not currently work. @value{GDBN} does not support
16512 entering expressions, printing values, or similar features using Pascal
16515 The Pascal-specific command @code{set print pascal_static-members}
16516 controls whether static members of Pascal objects are displayed.
16517 @xref{Print Settings, pascal_static-members}.
16522 @value{GDBN} supports the @url{https://www.rust-lang.org/, Rust
16523 Programming Language}. Type- and value-printing, and expression
16524 parsing, are reasonably complete. However, there are a few
16525 peculiarities and holes to be aware of.
16529 Linespecs (@pxref{Specify Location}) are never relative to the current
16530 crate. Instead, they act as if there were a global namespace of
16531 crates, somewhat similar to the way @code{extern crate} behaves.
16533 That is, if @value{GDBN} is stopped at a breakpoint in a function in
16534 crate @samp{A}, module @samp{B}, then @code{break B::f} will attempt
16535 to set a breakpoint in a function named @samp{f} in a crate named
16538 As a consequence of this approach, linespecs also cannot refer to
16539 items using @samp{self::} or @samp{super::}.
16542 Because @value{GDBN} implements Rust name-lookup semantics in
16543 expressions, it will sometimes prepend the current crate to a name.
16544 For example, if @value{GDBN} is stopped at a breakpoint in the crate
16545 @samp{K}, then @code{print ::x::y} will try to find the symbol
16548 However, since it is useful to be able to refer to other crates when
16549 debugging, @value{GDBN} provides the @code{extern} extension to
16550 circumvent this. To use the extension, just put @code{extern} before
16551 a path expression to refer to the otherwise unavailable ``global''
16554 In the above example, if you wanted to refer to the symbol @samp{y} in
16555 the crate @samp{x}, you would use @code{print extern x::y}.
16558 The Rust expression evaluator does not support ``statement-like''
16559 expressions such as @code{if} or @code{match}, or lambda expressions.
16562 Tuple expressions are not implemented.
16565 The Rust expression evaluator does not currently implement the
16566 @code{Drop} trait. Objects that may be created by the evaluator will
16567 never be destroyed.
16570 @value{GDBN} does not implement type inference for generics. In order
16571 to call generic functions or otherwise refer to generic items, you
16572 will have to specify the type parameters manually.
16575 @value{GDBN} currently uses the C@t{++} demangler for Rust. In most
16576 cases this does not cause any problems. However, in an expression
16577 context, completing a generic function name will give syntactically
16578 invalid results. This happens because Rust requires the @samp{::}
16579 operator between the function name and its generic arguments. For
16580 example, @value{GDBN} might provide a completion like
16581 @code{crate::f<u32>}, where the parser would require
16582 @code{crate::f::<u32>}.
16585 As of this writing, the Rust compiler (version 1.8) has a few holes in
16586 the debugging information it generates. These holes prevent certain
16587 features from being implemented by @value{GDBN}:
16591 Method calls cannot be made via traits.
16594 Operator overloading is not implemented.
16597 When debugging in a monomorphized function, you cannot use the generic
16601 The type @code{Self} is not available.
16604 @code{use} statements are not available, so some names may not be
16605 available in the crate.
16610 @subsection Modula-2
16612 @cindex Modula-2, @value{GDBN} support
16614 The extensions made to @value{GDBN} to support Modula-2 only support
16615 output from the @sc{gnu} Modula-2 compiler (which is currently being
16616 developed). Other Modula-2 compilers are not currently supported, and
16617 attempting to debug executables produced by them is most likely
16618 to give an error as @value{GDBN} reads in the executable's symbol
16621 @cindex expressions in Modula-2
16623 * M2 Operators:: Built-in operators
16624 * Built-In Func/Proc:: Built-in functions and procedures
16625 * M2 Constants:: Modula-2 constants
16626 * M2 Types:: Modula-2 types
16627 * M2 Defaults:: Default settings for Modula-2
16628 * Deviations:: Deviations from standard Modula-2
16629 * M2 Checks:: Modula-2 type and range checks
16630 * M2 Scope:: The scope operators @code{::} and @code{.}
16631 * GDB/M2:: @value{GDBN} and Modula-2
16635 @subsubsection Operators
16636 @cindex Modula-2 operators
16638 Operators must be defined on values of specific types. For instance,
16639 @code{+} is defined on numbers, but not on structures. Operators are
16640 often defined on groups of types. For the purposes of Modula-2, the
16641 following definitions hold:
16646 @emph{Integral types} consist of @code{INTEGER}, @code{CARDINAL}, and
16650 @emph{Character types} consist of @code{CHAR} and its subranges.
16653 @emph{Floating-point types} consist of @code{REAL}.
16656 @emph{Pointer types} consist of anything declared as @code{POINTER TO
16660 @emph{Scalar types} consist of all of the above.
16663 @emph{Set types} consist of @code{SET} and @code{BITSET} types.
16666 @emph{Boolean types} consist of @code{BOOLEAN}.
16670 The following operators are supported, and appear in order of
16671 increasing precedence:
16675 Function argument or array index separator.
16678 Assignment. The value of @var{var} @code{:=} @var{value} is
16682 Less than, greater than on integral, floating-point, or enumerated
16686 Less than or equal to, greater than or equal to
16687 on integral, floating-point and enumerated types, or set inclusion on
16688 set types. Same precedence as @code{<}.
16690 @item =@r{, }<>@r{, }#
16691 Equality and two ways of expressing inequality, valid on scalar types.
16692 Same precedence as @code{<}. In @value{GDBN} scripts, only @code{<>} is
16693 available for inequality, since @code{#} conflicts with the script
16697 Set membership. Defined on set types and the types of their members.
16698 Same precedence as @code{<}.
16701 Boolean disjunction. Defined on boolean types.
16704 Boolean conjunction. Defined on boolean types.
16707 The @value{GDBN} ``artificial array'' operator (@pxref{Expressions, ,Expressions}).
16710 Addition and subtraction on integral and floating-point types, or union
16711 and difference on set types.
16714 Multiplication on integral and floating-point types, or set intersection
16718 Division on floating-point types, or symmetric set difference on set
16719 types. Same precedence as @code{*}.
16722 Integer division and remainder. Defined on integral types. Same
16723 precedence as @code{*}.
16726 Negative. Defined on @code{INTEGER} and @code{REAL} data.
16729 Pointer dereferencing. Defined on pointer types.
16732 Boolean negation. Defined on boolean types. Same precedence as
16736 @code{RECORD} field selector. Defined on @code{RECORD} data. Same
16737 precedence as @code{^}.
16740 Array indexing. Defined on @code{ARRAY} data. Same precedence as @code{^}.
16743 Procedure argument list. Defined on @code{PROCEDURE} objects. Same precedence
16747 @value{GDBN} and Modula-2 scope operators.
16751 @emph{Warning:} Set expressions and their operations are not yet supported, so @value{GDBN}
16752 treats the use of the operator @code{IN}, or the use of operators
16753 @code{+}, @code{-}, @code{*}, @code{/}, @code{=}, , @code{<>}, @code{#},
16754 @code{<=}, and @code{>=} on sets as an error.
16758 @node Built-In Func/Proc
16759 @subsubsection Built-in Functions and Procedures
16760 @cindex Modula-2 built-ins
16762 Modula-2 also makes available several built-in procedures and functions.
16763 In describing these, the following metavariables are used:
16768 represents an @code{ARRAY} variable.
16771 represents a @code{CHAR} constant or variable.
16774 represents a variable or constant of integral type.
16777 represents an identifier that belongs to a set. Generally used in the
16778 same function with the metavariable @var{s}. The type of @var{s} should
16779 be @code{SET OF @var{mtype}} (where @var{mtype} is the type of @var{m}).
16782 represents a variable or constant of integral or floating-point type.
16785 represents a variable or constant of floating-point type.
16791 represents a variable.
16794 represents a variable or constant of one of many types. See the
16795 explanation of the function for details.
16798 All Modula-2 built-in procedures also return a result, described below.
16802 Returns the absolute value of @var{n}.
16805 If @var{c} is a lower case letter, it returns its upper case
16806 equivalent, otherwise it returns its argument.
16809 Returns the character whose ordinal value is @var{i}.
16812 Decrements the value in the variable @var{v} by one. Returns the new value.
16814 @item DEC(@var{v},@var{i})
16815 Decrements the value in the variable @var{v} by @var{i}. Returns the
16818 @item EXCL(@var{m},@var{s})
16819 Removes the element @var{m} from the set @var{s}. Returns the new
16822 @item FLOAT(@var{i})
16823 Returns the floating point equivalent of the integer @var{i}.
16825 @item HIGH(@var{a})
16826 Returns the index of the last member of @var{a}.
16829 Increments the value in the variable @var{v} by one. Returns the new value.
16831 @item INC(@var{v},@var{i})
16832 Increments the value in the variable @var{v} by @var{i}. Returns the
16835 @item INCL(@var{m},@var{s})
16836 Adds the element @var{m} to the set @var{s} if it is not already
16837 there. Returns the new set.
16840 Returns the maximum value of the type @var{t}.
16843 Returns the minimum value of the type @var{t}.
16846 Returns boolean TRUE if @var{i} is an odd number.
16849 Returns the ordinal value of its argument. For example, the ordinal
16850 value of a character is its @sc{ascii} value (on machines supporting
16851 the @sc{ascii} character set). The argument @var{x} must be of an
16852 ordered type, which include integral, character and enumerated types.
16854 @item SIZE(@var{x})
16855 Returns the size of its argument. The argument @var{x} can be a
16856 variable or a type.
16858 @item TRUNC(@var{r})
16859 Returns the integral part of @var{r}.
16861 @item TSIZE(@var{x})
16862 Returns the size of its argument. The argument @var{x} can be a
16863 variable or a type.
16865 @item VAL(@var{t},@var{i})
16866 Returns the member of the type @var{t} whose ordinal value is @var{i}.
16870 @emph{Warning:} Sets and their operations are not yet supported, so
16871 @value{GDBN} treats the use of procedures @code{INCL} and @code{EXCL} as
16875 @cindex Modula-2 constants
16877 @subsubsection Constants
16879 @value{GDBN} allows you to express the constants of Modula-2 in the following
16885 Integer constants are simply a sequence of digits. When used in an
16886 expression, a constant is interpreted to be type-compatible with the
16887 rest of the expression. Hexadecimal integers are specified by a
16888 trailing @samp{H}, and octal integers by a trailing @samp{B}.
16891 Floating point constants appear as a sequence of digits, followed by a
16892 decimal point and another sequence of digits. An optional exponent can
16893 then be specified, in the form @samp{E@r{[}+@r{|}-@r{]}@var{nnn}}, where
16894 @samp{@r{[}+@r{|}-@r{]}@var{nnn}} is the desired exponent. All of the
16895 digits of the floating point constant must be valid decimal (base 10)
16899 Character constants consist of a single character enclosed by a pair of
16900 like quotes, either single (@code{'}) or double (@code{"}). They may
16901 also be expressed by their ordinal value (their @sc{ascii} value, usually)
16902 followed by a @samp{C}.
16905 String constants consist of a sequence of characters enclosed by a
16906 pair of like quotes, either single (@code{'}) or double (@code{"}).
16907 Escape sequences in the style of C are also allowed. @xref{C
16908 Constants, ,C and C@t{++} Constants}, for a brief explanation of escape
16912 Enumerated constants consist of an enumerated identifier.
16915 Boolean constants consist of the identifiers @code{TRUE} and
16919 Pointer constants consist of integral values only.
16922 Set constants are not yet supported.
16926 @subsubsection Modula-2 Types
16927 @cindex Modula-2 types
16929 Currently @value{GDBN} can print the following data types in Modula-2
16930 syntax: array types, record types, set types, pointer types, procedure
16931 types, enumerated types, subrange types and base types. You can also
16932 print the contents of variables declared using these type.
16933 This section gives a number of simple source code examples together with
16934 sample @value{GDBN} sessions.
16936 The first example contains the following section of code:
16945 and you can request @value{GDBN} to interrogate the type and value of
16946 @code{r} and @code{s}.
16949 (@value{GDBP}) print s
16951 (@value{GDBP}) ptype s
16953 (@value{GDBP}) print r
16955 (@value{GDBP}) ptype r
16960 Likewise if your source code declares @code{s} as:
16964 s: SET ['A'..'Z'] ;
16968 then you may query the type of @code{s} by:
16971 (@value{GDBP}) ptype s
16972 type = SET ['A'..'Z']
16976 Note that at present you cannot interactively manipulate set
16977 expressions using the debugger.
16979 The following example shows how you might declare an array in Modula-2
16980 and how you can interact with @value{GDBN} to print its type and contents:
16984 s: ARRAY [-10..10] OF CHAR ;
16988 (@value{GDBP}) ptype s
16989 ARRAY [-10..10] OF CHAR
16992 Note that the array handling is not yet complete and although the type
16993 is printed correctly, expression handling still assumes that all
16994 arrays have a lower bound of zero and not @code{-10} as in the example
16997 Here are some more type related Modula-2 examples:
17001 colour = (blue, red, yellow, green) ;
17002 t = [blue..yellow] ;
17010 The @value{GDBN} interaction shows how you can query the data type
17011 and value of a variable.
17014 (@value{GDBP}) print s
17016 (@value{GDBP}) ptype t
17017 type = [blue..yellow]
17021 In this example a Modula-2 array is declared and its contents
17022 displayed. Observe that the contents are written in the same way as
17023 their @code{C} counterparts.
17027 s: ARRAY [1..5] OF CARDINAL ;
17033 (@value{GDBP}) print s
17034 $1 = @{1, 0, 0, 0, 0@}
17035 (@value{GDBP}) ptype s
17036 type = ARRAY [1..5] OF CARDINAL
17039 The Modula-2 language interface to @value{GDBN} also understands
17040 pointer types as shown in this example:
17044 s: POINTER TO ARRAY [1..5] OF CARDINAL ;
17051 and you can request that @value{GDBN} describes the type of @code{s}.
17054 (@value{GDBP}) ptype s
17055 type = POINTER TO ARRAY [1..5] OF CARDINAL
17058 @value{GDBN} handles compound types as we can see in this example.
17059 Here we combine array types, record types, pointer types and subrange
17070 myarray = ARRAY myrange OF CARDINAL ;
17071 myrange = [-2..2] ;
17073 s: POINTER TO ARRAY myrange OF foo ;
17077 and you can ask @value{GDBN} to describe the type of @code{s} as shown
17081 (@value{GDBP}) ptype s
17082 type = POINTER TO ARRAY [-2..2] OF foo = RECORD
17085 f3 : ARRAY [-2..2] OF CARDINAL;
17090 @subsubsection Modula-2 Defaults
17091 @cindex Modula-2 defaults
17093 If type and range checking are set automatically by @value{GDBN}, they
17094 both default to @code{on} whenever the working language changes to
17095 Modula-2. This happens regardless of whether you or @value{GDBN}
17096 selected the working language.
17098 If you allow @value{GDBN} to set the language automatically, then entering
17099 code compiled from a file whose name ends with @file{.mod} sets the
17100 working language to Modula-2. @xref{Automatically, ,Having @value{GDBN}
17101 Infer the Source Language}, for further details.
17104 @subsubsection Deviations from Standard Modula-2
17105 @cindex Modula-2, deviations from
17107 A few changes have been made to make Modula-2 programs easier to debug.
17108 This is done primarily via loosening its type strictness:
17112 Unlike in standard Modula-2, pointer constants can be formed by
17113 integers. This allows you to modify pointer variables during
17114 debugging. (In standard Modula-2, the actual address contained in a
17115 pointer variable is hidden from you; it can only be modified
17116 through direct assignment to another pointer variable or expression that
17117 returned a pointer.)
17120 C escape sequences can be used in strings and characters to represent
17121 non-printable characters. @value{GDBN} prints out strings with these
17122 escape sequences embedded. Single non-printable characters are
17123 printed using the @samp{CHR(@var{nnn})} format.
17126 The assignment operator (@code{:=}) returns the value of its right-hand
17130 All built-in procedures both modify @emph{and} return their argument.
17134 @subsubsection Modula-2 Type and Range Checks
17135 @cindex Modula-2 checks
17138 @emph{Warning:} in this release, @value{GDBN} does not yet perform type or
17141 @c FIXME remove warning when type/range checks added
17143 @value{GDBN} considers two Modula-2 variables type equivalent if:
17147 They are of types that have been declared equivalent via a @code{TYPE
17148 @var{t1} = @var{t2}} statement
17151 They have been declared on the same line. (Note: This is true of the
17152 @sc{gnu} Modula-2 compiler, but it may not be true of other compilers.)
17155 As long as type checking is enabled, any attempt to combine variables
17156 whose types are not equivalent is an error.
17158 Range checking is done on all mathematical operations, assignment, array
17159 index bounds, and all built-in functions and procedures.
17162 @subsubsection The Scope Operators @code{::} and @code{.}
17164 @cindex @code{.}, Modula-2 scope operator
17165 @cindex colon, doubled as scope operator
17167 @vindex colon-colon@r{, in Modula-2}
17168 @c Info cannot handle :: but TeX can.
17171 @vindex ::@r{, in Modula-2}
17174 There are a few subtle differences between the Modula-2 scope operator
17175 (@code{.}) and the @value{GDBN} scope operator (@code{::}). The two have
17180 @var{module} . @var{id}
17181 @var{scope} :: @var{id}
17185 where @var{scope} is the name of a module or a procedure,
17186 @var{module} the name of a module, and @var{id} is any declared
17187 identifier within your program, except another module.
17189 Using the @code{::} operator makes @value{GDBN} search the scope
17190 specified by @var{scope} for the identifier @var{id}. If it is not
17191 found in the specified scope, then @value{GDBN} searches all scopes
17192 enclosing the one specified by @var{scope}.
17194 Using the @code{.} operator makes @value{GDBN} search the current scope for
17195 the identifier specified by @var{id} that was imported from the
17196 definition module specified by @var{module}. With this operator, it is
17197 an error if the identifier @var{id} was not imported from definition
17198 module @var{module}, or if @var{id} is not an identifier in
17202 @subsubsection @value{GDBN} and Modula-2
17204 Some @value{GDBN} commands have little use when debugging Modula-2 programs.
17205 Five subcommands of @code{set print} and @code{show print} apply
17206 specifically to C and C@t{++}: @samp{vtbl}, @samp{demangle},
17207 @samp{asm-demangle}, @samp{object}, and @samp{union}. The first four
17208 apply to C@t{++}, and the last to the C @code{union} type, which has no direct
17209 analogue in Modula-2.
17211 The @code{@@} operator (@pxref{Expressions, ,Expressions}), while available
17212 with any language, is not useful with Modula-2. Its
17213 intent is to aid the debugging of @dfn{dynamic arrays}, which cannot be
17214 created in Modula-2 as they can in C or C@t{++}. However, because an
17215 address can be specified by an integral constant, the construct
17216 @samp{@{@var{type}@}@var{adrexp}} is still useful.
17218 @cindex @code{#} in Modula-2
17219 In @value{GDBN} scripts, the Modula-2 inequality operator @code{#} is
17220 interpreted as the beginning of a comment. Use @code{<>} instead.
17226 The extensions made to @value{GDBN} for Ada only support
17227 output from the @sc{gnu} Ada (GNAT) compiler.
17228 Other Ada compilers are not currently supported, and
17229 attempting to debug executables produced by them is most likely
17233 @cindex expressions in Ada
17235 * Ada Mode Intro:: General remarks on the Ada syntax
17236 and semantics supported by Ada mode
17238 * Omissions from Ada:: Restrictions on the Ada expression syntax.
17239 * Additions to Ada:: Extensions of the Ada expression syntax.
17240 * Overloading support for Ada:: Support for expressions involving overloaded
17242 * Stopping Before Main Program:: Debugging the program during elaboration.
17243 * Ada Exceptions:: Ada Exceptions
17244 * Ada Tasks:: Listing and setting breakpoints in tasks.
17245 * Ada Tasks and Core Files:: Tasking Support when Debugging Core Files
17246 * Ravenscar Profile:: Tasking Support when using the Ravenscar
17248 * Ada Settings:: New settable GDB parameters for Ada.
17249 * Ada Glitches:: Known peculiarities of Ada mode.
17252 @node Ada Mode Intro
17253 @subsubsection Introduction
17254 @cindex Ada mode, general
17256 The Ada mode of @value{GDBN} supports a fairly large subset of Ada expression
17257 syntax, with some extensions.
17258 The philosophy behind the design of this subset is
17262 That @value{GDBN} should provide basic literals and access to operations for
17263 arithmetic, dereferencing, field selection, indexing, and subprogram calls,
17264 leaving more sophisticated computations to subprograms written into the
17265 program (which therefore may be called from @value{GDBN}).
17268 That type safety and strict adherence to Ada language restrictions
17269 are not particularly important to the @value{GDBN} user.
17272 That brevity is important to the @value{GDBN} user.
17275 Thus, for brevity, the debugger acts as if all names declared in
17276 user-written packages are directly visible, even if they are not visible
17277 according to Ada rules, thus making it unnecessary to fully qualify most
17278 names with their packages, regardless of context. Where this causes
17279 ambiguity, @value{GDBN} asks the user's intent.
17281 The debugger will start in Ada mode if it detects an Ada main program.
17282 As for other languages, it will enter Ada mode when stopped in a program that
17283 was translated from an Ada source file.
17285 While in Ada mode, you may use `@t{--}' for comments. This is useful
17286 mostly for documenting command files. The standard @value{GDBN} comment
17287 (@samp{#}) still works at the beginning of a line in Ada mode, but not in the
17288 middle (to allow based literals).
17290 @node Omissions from Ada
17291 @subsubsection Omissions from Ada
17292 @cindex Ada, omissions from
17294 Here are the notable omissions from the subset:
17298 Only a subset of the attributes are supported:
17302 @t{'First}, @t{'Last}, and @t{'Length}
17303 on array objects (not on types and subtypes).
17306 @t{'Min} and @t{'Max}.
17309 @t{'Pos} and @t{'Val}.
17315 @t{'Range} on array objects (not subtypes), but only as the right
17316 operand of the membership (@code{in}) operator.
17319 @t{'Access}, @t{'Unchecked_Access}, and
17320 @t{'Unrestricted_Access} (a GNAT extension).
17328 @code{Characters.Latin_1} are not available and
17329 concatenation is not implemented. Thus, escape characters in strings are
17330 not currently available.
17333 Equality tests (@samp{=} and @samp{/=}) on arrays test for bitwise
17334 equality of representations. They will generally work correctly
17335 for strings and arrays whose elements have integer or enumeration types.
17336 They may not work correctly for arrays whose element
17337 types have user-defined equality, for arrays of real values
17338 (in particular, IEEE-conformant floating point, because of negative
17339 zeroes and NaNs), and for arrays whose elements contain unused bits with
17340 indeterminate values.
17343 The other component-by-component array operations (@code{and}, @code{or},
17344 @code{xor}, @code{not}, and relational tests other than equality)
17345 are not implemented.
17348 @cindex array aggregates (Ada)
17349 @cindex record aggregates (Ada)
17350 @cindex aggregates (Ada)
17351 There is limited support for array and record aggregates. They are
17352 permitted only on the right sides of assignments, as in these examples:
17355 (@value{GDBP}) set An_Array := (1, 2, 3, 4, 5, 6)
17356 (@value{GDBP}) set An_Array := (1, others => 0)
17357 (@value{GDBP}) set An_Array := (0|4 => 1, 1..3 => 2, 5 => 6)
17358 (@value{GDBP}) set A_2D_Array := ((1, 2, 3), (4, 5, 6), (7, 8, 9))
17359 (@value{GDBP}) set A_Record := (1, "Peter", True);
17360 (@value{GDBP}) set A_Record := (Name => "Peter", Id => 1, Alive => True)
17364 discriminant's value by assigning an aggregate has an
17365 undefined effect if that discriminant is used within the record.
17366 However, you can first modify discriminants by directly assigning to
17367 them (which normally would not be allowed in Ada), and then performing an
17368 aggregate assignment. For example, given a variable @code{A_Rec}
17369 declared to have a type such as:
17372 type Rec (Len : Small_Integer := 0) is record
17374 Vals : IntArray (1 .. Len);
17378 you can assign a value with a different size of @code{Vals} with two
17382 (@value{GDBP}) set A_Rec.Len := 4
17383 (@value{GDBP}) set A_Rec := (Id => 42, Vals => (1, 2, 3, 4))
17386 As this example also illustrates, @value{GDBN} is very loose about the usual
17387 rules concerning aggregates. You may leave out some of the
17388 components of an array or record aggregate (such as the @code{Len}
17389 component in the assignment to @code{A_Rec} above); they will retain their
17390 original values upon assignment. You may freely use dynamic values as
17391 indices in component associations. You may even use overlapping or
17392 redundant component associations, although which component values are
17393 assigned in such cases is not defined.
17396 Calls to dispatching subprograms are not implemented.
17399 The overloading algorithm is much more limited (i.e., less selective)
17400 than that of real Ada. It makes only limited use of the context in
17401 which a subexpression appears to resolve its meaning, and it is much
17402 looser in its rules for allowing type matches. As a result, some
17403 function calls will be ambiguous, and the user will be asked to choose
17404 the proper resolution.
17407 The @code{new} operator is not implemented.
17410 Entry calls are not implemented.
17413 Aside from printing, arithmetic operations on the native VAX floating-point
17414 formats are not supported.
17417 It is not possible to slice a packed array.
17420 The names @code{True} and @code{False}, when not part of a qualified name,
17421 are interpreted as if implicitly prefixed by @code{Standard}, regardless of
17423 Should your program
17424 redefine these names in a package or procedure (at best a dubious practice),
17425 you will have to use fully qualified names to access their new definitions.
17428 @node Additions to Ada
17429 @subsubsection Additions to Ada
17430 @cindex Ada, deviations from
17432 As it does for other languages, @value{GDBN} makes certain generic
17433 extensions to Ada (@pxref{Expressions}):
17437 If the expression @var{E} is a variable residing in memory (typically
17438 a local variable or array element) and @var{N} is a positive integer,
17439 then @code{@var{E}@@@var{N}} displays the values of @var{E} and the
17440 @var{N}-1 adjacent variables following it in memory as an array. In
17441 Ada, this operator is generally not necessary, since its prime use is
17442 in displaying parts of an array, and slicing will usually do this in
17443 Ada. However, there are occasional uses when debugging programs in
17444 which certain debugging information has been optimized away.
17447 @code{@var{B}::@var{var}} means ``the variable named @var{var} that
17448 appears in function or file @var{B}.'' When @var{B} is a file name,
17449 you must typically surround it in single quotes.
17452 The expression @code{@{@var{type}@} @var{addr}} means ``the variable of type
17453 @var{type} that appears at address @var{addr}.''
17456 A name starting with @samp{$} is a convenience variable
17457 (@pxref{Convenience Vars}) or a machine register (@pxref{Registers}).
17460 In addition, @value{GDBN} provides a few other shortcuts and outright
17461 additions specific to Ada:
17465 The assignment statement is allowed as an expression, returning
17466 its right-hand operand as its value. Thus, you may enter
17469 (@value{GDBP}) set x := y + 3
17470 (@value{GDBP}) print A(tmp := y + 1)
17474 The semicolon is allowed as an ``operator,'' returning as its value
17475 the value of its right-hand operand.
17476 This allows, for example,
17477 complex conditional breaks:
17480 (@value{GDBP}) break f
17481 (@value{GDBP}) condition 1 (report(i); k += 1; A(k) > 100)
17485 Rather than use catenation and symbolic character names to introduce special
17486 characters into strings, one may instead use a special bracket notation,
17487 which is also used to print strings. A sequence of characters of the form
17488 @samp{["@var{XX}"]} within a string or character literal denotes the
17489 (single) character whose numeric encoding is @var{XX} in hexadecimal. The
17490 sequence of characters @samp{["""]} also denotes a single quotation mark
17491 in strings. For example,
17493 "One line.["0a"]Next line.["0a"]"
17496 contains an ASCII newline character (@code{Ada.Characters.Latin_1.LF})
17500 The subtype used as a prefix for the attributes @t{'Pos}, @t{'Min}, and
17501 @t{'Max} is optional (and is ignored in any case). For example, it is valid
17505 (@value{GDBP}) print 'max(x, y)
17509 When printing arrays, @value{GDBN} uses positional notation when the
17510 array has a lower bound of 1, and uses a modified named notation otherwise.
17511 For example, a one-dimensional array of three integers with a lower bound
17512 of 3 might print as
17519 That is, in contrast to valid Ada, only the first component has a @code{=>}
17523 You may abbreviate attributes in expressions with any unique,
17524 multi-character subsequence of
17525 their names (an exact match gets preference).
17526 For example, you may use @t{a'len}, @t{a'gth}, or @t{a'lh}
17527 in place of @t{a'length}.
17530 @cindex quoting Ada internal identifiers
17531 Since Ada is case-insensitive, the debugger normally maps identifiers you type
17532 to lower case. The GNAT compiler uses upper-case characters for
17533 some of its internal identifiers, which are normally of no interest to users.
17534 For the rare occasions when you actually have to look at them,
17535 enclose them in angle brackets to avoid the lower-case mapping.
17538 (@value{GDBP}) print <JMPBUF_SAVE>[0]
17542 Printing an object of class-wide type or dereferencing an
17543 access-to-class-wide value will display all the components of the object's
17544 specific type (as indicated by its run-time tag). Likewise, component
17545 selection on such a value will operate on the specific type of the
17550 @node Overloading support for Ada
17551 @subsubsection Overloading support for Ada
17552 @cindex overloading, Ada
17554 The debugger supports limited overloading. Given a subprogram call in which
17555 the function symbol has multiple definitions, it will use the number of
17556 actual parameters and some information about their types to attempt to narrow
17557 the set of definitions. It also makes very limited use of context, preferring
17558 procedures to functions in the context of the @code{call} command, and
17559 functions to procedures elsewhere.
17561 If, after narrowing, the set of matching definitions still contains more than
17562 one definition, @value{GDBN} will display a menu to query which one it should
17566 (@value{GDBP}) print f(1)
17567 Multiple matches for f
17569 [1] foo.f (integer) return boolean at foo.adb:23
17570 [2] foo.f (foo.new_integer) return boolean at foo.adb:28
17574 In this case, just select one menu entry either to cancel expression evaluation
17575 (type @kbd{0} and press @key{RET}) or to continue evaluation with a specific
17576 instance (type the corresponding number and press @key{RET}).
17578 Here are a couple of commands to customize @value{GDBN}'s behavior in this
17583 @kindex set ada print-signatures
17584 @item set ada print-signatures
17585 Control whether parameter types and return types are displayed in overloads
17586 selection menus. It is @code{on} by default.
17587 @xref{Overloading support for Ada}.
17589 @kindex show ada print-signatures
17590 @item show ada print-signatures
17591 Show the current setting for displaying parameter types and return types in
17592 overloads selection menu.
17593 @xref{Overloading support for Ada}.
17597 @node Stopping Before Main Program
17598 @subsubsection Stopping at the Very Beginning
17600 @cindex breakpointing Ada elaboration code
17601 It is sometimes necessary to debug the program during elaboration, and
17602 before reaching the main procedure.
17603 As defined in the Ada Reference
17604 Manual, the elaboration code is invoked from a procedure called
17605 @code{adainit}. To run your program up to the beginning of
17606 elaboration, simply use the following two commands:
17607 @code{tbreak adainit} and @code{run}.
17609 @node Ada Exceptions
17610 @subsubsection Ada Exceptions
17612 A command is provided to list all Ada exceptions:
17615 @kindex info exceptions
17616 @item info exceptions
17617 @itemx info exceptions @var{regexp}
17618 The @code{info exceptions} command allows you to list all Ada exceptions
17619 defined within the program being debugged, as well as their addresses.
17620 With a regular expression, @var{regexp}, as argument, only those exceptions
17621 whose names match @var{regexp} are listed.
17624 Below is a small example, showing how the command can be used, first
17625 without argument, and next with a regular expression passed as an
17629 (@value{GDBP}) info exceptions
17630 All defined Ada exceptions:
17631 constraint_error: 0x613da0
17632 program_error: 0x613d20
17633 storage_error: 0x613ce0
17634 tasking_error: 0x613ca0
17635 const.aint_global_e: 0x613b00
17636 (@value{GDBP}) info exceptions const.aint
17637 All Ada exceptions matching regular expression "const.aint":
17638 constraint_error: 0x613da0
17639 const.aint_global_e: 0x613b00
17642 It is also possible to ask @value{GDBN} to stop your program's execution
17643 when an exception is raised. For more details, see @ref{Set Catchpoints}.
17646 @subsubsection Extensions for Ada Tasks
17647 @cindex Ada, tasking
17649 Support for Ada tasks is analogous to that for threads (@pxref{Threads}).
17650 @value{GDBN} provides the following task-related commands:
17655 This command shows a list of current Ada tasks, as in the following example:
17662 (@value{GDBP}) info tasks
17663 ID TID P-ID Pri State Name
17664 1 8088000 0 15 Child Activation Wait main_task
17665 2 80a4000 1 15 Accept Statement b
17666 3 809a800 1 15 Child Activation Wait a
17667 * 4 80ae800 3 15 Runnable c
17672 In this listing, the asterisk before the last task indicates it to be the
17673 task currently being inspected.
17677 Represents @value{GDBN}'s internal task number.
17683 The parent's task ID (@value{GDBN}'s internal task number).
17686 The base priority of the task.
17689 Current state of the task.
17693 The task has been created but has not been activated. It cannot be
17697 The task is not blocked for any reason known to Ada. (It may be waiting
17698 for a mutex, though.) It is conceptually "executing" in normal mode.
17701 The task is terminated, in the sense of ARM 9.3 (5). Any dependents
17702 that were waiting on terminate alternatives have been awakened and have
17703 terminated themselves.
17705 @item Child Activation Wait
17706 The task is waiting for created tasks to complete activation.
17708 @item Accept Statement
17709 The task is waiting on an accept or selective wait statement.
17711 @item Waiting on entry call
17712 The task is waiting on an entry call.
17714 @item Async Select Wait
17715 The task is waiting to start the abortable part of an asynchronous
17719 The task is waiting on a select statement with only a delay
17722 @item Child Termination Wait
17723 The task is sleeping having completed a master within itself, and is
17724 waiting for the tasks dependent on that master to become terminated or
17725 waiting on a terminate Phase.
17727 @item Wait Child in Term Alt
17728 The task is sleeping waiting for tasks on terminate alternatives to
17729 finish terminating.
17731 @item Accepting RV with @var{taskno}
17732 The task is accepting a rendez-vous with the task @var{taskno}.
17736 Name of the task in the program.
17740 @kindex info task @var{taskno}
17741 @item info task @var{taskno}
17742 This command shows detailled informations on the specified task, as in
17743 the following example:
17748 (@value{GDBP}) info tasks
17749 ID TID P-ID Pri State Name
17750 1 8077880 0 15 Child Activation Wait main_task
17751 * 2 807c468 1 15 Runnable task_1
17752 (@value{GDBP}) info task 2
17753 Ada Task: 0x807c468
17757 Parent: 1 (main_task)
17763 @kindex task@r{ (Ada)}
17764 @cindex current Ada task ID
17765 This command prints the ID of the current task.
17771 (@value{GDBP}) info tasks
17772 ID TID P-ID Pri State Name
17773 1 8077870 0 15 Child Activation Wait main_task
17774 * 2 807c458 1 15 Runnable t
17775 (@value{GDBP}) task
17776 [Current task is 2]
17779 @item task @var{taskno}
17780 @cindex Ada task switching
17781 This command is like the @code{thread @var{thread-id}}
17782 command (@pxref{Threads}). It switches the context of debugging
17783 from the current task to the given task.
17789 (@value{GDBP}) info tasks
17790 ID TID P-ID Pri State Name
17791 1 8077870 0 15 Child Activation Wait main_task
17792 * 2 807c458 1 15 Runnable t
17793 (@value{GDBP}) task 1
17794 [Switching to task 1]
17795 #0 0x8067726 in pthread_cond_wait ()
17797 #0 0x8067726 in pthread_cond_wait ()
17798 #1 0x8056714 in system.os_interface.pthread_cond_wait ()
17799 #2 0x805cb63 in system.task_primitives.operations.sleep ()
17800 #3 0x806153e in system.tasking.stages.activate_tasks ()
17801 #4 0x804aacc in un () at un.adb:5
17804 @item break @var{location} task @var{taskno}
17805 @itemx break @var{location} task @var{taskno} if @dots{}
17806 @cindex breakpoints and tasks, in Ada
17807 @cindex task breakpoints, in Ada
17808 @kindex break @dots{} task @var{taskno}@r{ (Ada)}
17809 These commands are like the @code{break @dots{} thread @dots{}}
17810 command (@pxref{Thread Stops}). The
17811 @var{location} argument specifies source lines, as described
17812 in @ref{Specify Location}.
17814 Use the qualifier @samp{task @var{taskno}} with a breakpoint command
17815 to specify that you only want @value{GDBN} to stop the program when a
17816 particular Ada task reaches this breakpoint. The @var{taskno} is one of the
17817 numeric task identifiers assigned by @value{GDBN}, shown in the first
17818 column of the @samp{info tasks} display.
17820 If you do not specify @samp{task @var{taskno}} when you set a
17821 breakpoint, the breakpoint applies to @emph{all} tasks of your
17824 You can use the @code{task} qualifier on conditional breakpoints as
17825 well; in this case, place @samp{task @var{taskno}} before the
17826 breakpoint condition (before the @code{if}).
17834 (@value{GDBP}) info tasks
17835 ID TID P-ID Pri State Name
17836 1 140022020 0 15 Child Activation Wait main_task
17837 2 140045060 1 15 Accept/Select Wait t2
17838 3 140044840 1 15 Runnable t1
17839 * 4 140056040 1 15 Runnable t3
17840 (@value{GDBP}) b 15 task 2
17841 Breakpoint 5 at 0x120044cb0: file test_task_debug.adb, line 15.
17842 (@value{GDBP}) cont
17847 Breakpoint 5, test_task_debug () at test_task_debug.adb:15
17849 (@value{GDBP}) info tasks
17850 ID TID P-ID Pri State Name
17851 1 140022020 0 15 Child Activation Wait main_task
17852 * 2 140045060 1 15 Runnable t2
17853 3 140044840 1 15 Runnable t1
17854 4 140056040 1 15 Delay Sleep t3
17858 @node Ada Tasks and Core Files
17859 @subsubsection Tasking Support when Debugging Core Files
17860 @cindex Ada tasking and core file debugging
17862 When inspecting a core file, as opposed to debugging a live program,
17863 tasking support may be limited or even unavailable, depending on
17864 the platform being used.
17865 For instance, on x86-linux, the list of tasks is available, but task
17866 switching is not supported.
17868 On certain platforms, the debugger needs to perform some
17869 memory writes in order to provide Ada tasking support. When inspecting
17870 a core file, this means that the core file must be opened with read-write
17871 privileges, using the command @samp{"set write on"} (@pxref{Patching}).
17872 Under these circumstances, you should make a backup copy of the core
17873 file before inspecting it with @value{GDBN}.
17875 @node Ravenscar Profile
17876 @subsubsection Tasking Support when using the Ravenscar Profile
17877 @cindex Ravenscar Profile
17879 The @dfn{Ravenscar Profile} is a subset of the Ada tasking features,
17880 specifically designed for systems with safety-critical real-time
17884 @kindex set ravenscar task-switching on
17885 @cindex task switching with program using Ravenscar Profile
17886 @item set ravenscar task-switching on
17887 Allows task switching when debugging a program that uses the Ravenscar
17888 Profile. This is the default.
17890 @kindex set ravenscar task-switching off
17891 @item set ravenscar task-switching off
17892 Turn off task switching when debugging a program that uses the Ravenscar
17893 Profile. This is mostly intended to disable the code that adds support
17894 for the Ravenscar Profile, in case a bug in either @value{GDBN} or in
17895 the Ravenscar runtime is preventing @value{GDBN} from working properly.
17896 To be effective, this command should be run before the program is started.
17898 @kindex show ravenscar task-switching
17899 @item show ravenscar task-switching
17900 Show whether it is possible to switch from task to task in a program
17901 using the Ravenscar Profile.
17906 @subsubsection Ada Settings
17907 @cindex Ada settings
17910 @kindex set varsize-limit
17911 @item set varsize-limit @var{size}
17912 Prevent @value{GDBN} from attempting to evaluate objects whose size
17913 is above the given limit (@var{size}) when those sizes are computed
17914 from run-time quantities. This is typically the case when the object
17915 has a variable size, such as an array whose bounds are not known at
17916 compile time for example. Setting @var{size} to @code{unlimited}
17917 removes the size limitation. By default, the limit is about 65KB.
17919 The purpose of having such a limit is to prevent @value{GDBN} from
17920 trying to grab enormous chunks of virtual memory when asked to evaluate
17921 a quantity whose bounds have been corrupted or have not yet been fully
17922 initialized. The limit applies to the results of some subexpressions
17923 as well as to complete expressions. For example, an expression denoting
17924 a simple integer component, such as @code{x.y.z}, may fail if the size of
17925 @code{x.y} is variable and exceeds @code{size}. On the other hand,
17926 @value{GDBN} is sometimes clever; the expression @code{A(i)}, where
17927 @code{A} is an array variable with non-constant size, will generally
17928 succeed regardless of the bounds on @code{A}, as long as the component
17929 size is less than @var{size}.
17931 @kindex show varsize-limit
17932 @item show varsize-limit
17933 Show the limit on types whose size is determined by run-time quantities.
17937 @subsubsection Known Peculiarities of Ada Mode
17938 @cindex Ada, problems
17940 Besides the omissions listed previously (@pxref{Omissions from Ada}),
17941 we know of several problems with and limitations of Ada mode in
17943 some of which will be fixed with planned future releases of the debugger
17944 and the GNU Ada compiler.
17948 Static constants that the compiler chooses not to materialize as objects in
17949 storage are invisible to the debugger.
17952 Named parameter associations in function argument lists are ignored (the
17953 argument lists are treated as positional).
17956 Many useful library packages are currently invisible to the debugger.
17959 Fixed-point arithmetic, conversions, input, and output is carried out using
17960 floating-point arithmetic, and may give results that only approximate those on
17964 The GNAT compiler never generates the prefix @code{Standard} for any of
17965 the standard symbols defined by the Ada language. @value{GDBN} knows about
17966 this: it will strip the prefix from names when you use it, and will never
17967 look for a name you have so qualified among local symbols, nor match against
17968 symbols in other packages or subprograms. If you have
17969 defined entities anywhere in your program other than parameters and
17970 local variables whose simple names match names in @code{Standard},
17971 GNAT's lack of qualification here can cause confusion. When this happens,
17972 you can usually resolve the confusion
17973 by qualifying the problematic names with package
17974 @code{Standard} explicitly.
17977 Older versions of the compiler sometimes generate erroneous debugging
17978 information, resulting in the debugger incorrectly printing the value
17979 of affected entities. In some cases, the debugger is able to work
17980 around an issue automatically. In other cases, the debugger is able
17981 to work around the issue, but the work-around has to be specifically
17984 @kindex set ada trust-PAD-over-XVS
17985 @kindex show ada trust-PAD-over-XVS
17988 @item set ada trust-PAD-over-XVS on
17989 Configure GDB to strictly follow the GNAT encoding when computing the
17990 value of Ada entities, particularly when @code{PAD} and @code{PAD___XVS}
17991 types are involved (see @code{ada/exp_dbug.ads} in the GCC sources for
17992 a complete description of the encoding used by the GNAT compiler).
17993 This is the default.
17995 @item set ada trust-PAD-over-XVS off
17996 This is related to the encoding using by the GNAT compiler. If @value{GDBN}
17997 sometimes prints the wrong value for certain entities, changing @code{ada
17998 trust-PAD-over-XVS} to @code{off} activates a work-around which may fix
17999 the issue. It is always safe to set @code{ada trust-PAD-over-XVS} to
18000 @code{off}, but this incurs a slight performance penalty, so it is
18001 recommended to leave this setting to @code{on} unless necessary.
18005 @cindex GNAT descriptive types
18006 @cindex GNAT encoding
18007 Internally, the debugger also relies on the compiler following a number
18008 of conventions known as the @samp{GNAT Encoding}, all documented in
18009 @file{gcc/ada/exp_dbug.ads} in the GCC sources. This encoding describes
18010 how the debugging information should be generated for certain types.
18011 In particular, this convention makes use of @dfn{descriptive types},
18012 which are artificial types generated purely to help the debugger.
18014 These encodings were defined at a time when the debugging information
18015 format used was not powerful enough to describe some of the more complex
18016 types available in Ada. Since DWARF allows us to express nearly all
18017 Ada features, the long-term goal is to slowly replace these descriptive
18018 types by their pure DWARF equivalent. To facilitate that transition,
18019 a new maintenance option is available to force the debugger to ignore
18020 those descriptive types. It allows the user to quickly evaluate how
18021 well @value{GDBN} works without them.
18025 @kindex maint ada set ignore-descriptive-types
18026 @item maintenance ada set ignore-descriptive-types [on|off]
18027 Control whether the debugger should ignore descriptive types.
18028 The default is not to ignore descriptives types (@code{off}).
18030 @kindex maint ada show ignore-descriptive-types
18031 @item maintenance ada show ignore-descriptive-types
18032 Show if descriptive types are ignored by @value{GDBN}.
18036 @node Unsupported Languages
18037 @section Unsupported Languages
18039 @cindex unsupported languages
18040 @cindex minimal language
18041 In addition to the other fully-supported programming languages,
18042 @value{GDBN} also provides a pseudo-language, called @code{minimal}.
18043 It does not represent a real programming language, but provides a set
18044 of capabilities close to what the C or assembly languages provide.
18045 This should allow most simple operations to be performed while debugging
18046 an application that uses a language currently not supported by @value{GDBN}.
18048 If the language is set to @code{auto}, @value{GDBN} will automatically
18049 select this language if the current frame corresponds to an unsupported
18053 @chapter Examining the Symbol Table
18055 The commands described in this chapter allow you to inquire about the
18056 symbols (names of variables, functions and types) defined in your
18057 program. This information is inherent in the text of your program and
18058 does not change as your program executes. @value{GDBN} finds it in your
18059 program's symbol table, in the file indicated when you started @value{GDBN}
18060 (@pxref{File Options, ,Choosing Files}), or by one of the
18061 file-management commands (@pxref{Files, ,Commands to Specify Files}).
18063 @cindex symbol names
18064 @cindex names of symbols
18065 @cindex quoting names
18066 @anchor{quoting names}
18067 Occasionally, you may need to refer to symbols that contain unusual
18068 characters, which @value{GDBN} ordinarily treats as word delimiters. The
18069 most frequent case is in referring to static variables in other
18070 source files (@pxref{Variables,,Program Variables}). File names
18071 are recorded in object files as debugging symbols, but @value{GDBN} would
18072 ordinarily parse a typical file name, like @file{foo.c}, as the three words
18073 @samp{foo} @samp{.} @samp{c}. To allow @value{GDBN} to recognize
18074 @samp{foo.c} as a single symbol, enclose it in single quotes; for example,
18081 looks up the value of @code{x} in the scope of the file @file{foo.c}.
18084 @cindex case-insensitive symbol names
18085 @cindex case sensitivity in symbol names
18086 @kindex set case-sensitive
18087 @item set case-sensitive on
18088 @itemx set case-sensitive off
18089 @itemx set case-sensitive auto
18090 Normally, when @value{GDBN} looks up symbols, it matches their names
18091 with case sensitivity determined by the current source language.
18092 Occasionally, you may wish to control that. The command @code{set
18093 case-sensitive} lets you do that by specifying @code{on} for
18094 case-sensitive matches or @code{off} for case-insensitive ones. If
18095 you specify @code{auto}, case sensitivity is reset to the default
18096 suitable for the source language. The default is case-sensitive
18097 matches for all languages except for Fortran, for which the default is
18098 case-insensitive matches.
18100 @kindex show case-sensitive
18101 @item show case-sensitive
18102 This command shows the current setting of case sensitivity for symbols
18105 @kindex set print type methods
18106 @item set print type methods
18107 @itemx set print type methods on
18108 @itemx set print type methods off
18109 Normally, when @value{GDBN} prints a class, it displays any methods
18110 declared in that class. You can control this behavior either by
18111 passing the appropriate flag to @code{ptype}, or using @command{set
18112 print type methods}. Specifying @code{on} will cause @value{GDBN} to
18113 display the methods; this is the default. Specifying @code{off} will
18114 cause @value{GDBN} to omit the methods.
18116 @kindex show print type methods
18117 @item show print type methods
18118 This command shows the current setting of method display when printing
18121 @kindex set print type nested-type-limit
18122 @item set print type nested-type-limit @var{limit}
18123 @itemx set print type nested-type-limit unlimited
18124 Set the limit of displayed nested types that the type printer will
18125 show. A @var{limit} of @code{unlimited} or @code{-1} will show all
18126 nested definitions. By default, the type printer will not show any nested
18127 types defined in classes.
18129 @kindex show print type nested-type-limit
18130 @item show print type nested-type-limit
18131 This command shows the current display limit of nested types when
18134 @kindex set print type typedefs
18135 @item set print type typedefs
18136 @itemx set print type typedefs on
18137 @itemx set print type typedefs off
18139 Normally, when @value{GDBN} prints a class, it displays any typedefs
18140 defined in that class. You can control this behavior either by
18141 passing the appropriate flag to @code{ptype}, or using @command{set
18142 print type typedefs}. Specifying @code{on} will cause @value{GDBN} to
18143 display the typedef definitions; this is the default. Specifying
18144 @code{off} will cause @value{GDBN} to omit the typedef definitions.
18145 Note that this controls whether the typedef definition itself is
18146 printed, not whether typedef names are substituted when printing other
18149 @kindex show print type typedefs
18150 @item show print type typedefs
18151 This command shows the current setting of typedef display when
18154 @kindex info address
18155 @cindex address of a symbol
18156 @item info address @var{symbol}
18157 Describe where the data for @var{symbol} is stored. For a register
18158 variable, this says which register it is kept in. For a non-register
18159 local variable, this prints the stack-frame offset at which the variable
18162 Note the contrast with @samp{print &@var{symbol}}, which does not work
18163 at all for a register variable, and for a stack local variable prints
18164 the exact address of the current instantiation of the variable.
18166 @kindex info symbol
18167 @cindex symbol from address
18168 @cindex closest symbol and offset for an address
18169 @item info symbol @var{addr}
18170 Print the name of a symbol which is stored at the address @var{addr}.
18171 If no symbol is stored exactly at @var{addr}, @value{GDBN} prints the
18172 nearest symbol and an offset from it:
18175 (@value{GDBP}) info symbol 0x54320
18176 _initialize_vx + 396 in section .text
18180 This is the opposite of the @code{info address} command. You can use
18181 it to find out the name of a variable or a function given its address.
18183 For dynamically linked executables, the name of executable or shared
18184 library containing the symbol is also printed:
18187 (@value{GDBP}) info symbol 0x400225
18188 _start + 5 in section .text of /tmp/a.out
18189 (@value{GDBP}) info symbol 0x2aaaac2811cf
18190 __read_nocancel + 6 in section .text of /usr/lib64/libc.so.6
18195 @item demangle @r{[}-l @var{language}@r{]} @r{[}@var{--}@r{]} @var{name}
18196 Demangle @var{name}.
18197 If @var{language} is provided it is the name of the language to demangle
18198 @var{name} in. Otherwise @var{name} is demangled in the current language.
18200 The @samp{--} option specifies the end of options,
18201 and is useful when @var{name} begins with a dash.
18203 The parameter @code{demangle-style} specifies how to interpret the kind
18204 of mangling used. @xref{Print Settings}.
18207 @item whatis[/@var{flags}] [@var{arg}]
18208 Print the data type of @var{arg}, which can be either an expression
18209 or a name of a data type. With no argument, print the data type of
18210 @code{$}, the last value in the value history.
18212 If @var{arg} is an expression (@pxref{Expressions, ,Expressions}), it
18213 is not actually evaluated, and any side-effecting operations (such as
18214 assignments or function calls) inside it do not take place.
18216 If @var{arg} is a variable or an expression, @code{whatis} prints its
18217 literal type as it is used in the source code. If the type was
18218 defined using a @code{typedef}, @code{whatis} will @emph{not} print
18219 the data type underlying the @code{typedef}. If the type of the
18220 variable or the expression is a compound data type, such as
18221 @code{struct} or @code{class}, @code{whatis} never prints their
18222 fields or methods. It just prints the @code{struct}/@code{class}
18223 name (a.k.a.@: its @dfn{tag}). If you want to see the members of
18224 such a compound data type, use @code{ptype}.
18226 If @var{arg} is a type name that was defined using @code{typedef},
18227 @code{whatis} @dfn{unrolls} only one level of that @code{typedef}.
18228 Unrolling means that @code{whatis} will show the underlying type used
18229 in the @code{typedef} declaration of @var{arg}. However, if that
18230 underlying type is also a @code{typedef}, @code{whatis} will not
18233 For C code, the type names may also have the form @samp{class
18234 @var{class-name}}, @samp{struct @var{struct-tag}}, @samp{union
18235 @var{union-tag}} or @samp{enum @var{enum-tag}}.
18237 @var{flags} can be used to modify how the type is displayed.
18238 Available flags are:
18242 Display in ``raw'' form. Normally, @value{GDBN} substitutes template
18243 parameters and typedefs defined in a class when printing the class'
18244 members. The @code{/r} flag disables this.
18247 Do not print methods defined in the class.
18250 Print methods defined in the class. This is the default, but the flag
18251 exists in case you change the default with @command{set print type methods}.
18254 Do not print typedefs defined in the class. Note that this controls
18255 whether the typedef definition itself is printed, not whether typedef
18256 names are substituted when printing other types.
18259 Print typedefs defined in the class. This is the default, but the flag
18260 exists in case you change the default with @command{set print type typedefs}.
18263 Print the offsets and sizes of fields in a struct, similar to what the
18264 @command{pahole} tool does. This option implies the @code{/tm} flags.
18266 For example, given the following declarations:
18302 Issuing a @kbd{ptype /o struct tuv} command would print:
18305 (@value{GDBP}) ptype /o struct tuv
18306 /* offset | size */ type = struct tuv @{
18307 /* 0 | 4 */ int a1;
18308 /* XXX 4-byte hole */
18309 /* 8 | 8 */ char *a2;
18310 /* 16 | 4 */ int a3;
18312 /* total size (bytes): 24 */
18316 Notice the format of the first column of comments. There, you can
18317 find two parts separated by the @samp{|} character: the @emph{offset},
18318 which indicates where the field is located inside the struct, in
18319 bytes, and the @emph{size} of the field. Another interesting line is
18320 the marker of a @emph{hole} in the struct, indicating that it may be
18321 possible to pack the struct and make it use less space by reorganizing
18324 It is also possible to print offsets inside an union:
18327 (@value{GDBP}) ptype /o union qwe
18328 /* offset | size */ type = union qwe @{
18329 /* 24 */ struct tuv @{
18330 /* 0 | 4 */ int a1;
18331 /* XXX 4-byte hole */
18332 /* 8 | 8 */ char *a2;
18333 /* 16 | 4 */ int a3;
18335 /* total size (bytes): 24 */
18337 /* 40 */ struct xyz @{
18338 /* 0 | 4 */ int f1;
18339 /* 4 | 1 */ char f2;
18340 /* XXX 3-byte hole */
18341 /* 8 | 8 */ void *f3;
18342 /* 16 | 24 */ struct tuv @{
18343 /* 16 | 4 */ int a1;
18344 /* XXX 4-byte hole */
18345 /* 24 | 8 */ char *a2;
18346 /* 32 | 4 */ int a3;
18348 /* total size (bytes): 24 */
18351 /* total size (bytes): 40 */
18354 /* total size (bytes): 40 */
18358 In this case, since @code{struct tuv} and @code{struct xyz} occupy the
18359 same space (because we are dealing with an union), the offset is not
18360 printed for them. However, you can still examine the offset of each
18361 of these structures' fields.
18363 Another useful scenario is printing the offsets of a struct containing
18367 (@value{GDBP}) ptype /o struct tyu
18368 /* offset | size */ type = struct tyu @{
18369 /* 0:31 | 4 */ int a1 : 1;
18370 /* 0:28 | 4 */ int a2 : 3;
18371 /* 0: 5 | 4 */ int a3 : 23;
18372 /* 3: 3 | 1 */ signed char a4 : 2;
18373 /* XXX 3-bit hole */
18374 /* XXX 4-byte hole */
18375 /* 8 | 8 */ int64_t a5;
18376 /* 16: 0 | 4 */ int a6 : 5;
18377 /* 16: 5 | 8 */ int64_t a7 : 3;
18378 "/* XXX 7-byte padding */
18380 /* total size (bytes): 24 */
18384 Note how the offset information is now extended to also include the
18385 first bit of the bitfield.
18389 @item ptype[/@var{flags}] [@var{arg}]
18390 @code{ptype} accepts the same arguments as @code{whatis}, but prints a
18391 detailed description of the type, instead of just the name of the type.
18392 @xref{Expressions, ,Expressions}.
18394 Contrary to @code{whatis}, @code{ptype} always unrolls any
18395 @code{typedef}s in its argument declaration, whether the argument is
18396 a variable, expression, or a data type. This means that @code{ptype}
18397 of a variable or an expression will not print literally its type as
18398 present in the source code---use @code{whatis} for that. @code{typedef}s at
18399 the pointer or reference targets are also unrolled. Only @code{typedef}s of
18400 fields, methods and inner @code{class typedef}s of @code{struct}s,
18401 @code{class}es and @code{union}s are not unrolled even with @code{ptype}.
18403 For example, for this variable declaration:
18406 typedef double real_t;
18407 struct complex @{ real_t real; double imag; @};
18408 typedef struct complex complex_t;
18410 real_t *real_pointer_var;
18414 the two commands give this output:
18418 (@value{GDBP}) whatis var
18420 (@value{GDBP}) ptype var
18421 type = struct complex @{
18425 (@value{GDBP}) whatis complex_t
18426 type = struct complex
18427 (@value{GDBP}) whatis struct complex
18428 type = struct complex
18429 (@value{GDBP}) ptype struct complex
18430 type = struct complex @{
18434 (@value{GDBP}) whatis real_pointer_var
18436 (@value{GDBP}) ptype real_pointer_var
18442 As with @code{whatis}, using @code{ptype} without an argument refers to
18443 the type of @code{$}, the last value in the value history.
18445 @cindex incomplete type
18446 Sometimes, programs use opaque data types or incomplete specifications
18447 of complex data structure. If the debug information included in the
18448 program does not allow @value{GDBN} to display a full declaration of
18449 the data type, it will say @samp{<incomplete type>}. For example,
18450 given these declarations:
18454 struct foo *fooptr;
18458 but no definition for @code{struct foo} itself, @value{GDBN} will say:
18461 (@value{GDBP}) ptype foo
18462 $1 = <incomplete type>
18466 ``Incomplete type'' is C terminology for data types that are not
18467 completely specified.
18469 @cindex unknown type
18470 Othertimes, information about a variable's type is completely absent
18471 from the debug information included in the program. This most often
18472 happens when the program or library where the variable is defined
18473 includes no debug information at all. @value{GDBN} knows the variable
18474 exists from inspecting the linker/loader symbol table (e.g., the ELF
18475 dynamic symbol table), but such symbols do not contain type
18476 information. Inspecting the type of a (global) variable for which
18477 @value{GDBN} has no type information shows:
18480 (@value{GDBP}) ptype var
18481 type = <data variable, no debug info>
18484 @xref{Variables, no debug info variables}, for how to print the values
18488 @item info types [-q] [@var{regexp}]
18489 Print a brief description of all types whose names match the regular
18490 expression @var{regexp} (or all types in your program, if you supply
18491 no argument). Each complete typename is matched as though it were a
18492 complete line; thus, @samp{i type value} gives information on all
18493 types in your program whose names include the string @code{value}, but
18494 @samp{i type ^value$} gives information only on types whose complete
18495 name is @code{value}.
18497 In programs using different languages, @value{GDBN} chooses the syntax
18498 to print the type description according to the
18499 @samp{set language} value: using @samp{set language auto}
18500 (see @ref{Automatically, ,Set Language Automatically}) means to use the
18501 language of the type, other values mean to use
18502 the manually specified language (see @ref{Manually, ,Set Language Manually}).
18504 This command differs from @code{ptype} in two ways: first, like
18505 @code{whatis}, it does not print a detailed description; second, it
18506 lists all source files and line numbers where a type is defined.
18508 The output from @samp{into types} is proceeded with a header line
18509 describing what types are being listed. The optional flag @samp{-q},
18510 which stands for @samp{quiet}, disables printing this header
18513 @kindex info type-printers
18514 @item info type-printers
18515 Versions of @value{GDBN} that ship with Python scripting enabled may
18516 have ``type printers'' available. When using @command{ptype} or
18517 @command{whatis}, these printers are consulted when the name of a type
18518 is needed. @xref{Type Printing API}, for more information on writing
18521 @code{info type-printers} displays all the available type printers.
18523 @kindex enable type-printer
18524 @kindex disable type-printer
18525 @item enable type-printer @var{name}@dots{}
18526 @item disable type-printer @var{name}@dots{}
18527 These commands can be used to enable or disable type printers.
18530 @cindex local variables
18531 @item info scope @var{location}
18532 List all the variables local to a particular scope. This command
18533 accepts a @var{location} argument---a function name, a source line, or
18534 an address preceded by a @samp{*}, and prints all the variables local
18535 to the scope defined by that location. (@xref{Specify Location}, for
18536 details about supported forms of @var{location}.) For example:
18539 (@value{GDBP}) @b{info scope command_line_handler}
18540 Scope for command_line_handler:
18541 Symbol rl is an argument at stack/frame offset 8, length 4.
18542 Symbol linebuffer is in static storage at address 0x150a18, length 4.
18543 Symbol linelength is in static storage at address 0x150a1c, length 4.
18544 Symbol p is a local variable in register $esi, length 4.
18545 Symbol p1 is a local variable in register $ebx, length 4.
18546 Symbol nline is a local variable in register $edx, length 4.
18547 Symbol repeat is a local variable at frame offset -8, length 4.
18551 This command is especially useful for determining what data to collect
18552 during a @dfn{trace experiment}, see @ref{Tracepoint Actions,
18555 @kindex info source
18557 Show information about the current source file---that is, the source file for
18558 the function containing the current point of execution:
18561 the name of the source file, and the directory containing it,
18563 the directory it was compiled in,
18565 its length, in lines,
18567 which programming language it is written in,
18569 if the debug information provides it, the program that compiled the file
18570 (which may include, e.g., the compiler version and command line arguments),
18572 whether the executable includes debugging information for that file, and
18573 if so, what format the information is in (e.g., STABS, Dwarf 2, etc.), and
18575 whether the debugging information includes information about
18576 preprocessor macros.
18580 @kindex info sources
18582 Print the names of all source files in your program for which there is
18583 debugging information, organized into two lists: files whose symbols
18584 have already been read, and files whose symbols will be read when needed.
18586 @kindex info functions
18587 @item info functions [-q]
18588 Print the names and data types of all defined functions.
18589 Similarly to @samp{info types}, this command groups its output by source
18590 files and annotates each function definition with its source line
18593 In programs using different languages, @value{GDBN} chooses the syntax
18594 to print the function name and type according to the
18595 @samp{set language} value: using @samp{set language auto}
18596 (see @ref{Automatically, ,Set Language Automatically}) means to use the
18597 language of the function, other values mean to use
18598 the manually specified language (see @ref{Manually, ,Set Language Manually}).
18600 The optional flag @samp{-q}, which stands for @samp{quiet}, disables
18601 printing header information and messages explaining why no functions
18604 @item info functions [-q] [-t @var{type_regexp}] [@var{regexp}]
18605 Like @samp{info functions}, but only print the names and data types
18606 of the functions selected with the provided regexp(s).
18608 If @var{regexp} is provided, print only the functions whose names
18609 match the regular expression @var{regexp}.
18610 Thus, @samp{info fun step} finds all functions whose
18611 names include @code{step}; @samp{info fun ^step} finds those whose names
18612 start with @code{step}. If a function name contains characters that
18613 conflict with the regular expression language (e.g.@:
18614 @samp{operator*()}), they may be quoted with a backslash.
18616 If @var{type_regexp} is provided, print only the functions whose
18617 types, as printed by the @code{whatis} command, match
18618 the regular expression @var{type_regexp}.
18619 If @var{type_regexp} contains space(s), it should be enclosed in
18620 quote characters. If needed, use backslash to escape the meaning
18621 of special characters or quotes.
18622 Thus, @samp{info fun -t '^int ('} finds the functions that return
18623 an integer; @samp{info fun -t '(.*int.*'} finds the functions that
18624 have an argument type containing int; @samp{info fun -t '^int (' ^step}
18625 finds the functions whose names start with @code{step} and that return
18628 If both @var{regexp} and @var{type_regexp} are provided, a function
18629 is printed only if its name matches @var{regexp} and its type matches
18633 @kindex info variables
18634 @item info variables [-q]
18635 Print the names and data types of all variables that are defined
18636 outside of functions (i.e.@: excluding local variables).
18637 The printed variables are grouped by source files and annotated with
18638 their respective source line numbers.
18640 In programs using different languages, @value{GDBN} chooses the syntax
18641 to print the variable name and type according to the
18642 @samp{set language} value: using @samp{set language auto}
18643 (see @ref{Automatically, ,Set Language Automatically}) means to use the
18644 language of the variable, other values mean to use
18645 the manually specified language (see @ref{Manually, ,Set Language Manually}).
18647 The optional flag @samp{-q}, which stands for @samp{quiet}, disables
18648 printing header information and messages explaining why no variables
18651 @item info variables [-q] [-t @var{type_regexp}] [@var{regexp}]
18652 Like @kbd{info variables}, but only print the variables selected
18653 with the provided regexp(s).
18655 If @var{regexp} is provided, print only the variables whose names
18656 match the regular expression @var{regexp}.
18658 If @var{type_regexp} is provided, print only the variables whose
18659 types, as printed by the @code{whatis} command, match
18660 the regular expression @var{type_regexp}.
18661 If @var{type_regexp} contains space(s), it should be enclosed in
18662 quote characters. If needed, use backslash to escape the meaning
18663 of special characters or quotes.
18665 If both @var{regexp} and @var{type_regexp} are provided, an argument
18666 is printed only if its name matches @var{regexp} and its type matches
18669 @kindex info classes
18670 @cindex Objective-C, classes and selectors
18672 @itemx info classes @var{regexp}
18673 Display all Objective-C classes in your program, or
18674 (with the @var{regexp} argument) all those matching a particular regular
18677 @kindex info selectors
18678 @item info selectors
18679 @itemx info selectors @var{regexp}
18680 Display all Objective-C selectors in your program, or
18681 (with the @var{regexp} argument) all those matching a particular regular
18685 This was never implemented.
18686 @kindex info methods
18688 @itemx info methods @var{regexp}
18689 The @code{info methods} command permits the user to examine all defined
18690 methods within C@t{++} program, or (with the @var{regexp} argument) a
18691 specific set of methods found in the various C@t{++} classes. Many
18692 C@t{++} classes provide a large number of methods. Thus, the output
18693 from the @code{ptype} command can be overwhelming and hard to use. The
18694 @code{info-methods} command filters the methods, printing only those
18695 which match the regular-expression @var{regexp}.
18698 @cindex opaque data types
18699 @kindex set opaque-type-resolution
18700 @item set opaque-type-resolution on
18701 Tell @value{GDBN} to resolve opaque types. An opaque type is a type
18702 declared as a pointer to a @code{struct}, @code{class}, or
18703 @code{union}---for example, @code{struct MyType *}---that is used in one
18704 source file although the full declaration of @code{struct MyType} is in
18705 another source file. The default is on.
18707 A change in the setting of this subcommand will not take effect until
18708 the next time symbols for a file are loaded.
18710 @item set opaque-type-resolution off
18711 Tell @value{GDBN} not to resolve opaque types. In this case, the type
18712 is printed as follows:
18714 @{<no data fields>@}
18717 @kindex show opaque-type-resolution
18718 @item show opaque-type-resolution
18719 Show whether opaque types are resolved or not.
18721 @kindex set print symbol-loading
18722 @cindex print messages when symbols are loaded
18723 @item set print symbol-loading
18724 @itemx set print symbol-loading full
18725 @itemx set print symbol-loading brief
18726 @itemx set print symbol-loading off
18727 The @code{set print symbol-loading} command allows you to control the
18728 printing of messages when @value{GDBN} loads symbol information.
18729 By default a message is printed for the executable and one for each
18730 shared library, and normally this is what you want. However, when
18731 debugging apps with large numbers of shared libraries these messages
18733 When set to @code{brief} a message is printed for each executable,
18734 and when @value{GDBN} loads a collection of shared libraries at once
18735 it will only print one message regardless of the number of shared
18736 libraries. When set to @code{off} no messages are printed.
18738 @kindex show print symbol-loading
18739 @item show print symbol-loading
18740 Show whether messages will be printed when a @value{GDBN} command
18741 entered from the keyboard causes symbol information to be loaded.
18743 @kindex maint print symbols
18744 @cindex symbol dump
18745 @kindex maint print psymbols
18746 @cindex partial symbol dump
18747 @kindex maint print msymbols
18748 @cindex minimal symbol dump
18749 @item maint print symbols @r{[}-pc @var{address}@r{]} @r{[}@var{filename}@r{]}
18750 @itemx maint print symbols @r{[}-objfile @var{objfile}@r{]} @r{[}-source @var{source}@r{]} @r{[}--@r{]} @r{[}@var{filename}@r{]}
18751 @itemx maint print psymbols @r{[}-objfile @var{objfile}@r{]} @r{[}-pc @var{address}@r{]} @r{[}--@r{]} @r{[}@var{filename}@r{]}
18752 @itemx maint print psymbols @r{[}-objfile @var{objfile}@r{]} @r{[}-source @var{source}@r{]} @r{[}--@r{]} @r{[}@var{filename}@r{]}
18753 @itemx maint print msymbols @r{[}-objfile @var{objfile}@r{]} @r{[}--@r{]} @r{[}@var{filename}@r{]}
18754 Write a dump of debugging symbol data into the file @var{filename} or
18755 the terminal if @var{filename} is unspecified.
18756 If @code{-objfile @var{objfile}} is specified, only dump symbols for
18758 If @code{-pc @var{address}} is specified, only dump symbols for the file
18759 with code at that address. Note that @var{address} may be a symbol like
18761 If @code{-source @var{source}} is specified, only dump symbols for that
18764 These commands are used to debug the @value{GDBN} symbol-reading code.
18765 These commands do not modify internal @value{GDBN} state, therefore
18766 @samp{maint print symbols} will only print symbols for already expanded symbol
18768 You can use the command @code{info sources} to find out which files these are.
18769 If you use @samp{maint print psymbols} instead, the dump shows information
18770 about symbols that @value{GDBN} only knows partially---that is, symbols
18771 defined in files that @value{GDBN} has skimmed, but not yet read completely.
18772 Finally, @samp{maint print msymbols} just dumps ``minimal symbols'', e.g.,
18775 @xref{Files, ,Commands to Specify Files}, for a discussion of how
18776 @value{GDBN} reads symbols (in the description of @code{symbol-file}).
18778 @kindex maint info symtabs
18779 @kindex maint info psymtabs
18780 @cindex listing @value{GDBN}'s internal symbol tables
18781 @cindex symbol tables, listing @value{GDBN}'s internal
18782 @cindex full symbol tables, listing @value{GDBN}'s internal
18783 @cindex partial symbol tables, listing @value{GDBN}'s internal
18784 @item maint info symtabs @r{[} @var{regexp} @r{]}
18785 @itemx maint info psymtabs @r{[} @var{regexp} @r{]}
18787 List the @code{struct symtab} or @code{struct partial_symtab}
18788 structures whose names match @var{regexp}. If @var{regexp} is not
18789 given, list them all. The output includes expressions which you can
18790 copy into a @value{GDBN} debugging this one to examine a particular
18791 structure in more detail. For example:
18794 (@value{GDBP}) maint info psymtabs dwarf2read
18795 @{ objfile /home/gnu/build/gdb/gdb
18796 ((struct objfile *) 0x82e69d0)
18797 @{ psymtab /home/gnu/src/gdb/dwarf2read.c
18798 ((struct partial_symtab *) 0x8474b10)
18801 text addresses 0x814d3c8 -- 0x8158074
18802 globals (* (struct partial_symbol **) 0x8507a08 @@ 9)
18803 statics (* (struct partial_symbol **) 0x40e95b78 @@ 2882)
18804 dependencies (none)
18807 (@value{GDBP}) maint info symtabs
18811 We see that there is one partial symbol table whose filename contains
18812 the string @samp{dwarf2read}, belonging to the @samp{gdb} executable;
18813 and we see that @value{GDBN} has not read in any symtabs yet at all.
18814 If we set a breakpoint on a function, that will cause @value{GDBN} to
18815 read the symtab for the compilation unit containing that function:
18818 (@value{GDBP}) break dwarf2_psymtab_to_symtab
18819 Breakpoint 1 at 0x814e5da: file /home/gnu/src/gdb/dwarf2read.c,
18821 (@value{GDBP}) maint info symtabs
18822 @{ objfile /home/gnu/build/gdb/gdb
18823 ((struct objfile *) 0x82e69d0)
18824 @{ symtab /home/gnu/src/gdb/dwarf2read.c
18825 ((struct symtab *) 0x86c1f38)
18828 blockvector ((struct blockvector *) 0x86c1bd0) (primary)
18829 linetable ((struct linetable *) 0x8370fa0)
18830 debugformat DWARF 2
18836 @kindex maint info line-table
18837 @cindex listing @value{GDBN}'s internal line tables
18838 @cindex line tables, listing @value{GDBN}'s internal
18839 @item maint info line-table @r{[} @var{regexp} @r{]}
18841 List the @code{struct linetable} from all @code{struct symtab}
18842 instances whose name matches @var{regexp}. If @var{regexp} is not
18843 given, list the @code{struct linetable} from all @code{struct symtab}.
18845 @kindex maint set symbol-cache-size
18846 @cindex symbol cache size
18847 @item maint set symbol-cache-size @var{size}
18848 Set the size of the symbol cache to @var{size}.
18849 The default size is intended to be good enough for debugging
18850 most applications. This option exists to allow for experimenting
18851 with different sizes.
18853 @kindex maint show symbol-cache-size
18854 @item maint show symbol-cache-size
18855 Show the size of the symbol cache.
18857 @kindex maint print symbol-cache
18858 @cindex symbol cache, printing its contents
18859 @item maint print symbol-cache
18860 Print the contents of the symbol cache.
18861 This is useful when debugging symbol cache issues.
18863 @kindex maint print symbol-cache-statistics
18864 @cindex symbol cache, printing usage statistics
18865 @item maint print symbol-cache-statistics
18866 Print symbol cache usage statistics.
18867 This helps determine how well the cache is being utilized.
18869 @kindex maint flush-symbol-cache
18870 @cindex symbol cache, flushing
18871 @item maint flush-symbol-cache
18872 Flush the contents of the symbol cache, all entries are removed.
18873 This command is useful when debugging the symbol cache.
18874 It is also useful when collecting performance data.
18879 @chapter Altering Execution
18881 Once you think you have found an error in your program, you might want to
18882 find out for certain whether correcting the apparent error would lead to
18883 correct results in the rest of the run. You can find the answer by
18884 experiment, using the @value{GDBN} features for altering execution of the
18887 For example, you can store new values into variables or memory
18888 locations, give your program a signal, restart it at a different
18889 address, or even return prematurely from a function.
18892 * Assignment:: Assignment to variables
18893 * Jumping:: Continuing at a different address
18894 * Signaling:: Giving your program a signal
18895 * Returning:: Returning from a function
18896 * Calling:: Calling your program's functions
18897 * Patching:: Patching your program
18898 * Compiling and Injecting Code:: Compiling and injecting code in @value{GDBN}
18902 @section Assignment to Variables
18905 @cindex setting variables
18906 To alter the value of a variable, evaluate an assignment expression.
18907 @xref{Expressions, ,Expressions}. For example,
18914 stores the value 4 into the variable @code{x}, and then prints the
18915 value of the assignment expression (which is 4).
18916 @xref{Languages, ,Using @value{GDBN} with Different Languages}, for more
18917 information on operators in supported languages.
18919 @kindex set variable
18920 @cindex variables, setting
18921 If you are not interested in seeing the value of the assignment, use the
18922 @code{set} command instead of the @code{print} command. @code{set} is
18923 really the same as @code{print} except that the expression's value is
18924 not printed and is not put in the value history (@pxref{Value History,
18925 ,Value History}). The expression is evaluated only for its effects.
18927 If the beginning of the argument string of the @code{set} command
18928 appears identical to a @code{set} subcommand, use the @code{set
18929 variable} command instead of just @code{set}. This command is identical
18930 to @code{set} except for its lack of subcommands. For example, if your
18931 program has a variable @code{width}, you get an error if you try to set
18932 a new value with just @samp{set width=13}, because @value{GDBN} has the
18933 command @code{set width}:
18936 (@value{GDBP}) whatis width
18938 (@value{GDBP}) p width
18940 (@value{GDBP}) set width=47
18941 Invalid syntax in expression.
18945 The invalid expression, of course, is @samp{=47}. In
18946 order to actually set the program's variable @code{width}, use
18949 (@value{GDBP}) set var width=47
18952 Because the @code{set} command has many subcommands that can conflict
18953 with the names of program variables, it is a good idea to use the
18954 @code{set variable} command instead of just @code{set}. For example, if
18955 your program has a variable @code{g}, you run into problems if you try
18956 to set a new value with just @samp{set g=4}, because @value{GDBN} has
18957 the command @code{set gnutarget}, abbreviated @code{set g}:
18961 (@value{GDBP}) whatis g
18965 (@value{GDBP}) set g=4
18969 The program being debugged has been started already.
18970 Start it from the beginning? (y or n) y
18971 Starting program: /home/smith/cc_progs/a.out
18972 "/home/smith/cc_progs/a.out": can't open to read symbols:
18973 Invalid bfd target.
18974 (@value{GDBP}) show g
18975 The current BFD target is "=4".
18980 The program variable @code{g} did not change, and you silently set the
18981 @code{gnutarget} to an invalid value. In order to set the variable
18985 (@value{GDBP}) set var g=4
18988 @value{GDBN} allows more implicit conversions in assignments than C; you can
18989 freely store an integer value into a pointer variable or vice versa,
18990 and you can convert any structure to any other structure that is the
18991 same length or shorter.
18992 @comment FIXME: how do structs align/pad in these conversions?
18993 @comment /doc@cygnus.com 18dec1990
18995 To store values into arbitrary places in memory, use the @samp{@{@dots{}@}}
18996 construct to generate a value of specified type at a specified address
18997 (@pxref{Expressions, ,Expressions}). For example, @code{@{int@}0x83040} refers
18998 to memory location @code{0x83040} as an integer (which implies a certain size
18999 and representation in memory), and
19002 set @{int@}0x83040 = 4
19006 stores the value 4 into that memory location.
19009 @section Continuing at a Different Address
19011 Ordinarily, when you continue your program, you do so at the place where
19012 it stopped, with the @code{continue} command. You can instead continue at
19013 an address of your own choosing, with the following commands:
19017 @kindex j @r{(@code{jump})}
19018 @item jump @var{location}
19019 @itemx j @var{location}
19020 Resume execution at @var{location}. Execution stops again immediately
19021 if there is a breakpoint there. @xref{Specify Location}, for a description
19022 of the different forms of @var{location}. It is common
19023 practice to use the @code{tbreak} command in conjunction with
19024 @code{jump}. @xref{Set Breaks, ,Setting Breakpoints}.
19026 The @code{jump} command does not change the current stack frame, or
19027 the stack pointer, or the contents of any memory location or any
19028 register other than the program counter. If @var{location} is in
19029 a different function from the one currently executing, the results may
19030 be bizarre if the two functions expect different patterns of arguments or
19031 of local variables. For this reason, the @code{jump} command requests
19032 confirmation if the specified line is not in the function currently
19033 executing. However, even bizarre results are predictable if you are
19034 well acquainted with the machine-language code of your program.
19037 On many systems, you can get much the same effect as the @code{jump}
19038 command by storing a new value into the register @code{$pc}. The
19039 difference is that this does not start your program running; it only
19040 changes the address of where it @emph{will} run when you continue. For
19048 makes the next @code{continue} command or stepping command execute at
19049 address @code{0x485}, rather than at the address where your program stopped.
19050 @xref{Continuing and Stepping, ,Continuing and Stepping}.
19052 The most common occasion to use the @code{jump} command is to back
19053 up---perhaps with more breakpoints set---over a portion of a program
19054 that has already executed, in order to examine its execution in more
19059 @section Giving your Program a Signal
19060 @cindex deliver a signal to a program
19064 @item signal @var{signal}
19065 Resume execution where your program is stopped, but immediately give it the
19066 signal @var{signal}. The @var{signal} can be the name or the number of a
19067 signal. For example, on many systems @code{signal 2} and @code{signal
19068 SIGINT} are both ways of sending an interrupt signal.
19070 Alternatively, if @var{signal} is zero, continue execution without
19071 giving a signal. This is useful when your program stopped on account of
19072 a signal and would ordinarily see the signal when resumed with the
19073 @code{continue} command; @samp{signal 0} causes it to resume without a
19076 @emph{Note:} When resuming a multi-threaded program, @var{signal} is
19077 delivered to the currently selected thread, not the thread that last
19078 reported a stop. This includes the situation where a thread was
19079 stopped due to a signal. So if you want to continue execution
19080 suppressing the signal that stopped a thread, you should select that
19081 same thread before issuing the @samp{signal 0} command. If you issue
19082 the @samp{signal 0} command with another thread as the selected one,
19083 @value{GDBN} detects that and asks for confirmation.
19085 Invoking the @code{signal} command is not the same as invoking the
19086 @code{kill} utility from the shell. Sending a signal with @code{kill}
19087 causes @value{GDBN} to decide what to do with the signal depending on
19088 the signal handling tables (@pxref{Signals}). The @code{signal} command
19089 passes the signal directly to your program.
19091 @code{signal} does not repeat when you press @key{RET} a second time
19092 after executing the command.
19094 @kindex queue-signal
19095 @item queue-signal @var{signal}
19096 Queue @var{signal} to be delivered immediately to the current thread
19097 when execution of the thread resumes. The @var{signal} can be the name or
19098 the number of a signal. For example, on many systems @code{signal 2} and
19099 @code{signal SIGINT} are both ways of sending an interrupt signal.
19100 The handling of the signal must be set to pass the signal to the program,
19101 otherwise @value{GDBN} will report an error.
19102 You can control the handling of signals from @value{GDBN} with the
19103 @code{handle} command (@pxref{Signals}).
19105 Alternatively, if @var{signal} is zero, any currently queued signal
19106 for the current thread is discarded and when execution resumes no signal
19107 will be delivered. This is useful when your program stopped on account
19108 of a signal and would ordinarily see the signal when resumed with the
19109 @code{continue} command.
19111 This command differs from the @code{signal} command in that the signal
19112 is just queued, execution is not resumed. And @code{queue-signal} cannot
19113 be used to pass a signal whose handling state has been set to @code{nopass}
19118 @xref{stepping into signal handlers}, for information on how stepping
19119 commands behave when the thread has a signal queued.
19122 @section Returning from a Function
19125 @cindex returning from a function
19128 @itemx return @var{expression}
19129 You can cancel execution of a function call with the @code{return}
19130 command. If you give an
19131 @var{expression} argument, its value is used as the function's return
19135 When you use @code{return}, @value{GDBN} discards the selected stack frame
19136 (and all frames within it). You can think of this as making the
19137 discarded frame return prematurely. If you wish to specify a value to
19138 be returned, give that value as the argument to @code{return}.
19140 This pops the selected stack frame (@pxref{Selection, ,Selecting a
19141 Frame}), and any other frames inside of it, leaving its caller as the
19142 innermost remaining frame. That frame becomes selected. The
19143 specified value is stored in the registers used for returning values
19146 The @code{return} command does not resume execution; it leaves the
19147 program stopped in the state that would exist if the function had just
19148 returned. In contrast, the @code{finish} command (@pxref{Continuing
19149 and Stepping, ,Continuing and Stepping}) resumes execution until the
19150 selected stack frame returns naturally.
19152 @value{GDBN} needs to know how the @var{expression} argument should be set for
19153 the inferior. The concrete registers assignment depends on the OS ABI and the
19154 type being returned by the selected stack frame. For example it is common for
19155 OS ABI to return floating point values in FPU registers while integer values in
19156 CPU registers. Still some ABIs return even floating point values in CPU
19157 registers. Larger integer widths (such as @code{long long int}) also have
19158 specific placement rules. @value{GDBN} already knows the OS ABI from its
19159 current target so it needs to find out also the type being returned to make the
19160 assignment into the right register(s).
19162 Normally, the selected stack frame has debug info. @value{GDBN} will always
19163 use the debug info instead of the implicit type of @var{expression} when the
19164 debug info is available. For example, if you type @kbd{return -1}, and the
19165 function in the current stack frame is declared to return a @code{long long
19166 int}, @value{GDBN} transparently converts the implicit @code{int} value of -1
19167 into a @code{long long int}:
19170 Breakpoint 1, func () at gdb.base/return-nodebug.c:29
19172 (@value{GDBP}) return -1
19173 Make func return now? (y or n) y
19174 #0 0x004004f6 in main () at gdb.base/return-nodebug.c:43
19175 43 printf ("result=%lld\n", func ());
19179 However, if the selected stack frame does not have a debug info, e.g., if the
19180 function was compiled without debug info, @value{GDBN} has to find out the type
19181 to return from user. Specifying a different type by mistake may set the value
19182 in different inferior registers than the caller code expects. For example,
19183 typing @kbd{return -1} with its implicit type @code{int} would set only a part
19184 of a @code{long long int} result for a debug info less function (on 32-bit
19185 architectures). Therefore the user is required to specify the return type by
19186 an appropriate cast explicitly:
19189 Breakpoint 2, 0x0040050b in func ()
19190 (@value{GDBP}) return -1
19191 Return value type not available for selected stack frame.
19192 Please use an explicit cast of the value to return.
19193 (@value{GDBP}) return (long long int) -1
19194 Make selected stack frame return now? (y or n) y
19195 #0 0x00400526 in main ()
19200 @section Calling Program Functions
19203 @cindex calling functions
19204 @cindex inferior functions, calling
19205 @item print @var{expr}
19206 Evaluate the expression @var{expr} and display the resulting value.
19207 The expression may include calls to functions in the program being
19211 @item call @var{expr}
19212 Evaluate the expression @var{expr} without displaying @code{void}
19215 You can use this variant of the @code{print} command if you want to
19216 execute a function from your program that does not return anything
19217 (a.k.a.@: @dfn{a void function}), but without cluttering the output
19218 with @code{void} returned values that @value{GDBN} will otherwise
19219 print. If the result is not void, it is printed and saved in the
19223 It is possible for the function you call via the @code{print} or
19224 @code{call} command to generate a signal (e.g., if there's a bug in
19225 the function, or if you passed it incorrect arguments). What happens
19226 in that case is controlled by the @code{set unwindonsignal} command.
19228 Similarly, with a C@t{++} program it is possible for the function you
19229 call via the @code{print} or @code{call} command to generate an
19230 exception that is not handled due to the constraints of the dummy
19231 frame. In this case, any exception that is raised in the frame, but has
19232 an out-of-frame exception handler will not be found. GDB builds a
19233 dummy-frame for the inferior function call, and the unwinder cannot
19234 seek for exception handlers outside of this dummy-frame. What happens
19235 in that case is controlled by the
19236 @code{set unwind-on-terminating-exception} command.
19239 @item set unwindonsignal
19240 @kindex set unwindonsignal
19241 @cindex unwind stack in called functions
19242 @cindex call dummy stack unwinding
19243 Set unwinding of the stack if a signal is received while in a function
19244 that @value{GDBN} called in the program being debugged. If set to on,
19245 @value{GDBN} unwinds the stack it created for the call and restores
19246 the context to what it was before the call. If set to off (the
19247 default), @value{GDBN} stops in the frame where the signal was
19250 @item show unwindonsignal
19251 @kindex show unwindonsignal
19252 Show the current setting of stack unwinding in the functions called by
19255 @item set unwind-on-terminating-exception
19256 @kindex set unwind-on-terminating-exception
19257 @cindex unwind stack in called functions with unhandled exceptions
19258 @cindex call dummy stack unwinding on unhandled exception.
19259 Set unwinding of the stack if a C@t{++} exception is raised, but left
19260 unhandled while in a function that @value{GDBN} called in the program being
19261 debugged. If set to on (the default), @value{GDBN} unwinds the stack
19262 it created for the call and restores the context to what it was before
19263 the call. If set to off, @value{GDBN} the exception is delivered to
19264 the default C@t{++} exception handler and the inferior terminated.
19266 @item show unwind-on-terminating-exception
19267 @kindex show unwind-on-terminating-exception
19268 Show the current setting of stack unwinding in the functions called by
19271 @item set may-call-functions
19272 @kindex set may-call-functions
19273 @cindex disabling calling functions in the program
19274 @cindex calling functions in the program, disabling
19275 Set permission to call functions in the program.
19276 This controls whether @value{GDBN} will attempt to call functions in
19277 the program, such as with expressions in the @code{print} command. It
19278 defaults to @code{on}.
19280 To call a function in the program, @value{GDBN} has to temporarily
19281 modify the state of the inferior. This has potentially undesired side
19282 effects. Also, having @value{GDBN} call nested functions is likely to
19283 be erroneous and may even crash the program being debugged. You can
19284 avoid such hazards by forbidding @value{GDBN} from calling functions
19285 in the program being debugged. If calling functions in the program
19286 is forbidden, GDB will throw an error when a command (such as printing
19287 an expression) starts a function call in the program.
19289 @item show may-call-functions
19290 @kindex show may-call-functions
19291 Show permission to call functions in the program.
19295 @subsection Calling functions with no debug info
19297 @cindex no debug info functions
19298 Sometimes, a function you wish to call is missing debug information.
19299 In such case, @value{GDBN} does not know the type of the function,
19300 including the types of the function's parameters. To avoid calling
19301 the inferior function incorrectly, which could result in the called
19302 function functioning erroneously and even crash, @value{GDBN} refuses
19303 to call the function unless you tell it the type of the function.
19305 For prototyped (i.e.@: ANSI/ISO style) functions, there are two ways
19306 to do that. The simplest is to cast the call to the function's
19307 declared return type. For example:
19310 (@value{GDBP}) p getenv ("PATH")
19311 'getenv' has unknown return type; cast the call to its declared return type
19312 (@value{GDBP}) p (char *) getenv ("PATH")
19313 $1 = 0x7fffffffe7ba "/usr/local/bin:/"...
19316 Casting the return type of a no-debug function is equivalent to
19317 casting the function to a pointer to a prototyped function that has a
19318 prototype that matches the types of the passed-in arguments, and
19319 calling that. I.e., the call above is equivalent to:
19322 (@value{GDBP}) p ((char * (*) (const char *)) getenv) ("PATH")
19326 and given this prototyped C or C++ function with float parameters:
19329 float multiply (float v1, float v2) @{ return v1 * v2; @}
19333 these calls are equivalent:
19336 (@value{GDBP}) p (float) multiply (2.0f, 3.0f)
19337 (@value{GDBP}) p ((float (*) (float, float)) multiply) (2.0f, 3.0f)
19340 If the function you wish to call is declared as unprototyped (i.e.@:
19341 old K&R style), you must use the cast-to-function-pointer syntax, so
19342 that @value{GDBN} knows that it needs to apply default argument
19343 promotions (promote float arguments to double). @xref{ABI, float
19344 promotion}. For example, given this unprototyped C function with
19345 float parameters, and no debug info:
19349 multiply_noproto (v1, v2)
19357 you call it like this:
19360 (@value{GDBP}) p ((float (*) ()) multiply_noproto) (2.0f, 3.0f)
19364 @section Patching Programs
19366 @cindex patching binaries
19367 @cindex writing into executables
19368 @cindex writing into corefiles
19370 By default, @value{GDBN} opens the file containing your program's
19371 executable code (or the corefile) read-only. This prevents accidental
19372 alterations to machine code; but it also prevents you from intentionally
19373 patching your program's binary.
19375 If you'd like to be able to patch the binary, you can specify that
19376 explicitly with the @code{set write} command. For example, you might
19377 want to turn on internal debugging flags, or even to make emergency
19383 @itemx set write off
19384 If you specify @samp{set write on}, @value{GDBN} opens executable and
19385 core files for both reading and writing; if you specify @kbd{set write
19386 off} (the default), @value{GDBN} opens them read-only.
19388 If you have already loaded a file, you must load it again (using the
19389 @code{exec-file} or @code{core-file} command) after changing @code{set
19390 write}, for your new setting to take effect.
19394 Display whether executable files and core files are opened for writing
19395 as well as reading.
19398 @node Compiling and Injecting Code
19399 @section Compiling and injecting code in @value{GDBN}
19400 @cindex injecting code
19401 @cindex writing into executables
19402 @cindex compiling code
19404 @value{GDBN} supports on-demand compilation and code injection into
19405 programs running under @value{GDBN}. GCC 5.0 or higher built with
19406 @file{libcc1.so} must be installed for this functionality to be enabled.
19407 This functionality is implemented with the following commands.
19410 @kindex compile code
19411 @item compile code @var{source-code}
19412 @itemx compile code -raw @var{--} @var{source-code}
19413 Compile @var{source-code} with the compiler language found as the current
19414 language in @value{GDBN} (@pxref{Languages}). If compilation and
19415 injection is not supported with the current language specified in
19416 @value{GDBN}, or the compiler does not support this feature, an error
19417 message will be printed. If @var{source-code} compiles and links
19418 successfully, @value{GDBN} will load the object-code emitted,
19419 and execute it within the context of the currently selected inferior.
19420 It is important to note that the compiled code is executed immediately.
19421 After execution, the compiled code is removed from @value{GDBN} and any
19422 new types or variables you have defined will be deleted.
19424 The command allows you to specify @var{source-code} in two ways.
19425 The simplest method is to provide a single line of code to the command.
19429 compile code printf ("hello world\n");
19432 If you specify options on the command line as well as source code, they
19433 may conflict. The @samp{--} delimiter can be used to separate options
19434 from actual source code. E.g.:
19437 compile code -r -- printf ("hello world\n");
19440 Alternatively you can enter source code as multiple lines of text. To
19441 enter this mode, invoke the @samp{compile code} command without any text
19442 following the command. This will start the multiple-line editor and
19443 allow you to type as many lines of source code as required. When you
19444 have completed typing, enter @samp{end} on its own line to exit the
19449 >printf ("hello\n");
19450 >printf ("world\n");
19454 Specifying @samp{-raw}, prohibits @value{GDBN} from wrapping the
19455 provided @var{source-code} in a callable scope. In this case, you must
19456 specify the entry point of the code by defining a function named
19457 @code{_gdb_expr_}. The @samp{-raw} code cannot access variables of the
19458 inferior. Using @samp{-raw} option may be needed for example when
19459 @var{source-code} requires @samp{#include} lines which may conflict with
19460 inferior symbols otherwise.
19462 @kindex compile file
19463 @item compile file @var{filename}
19464 @itemx compile file -raw @var{filename}
19465 Like @code{compile code}, but take the source code from @var{filename}.
19468 compile file /home/user/example.c
19473 @item compile print [[@var{options}] --] @var{expr}
19474 @itemx compile print [[@var{options}] --] /@var{f} @var{expr}
19475 Compile and execute @var{expr} with the compiler language found as the
19476 current language in @value{GDBN} (@pxref{Languages}). By default the
19477 value of @var{expr} is printed in a format appropriate to its data type;
19478 you can choose a different format by specifying @samp{/@var{f}}, where
19479 @var{f} is a letter specifying the format; see @ref{Output Formats,,Output
19480 Formats}. The @code{compile print} command accepts the same options
19481 as the @code{print} command; see @ref{print options}.
19483 @item compile print [[@var{options}] --]
19484 @itemx compile print [[@var{options}] --] /@var{f}
19485 @cindex reprint the last value
19486 Alternatively you can enter the expression (source code producing it) as
19487 multiple lines of text. To enter this mode, invoke the @samp{compile print}
19488 command without any text following the command. This will start the
19489 multiple-line editor.
19493 The process of compiling and injecting the code can be inspected using:
19496 @anchor{set debug compile}
19497 @item set debug compile
19498 @cindex compile command debugging info
19499 Turns on or off display of @value{GDBN} process of compiling and
19500 injecting the code. The default is off.
19502 @item show debug compile
19503 Displays the current state of displaying @value{GDBN} process of
19504 compiling and injecting the code.
19506 @anchor{set debug compile-cplus-types}
19507 @item set debug compile-cplus-types
19508 @cindex compile C@t{++} type conversion
19509 Turns on or off the display of C@t{++} type conversion debugging information.
19510 The default is off.
19512 @item show debug compile-cplus-types
19513 Displays the current state of displaying debugging information for
19514 C@t{++} type conversion.
19517 @subsection Compilation options for the @code{compile} command
19519 @value{GDBN} needs to specify the right compilation options for the code
19520 to be injected, in part to make its ABI compatible with the inferior
19521 and in part to make the injected code compatible with @value{GDBN}'s
19525 The options used, in increasing precedence:
19528 @item target architecture and OS options (@code{gdbarch})
19529 These options depend on target processor type and target operating
19530 system, usually they specify at least 32-bit (@code{-m32}) or 64-bit
19531 (@code{-m64}) compilation option.
19533 @item compilation options recorded in the target
19534 @value{NGCC} (since version 4.7) stores the options used for compilation
19535 into @code{DW_AT_producer} part of DWARF debugging information according
19536 to the @value{NGCC} option @code{-grecord-gcc-switches}. One has to
19537 explicitly specify @code{-g} during inferior compilation otherwise
19538 @value{NGCC} produces no DWARF. This feature is only relevant for
19539 platforms where @code{-g} produces DWARF by default, otherwise one may
19540 try to enforce DWARF by using @code{-gdwarf-4}.
19542 @item compilation options set by @code{set compile-args}
19546 You can override compilation options using the following command:
19549 @item set compile-args
19550 @cindex compile command options override
19551 Set compilation options used for compiling and injecting code with the
19552 @code{compile} commands. These options override any conflicting ones
19553 from the target architecture and/or options stored during inferior
19556 @item show compile-args
19557 Displays the current state of compilation options override.
19558 This does not show all the options actually used during compilation,
19559 use @ref{set debug compile} for that.
19562 @subsection Caveats when using the @code{compile} command
19564 There are a few caveats to keep in mind when using the @code{compile}
19565 command. As the caveats are different per language, the table below
19566 highlights specific issues on a per language basis.
19569 @item C code examples and caveats
19570 When the language in @value{GDBN} is set to @samp{C}, the compiler will
19571 attempt to compile the source code with a @samp{C} compiler. The source
19572 code provided to the @code{compile} command will have much the same
19573 access to variables and types as it normally would if it were part of
19574 the program currently being debugged in @value{GDBN}.
19576 Below is a sample program that forms the basis of the examples that
19577 follow. This program has been compiled and loaded into @value{GDBN},
19578 much like any other normal debugging session.
19581 void function1 (void)
19584 printf ("function 1\n");
19587 void function2 (void)
19602 For the purposes of the examples in this section, the program above has
19603 been compiled, loaded into @value{GDBN}, stopped at the function
19604 @code{main}, and @value{GDBN} is awaiting input from the user.
19606 To access variables and types for any program in @value{GDBN}, the
19607 program must be compiled and packaged with debug information. The
19608 @code{compile} command is not an exception to this rule. Without debug
19609 information, you can still use the @code{compile} command, but you will
19610 be very limited in what variables and types you can access.
19612 So with that in mind, the example above has been compiled with debug
19613 information enabled. The @code{compile} command will have access to
19614 all variables and types (except those that may have been optimized
19615 out). Currently, as @value{GDBN} has stopped the program in the
19616 @code{main} function, the @code{compile} command would have access to
19617 the variable @code{k}. You could invoke the @code{compile} command
19618 and type some source code to set the value of @code{k}. You can also
19619 read it, or do anything with that variable you would normally do in
19620 @code{C}. Be aware that changes to inferior variables in the
19621 @code{compile} command are persistent. In the following example:
19624 compile code k = 3;
19628 the variable @code{k} is now 3. It will retain that value until
19629 something else in the example program changes it, or another
19630 @code{compile} command changes it.
19632 Normal scope and access rules apply to source code compiled and
19633 injected by the @code{compile} command. In the example, the variables
19634 @code{j} and @code{k} are not accessible yet, because the program is
19635 currently stopped in the @code{main} function, where these variables
19636 are not in scope. Therefore, the following command
19639 compile code j = 3;
19643 will result in a compilation error message.
19645 Once the program is continued, execution will bring these variables in
19646 scope, and they will become accessible; then the code you specify via
19647 the @code{compile} command will be able to access them.
19649 You can create variables and types with the @code{compile} command as
19650 part of your source code. Variables and types that are created as part
19651 of the @code{compile} command are not visible to the rest of the program for
19652 the duration of its run. This example is valid:
19655 compile code int ff = 5; printf ("ff is %d\n", ff);
19658 However, if you were to type the following into @value{GDBN} after that
19659 command has completed:
19662 compile code printf ("ff is %d\n'', ff);
19666 a compiler error would be raised as the variable @code{ff} no longer
19667 exists. Object code generated and injected by the @code{compile}
19668 command is removed when its execution ends. Caution is advised
19669 when assigning to program variables values of variables created by the
19670 code submitted to the @code{compile} command. This example is valid:
19673 compile code int ff = 5; k = ff;
19676 The value of the variable @code{ff} is assigned to @code{k}. The variable
19677 @code{k} does not require the existence of @code{ff} to maintain the value
19678 it has been assigned. However, pointers require particular care in
19679 assignment. If the source code compiled with the @code{compile} command
19680 changed the address of a pointer in the example program, perhaps to a
19681 variable created in the @code{compile} command, that pointer would point
19682 to an invalid location when the command exits. The following example
19683 would likely cause issues with your debugged program:
19686 compile code int ff = 5; p = &ff;
19689 In this example, @code{p} would point to @code{ff} when the
19690 @code{compile} command is executing the source code provided to it.
19691 However, as variables in the (example) program persist with their
19692 assigned values, the variable @code{p} would point to an invalid
19693 location when the command exists. A general rule should be followed
19694 in that you should either assign @code{NULL} to any assigned pointers,
19695 or restore a valid location to the pointer before the command exits.
19697 Similar caution must be exercised with any structs, unions, and typedefs
19698 defined in @code{compile} command. Types defined in the @code{compile}
19699 command will no longer be available in the next @code{compile} command.
19700 Therefore, if you cast a variable to a type defined in the
19701 @code{compile} command, care must be taken to ensure that any future
19702 need to resolve the type can be achieved.
19705 (gdb) compile code static struct a @{ int a; @} v = @{ 42 @}; argv = &v;
19706 (gdb) compile code printf ("%d\n", ((struct a *) argv)->a);
19707 gdb command line:1:36: error: dereferencing pointer to incomplete type ‘struct a’
19708 Compilation failed.
19709 (gdb) compile code struct a @{ int a; @}; printf ("%d\n", ((struct a *) argv)->a);
19713 Variables that have been optimized away by the compiler are not
19714 accessible to the code submitted to the @code{compile} command.
19715 Access to those variables will generate a compiler error which @value{GDBN}
19716 will print to the console.
19719 @subsection Compiler search for the @code{compile} command
19721 @value{GDBN} needs to find @value{NGCC} for the inferior being debugged
19722 which may not be obvious for remote targets of different architecture
19723 than where @value{GDBN} is running. Environment variable @code{PATH} on
19724 @value{GDBN} host is searched for @value{NGCC} binary matching the
19725 target architecture and operating system. This search can be overriden
19726 by @code{set compile-gcc} @value{GDBN} command below. @code{PATH} is
19727 taken from shell that executed @value{GDBN}, it is not the value set by
19728 @value{GDBN} command @code{set environment}). @xref{Environment}.
19731 Specifically @code{PATH} is searched for binaries matching regular expression
19732 @code{@var{arch}(-[^-]*)?-@var{os}-gcc} according to the inferior target being
19733 debugged. @var{arch} is processor name --- multiarch is supported, so for
19734 example both @code{i386} and @code{x86_64} targets look for pattern
19735 @code{(x86_64|i.86)} and both @code{s390} and @code{s390x} targets look
19736 for pattern @code{s390x?}. @var{os} is currently supported only for
19737 pattern @code{linux(-gnu)?}.
19739 On Posix hosts the compiler driver @value{GDBN} needs to find also
19740 shared library @file{libcc1.so} from the compiler. It is searched in
19741 default shared library search path (overridable with usual environment
19742 variable @code{LD_LIBRARY_PATH}), unrelated to @code{PATH} or @code{set
19743 compile-gcc} settings. Contrary to it @file{libcc1plugin.so} is found
19744 according to the installation of the found compiler --- as possibly
19745 specified by the @code{set compile-gcc} command.
19748 @item set compile-gcc
19749 @cindex compile command driver filename override
19750 Set compilation command used for compiling and injecting code with the
19751 @code{compile} commands. If this option is not set (it is set to
19752 an empty string), the search described above will occur --- that is the
19755 @item show compile-gcc
19756 Displays the current compile command @value{NGCC} driver filename.
19757 If set, it is the main command @command{gcc}, found usually for example
19758 under name @file{x86_64-linux-gnu-gcc}.
19762 @chapter @value{GDBN} Files
19764 @value{GDBN} needs to know the file name of the program to be debugged,
19765 both in order to read its symbol table and in order to start your
19766 program. To debug a core dump of a previous run, you must also tell
19767 @value{GDBN} the name of the core dump file.
19770 * Files:: Commands to specify files
19771 * File Caching:: Information about @value{GDBN}'s file caching
19772 * Separate Debug Files:: Debugging information in separate files
19773 * MiniDebugInfo:: Debugging information in a special section
19774 * Index Files:: Index files speed up GDB
19775 * Symbol Errors:: Errors reading symbol files
19776 * Data Files:: GDB data files
19780 @section Commands to Specify Files
19782 @cindex symbol table
19783 @cindex core dump file
19785 You may want to specify executable and core dump file names. The usual
19786 way to do this is at start-up time, using the arguments to
19787 @value{GDBN}'s start-up commands (@pxref{Invocation, , Getting In and
19788 Out of @value{GDBN}}).
19790 Occasionally it is necessary to change to a different file during a
19791 @value{GDBN} session. Or you may run @value{GDBN} and forget to
19792 specify a file you want to use. Or you are debugging a remote target
19793 via @code{gdbserver} (@pxref{Server, file, Using the @code{gdbserver}
19794 Program}). In these situations the @value{GDBN} commands to specify
19795 new files are useful.
19798 @cindex executable file
19800 @item file @var{filename}
19801 Use @var{filename} as the program to be debugged. It is read for its
19802 symbols and for the contents of pure memory. It is also the program
19803 executed when you use the @code{run} command. If you do not specify a
19804 directory and the file is not found in the @value{GDBN} working directory,
19805 @value{GDBN} uses the environment variable @code{PATH} as a list of
19806 directories to search, just as the shell does when looking for a program
19807 to run. You can change the value of this variable, for both @value{GDBN}
19808 and your program, using the @code{path} command.
19810 @cindex unlinked object files
19811 @cindex patching object files
19812 You can load unlinked object @file{.o} files into @value{GDBN} using
19813 the @code{file} command. You will not be able to ``run'' an object
19814 file, but you can disassemble functions and inspect variables. Also,
19815 if the underlying BFD functionality supports it, you could use
19816 @kbd{gdb -write} to patch object files using this technique. Note
19817 that @value{GDBN} can neither interpret nor modify relocations in this
19818 case, so branches and some initialized variables will appear to go to
19819 the wrong place. But this feature is still handy from time to time.
19822 @code{file} with no argument makes @value{GDBN} discard any information it
19823 has on both executable file and the symbol table.
19826 @item exec-file @r{[} @var{filename} @r{]}
19827 Specify that the program to be run (but not the symbol table) is found
19828 in @var{filename}. @value{GDBN} searches the environment variable @code{PATH}
19829 if necessary to locate your program. Omitting @var{filename} means to
19830 discard information on the executable file.
19832 @kindex symbol-file
19833 @item symbol-file @r{[} @var{filename} @r{[} -o @var{offset} @r{]]}
19834 Read symbol table information from file @var{filename}. @code{PATH} is
19835 searched when necessary. Use the @code{file} command to get both symbol
19836 table and program to run from the same file.
19838 If an optional @var{offset} is specified, it is added to the start
19839 address of each section in the symbol file. This is useful if the
19840 program is relocated at runtime, such as the Linux kernel with kASLR
19843 @code{symbol-file} with no argument clears out @value{GDBN} information on your
19844 program's symbol table.
19846 The @code{symbol-file} command causes @value{GDBN} to forget the contents of
19847 some breakpoints and auto-display expressions. This is because they may
19848 contain pointers to the internal data recording symbols and data types,
19849 which are part of the old symbol table data being discarded inside
19852 @code{symbol-file} does not repeat if you press @key{RET} again after
19855 When @value{GDBN} is configured for a particular environment, it
19856 understands debugging information in whatever format is the standard
19857 generated for that environment; you may use either a @sc{gnu} compiler, or
19858 other compilers that adhere to the local conventions.
19859 Best results are usually obtained from @sc{gnu} compilers; for example,
19860 using @code{@value{NGCC}} you can generate debugging information for
19863 For most kinds of object files, with the exception of old SVR3 systems
19864 using COFF, the @code{symbol-file} command does not normally read the
19865 symbol table in full right away. Instead, it scans the symbol table
19866 quickly to find which source files and which symbols are present. The
19867 details are read later, one source file at a time, as they are needed.
19869 The purpose of this two-stage reading strategy is to make @value{GDBN}
19870 start up faster. For the most part, it is invisible except for
19871 occasional pauses while the symbol table details for a particular source
19872 file are being read. (The @code{set verbose} command can turn these
19873 pauses into messages if desired. @xref{Messages/Warnings, ,Optional
19874 Warnings and Messages}.)
19876 We have not implemented the two-stage strategy for COFF yet. When the
19877 symbol table is stored in COFF format, @code{symbol-file} reads the
19878 symbol table data in full right away. Note that ``stabs-in-COFF''
19879 still does the two-stage strategy, since the debug info is actually
19883 @cindex reading symbols immediately
19884 @cindex symbols, reading immediately
19885 @item symbol-file @r{[} -readnow @r{]} @var{filename}
19886 @itemx file @r{[} -readnow @r{]} @var{filename}
19887 You can override the @value{GDBN} two-stage strategy for reading symbol
19888 tables by using the @samp{-readnow} option with any of the commands that
19889 load symbol table information, if you want to be sure @value{GDBN} has the
19890 entire symbol table available.
19892 @cindex @code{-readnever}, option for symbol-file command
19893 @cindex never read symbols
19894 @cindex symbols, never read
19895 @item symbol-file @r{[} -readnever @r{]} @var{filename}
19896 @itemx file @r{[} -readnever @r{]} @var{filename}
19897 You can instruct @value{GDBN} to never read the symbolic information
19898 contained in @var{filename} by using the @samp{-readnever} option.
19899 @xref{--readnever}.
19901 @c FIXME: for now no mention of directories, since this seems to be in
19902 @c flux. 13mar1992 status is that in theory GDB would look either in
19903 @c current dir or in same dir as myprog; but issues like competing
19904 @c GDB's, or clutter in system dirs, mean that in practice right now
19905 @c only current dir is used. FFish says maybe a special GDB hierarchy
19906 @c (eg rooted in val of env var GDBSYMS) could exist for mappable symbol
19910 @item core-file @r{[}@var{filename}@r{]}
19912 Specify the whereabouts of a core dump file to be used as the ``contents
19913 of memory''. Traditionally, core files contain only some parts of the
19914 address space of the process that generated them; @value{GDBN} can access the
19915 executable file itself for other parts.
19917 @code{core-file} with no argument specifies that no core file is
19920 Note that the core file is ignored when your program is actually running
19921 under @value{GDBN}. So, if you have been running your program and you
19922 wish to debug a core file instead, you must kill the subprocess in which
19923 the program is running. To do this, use the @code{kill} command
19924 (@pxref{Kill Process, ,Killing the Child Process}).
19926 @kindex add-symbol-file
19927 @cindex dynamic linking
19928 @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{]}
19929 The @code{add-symbol-file} command reads additional symbol table
19930 information from the file @var{filename}. You would use this command
19931 when @var{filename} has been dynamically loaded (by some other means)
19932 into the program that is running. The @var{textaddress} parameter gives
19933 the memory address at which the file's text section has been loaded.
19934 You can additionally specify the base address of other sections using
19935 an arbitrary number of @samp{-s @var{section} @var{address}} pairs.
19936 If a section is omitted, @value{GDBN} will use its default addresses
19937 as found in @var{filename}. Any @var{address} or @var{textaddress}
19938 can be given as an expression.
19940 If an optional @var{offset} is specified, it is added to the start
19941 address of each section, except those for which the address was
19942 specified explicitly.
19944 The symbol table of the file @var{filename} is added to the symbol table
19945 originally read with the @code{symbol-file} command. You can use the
19946 @code{add-symbol-file} command any number of times; the new symbol data
19947 thus read is kept in addition to the old.
19949 Changes can be reverted using the command @code{remove-symbol-file}.
19951 @cindex relocatable object files, reading symbols from
19952 @cindex object files, relocatable, reading symbols from
19953 @cindex reading symbols from relocatable object files
19954 @cindex symbols, reading from relocatable object files
19955 @cindex @file{.o} files, reading symbols from
19956 Although @var{filename} is typically a shared library file, an
19957 executable file, or some other object file which has been fully
19958 relocated for loading into a process, you can also load symbolic
19959 information from relocatable @file{.o} files, as long as:
19963 the file's symbolic information refers only to linker symbols defined in
19964 that file, not to symbols defined by other object files,
19966 every section the file's symbolic information refers to has actually
19967 been loaded into the inferior, as it appears in the file, and
19969 you can determine the address at which every section was loaded, and
19970 provide these to the @code{add-symbol-file} command.
19974 Some embedded operating systems, like Sun Chorus and VxWorks, can load
19975 relocatable files into an already running program; such systems
19976 typically make the requirements above easy to meet. However, it's
19977 important to recognize that many native systems use complex link
19978 procedures (@code{.linkonce} section factoring and C@t{++} constructor table
19979 assembly, for example) that make the requirements difficult to meet. In
19980 general, one cannot assume that using @code{add-symbol-file} to read a
19981 relocatable object file's symbolic information will have the same effect
19982 as linking the relocatable object file into the program in the normal
19985 @code{add-symbol-file} does not repeat if you press @key{RET} after using it.
19987 @kindex remove-symbol-file
19988 @item remove-symbol-file @var{filename}
19989 @item remove-symbol-file -a @var{address}
19990 Remove a symbol file added via the @code{add-symbol-file} command. The
19991 file to remove can be identified by its @var{filename} or by an @var{address}
19992 that lies within the boundaries of this symbol file in memory. Example:
19995 (gdb) add-symbol-file /home/user/gdb/mylib.so 0x7ffff7ff9480
19996 add symbol table from file "/home/user/gdb/mylib.so" at
19997 .text_addr = 0x7ffff7ff9480
19999 Reading symbols from /home/user/gdb/mylib.so...done.
20000 (gdb) remove-symbol-file -a 0x7ffff7ff9480
20001 Remove symbol table from file "/home/user/gdb/mylib.so"? (y or n) y
20006 @code{remove-symbol-file} does not repeat if you press @key{RET} after using it.
20008 @kindex add-symbol-file-from-memory
20009 @cindex @code{syscall DSO}
20010 @cindex load symbols from memory
20011 @item add-symbol-file-from-memory @var{address}
20012 Load symbols from the given @var{address} in a dynamically loaded
20013 object file whose image is mapped directly into the inferior's memory.
20014 For example, the Linux kernel maps a @code{syscall DSO} into each
20015 process's address space; this DSO provides kernel-specific code for
20016 some system calls. The argument can be any expression whose
20017 evaluation yields the address of the file's shared object file header.
20018 For this command to work, you must have used @code{symbol-file} or
20019 @code{exec-file} commands in advance.
20022 @item section @var{section} @var{addr}
20023 The @code{section} command changes the base address of the named
20024 @var{section} of the exec file to @var{addr}. This can be used if the
20025 exec file does not contain section addresses, (such as in the
20026 @code{a.out} format), or when the addresses specified in the file
20027 itself are wrong. Each section must be changed separately. The
20028 @code{info files} command, described below, lists all the sections and
20032 @kindex info target
20035 @code{info files} and @code{info target} are synonymous; both print the
20036 current target (@pxref{Targets, ,Specifying a Debugging Target}),
20037 including the names of the executable and core dump files currently in
20038 use by @value{GDBN}, and the files from which symbols were loaded. The
20039 command @code{help target} lists all possible targets rather than
20042 @kindex maint info sections
20043 @item maint info sections
20044 Another command that can give you extra information about program sections
20045 is @code{maint info sections}. In addition to the section information
20046 displayed by @code{info files}, this command displays the flags and file
20047 offset of each section in the executable and core dump files. In addition,
20048 @code{maint info sections} provides the following command options (which
20049 may be arbitrarily combined):
20053 Display sections for all loaded object files, including shared libraries.
20054 @item @var{sections}
20055 Display info only for named @var{sections}.
20056 @item @var{section-flags}
20057 Display info only for sections for which @var{section-flags} are true.
20058 The section flags that @value{GDBN} currently knows about are:
20061 Section will have space allocated in the process when loaded.
20062 Set for all sections except those containing debug information.
20064 Section will be loaded from the file into the child process memory.
20065 Set for pre-initialized code and data, clear for @code{.bss} sections.
20067 Section needs to be relocated before loading.
20069 Section cannot be modified by the child process.
20071 Section contains executable code only.
20073 Section contains data only (no executable code).
20075 Section will reside in ROM.
20077 Section contains data for constructor/destructor lists.
20079 Section is not empty.
20081 An instruction to the linker to not output the section.
20082 @item COFF_SHARED_LIBRARY
20083 A notification to the linker that the section contains
20084 COFF shared library information.
20086 Section contains common symbols.
20089 @kindex set trust-readonly-sections
20090 @cindex read-only sections
20091 @item set trust-readonly-sections on
20092 Tell @value{GDBN} that readonly sections in your object file
20093 really are read-only (i.e.@: that their contents will not change).
20094 In that case, @value{GDBN} can fetch values from these sections
20095 out of the object file, rather than from the target program.
20096 For some targets (notably embedded ones), this can be a significant
20097 enhancement to debugging performance.
20099 The default is off.
20101 @item set trust-readonly-sections off
20102 Tell @value{GDBN} not to trust readonly sections. This means that
20103 the contents of the section might change while the program is running,
20104 and must therefore be fetched from the target when needed.
20106 @item show trust-readonly-sections
20107 Show the current setting of trusting readonly sections.
20110 All file-specifying commands allow both absolute and relative file names
20111 as arguments. @value{GDBN} always converts the file name to an absolute file
20112 name and remembers it that way.
20114 @cindex shared libraries
20115 @anchor{Shared Libraries}
20116 @value{GDBN} supports @sc{gnu}/Linux, MS-Windows, SunOS,
20117 Darwin/Mach-O, SVr4, IBM RS/6000 AIX, QNX Neutrino, FDPIC (FR-V), and
20118 DSBT (TIC6X) shared libraries.
20120 On MS-Windows @value{GDBN} must be linked with the Expat library to support
20121 shared libraries. @xref{Expat}.
20123 @value{GDBN} automatically loads symbol definitions from shared libraries
20124 when you use the @code{run} command, or when you examine a core file.
20125 (Before you issue the @code{run} command, @value{GDBN} does not understand
20126 references to a function in a shared library, however---unless you are
20127 debugging a core file).
20129 @c FIXME: some @value{GDBN} release may permit some refs to undef
20130 @c FIXME...symbols---eg in a break cmd---assuming they are from a shared
20131 @c FIXME...lib; check this from time to time when updating manual
20133 There are times, however, when you may wish to not automatically load
20134 symbol definitions from shared libraries, such as when they are
20135 particularly large or there are many of them.
20137 To control the automatic loading of shared library symbols, use the
20141 @kindex set auto-solib-add
20142 @item set auto-solib-add @var{mode}
20143 If @var{mode} is @code{on}, symbols from all shared object libraries
20144 will be loaded automatically when the inferior begins execution, you
20145 attach to an independently started inferior, or when the dynamic linker
20146 informs @value{GDBN} that a new library has been loaded. If @var{mode}
20147 is @code{off}, symbols must be loaded manually, using the
20148 @code{sharedlibrary} command. The default value is @code{on}.
20150 @cindex memory used for symbol tables
20151 If your program uses lots of shared libraries with debug info that
20152 takes large amounts of memory, you can decrease the @value{GDBN}
20153 memory footprint by preventing it from automatically loading the
20154 symbols from shared libraries. To that end, type @kbd{set
20155 auto-solib-add off} before running the inferior, then load each
20156 library whose debug symbols you do need with @kbd{sharedlibrary
20157 @var{regexp}}, where @var{regexp} is a regular expression that matches
20158 the libraries whose symbols you want to be loaded.
20160 @kindex show auto-solib-add
20161 @item show auto-solib-add
20162 Display the current autoloading mode.
20165 @cindex load shared library
20166 To explicitly load shared library symbols, use the @code{sharedlibrary}
20170 @kindex info sharedlibrary
20172 @item info share @var{regex}
20173 @itemx info sharedlibrary @var{regex}
20174 Print the names of the shared libraries which are currently loaded
20175 that match @var{regex}. If @var{regex} is omitted then print
20176 all shared libraries that are loaded.
20179 @item info dll @var{regex}
20180 This is an alias of @code{info sharedlibrary}.
20182 @kindex sharedlibrary
20184 @item sharedlibrary @var{regex}
20185 @itemx share @var{regex}
20186 Load shared object library symbols for files matching a
20187 Unix regular expression.
20188 As with files loaded automatically, it only loads shared libraries
20189 required by your program for a core file or after typing @code{run}. If
20190 @var{regex} is omitted all shared libraries required by your program are
20193 @item nosharedlibrary
20194 @kindex nosharedlibrary
20195 @cindex unload symbols from shared libraries
20196 Unload all shared object library symbols. This discards all symbols
20197 that have been loaded from all shared libraries. Symbols from shared
20198 libraries that were loaded by explicit user requests are not
20202 Sometimes you may wish that @value{GDBN} stops and gives you control
20203 when any of shared library events happen. The best way to do this is
20204 to use @code{catch load} and @code{catch unload} (@pxref{Set
20207 @value{GDBN} also supports the the @code{set stop-on-solib-events}
20208 command for this. This command exists for historical reasons. It is
20209 less useful than setting a catchpoint, because it does not allow for
20210 conditions or commands as a catchpoint does.
20213 @item set stop-on-solib-events
20214 @kindex set stop-on-solib-events
20215 This command controls whether @value{GDBN} should give you control
20216 when the dynamic linker notifies it about some shared library event.
20217 The most common event of interest is loading or unloading of a new
20220 @item show stop-on-solib-events
20221 @kindex show stop-on-solib-events
20222 Show whether @value{GDBN} stops and gives you control when shared
20223 library events happen.
20226 Shared libraries are also supported in many cross or remote debugging
20227 configurations. @value{GDBN} needs to have access to the target's libraries;
20228 this can be accomplished either by providing copies of the libraries
20229 on the host system, or by asking @value{GDBN} to automatically retrieve the
20230 libraries from the target. If copies of the target libraries are
20231 provided, they need to be the same as the target libraries, although the
20232 copies on the target can be stripped as long as the copies on the host are
20235 @cindex where to look for shared libraries
20236 For remote debugging, you need to tell @value{GDBN} where the target
20237 libraries are, so that it can load the correct copies---otherwise, it
20238 may try to load the host's libraries. @value{GDBN} has two variables
20239 to specify the search directories for target libraries.
20242 @cindex prefix for executable and shared library file names
20243 @cindex system root, alternate
20244 @kindex set solib-absolute-prefix
20245 @kindex set sysroot
20246 @item set sysroot @var{path}
20247 Use @var{path} as the system root for the program being debugged. Any
20248 absolute shared library paths will be prefixed with @var{path}; many
20249 runtime loaders store the absolute paths to the shared library in the
20250 target program's memory. When starting processes remotely, and when
20251 attaching to already-running processes (local or remote), their
20252 executable filenames will be prefixed with @var{path} if reported to
20253 @value{GDBN} as absolute by the operating system. If you use
20254 @code{set sysroot} to find executables and shared libraries, they need
20255 to be laid out in the same way that they are on the target, with
20256 e.g.@: a @file{/bin}, @file{/lib} and @file{/usr/lib} hierarchy under
20259 If @var{path} starts with the sequence @file{target:} and the target
20260 system is remote then @value{GDBN} will retrieve the target binaries
20261 from the remote system. This is only supported when using a remote
20262 target that supports the @code{remote get} command (@pxref{File
20263 Transfer,,Sending files to a remote system}). The part of @var{path}
20264 following the initial @file{target:} (if present) is used as system
20265 root prefix on the remote file system. If @var{path} starts with the
20266 sequence @file{remote:} this is converted to the sequence
20267 @file{target:} by @code{set sysroot}@footnote{Historically the
20268 functionality to retrieve binaries from the remote system was
20269 provided by prefixing @var{path} with @file{remote:}}. If you want
20270 to specify a local system root using a directory that happens to be
20271 named @file{target:} or @file{remote:}, you need to use some
20272 equivalent variant of the name like @file{./target:}.
20274 For targets with an MS-DOS based filesystem, such as MS-Windows and
20275 SymbianOS, @value{GDBN} tries prefixing a few variants of the target
20276 absolute file name with @var{path}. But first, on Unix hosts,
20277 @value{GDBN} converts all backslash directory separators into forward
20278 slashes, because the backslash is not a directory separator on Unix:
20281 c:\foo\bar.dll @result{} c:/foo/bar.dll
20284 Then, @value{GDBN} attempts prefixing the target file name with
20285 @var{path}, and looks for the resulting file name in the host file
20289 c:/foo/bar.dll @result{} /path/to/sysroot/c:/foo/bar.dll
20292 If that does not find the binary, @value{GDBN} tries removing
20293 the @samp{:} character from the drive spec, both for convenience, and,
20294 for the case of the host file system not supporting file names with
20298 c:/foo/bar.dll @result{} /path/to/sysroot/c/foo/bar.dll
20301 This makes it possible to have a system root that mirrors a target
20302 with more than one drive. E.g., you may want to setup your local
20303 copies of the target system shared libraries like so (note @samp{c} vs
20307 @file{/path/to/sysroot/c/sys/bin/foo.dll}
20308 @file{/path/to/sysroot/c/sys/bin/bar.dll}
20309 @file{/path/to/sysroot/z/sys/bin/bar.dll}
20313 and point the system root at @file{/path/to/sysroot}, so that
20314 @value{GDBN} can find the correct copies of both
20315 @file{c:\sys\bin\foo.dll}, and @file{z:\sys\bin\bar.dll}.
20317 If that still does not find the binary, @value{GDBN} tries
20318 removing the whole drive spec from the target file name:
20321 c:/foo/bar.dll @result{} /path/to/sysroot/foo/bar.dll
20324 This last lookup makes it possible to not care about the drive name,
20325 if you don't want or need to.
20327 The @code{set solib-absolute-prefix} command is an alias for @code{set
20330 @cindex default system root
20331 @cindex @samp{--with-sysroot}
20332 You can set the default system root by using the configure-time
20333 @samp{--with-sysroot} option. If the system root is inside
20334 @value{GDBN}'s configured binary prefix (set with @samp{--prefix} or
20335 @samp{--exec-prefix}), then the default system root will be updated
20336 automatically if the installed @value{GDBN} is moved to a new
20339 @kindex show sysroot
20341 Display the current executable and shared library prefix.
20343 @kindex set solib-search-path
20344 @item set solib-search-path @var{path}
20345 If this variable is set, @var{path} is a colon-separated list of
20346 directories to search for shared libraries. @samp{solib-search-path}
20347 is used after @samp{sysroot} fails to locate the library, or if the
20348 path to the library is relative instead of absolute. If you want to
20349 use @samp{solib-search-path} instead of @samp{sysroot}, be sure to set
20350 @samp{sysroot} to a nonexistent directory to prevent @value{GDBN} from
20351 finding your host's libraries. @samp{sysroot} is preferred; setting
20352 it to a nonexistent directory may interfere with automatic loading
20353 of shared library symbols.
20355 @kindex show solib-search-path
20356 @item show solib-search-path
20357 Display the current shared library search path.
20359 @cindex DOS file-name semantics of file names.
20360 @kindex set target-file-system-kind (unix|dos-based|auto)
20361 @kindex show target-file-system-kind
20362 @item set target-file-system-kind @var{kind}
20363 Set assumed file system kind for target reported file names.
20365 Shared library file names as reported by the target system may not
20366 make sense as is on the system @value{GDBN} is running on. For
20367 example, when remote debugging a target that has MS-DOS based file
20368 system semantics, from a Unix host, the target may be reporting to
20369 @value{GDBN} a list of loaded shared libraries with file names such as
20370 @file{c:\Windows\kernel32.dll}. On Unix hosts, there's no concept of
20371 drive letters, so the @samp{c:\} prefix is not normally understood as
20372 indicating an absolute file name, and neither is the backslash
20373 normally considered a directory separator character. In that case,
20374 the native file system would interpret this whole absolute file name
20375 as a relative file name with no directory components. This would make
20376 it impossible to point @value{GDBN} at a copy of the remote target's
20377 shared libraries on the host using @code{set sysroot}, and impractical
20378 with @code{set solib-search-path}. Setting
20379 @code{target-file-system-kind} to @code{dos-based} tells @value{GDBN}
20380 to interpret such file names similarly to how the target would, and to
20381 map them to file names valid on @value{GDBN}'s native file system
20382 semantics. The value of @var{kind} can be @code{"auto"}, in addition
20383 to one of the supported file system kinds. In that case, @value{GDBN}
20384 tries to determine the appropriate file system variant based on the
20385 current target's operating system (@pxref{ABI, ,Configuring the
20386 Current ABI}). The supported file system settings are:
20390 Instruct @value{GDBN} to assume the target file system is of Unix
20391 kind. Only file names starting the forward slash (@samp{/}) character
20392 are considered absolute, and the directory separator character is also
20396 Instruct @value{GDBN} to assume the target file system is DOS based.
20397 File names starting with either a forward slash, or a drive letter
20398 followed by a colon (e.g., @samp{c:}), are considered absolute, and
20399 both the slash (@samp{/}) and the backslash (@samp{\\}) characters are
20400 considered directory separators.
20403 Instruct @value{GDBN} to use the file system kind associated with the
20404 target operating system (@pxref{ABI, ,Configuring the Current ABI}).
20405 This is the default.
20409 @cindex file name canonicalization
20410 @cindex base name differences
20411 When processing file names provided by the user, @value{GDBN}
20412 frequently needs to compare them to the file names recorded in the
20413 program's debug info. Normally, @value{GDBN} compares just the
20414 @dfn{base names} of the files as strings, which is reasonably fast
20415 even for very large programs. (The base name of a file is the last
20416 portion of its name, after stripping all the leading directories.)
20417 This shortcut in comparison is based upon the assumption that files
20418 cannot have more than one base name. This is usually true, but
20419 references to files that use symlinks or similar filesystem
20420 facilities violate that assumption. If your program records files
20421 using such facilities, or if you provide file names to @value{GDBN}
20422 using symlinks etc., you can set @code{basenames-may-differ} to
20423 @code{true} to instruct @value{GDBN} to completely canonicalize each
20424 pair of file names it needs to compare. This will make file-name
20425 comparisons accurate, but at a price of a significant slowdown.
20428 @item set basenames-may-differ
20429 @kindex set basenames-may-differ
20430 Set whether a source file may have multiple base names.
20432 @item show basenames-may-differ
20433 @kindex show basenames-may-differ
20434 Show whether a source file may have multiple base names.
20438 @section File Caching
20439 @cindex caching of opened files
20440 @cindex caching of bfd objects
20442 To speed up file loading, and reduce memory usage, @value{GDBN} will
20443 reuse the @code{bfd} objects used to track open files. @xref{Top, ,
20444 BFD, bfd, The Binary File Descriptor Library}. The following commands
20445 allow visibility and control of the caching behavior.
20448 @kindex maint info bfds
20449 @item maint info bfds
20450 This prints information about each @code{bfd} object that is known to
20453 @kindex maint set bfd-sharing
20454 @kindex maint show bfd-sharing
20455 @kindex bfd caching
20456 @item maint set bfd-sharing
20457 @item maint show bfd-sharing
20458 Control whether @code{bfd} objects can be shared. When sharing is
20459 enabled @value{GDBN} reuses already open @code{bfd} objects rather
20460 than reopening the same file. Turning sharing off does not cause
20461 already shared @code{bfd} objects to be unshared, but all future files
20462 that are opened will create a new @code{bfd} object. Similarly,
20463 re-enabling sharing does not cause multiple existing @code{bfd}
20464 objects to be collapsed into a single shared @code{bfd} object.
20466 @kindex set debug bfd-cache @var{level}
20467 @kindex bfd caching
20468 @item set debug bfd-cache @var{level}
20469 Turns on debugging of the bfd cache, setting the level to @var{level}.
20471 @kindex show debug bfd-cache
20472 @kindex bfd caching
20473 @item show debug bfd-cache
20474 Show the current debugging level of the bfd cache.
20477 @node Separate Debug Files
20478 @section Debugging Information in Separate Files
20479 @cindex separate debugging information files
20480 @cindex debugging information in separate files
20481 @cindex @file{.debug} subdirectories
20482 @cindex debugging information directory, global
20483 @cindex global debugging information directories
20484 @cindex build ID, and separate debugging files
20485 @cindex @file{.build-id} directory
20487 @value{GDBN} allows you to put a program's debugging information in a
20488 file separate from the executable itself, in a way that allows
20489 @value{GDBN} to find and load the debugging information automatically.
20490 Since debugging information can be very large---sometimes larger
20491 than the executable code itself---some systems distribute debugging
20492 information for their executables in separate files, which users can
20493 install only when they need to debug a problem.
20495 @value{GDBN} supports two ways of specifying the separate debug info
20500 The executable contains a @dfn{debug link} that specifies the name of
20501 the separate debug info file. The separate debug file's name is
20502 usually @file{@var{executable}.debug}, where @var{executable} is the
20503 name of the corresponding executable file without leading directories
20504 (e.g., @file{ls.debug} for @file{/usr/bin/ls}). In addition, the
20505 debug link specifies a 32-bit @dfn{Cyclic Redundancy Check} (CRC)
20506 checksum for the debug file, which @value{GDBN} uses to validate that
20507 the executable and the debug file came from the same build.
20510 The executable contains a @dfn{build ID}, a unique bit string that is
20511 also present in the corresponding debug info file. (This is supported
20512 only on some operating systems, when using the ELF or PE file formats
20513 for binary files and the @sc{gnu} Binutils.) For more details about
20514 this feature, see the description of the @option{--build-id}
20515 command-line option in @ref{Options, , Command Line Options, ld,
20516 The GNU Linker}. The debug info file's name is not specified
20517 explicitly by the build ID, but can be computed from the build ID, see
20521 Depending on the way the debug info file is specified, @value{GDBN}
20522 uses two different methods of looking for the debug file:
20526 For the ``debug link'' method, @value{GDBN} looks up the named file in
20527 the directory of the executable file, then in a subdirectory of that
20528 directory named @file{.debug}, and finally under each one of the
20529 global debug directories, in a subdirectory whose name is identical to
20530 the leading directories of the executable's absolute file name. (On
20531 MS-Windows/MS-DOS, the drive letter of the executable's leading
20532 directories is converted to a one-letter subdirectory, i.e.@:
20533 @file{d:/usr/bin/} is converted to @file{/d/usr/bin/}, because Windows
20534 filesystems disallow colons in file names.)
20537 For the ``build ID'' method, @value{GDBN} looks in the
20538 @file{.build-id} subdirectory of each one of the global debug directories for
20539 a file named @file{@var{nn}/@var{nnnnnnnn}.debug}, where @var{nn} are the
20540 first 2 hex characters of the build ID bit string, and @var{nnnnnnnn}
20541 are the rest of the bit string. (Real build ID strings are 32 or more
20542 hex characters, not 10.)
20545 So, for example, suppose you ask @value{GDBN} to debug
20546 @file{/usr/bin/ls}, which has a debug link that specifies the
20547 file @file{ls.debug}, and a build ID whose value in hex is
20548 @code{abcdef1234}. If the list of the global debug directories includes
20549 @file{/usr/lib/debug}, then @value{GDBN} will look for the following
20550 debug information files, in the indicated order:
20554 @file{/usr/lib/debug/.build-id/ab/cdef1234.debug}
20556 @file{/usr/bin/ls.debug}
20558 @file{/usr/bin/.debug/ls.debug}
20560 @file{/usr/lib/debug/usr/bin/ls.debug}.
20563 @anchor{debug-file-directory}
20564 Global debugging info directories default to what is set by @value{GDBN}
20565 configure option @option{--with-separate-debug-dir}. During @value{GDBN} run
20566 you can also set the global debugging info directories, and view the list
20567 @value{GDBN} is currently using.
20571 @kindex set debug-file-directory
20572 @item set debug-file-directory @var{directories}
20573 Set the directories which @value{GDBN} searches for separate debugging
20574 information files to @var{directory}. Multiple path components can be set
20575 concatenating them by a path separator.
20577 @kindex show debug-file-directory
20578 @item show debug-file-directory
20579 Show the directories @value{GDBN} searches for separate debugging
20584 @cindex @code{.gnu_debuglink} sections
20585 @cindex debug link sections
20586 A debug link is a special section of the executable file named
20587 @code{.gnu_debuglink}. The section must contain:
20591 A filename, with any leading directory components removed, followed by
20594 zero to three bytes of padding, as needed to reach the next four-byte
20595 boundary within the section, and
20597 a four-byte CRC checksum, stored in the same endianness used for the
20598 executable file itself. The checksum is computed on the debugging
20599 information file's full contents by the function given below, passing
20600 zero as the @var{crc} argument.
20603 Any executable file format can carry a debug link, as long as it can
20604 contain a section named @code{.gnu_debuglink} with the contents
20607 @cindex @code{.note.gnu.build-id} sections
20608 @cindex build ID sections
20609 The build ID is a special section in the executable file (and in other
20610 ELF binary files that @value{GDBN} may consider). This section is
20611 often named @code{.note.gnu.build-id}, but that name is not mandatory.
20612 It contains unique identification for the built files---the ID remains
20613 the same across multiple builds of the same build tree. The default
20614 algorithm SHA1 produces 160 bits (40 hexadecimal characters) of the
20615 content for the build ID string. The same section with an identical
20616 value is present in the original built binary with symbols, in its
20617 stripped variant, and in the separate debugging information file.
20619 The debugging information file itself should be an ordinary
20620 executable, containing a full set of linker symbols, sections, and
20621 debugging information. The sections of the debugging information file
20622 should have the same names, addresses, and sizes as the original file,
20623 but they need not contain any data---much like a @code{.bss} section
20624 in an ordinary executable.
20626 The @sc{gnu} binary utilities (Binutils) package includes the
20627 @samp{objcopy} utility that can produce
20628 the separated executable / debugging information file pairs using the
20629 following commands:
20632 @kbd{objcopy --only-keep-debug foo foo.debug}
20637 These commands remove the debugging
20638 information from the executable file @file{foo} and place it in the file
20639 @file{foo.debug}. You can use the first, second or both methods to link the
20644 The debug link method needs the following additional command to also leave
20645 behind a debug link in @file{foo}:
20648 @kbd{objcopy --add-gnu-debuglink=foo.debug foo}
20651 Ulrich Drepper's @file{elfutils} package, starting with version 0.53, contains
20652 a version of the @code{strip} command such that the command @kbd{strip foo -f
20653 foo.debug} has the same functionality as the two @code{objcopy} commands and
20654 the @code{ln -s} command above, together.
20657 Build ID gets embedded into the main executable using @code{ld --build-id} or
20658 the @value{NGCC} counterpart @code{gcc -Wl,--build-id}. Build ID support plus
20659 compatibility fixes for debug files separation are present in @sc{gnu} binary
20660 utilities (Binutils) package since version 2.18.
20665 @cindex CRC algorithm definition
20666 The CRC used in @code{.gnu_debuglink} is the CRC-32 defined in
20667 IEEE 802.3 using the polynomial:
20669 @c TexInfo requires naked braces for multi-digit exponents for Tex
20670 @c output, but this causes HTML output to barf. HTML has to be set using
20671 @c raw commands. So we end up having to specify this equation in 2
20676 <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>
20677 + <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
20683 @math{x^{32} + x^{26} + x^{23} + x^{22} + x^{16} + x^{12} + x^{11}}
20684 @math{+ x^{10} + x^8 + x^7 + x^5 + x^4 + x^2 + x + 1}
20688 The function is computed byte at a time, taking the least
20689 significant bit of each byte first. The initial pattern
20690 @code{0xffffffff} is used, to ensure leading zeros affect the CRC and
20691 the final result is inverted to ensure trailing zeros also affect the
20694 @emph{Note:} This is the same CRC polynomial as used in handling the
20695 @dfn{Remote Serial Protocol} @code{qCRC} packet (@pxref{qCRC packet}).
20696 However in the case of the Remote Serial Protocol, the CRC is computed
20697 @emph{most} significant bit first, and the result is not inverted, so
20698 trailing zeros have no effect on the CRC value.
20700 To complete the description, we show below the code of the function
20701 which produces the CRC used in @code{.gnu_debuglink}. Inverting the
20702 initially supplied @code{crc} argument means that an initial call to
20703 this function passing in zero will start computing the CRC using
20706 @kindex gnu_debuglink_crc32
20709 gnu_debuglink_crc32 (unsigned long crc,
20710 unsigned char *buf, size_t len)
20712 static const unsigned long crc32_table[256] =
20714 0x00000000, 0x77073096, 0xee0e612c, 0x990951ba, 0x076dc419,
20715 0x706af48f, 0xe963a535, 0x9e6495a3, 0x0edb8832, 0x79dcb8a4,
20716 0xe0d5e91e, 0x97d2d988, 0x09b64c2b, 0x7eb17cbd, 0xe7b82d07,
20717 0x90bf1d91, 0x1db71064, 0x6ab020f2, 0xf3b97148, 0x84be41de,
20718 0x1adad47d, 0x6ddde4eb, 0xf4d4b551, 0x83d385c7, 0x136c9856,
20719 0x646ba8c0, 0xfd62f97a, 0x8a65c9ec, 0x14015c4f, 0x63066cd9,
20720 0xfa0f3d63, 0x8d080df5, 0x3b6e20c8, 0x4c69105e, 0xd56041e4,
20721 0xa2677172, 0x3c03e4d1, 0x4b04d447, 0xd20d85fd, 0xa50ab56b,
20722 0x35b5a8fa, 0x42b2986c, 0xdbbbc9d6, 0xacbcf940, 0x32d86ce3,
20723 0x45df5c75, 0xdcd60dcf, 0xabd13d59, 0x26d930ac, 0x51de003a,
20724 0xc8d75180, 0xbfd06116, 0x21b4f4b5, 0x56b3c423, 0xcfba9599,
20725 0xb8bda50f, 0x2802b89e, 0x5f058808, 0xc60cd9b2, 0xb10be924,
20726 0x2f6f7c87, 0x58684c11, 0xc1611dab, 0xb6662d3d, 0x76dc4190,
20727 0x01db7106, 0x98d220bc, 0xefd5102a, 0x71b18589, 0x06b6b51f,
20728 0x9fbfe4a5, 0xe8b8d433, 0x7807c9a2, 0x0f00f934, 0x9609a88e,
20729 0xe10e9818, 0x7f6a0dbb, 0x086d3d2d, 0x91646c97, 0xe6635c01,
20730 0x6b6b51f4, 0x1c6c6162, 0x856530d8, 0xf262004e, 0x6c0695ed,
20731 0x1b01a57b, 0x8208f4c1, 0xf50fc457, 0x65b0d9c6, 0x12b7e950,
20732 0x8bbeb8ea, 0xfcb9887c, 0x62dd1ddf, 0x15da2d49, 0x8cd37cf3,
20733 0xfbd44c65, 0x4db26158, 0x3ab551ce, 0xa3bc0074, 0xd4bb30e2,
20734 0x4adfa541, 0x3dd895d7, 0xa4d1c46d, 0xd3d6f4fb, 0x4369e96a,
20735 0x346ed9fc, 0xad678846, 0xda60b8d0, 0x44042d73, 0x33031de5,
20736 0xaa0a4c5f, 0xdd0d7cc9, 0x5005713c, 0x270241aa, 0xbe0b1010,
20737 0xc90c2086, 0x5768b525, 0x206f85b3, 0xb966d409, 0xce61e49f,
20738 0x5edef90e, 0x29d9c998, 0xb0d09822, 0xc7d7a8b4, 0x59b33d17,
20739 0x2eb40d81, 0xb7bd5c3b, 0xc0ba6cad, 0xedb88320, 0x9abfb3b6,
20740 0x03b6e20c, 0x74b1d29a, 0xead54739, 0x9dd277af, 0x04db2615,
20741 0x73dc1683, 0xe3630b12, 0x94643b84, 0x0d6d6a3e, 0x7a6a5aa8,
20742 0xe40ecf0b, 0x9309ff9d, 0x0a00ae27, 0x7d079eb1, 0xf00f9344,
20743 0x8708a3d2, 0x1e01f268, 0x6906c2fe, 0xf762575d, 0x806567cb,
20744 0x196c3671, 0x6e6b06e7, 0xfed41b76, 0x89d32be0, 0x10da7a5a,
20745 0x67dd4acc, 0xf9b9df6f, 0x8ebeeff9, 0x17b7be43, 0x60b08ed5,
20746 0xd6d6a3e8, 0xa1d1937e, 0x38d8c2c4, 0x4fdff252, 0xd1bb67f1,
20747 0xa6bc5767, 0x3fb506dd, 0x48b2364b, 0xd80d2bda, 0xaf0a1b4c,
20748 0x36034af6, 0x41047a60, 0xdf60efc3, 0xa867df55, 0x316e8eef,
20749 0x4669be79, 0xcb61b38c, 0xbc66831a, 0x256fd2a0, 0x5268e236,
20750 0xcc0c7795, 0xbb0b4703, 0x220216b9, 0x5505262f, 0xc5ba3bbe,
20751 0xb2bd0b28, 0x2bb45a92, 0x5cb36a04, 0xc2d7ffa7, 0xb5d0cf31,
20752 0x2cd99e8b, 0x5bdeae1d, 0x9b64c2b0, 0xec63f226, 0x756aa39c,
20753 0x026d930a, 0x9c0906a9, 0xeb0e363f, 0x72076785, 0x05005713,
20754 0x95bf4a82, 0xe2b87a14, 0x7bb12bae, 0x0cb61b38, 0x92d28e9b,
20755 0xe5d5be0d, 0x7cdcefb7, 0x0bdbdf21, 0x86d3d2d4, 0xf1d4e242,
20756 0x68ddb3f8, 0x1fda836e, 0x81be16cd, 0xf6b9265b, 0x6fb077e1,
20757 0x18b74777, 0x88085ae6, 0xff0f6a70, 0x66063bca, 0x11010b5c,
20758 0x8f659eff, 0xf862ae69, 0x616bffd3, 0x166ccf45, 0xa00ae278,
20759 0xd70dd2ee, 0x4e048354, 0x3903b3c2, 0xa7672661, 0xd06016f7,
20760 0x4969474d, 0x3e6e77db, 0xaed16a4a, 0xd9d65adc, 0x40df0b66,
20761 0x37d83bf0, 0xa9bcae53, 0xdebb9ec5, 0x47b2cf7f, 0x30b5ffe9,
20762 0xbdbdf21c, 0xcabac28a, 0x53b39330, 0x24b4a3a6, 0xbad03605,
20763 0xcdd70693, 0x54de5729, 0x23d967bf, 0xb3667a2e, 0xc4614ab8,
20764 0x5d681b02, 0x2a6f2b94, 0xb40bbe37, 0xc30c8ea1, 0x5a05df1b,
20767 unsigned char *end;
20769 crc = ~crc & 0xffffffff;
20770 for (end = buf + len; buf < end; ++buf)
20771 crc = crc32_table[(crc ^ *buf) & 0xff] ^ (crc >> 8);
20772 return ~crc & 0xffffffff;
20777 This computation does not apply to the ``build ID'' method.
20779 @node MiniDebugInfo
20780 @section Debugging information in a special section
20781 @cindex separate debug sections
20782 @cindex @samp{.gnu_debugdata} section
20784 Some systems ship pre-built executables and libraries that have a
20785 special @samp{.gnu_debugdata} section. This feature is called
20786 @dfn{MiniDebugInfo}. This section holds an LZMA-compressed object and
20787 is used to supply extra symbols for backtraces.
20789 The intent of this section is to provide extra minimal debugging
20790 information for use in simple backtraces. It is not intended to be a
20791 replacement for full separate debugging information (@pxref{Separate
20792 Debug Files}). The example below shows the intended use; however,
20793 @value{GDBN} does not currently put restrictions on what sort of
20794 debugging information might be included in the section.
20796 @value{GDBN} has support for this extension. If the section exists,
20797 then it is used provided that no other source of debugging information
20798 can be found, and that @value{GDBN} was configured with LZMA support.
20800 This section can be easily created using @command{objcopy} and other
20801 standard utilities:
20804 # Extract the dynamic symbols from the main binary, there is no need
20805 # to also have these in the normal symbol table.
20806 nm -D @var{binary} --format=posix --defined-only \
20807 | awk '@{ print $1 @}' | sort > dynsyms
20809 # Extract all the text (i.e. function) symbols from the debuginfo.
20810 # (Note that we actually also accept "D" symbols, for the benefit
20811 # of platforms like PowerPC64 that use function descriptors.)
20812 nm @var{binary} --format=posix --defined-only \
20813 | awk '@{ if ($2 == "T" || $2 == "t" || $2 == "D") print $1 @}' \
20816 # Keep all the function symbols not already in the dynamic symbol
20818 comm -13 dynsyms funcsyms > keep_symbols
20820 # Separate full debug info into debug binary.
20821 objcopy --only-keep-debug @var{binary} debug
20823 # Copy the full debuginfo, keeping only a minimal set of symbols and
20824 # removing some unnecessary sections.
20825 objcopy -S --remove-section .gdb_index --remove-section .comment \
20826 --keep-symbols=keep_symbols debug mini_debuginfo
20828 # Drop the full debug info from the original binary.
20829 strip --strip-all -R .comment @var{binary}
20831 # Inject the compressed data into the .gnu_debugdata section of the
20834 objcopy --add-section .gnu_debugdata=mini_debuginfo.xz @var{binary}
20838 @section Index Files Speed Up @value{GDBN}
20839 @cindex index files
20840 @cindex @samp{.gdb_index} section
20842 When @value{GDBN} finds a symbol file, it scans the symbols in the
20843 file in order to construct an internal symbol table. This lets most
20844 @value{GDBN} operations work quickly---at the cost of a delay early
20845 on. For large programs, this delay can be quite lengthy, so
20846 @value{GDBN} provides a way to build an index, which speeds up
20849 For convenience, @value{GDBN} comes with a program,
20850 @command{gdb-add-index}, which can be used to add the index to a
20851 symbol file. It takes the symbol file as its only argument:
20854 $ gdb-add-index symfile
20857 @xref{gdb-add-index}.
20859 It is also possible to do the work manually. Here is what
20860 @command{gdb-add-index} does behind the curtains.
20862 The index is stored as a section in the symbol file. @value{GDBN} can
20863 write the index to a file, then you can put it into the symbol file
20864 using @command{objcopy}.
20866 To create an index file, use the @code{save gdb-index} command:
20869 @item save gdb-index [-dwarf-5] @var{directory}
20870 @kindex save gdb-index
20871 Create index files for all symbol files currently known by
20872 @value{GDBN}. For each known @var{symbol-file}, this command by
20873 default creates it produces a single file
20874 @file{@var{symbol-file}.gdb-index}. If you invoke this command with
20875 the @option{-dwarf-5} option, it produces 2 files:
20876 @file{@var{symbol-file}.debug_names} and
20877 @file{@var{symbol-file}.debug_str}. The files are created in the
20878 given @var{directory}.
20881 Once you have created an index file you can merge it into your symbol
20882 file, here named @file{symfile}, using @command{objcopy}:
20885 $ objcopy --add-section .gdb_index=symfile.gdb-index \
20886 --set-section-flags .gdb_index=readonly symfile symfile
20889 Or for @code{-dwarf-5}:
20892 $ objcopy --dump-section .debug_str=symfile.debug_str.new symfile
20893 $ cat symfile.debug_str >>symfile.debug_str.new
20894 $ objcopy --add-section .debug_names=symfile.gdb-index \
20895 --set-section-flags .debug_names=readonly \
20896 --update-section .debug_str=symfile.debug_str.new symfile symfile
20899 @value{GDBN} will normally ignore older versions of @file{.gdb_index}
20900 sections that have been deprecated. Usually they are deprecated because
20901 they are missing a new feature or have performance issues.
20902 To tell @value{GDBN} to use a deprecated index section anyway
20903 specify @code{set use-deprecated-index-sections on}.
20904 The default is @code{off}.
20905 This can speed up startup, but may result in some functionality being lost.
20906 @xref{Index Section Format}.
20908 @emph{Warning:} Setting @code{use-deprecated-index-sections} to @code{on}
20909 must be done before gdb reads the file. The following will not work:
20912 $ gdb -ex "set use-deprecated-index-sections on" <program>
20915 Instead you must do, for example,
20918 $ gdb -iex "set use-deprecated-index-sections on" <program>
20921 There are currently some limitation on indices. They only work when
20922 for DWARF debugging information, not stabs. And, they do not
20923 currently work for programs using Ada.
20925 @subsection Automatic symbol index cache
20927 @cindex automatic symbol index cache
20928 It is possible for @value{GDBN} to automatically save a copy of this index in a
20929 cache on disk and retrieve it from there when loading the same binary in the
20930 future. This feature can be turned on with @kbd{set index-cache on}. The
20931 following commands can be used to tweak the behavior of the index cache.
20935 @kindex set index-cache
20936 @item set index-cache on
20937 @itemx set index-cache off
20938 Enable or disable the use of the symbol index cache.
20940 @item set index-cache directory @var{directory}
20941 @kindex show index-cache
20942 @itemx show index-cache directory
20943 Set/show the directory where index files will be saved.
20945 The default value for this directory depends on the host platform. On
20946 most systems, the index is cached in the @file{gdb} subdirectory of
20947 the directory pointed to by the @env{XDG_CACHE_HOME} environment
20948 variable, if it is defined, else in the @file{.cache/gdb} subdirectory
20949 of your home directory. However, on some systems, the default may
20950 differ according to local convention.
20952 There is no limit on the disk space used by index cache. It is perfectly safe
20953 to delete the content of that directory to free up disk space.
20955 @item show index-cache stats
20956 Print the number of cache hits and misses since the launch of @value{GDBN}.
20960 @node Symbol Errors
20961 @section Errors Reading Symbol Files
20963 While reading a symbol file, @value{GDBN} occasionally encounters problems,
20964 such as symbol types it does not recognize, or known bugs in compiler
20965 output. By default, @value{GDBN} does not notify you of such problems, since
20966 they are relatively common and primarily of interest to people
20967 debugging compilers. If you are interested in seeing information
20968 about ill-constructed symbol tables, you can either ask @value{GDBN} to print
20969 only one message about each such type of problem, no matter how many
20970 times the problem occurs; or you can ask @value{GDBN} to print more messages,
20971 to see how many times the problems occur, with the @code{set
20972 complaints} command (@pxref{Messages/Warnings, ,Optional Warnings and
20975 The messages currently printed, and their meanings, include:
20978 @item inner block not inside outer block in @var{symbol}
20980 The symbol information shows where symbol scopes begin and end
20981 (such as at the start of a function or a block of statements). This
20982 error indicates that an inner scope block is not fully contained
20983 in its outer scope blocks.
20985 @value{GDBN} circumvents the problem by treating the inner block as if it had
20986 the same scope as the outer block. In the error message, @var{symbol}
20987 may be shown as ``@code{(don't know)}'' if the outer block is not a
20990 @item block at @var{address} out of order
20992 The symbol information for symbol scope blocks should occur in
20993 order of increasing addresses. This error indicates that it does not
20996 @value{GDBN} does not circumvent this problem, and has trouble
20997 locating symbols in the source file whose symbols it is reading. (You
20998 can often determine what source file is affected by specifying
20999 @code{set verbose on}. @xref{Messages/Warnings, ,Optional Warnings and
21002 @item bad block start address patched
21004 The symbol information for a symbol scope block has a start address
21005 smaller than the address of the preceding source line. This is known
21006 to occur in the SunOS 4.1.1 (and earlier) C compiler.
21008 @value{GDBN} circumvents the problem by treating the symbol scope block as
21009 starting on the previous source line.
21011 @item bad string table offset in symbol @var{n}
21014 Symbol number @var{n} contains a pointer into the string table which is
21015 larger than the size of the string table.
21017 @value{GDBN} circumvents the problem by considering the symbol to have the
21018 name @code{foo}, which may cause other problems if many symbols end up
21021 @item unknown symbol type @code{0x@var{nn}}
21023 The symbol information contains new data types that @value{GDBN} does
21024 not yet know how to read. @code{0x@var{nn}} is the symbol type of the
21025 uncomprehended information, in hexadecimal.
21027 @value{GDBN} circumvents the error by ignoring this symbol information.
21028 This usually allows you to debug your program, though certain symbols
21029 are not accessible. If you encounter such a problem and feel like
21030 debugging it, you can debug @code{@value{GDBP}} with itself, breakpoint
21031 on @code{complain}, then go up to the function @code{read_dbx_symtab}
21032 and examine @code{*bufp} to see the symbol.
21034 @item stub type has NULL name
21036 @value{GDBN} could not find the full definition for a struct or class.
21038 @item const/volatile indicator missing (ok if using g++ v1.x), got@dots{}
21039 The symbol information for a C@t{++} member function is missing some
21040 information that recent versions of the compiler should have output for
21043 @item info mismatch between compiler and debugger
21045 @value{GDBN} could not parse a type specification output by the compiler.
21050 @section GDB Data Files
21052 @cindex prefix for data files
21053 @value{GDBN} will sometimes read an auxiliary data file. These files
21054 are kept in a directory known as the @dfn{data directory}.
21056 You can set the data directory's name, and view the name @value{GDBN}
21057 is currently using.
21060 @kindex set data-directory
21061 @item set data-directory @var{directory}
21062 Set the directory which @value{GDBN} searches for auxiliary data files
21063 to @var{directory}.
21065 @kindex show data-directory
21066 @item show data-directory
21067 Show the directory @value{GDBN} searches for auxiliary data files.
21070 @cindex default data directory
21071 @cindex @samp{--with-gdb-datadir}
21072 You can set the default data directory by using the configure-time
21073 @samp{--with-gdb-datadir} option. If the data directory is inside
21074 @value{GDBN}'s configured binary prefix (set with @samp{--prefix} or
21075 @samp{--exec-prefix}), then the default data directory will be updated
21076 automatically if the installed @value{GDBN} is moved to a new
21079 The data directory may also be specified with the
21080 @code{--data-directory} command line option.
21081 @xref{Mode Options}.
21084 @chapter Specifying a Debugging Target
21086 @cindex debugging target
21087 A @dfn{target} is the execution environment occupied by your program.
21089 Often, @value{GDBN} runs in the same host environment as your program;
21090 in that case, the debugging target is specified as a side effect when
21091 you use the @code{file} or @code{core} commands. When you need more
21092 flexibility---for example, running @value{GDBN} on a physically separate
21093 host, or controlling a standalone system over a serial port or a
21094 realtime system over a TCP/IP connection---you can use the @code{target}
21095 command to specify one of the target types configured for @value{GDBN}
21096 (@pxref{Target Commands, ,Commands for Managing Targets}).
21098 @cindex target architecture
21099 It is possible to build @value{GDBN} for several different @dfn{target
21100 architectures}. When @value{GDBN} is built like that, you can choose
21101 one of the available architectures with the @kbd{set architecture}
21105 @kindex set architecture
21106 @kindex show architecture
21107 @item set architecture @var{arch}
21108 This command sets the current target architecture to @var{arch}. The
21109 value of @var{arch} can be @code{"auto"}, in addition to one of the
21110 supported architectures.
21112 @item show architecture
21113 Show the current target architecture.
21115 @item set processor
21117 @kindex set processor
21118 @kindex show processor
21119 These are alias commands for, respectively, @code{set architecture}
21120 and @code{show architecture}.
21124 * Active Targets:: Active targets
21125 * Target Commands:: Commands for managing targets
21126 * Byte Order:: Choosing target byte order
21129 @node Active Targets
21130 @section Active Targets
21132 @cindex stacking targets
21133 @cindex active targets
21134 @cindex multiple targets
21136 There are multiple classes of targets such as: processes, executable files or
21137 recording sessions. Core files belong to the process class, making core file
21138 and process mutually exclusive. Otherwise, @value{GDBN} can work concurrently
21139 on multiple active targets, one in each class. This allows you to (for
21140 example) start a process and inspect its activity, while still having access to
21141 the executable file after the process finishes. Or if you start process
21142 recording (@pxref{Reverse Execution}) and @code{reverse-step} there, you are
21143 presented a virtual layer of the recording target, while the process target
21144 remains stopped at the chronologically last point of the process execution.
21146 Use the @code{core-file} and @code{exec-file} commands to select a new core
21147 file or executable target (@pxref{Files, ,Commands to Specify Files}). To
21148 specify as a target a process that is already running, use the @code{attach}
21149 command (@pxref{Attach, ,Debugging an Already-running Process}).
21151 @node Target Commands
21152 @section Commands for Managing Targets
21155 @item target @var{type} @var{parameters}
21156 Connects the @value{GDBN} host environment to a target machine or
21157 process. A target is typically a protocol for talking to debugging
21158 facilities. You use the argument @var{type} to specify the type or
21159 protocol of the target machine.
21161 Further @var{parameters} are interpreted by the target protocol, but
21162 typically include things like device names or host names to connect
21163 with, process numbers, and baud rates.
21165 The @code{target} command does not repeat if you press @key{RET} again
21166 after executing the command.
21168 @kindex help target
21170 Displays the names of all targets available. To display targets
21171 currently selected, use either @code{info target} or @code{info files}
21172 (@pxref{Files, ,Commands to Specify Files}).
21174 @item help target @var{name}
21175 Describe a particular target, including any parameters necessary to
21178 @kindex set gnutarget
21179 @item set gnutarget @var{args}
21180 @value{GDBN} uses its own library BFD to read your files. @value{GDBN}
21181 knows whether it is reading an @dfn{executable},
21182 a @dfn{core}, or a @dfn{.o} file; however, you can specify the file format
21183 with the @code{set gnutarget} command. Unlike most @code{target} commands,
21184 with @code{gnutarget} the @code{target} refers to a program, not a machine.
21187 @emph{Warning:} To specify a file format with @code{set gnutarget},
21188 you must know the actual BFD name.
21192 @xref{Files, , Commands to Specify Files}.
21194 @kindex show gnutarget
21195 @item show gnutarget
21196 Use the @code{show gnutarget} command to display what file format
21197 @code{gnutarget} is set to read. If you have not set @code{gnutarget},
21198 @value{GDBN} will determine the file format for each file automatically,
21199 and @code{show gnutarget} displays @samp{The current BFD target is "auto"}.
21202 @cindex common targets
21203 Here are some common targets (available, or not, depending on the GDB
21208 @item target exec @var{program}
21209 @cindex executable file target
21210 An executable file. @samp{target exec @var{program}} is the same as
21211 @samp{exec-file @var{program}}.
21213 @item target core @var{filename}
21214 @cindex core dump file target
21215 A core dump file. @samp{target core @var{filename}} is the same as
21216 @samp{core-file @var{filename}}.
21218 @item target remote @var{medium}
21219 @cindex remote target
21220 A remote system connected to @value{GDBN} via a serial line or network
21221 connection. This command tells @value{GDBN} to use its own remote
21222 protocol over @var{medium} for debugging. @xref{Remote Debugging}.
21224 For example, if you have a board connected to @file{/dev/ttya} on the
21225 machine running @value{GDBN}, you could say:
21228 target remote /dev/ttya
21231 @code{target remote} supports the @code{load} command. This is only
21232 useful if you have some other way of getting the stub to the target
21233 system, and you can put it somewhere in memory where it won't get
21234 clobbered by the download.
21236 @item target sim @r{[}@var{simargs}@r{]} @dots{}
21237 @cindex built-in simulator target
21238 Builtin CPU simulator. @value{GDBN} includes simulators for most architectures.
21246 works; however, you cannot assume that a specific memory map, device
21247 drivers, or even basic I/O is available, although some simulators do
21248 provide these. For info about any processor-specific simulator details,
21249 see the appropriate section in @ref{Embedded Processors, ,Embedded
21252 @item target native
21253 @cindex native target
21254 Setup for local/native process debugging. Useful to make the
21255 @code{run} command spawn native processes (likewise @code{attach},
21256 etc.@:) even when @code{set auto-connect-native-target} is @code{off}
21257 (@pxref{set auto-connect-native-target}).
21261 Different targets are available on different configurations of @value{GDBN};
21262 your configuration may have more or fewer targets.
21264 Many remote targets require you to download the executable's code once
21265 you've successfully established a connection. You may wish to control
21266 various aspects of this process.
21271 @kindex set hash@r{, for remote monitors}
21272 @cindex hash mark while downloading
21273 This command controls whether a hash mark @samp{#} is displayed while
21274 downloading a file to the remote monitor. If on, a hash mark is
21275 displayed after each S-record is successfully downloaded to the
21279 @kindex show hash@r{, for remote monitors}
21280 Show the current status of displaying the hash mark.
21282 @item set debug monitor
21283 @kindex set debug monitor
21284 @cindex display remote monitor communications
21285 Enable or disable display of communications messages between
21286 @value{GDBN} and the remote monitor.
21288 @item show debug monitor
21289 @kindex show debug monitor
21290 Show the current status of displaying communications between
21291 @value{GDBN} and the remote monitor.
21296 @kindex load @var{filename} @var{offset}
21297 @item load @var{filename} @var{offset}
21299 Depending on what remote debugging facilities are configured into
21300 @value{GDBN}, the @code{load} command may be available. Where it exists, it
21301 is meant to make @var{filename} (an executable) available for debugging
21302 on the remote system---by downloading, or dynamic linking, for example.
21303 @code{load} also records the @var{filename} symbol table in @value{GDBN}, like
21304 the @code{add-symbol-file} command.
21306 If your @value{GDBN} does not have a @code{load} command, attempting to
21307 execute it gets the error message ``@code{You can't do that when your
21308 target is @dots{}}''
21310 The file is loaded at whatever address is specified in the executable.
21311 For some object file formats, you can specify the load address when you
21312 link the program; for other formats, like a.out, the object file format
21313 specifies a fixed address.
21314 @c FIXME! This would be a good place for an xref to the GNU linker doc.
21316 It is also possible to tell @value{GDBN} to load the executable file at a
21317 specific offset described by the optional argument @var{offset}. When
21318 @var{offset} is provided, @var{filename} must also be provided.
21320 Depending on the remote side capabilities, @value{GDBN} may be able to
21321 load programs into flash memory.
21323 @code{load} does not repeat if you press @key{RET} again after using it.
21328 @kindex flash-erase
21330 @anchor{flash-erase}
21332 Erases all known flash memory regions on the target.
21337 @section Choosing Target Byte Order
21339 @cindex choosing target byte order
21340 @cindex target byte order
21342 Some types of processors, such as the @acronym{MIPS}, PowerPC, and Renesas SH,
21343 offer the ability to run either big-endian or little-endian byte
21344 orders. Usually the executable or symbol will include a bit to
21345 designate the endian-ness, and you will not need to worry about
21346 which to use. However, you may still find it useful to adjust
21347 @value{GDBN}'s idea of processor endian-ness manually.
21351 @item set endian big
21352 Instruct @value{GDBN} to assume the target is big-endian.
21354 @item set endian little
21355 Instruct @value{GDBN} to assume the target is little-endian.
21357 @item set endian auto
21358 Instruct @value{GDBN} to use the byte order associated with the
21362 Display @value{GDBN}'s current idea of the target byte order.
21366 If the @code{set endian auto} mode is in effect and no executable has
21367 been selected, then the endianness used is the last one chosen either
21368 by one of the @code{set endian big} and @code{set endian little}
21369 commands or by inferring from the last executable used. If no
21370 endianness has been previously chosen, then the default for this mode
21371 is inferred from the target @value{GDBN} has been built for, and is
21372 @code{little} if the name of the target CPU has an @code{el} suffix
21373 and @code{big} otherwise.
21375 Note that these commands merely adjust interpretation of symbolic
21376 data on the host, and that they have absolutely no effect on the
21380 @node Remote Debugging
21381 @chapter Debugging Remote Programs
21382 @cindex remote debugging
21384 If you are trying to debug a program running on a machine that cannot run
21385 @value{GDBN} in the usual way, it is often useful to use remote debugging.
21386 For example, you might use remote debugging on an operating system kernel,
21387 or on a small system which does not have a general purpose operating system
21388 powerful enough to run a full-featured debugger.
21390 Some configurations of @value{GDBN} have special serial or TCP/IP interfaces
21391 to make this work with particular debugging targets. In addition,
21392 @value{GDBN} comes with a generic serial protocol (specific to @value{GDBN},
21393 but not specific to any particular target system) which you can use if you
21394 write the remote stubs---the code that runs on the remote system to
21395 communicate with @value{GDBN}.
21397 Other remote targets may be available in your
21398 configuration of @value{GDBN}; use @code{help target} to list them.
21401 * Connecting:: Connecting to a remote target
21402 * File Transfer:: Sending files to a remote system
21403 * Server:: Using the gdbserver program
21404 * Remote Configuration:: Remote configuration
21405 * Remote Stub:: Implementing a remote stub
21409 @section Connecting to a Remote Target
21410 @cindex remote debugging, connecting
21411 @cindex @code{gdbserver}, connecting
21412 @cindex remote debugging, types of connections
21413 @cindex @code{gdbserver}, types of connections
21414 @cindex @code{gdbserver}, @code{target remote} mode
21415 @cindex @code{gdbserver}, @code{target extended-remote} mode
21417 This section describes how to connect to a remote target, including the
21418 types of connections and their differences, how to set up executable and
21419 symbol files on the host and target, and the commands used for
21420 connecting to and disconnecting from the remote target.
21422 @subsection Types of Remote Connections
21424 @value{GDBN} supports two types of remote connections, @code{target remote}
21425 mode and @code{target extended-remote} mode. Note that many remote targets
21426 support only @code{target remote} mode. There are several major
21427 differences between the two types of connections, enumerated here:
21431 @cindex remote debugging, detach and program exit
21432 @item Result of detach or program exit
21433 @strong{With target remote mode:} When the debugged program exits or you
21434 detach from it, @value{GDBN} disconnects from the target. When using
21435 @code{gdbserver}, @code{gdbserver} will exit.
21437 @strong{With target extended-remote mode:} When the debugged program exits or
21438 you detach from it, @value{GDBN} remains connected to the target, even
21439 though no program is running. You can rerun the program, attach to a
21440 running program, or use @code{monitor} commands specific to the target.
21442 When using @code{gdbserver} in this case, it does not exit unless it was
21443 invoked using the @option{--once} option. If the @option{--once} option
21444 was not used, you can ask @code{gdbserver} to exit using the
21445 @code{monitor exit} command (@pxref{Monitor Commands for gdbserver}).
21447 @item Specifying the program to debug
21448 For both connection types you use the @code{file} command to specify the
21449 program on the host system. If you are using @code{gdbserver} there are
21450 some differences in how to specify the location of the program on the
21453 @strong{With target remote mode:} You must either specify the program to debug
21454 on the @code{gdbserver} command line or use the @option{--attach} option
21455 (@pxref{Attaching to a program,,Attaching to a Running Program}).
21457 @cindex @option{--multi}, @code{gdbserver} option
21458 @strong{With target extended-remote mode:} You may specify the program to debug
21459 on the @code{gdbserver} command line, or you can load the program or attach
21460 to it using @value{GDBN} commands after connecting to @code{gdbserver}.
21462 @anchor{--multi Option in Types of Remote Connnections}
21463 You can start @code{gdbserver} without supplying an initial command to run
21464 or process ID to attach. To do this, use the @option{--multi} command line
21465 option. Then you can connect using @code{target extended-remote} and start
21466 the program you want to debug (see below for details on using the
21467 @code{run} command in this scenario). Note that the conditions under which
21468 @code{gdbserver} terminates depend on how @value{GDBN} connects to it
21469 (@code{target remote} or @code{target extended-remote}). The
21470 @option{--multi} option to @code{gdbserver} has no influence on that.
21472 @item The @code{run} command
21473 @strong{With target remote mode:} The @code{run} command is not
21474 supported. Once a connection has been established, you can use all
21475 the usual @value{GDBN} commands to examine and change data. The
21476 remote program is already running, so you can use commands like
21477 @kbd{step} and @kbd{continue}.
21479 @strong{With target extended-remote mode:} The @code{run} command is
21480 supported. The @code{run} command uses the value set by
21481 @code{set remote exec-file} (@pxref{set remote exec-file}) to select
21482 the program to run. Command line arguments are supported, except for
21483 wildcard expansion and I/O redirection (@pxref{Arguments}).
21485 If you specify the program to debug on the command line, then the
21486 @code{run} command is not required to start execution, and you can
21487 resume using commands like @kbd{step} and @kbd{continue} as with
21488 @code{target remote} mode.
21490 @anchor{Attaching in Types of Remote Connections}
21492 @strong{With target remote mode:} The @value{GDBN} command @code{attach} is
21493 not supported. To attach to a running program using @code{gdbserver}, you
21494 must use the @option{--attach} option (@pxref{Running gdbserver}).
21496 @strong{With target extended-remote mode:} To attach to a running program,
21497 you may use the @code{attach} command after the connection has been
21498 established. If you are using @code{gdbserver}, you may also invoke
21499 @code{gdbserver} using the @option{--attach} option
21500 (@pxref{Running gdbserver}).
21504 @anchor{Host and target files}
21505 @subsection Host and Target Files
21506 @cindex remote debugging, symbol files
21507 @cindex symbol files, remote debugging
21509 @value{GDBN}, running on the host, needs access to symbol and debugging
21510 information for your program running on the target. This requires
21511 access to an unstripped copy of your program, and possibly any associated
21512 symbol files. Note that this section applies equally to both @code{target
21513 remote} mode and @code{target extended-remote} mode.
21515 Some remote targets (@pxref{qXfer executable filename read}, and
21516 @pxref{Host I/O Packets}) allow @value{GDBN} to access program files over
21517 the same connection used to communicate with @value{GDBN}. With such a
21518 target, if the remote program is unstripped, the only command you need is
21519 @code{target remote} (or @code{target extended-remote}).
21521 If the remote program is stripped, or the target does not support remote
21522 program file access, start up @value{GDBN} using the name of the local
21523 unstripped copy of your program as the first argument, or use the
21524 @code{file} command. Use @code{set sysroot} to specify the location (on
21525 the host) of target libraries (unless your @value{GDBN} was compiled with
21526 the correct sysroot using @code{--with-sysroot}). Alternatively, you
21527 may use @code{set solib-search-path} to specify how @value{GDBN} locates
21530 The symbol file and target libraries must exactly match the executable
21531 and libraries on the target, with one exception: the files on the host
21532 system should not be stripped, even if the files on the target system
21533 are. Mismatched or missing files will lead to confusing results
21534 during debugging. On @sc{gnu}/Linux targets, mismatched or missing
21535 files may also prevent @code{gdbserver} from debugging multi-threaded
21538 @subsection Remote Connection Commands
21539 @cindex remote connection commands
21540 @value{GDBN} can communicate with the target over a serial line, a
21541 local Unix domain socket, or
21542 over an @acronym{IP} network using @acronym{TCP} or @acronym{UDP}. In
21543 each case, @value{GDBN} uses the same protocol for debugging your
21544 program; only the medium carrying the debugging packets varies. The
21545 @code{target remote} and @code{target extended-remote} commands
21546 establish a connection to the target. Both commands accept the same
21547 arguments, which indicate the medium to use:
21551 @item target remote @var{serial-device}
21552 @itemx target extended-remote @var{serial-device}
21553 @cindex serial line, @code{target remote}
21554 Use @var{serial-device} to communicate with the target. For example,
21555 to use a serial line connected to the device named @file{/dev/ttyb}:
21558 target remote /dev/ttyb
21561 If you're using a serial line, you may want to give @value{GDBN} the
21562 @samp{--baud} option, or use the @code{set serial baud} command
21563 (@pxref{Remote Configuration, set serial baud}) before the
21564 @code{target} command.
21566 @item target remote @var{local-socket}
21567 @itemx target extended-remote @var{local-socket}
21568 @cindex local socket, @code{target remote}
21569 @cindex Unix domain socket
21570 Use @var{local-socket} to communicate with the target. For example,
21571 to use a local Unix domain socket bound to the file system entry @file{/tmp/gdb-socket0}:
21574 target remote /tmp/gdb-socket0
21577 Note that this command has the same form as the command to connect
21578 to a serial line. @value{GDBN} will automatically determine which
21579 kind of file you have specified and will make the appropriate kind
21581 This feature is not available if the host system does not support
21582 Unix domain sockets.
21584 @item target remote @code{@var{host}:@var{port}}
21585 @itemx target remote @code{@var{[host]}:@var{port}}
21586 @itemx target remote @code{tcp:@var{host}:@var{port}}
21587 @itemx target remote @code{tcp:@var{[host]}:@var{port}}
21588 @itemx target remote @code{tcp4:@var{host}:@var{port}}
21589 @itemx target remote @code{tcp6:@var{host}:@var{port}}
21590 @itemx target remote @code{tcp6:@var{[host]}:@var{port}}
21591 @itemx target extended-remote @code{@var{host}:@var{port}}
21592 @itemx target extended-remote @code{@var{[host]}:@var{port}}
21593 @itemx target extended-remote @code{tcp:@var{host}:@var{port}}
21594 @itemx target extended-remote @code{tcp:@var{[host]}:@var{port}}
21595 @itemx target extended-remote @code{tcp4:@var{host}:@var{port}}
21596 @itemx target extended-remote @code{tcp6:@var{host}:@var{port}}
21597 @itemx target extended-remote @code{tcp6:@var{[host]}:@var{port}}
21598 @cindex @acronym{TCP} port, @code{target remote}
21599 Debug using a @acronym{TCP} connection to @var{port} on @var{host}.
21600 The @var{host} may be either a host name, a numeric @acronym{IPv4}
21601 address, or a numeric @acronym{IPv6} address (with or without the
21602 square brackets to separate the address from the port); @var{port}
21603 must be a decimal number. The @var{host} could be the target machine
21604 itself, if it is directly connected to the net, or it might be a
21605 terminal server which in turn has a serial line to the target.
21607 For example, to connect to port 2828 on a terminal server named
21611 target remote manyfarms:2828
21614 To connect to port 2828 on a terminal server whose address is
21615 @code{2001:0db8:85a3:0000:0000:8a2e:0370:7334}, you can either use the
21616 square bracket syntax:
21619 target remote [2001:0db8:85a3:0000:0000:8a2e:0370:7334]:2828
21623 or explicitly specify the @acronym{IPv6} protocol:
21626 target remote tcp6:2001:0db8:85a3:0000:0000:8a2e:0370:7334:2828
21629 This last example may be confusing to the reader, because there is no
21630 visible separation between the hostname and the port number.
21631 Therefore, we recommend the user to provide @acronym{IPv6} addresses
21632 using square brackets for clarity. However, it is important to
21633 mention that for @value{GDBN} there is no ambiguity: the number after
21634 the last colon is considered to be the port number.
21636 If your remote target is actually running on the same machine as your
21637 debugger session (e.g.@: a simulator for your target running on the
21638 same host), you can omit the hostname. For example, to connect to
21639 port 1234 on your local machine:
21642 target remote :1234
21646 Note that the colon is still required here.
21648 @item target remote @code{udp:@var{host}:@var{port}}
21649 @itemx target remote @code{udp:@var{[host]}:@var{port}}
21650 @itemx target remote @code{udp4:@var{host}:@var{port}}
21651 @itemx target remote @code{udp6:@var{[host]}:@var{port}}
21652 @itemx target extended-remote @code{udp:@var{host}:@var{port}}
21653 @itemx target extended-remote @code{udp:@var{host}:@var{port}}
21654 @itemx target extended-remote @code{udp:@var{[host]}:@var{port}}
21655 @itemx target extended-remote @code{udp4:@var{host}:@var{port}}
21656 @itemx target extended-remote @code{udp6:@var{host}:@var{port}}
21657 @itemx target extended-remote @code{udp6:@var{[host]}:@var{port}}
21658 @cindex @acronym{UDP} port, @code{target remote}
21659 Debug using @acronym{UDP} packets to @var{port} on @var{host}. For example, to
21660 connect to @acronym{UDP} port 2828 on a terminal server named @code{manyfarms}:
21663 target remote udp:manyfarms:2828
21666 When using a @acronym{UDP} connection for remote debugging, you should
21667 keep in mind that the `U' stands for ``Unreliable''. @acronym{UDP}
21668 can silently drop packets on busy or unreliable networks, which will
21669 cause havoc with your debugging session.
21671 @item target remote | @var{command}
21672 @itemx target extended-remote | @var{command}
21673 @cindex pipe, @code{target remote} to
21674 Run @var{command} in the background and communicate with it using a
21675 pipe. The @var{command} is a shell command, to be parsed and expanded
21676 by the system's command shell, @code{/bin/sh}; it should expect remote
21677 protocol packets on its standard input, and send replies on its
21678 standard output. You could use this to run a stand-alone simulator
21679 that speaks the remote debugging protocol, to make net connections
21680 using programs like @code{ssh}, or for other similar tricks.
21682 If @var{command} closes its standard output (perhaps by exiting),
21683 @value{GDBN} will try to send it a @code{SIGTERM} signal. (If the
21684 program has already exited, this will have no effect.)
21688 @cindex interrupting remote programs
21689 @cindex remote programs, interrupting
21690 Whenever @value{GDBN} is waiting for the remote program, if you type the
21691 interrupt character (often @kbd{Ctrl-c}), @value{GDBN} attempts to stop the
21692 program. This may or may not succeed, depending in part on the hardware
21693 and the serial drivers the remote system uses. If you type the
21694 interrupt character once again, @value{GDBN} displays this prompt:
21697 Interrupted while waiting for the program.
21698 Give up (and stop debugging it)? (y or n)
21701 In @code{target remote} mode, if you type @kbd{y}, @value{GDBN} abandons
21702 the remote debugging session. (If you decide you want to try again later,
21703 you can use @kbd{target remote} again to connect once more.) If you type
21704 @kbd{n}, @value{GDBN} goes back to waiting.
21706 In @code{target extended-remote} mode, typing @kbd{n} will leave
21707 @value{GDBN} connected to the target.
21710 @kindex detach (remote)
21712 When you have finished debugging the remote program, you can use the
21713 @code{detach} command to release it from @value{GDBN} control.
21714 Detaching from the target normally resumes its execution, but the results
21715 will depend on your particular remote stub. After the @code{detach}
21716 command in @code{target remote} mode, @value{GDBN} is free to connect to
21717 another target. In @code{target extended-remote} mode, @value{GDBN} is
21718 still connected to the target.
21722 The @code{disconnect} command closes the connection to the target, and
21723 the target is generally not resumed. It will wait for @value{GDBN}
21724 (this instance or another one) to connect and continue debugging. After
21725 the @code{disconnect} command, @value{GDBN} is again free to connect to
21728 @cindex send command to remote monitor
21729 @cindex extend @value{GDBN} for remote targets
21730 @cindex add new commands for external monitor
21732 @item monitor @var{cmd}
21733 This command allows you to send arbitrary commands directly to the
21734 remote monitor. Since @value{GDBN} doesn't care about the commands it
21735 sends like this, this command is the way to extend @value{GDBN}---you
21736 can add new commands that only the external monitor will understand
21740 @node File Transfer
21741 @section Sending files to a remote system
21742 @cindex remote target, file transfer
21743 @cindex file transfer
21744 @cindex sending files to remote systems
21746 Some remote targets offer the ability to transfer files over the same
21747 connection used to communicate with @value{GDBN}. This is convenient
21748 for targets accessible through other means, e.g.@: @sc{gnu}/Linux systems
21749 running @code{gdbserver} over a network interface. For other targets,
21750 e.g.@: embedded devices with only a single serial port, this may be
21751 the only way to upload or download files.
21753 Not all remote targets support these commands.
21757 @item remote put @var{hostfile} @var{targetfile}
21758 Copy file @var{hostfile} from the host system (the machine running
21759 @value{GDBN}) to @var{targetfile} on the target system.
21762 @item remote get @var{targetfile} @var{hostfile}
21763 Copy file @var{targetfile} from the target system to @var{hostfile}
21764 on the host system.
21766 @kindex remote delete
21767 @item remote delete @var{targetfile}
21768 Delete @var{targetfile} from the target system.
21773 @section Using the @code{gdbserver} Program
21776 @cindex remote connection without stubs
21777 @code{gdbserver} is a control program for Unix-like systems, which
21778 allows you to connect your program with a remote @value{GDBN} via
21779 @code{target remote} or @code{target extended-remote}---but without
21780 linking in the usual debugging stub.
21782 @code{gdbserver} is not a complete replacement for the debugging stubs,
21783 because it requires essentially the same operating-system facilities
21784 that @value{GDBN} itself does. In fact, a system that can run
21785 @code{gdbserver} to connect to a remote @value{GDBN} could also run
21786 @value{GDBN} locally! @code{gdbserver} is sometimes useful nevertheless,
21787 because it is a much smaller program than @value{GDBN} itself. It is
21788 also easier to port than all of @value{GDBN}, so you may be able to get
21789 started more quickly on a new system by using @code{gdbserver}.
21790 Finally, if you develop code for real-time systems, you may find that
21791 the tradeoffs involved in real-time operation make it more convenient to
21792 do as much development work as possible on another system, for example
21793 by cross-compiling. You can use @code{gdbserver} to make a similar
21794 choice for debugging.
21796 @value{GDBN} and @code{gdbserver} communicate via either a serial line
21797 or a TCP connection, using the standard @value{GDBN} remote serial
21801 @emph{Warning:} @code{gdbserver} does not have any built-in security.
21802 Do not run @code{gdbserver} connected to any public network; a
21803 @value{GDBN} connection to @code{gdbserver} provides access to the
21804 target system with the same privileges as the user running
21808 @anchor{Running gdbserver}
21809 @subsection Running @code{gdbserver}
21810 @cindex arguments, to @code{gdbserver}
21811 @cindex @code{gdbserver}, command-line arguments
21813 Run @code{gdbserver} on the target system. You need a copy of the
21814 program you want to debug, including any libraries it requires.
21815 @code{gdbserver} does not need your program's symbol table, so you can
21816 strip the program if necessary to save space. @value{GDBN} on the host
21817 system does all the symbol handling.
21819 To use the server, you must tell it how to communicate with @value{GDBN};
21820 the name of your program; and the arguments for your program. The usual
21824 target> gdbserver @var{comm} @var{program} [ @var{args} @dots{} ]
21827 @var{comm} is either a device name (to use a serial line), or a TCP
21828 hostname and portnumber, or @code{-} or @code{stdio} to use
21829 stdin/stdout of @code{gdbserver}.
21830 For example, to debug Emacs with the argument
21831 @samp{foo.txt} and communicate with @value{GDBN} over the serial port
21835 target> gdbserver /dev/com1 emacs foo.txt
21838 @code{gdbserver} waits passively for the host @value{GDBN} to communicate
21841 To use a TCP connection instead of a serial line:
21844 target> gdbserver host:2345 emacs foo.txt
21847 The only difference from the previous example is the first argument,
21848 specifying that you are communicating with the host @value{GDBN} via
21849 TCP. The @samp{host:2345} argument means that @code{gdbserver} is to
21850 expect a TCP connection from machine @samp{host} to local TCP port 2345.
21851 (Currently, the @samp{host} part is ignored.) You can choose any number
21852 you want for the port number as long as it does not conflict with any
21853 TCP ports already in use on the target system (for example, @code{23} is
21854 reserved for @code{telnet}).@footnote{If you choose a port number that
21855 conflicts with another service, @code{gdbserver} prints an error message
21856 and exits.} You must use the same port number with the host @value{GDBN}
21857 @code{target remote} command.
21859 The @code{stdio} connection is useful when starting @code{gdbserver}
21863 (gdb) target remote | ssh -T hostname gdbserver - hello
21866 The @samp{-T} option to ssh is provided because we don't need a remote pty,
21867 and we don't want escape-character handling. Ssh does this by default when
21868 a command is provided, the flag is provided to make it explicit.
21869 You could elide it if you want to.
21871 Programs started with stdio-connected gdbserver have @file{/dev/null} for
21872 @code{stdin}, and @code{stdout},@code{stderr} are sent back to gdb for
21873 display through a pipe connected to gdbserver.
21874 Both @code{stdout} and @code{stderr} use the same pipe.
21876 @anchor{Attaching to a program}
21877 @subsubsection Attaching to a Running Program
21878 @cindex attach to a program, @code{gdbserver}
21879 @cindex @option{--attach}, @code{gdbserver} option
21881 On some targets, @code{gdbserver} can also attach to running programs.
21882 This is accomplished via the @code{--attach} argument. The syntax is:
21885 target> gdbserver --attach @var{comm} @var{pid}
21888 @var{pid} is the process ID of a currently running process. It isn't
21889 necessary to point @code{gdbserver} at a binary for the running process.
21891 In @code{target extended-remote} mode, you can also attach using the
21892 @value{GDBN} attach command
21893 (@pxref{Attaching in Types of Remote Connections}).
21896 You can debug processes by name instead of process ID if your target has the
21897 @code{pidof} utility:
21900 target> gdbserver --attach @var{comm} `pidof @var{program}`
21903 In case more than one copy of @var{program} is running, or @var{program}
21904 has multiple threads, most versions of @code{pidof} support the
21905 @code{-s} option to only return the first process ID.
21907 @subsubsection TCP port allocation lifecycle of @code{gdbserver}
21909 This section applies only when @code{gdbserver} is run to listen on a TCP
21912 @code{gdbserver} normally terminates after all of its debugged processes have
21913 terminated in @kbd{target remote} mode. On the other hand, for @kbd{target
21914 extended-remote}, @code{gdbserver} stays running even with no processes left.
21915 @value{GDBN} normally terminates the spawned debugged process on its exit,
21916 which normally also terminates @code{gdbserver} in the @kbd{target remote}
21917 mode. Therefore, when the connection drops unexpectedly, and @value{GDBN}
21918 cannot ask @code{gdbserver} to kill its debugged processes, @code{gdbserver}
21919 stays running even in the @kbd{target remote} mode.
21921 When @code{gdbserver} stays running, @value{GDBN} can connect to it again later.
21922 Such reconnecting is useful for features like @ref{disconnected tracing}. For
21923 completeness, at most one @value{GDBN} can be connected at a time.
21925 @cindex @option{--once}, @code{gdbserver} option
21926 By default, @code{gdbserver} keeps the listening TCP port open, so that
21927 subsequent connections are possible. However, if you start @code{gdbserver}
21928 with the @option{--once} option, it will stop listening for any further
21929 connection attempts after connecting to the first @value{GDBN} session. This
21930 means no further connections to @code{gdbserver} will be possible after the
21931 first one. It also means @code{gdbserver} will terminate after the first
21932 connection with remote @value{GDBN} has closed, even for unexpectedly closed
21933 connections and even in the @kbd{target extended-remote} mode. The
21934 @option{--once} option allows reusing the same port number for connecting to
21935 multiple instances of @code{gdbserver} running on the same host, since each
21936 instance closes its port after the first connection.
21938 @anchor{Other Command-Line Arguments for gdbserver}
21939 @subsubsection Other Command-Line Arguments for @code{gdbserver}
21941 You can use the @option{--multi} option to start @code{gdbserver} without
21942 specifying a program to debug or a process to attach to. Then you can
21943 attach in @code{target extended-remote} mode and run or attach to a
21944 program. For more information,
21945 @pxref{--multi Option in Types of Remote Connnections}.
21947 @cindex @option{--debug}, @code{gdbserver} option
21948 The @option{--debug} option tells @code{gdbserver} to display extra
21949 status information about the debugging process.
21950 @cindex @option{--remote-debug}, @code{gdbserver} option
21951 The @option{--remote-debug} option tells @code{gdbserver} to display
21952 remote protocol debug output.
21953 @cindex @option{--debug-file}, @code{gdbserver} option
21954 @cindex @code{gdbserver}, send all debug output to a single file
21955 The @option{--debug-file=@var{filename}} option tells @code{gdbserver} to
21956 write any debug output to the given @var{filename}. These options are intended
21957 for @code{gdbserver} development and for bug reports to the developers.
21959 @cindex @option{--debug-format}, @code{gdbserver} option
21960 The @option{--debug-format=option1[,option2,...]} option tells
21961 @code{gdbserver} to include additional information in each output.
21962 Possible options are:
21966 Turn off all extra information in debugging output.
21968 Turn on all extra information in debugging output.
21970 Include a timestamp in each line of debugging output.
21973 Options are processed in order. Thus, for example, if @option{none}
21974 appears last then no additional information is added to debugging output.
21976 @cindex @option{--wrapper}, @code{gdbserver} option
21977 The @option{--wrapper} option specifies a wrapper to launch programs
21978 for debugging. The option should be followed by the name of the
21979 wrapper, then any command-line arguments to pass to the wrapper, then
21980 @kbd{--} indicating the end of the wrapper arguments.
21982 @code{gdbserver} runs the specified wrapper program with a combined
21983 command line including the wrapper arguments, then the name of the
21984 program to debug, then any arguments to the program. The wrapper
21985 runs until it executes your program, and then @value{GDBN} gains control.
21987 You can use any program that eventually calls @code{execve} with
21988 its arguments as a wrapper. Several standard Unix utilities do
21989 this, e.g.@: @code{env} and @code{nohup}. Any Unix shell script ending
21990 with @code{exec "$@@"} will also work.
21992 For example, you can use @code{env} to pass an environment variable to
21993 the debugged program, without setting the variable in @code{gdbserver}'s
21997 $ gdbserver --wrapper env LD_PRELOAD=libtest.so -- :2222 ./testprog
22000 @cindex @option{--selftest}
22001 The @option{--selftest} option runs the self tests in @code{gdbserver}:
22004 $ gdbserver --selftest
22005 Ran 2 unit tests, 0 failed
22008 These tests are disabled in release.
22009 @subsection Connecting to @code{gdbserver}
22011 The basic procedure for connecting to the remote target is:
22015 Run @value{GDBN} on the host system.
22018 Make sure you have the necessary symbol files
22019 (@pxref{Host and target files}).
22020 Load symbols for your application using the @code{file} command before you
22021 connect. Use @code{set sysroot} to locate target libraries (unless your
22022 @value{GDBN} was compiled with the correct sysroot using
22023 @code{--with-sysroot}).
22026 Connect to your target (@pxref{Connecting,,Connecting to a Remote Target}).
22027 For TCP connections, you must start up @code{gdbserver} prior to using
22028 the @code{target} command. Otherwise you may get an error whose
22029 text depends on the host system, but which usually looks something like
22030 @samp{Connection refused}. Don't use the @code{load}
22031 command in @value{GDBN} when using @code{target remote} mode, since the
22032 program is already on the target.
22036 @anchor{Monitor Commands for gdbserver}
22037 @subsection Monitor Commands for @code{gdbserver}
22038 @cindex monitor commands, for @code{gdbserver}
22040 During a @value{GDBN} session using @code{gdbserver}, you can use the
22041 @code{monitor} command to send special requests to @code{gdbserver}.
22042 Here are the available commands.
22046 List the available monitor commands.
22048 @item monitor set debug 0
22049 @itemx monitor set debug 1
22050 Disable or enable general debugging messages.
22052 @item monitor set remote-debug 0
22053 @itemx monitor set remote-debug 1
22054 Disable or enable specific debugging messages associated with the remote
22055 protocol (@pxref{Remote Protocol}).
22057 @item monitor set debug-file filename
22058 @itemx monitor set debug-file
22059 Send any debug output to the given file, or to stderr.
22061 @item monitor set debug-format option1@r{[},option2,...@r{]}
22062 Specify additional text to add to debugging messages.
22063 Possible options are:
22067 Turn off all extra information in debugging output.
22069 Turn on all extra information in debugging output.
22071 Include a timestamp in each line of debugging output.
22074 Options are processed in order. Thus, for example, if @option{none}
22075 appears last then no additional information is added to debugging output.
22077 @item monitor set libthread-db-search-path [PATH]
22078 @cindex gdbserver, search path for @code{libthread_db}
22079 When this command is issued, @var{path} is a colon-separated list of
22080 directories to search for @code{libthread_db} (@pxref{Threads,,set
22081 libthread-db-search-path}). If you omit @var{path},
22082 @samp{libthread-db-search-path} will be reset to its default value.
22084 The special entry @samp{$pdir} for @samp{libthread-db-search-path} is
22085 not supported in @code{gdbserver}.
22088 Tell gdbserver to exit immediately. This command should be followed by
22089 @code{disconnect} to close the debugging session. @code{gdbserver} will
22090 detach from any attached processes and kill any processes it created.
22091 Use @code{monitor exit} to terminate @code{gdbserver} at the end
22092 of a multi-process mode debug session.
22096 @subsection Tracepoints support in @code{gdbserver}
22097 @cindex tracepoints support in @code{gdbserver}
22099 On some targets, @code{gdbserver} supports tracepoints, fast
22100 tracepoints and static tracepoints.
22102 For fast or static tracepoints to work, a special library called the
22103 @dfn{in-process agent} (IPA), must be loaded in the inferior process.
22104 This library is built and distributed as an integral part of
22105 @code{gdbserver}. In addition, support for static tracepoints
22106 requires building the in-process agent library with static tracepoints
22107 support. At present, the UST (LTTng Userspace Tracer,
22108 @url{http://lttng.org/ust}) tracing engine is supported. This support
22109 is automatically available if UST development headers are found in the
22110 standard include path when @code{gdbserver} is built, or if
22111 @code{gdbserver} was explicitly configured using @option{--with-ust}
22112 to point at such headers. You can explicitly disable the support
22113 using @option{--with-ust=no}.
22115 There are several ways to load the in-process agent in your program:
22118 @item Specifying it as dependency at link time
22120 You can link your program dynamically with the in-process agent
22121 library. On most systems, this is accomplished by adding
22122 @code{-linproctrace} to the link command.
22124 @item Using the system's preloading mechanisms
22126 You can force loading the in-process agent at startup time by using
22127 your system's support for preloading shared libraries. Many Unixes
22128 support the concept of preloading user defined libraries. In most
22129 cases, you do that by specifying @code{LD_PRELOAD=libinproctrace.so}
22130 in the environment. See also the description of @code{gdbserver}'s
22131 @option{--wrapper} command line option.
22133 @item Using @value{GDBN} to force loading the agent at run time
22135 On some systems, you can force the inferior to load a shared library,
22136 by calling a dynamic loader function in the inferior that takes care
22137 of dynamically looking up and loading a shared library. On most Unix
22138 systems, the function is @code{dlopen}. You'll use the @code{call}
22139 command for that. For example:
22142 (@value{GDBP}) call dlopen ("libinproctrace.so", ...)
22145 Note that on most Unix systems, for the @code{dlopen} function to be
22146 available, the program needs to be linked with @code{-ldl}.
22149 On systems that have a userspace dynamic loader, like most Unix
22150 systems, when you connect to @code{gdbserver} using @code{target
22151 remote}, you'll find that the program is stopped at the dynamic
22152 loader's entry point, and no shared library has been loaded in the
22153 program's address space yet, including the in-process agent. In that
22154 case, before being able to use any of the fast or static tracepoints
22155 features, you need to let the loader run and load the shared
22156 libraries. The simplest way to do that is to run the program to the
22157 main procedure. E.g., if debugging a C or C@t{++} program, start
22158 @code{gdbserver} like so:
22161 $ gdbserver :9999 myprogram
22164 Start GDB and connect to @code{gdbserver} like so, and run to main:
22168 (@value{GDBP}) target remote myhost:9999
22169 0x00007f215893ba60 in ?? () from /lib64/ld-linux-x86-64.so.2
22170 (@value{GDBP}) b main
22171 (@value{GDBP}) continue
22174 The in-process tracing agent library should now be loaded into the
22175 process; you can confirm it with the @code{info sharedlibrary}
22176 command, which will list @file{libinproctrace.so} as loaded in the
22177 process. You are now ready to install fast tracepoints, list static
22178 tracepoint markers, probe static tracepoints markers, and start
22181 @node Remote Configuration
22182 @section Remote Configuration
22185 @kindex show remote
22186 This section documents the configuration options available when
22187 debugging remote programs. For the options related to the File I/O
22188 extensions of the remote protocol, see @ref{system,
22189 system-call-allowed}.
22192 @item set remoteaddresssize @var{bits}
22193 @cindex address size for remote targets
22194 @cindex bits in remote address
22195 Set the maximum size of address in a memory packet to the specified
22196 number of bits. @value{GDBN} will mask off the address bits above
22197 that number, when it passes addresses to the remote target. The
22198 default value is the number of bits in the target's address.
22200 @item show remoteaddresssize
22201 Show the current value of remote address size in bits.
22203 @item set serial baud @var{n}
22204 @cindex baud rate for remote targets
22205 Set the baud rate for the remote serial I/O to @var{n} baud. The
22206 value is used to set the speed of the serial port used for debugging
22209 @item show serial baud
22210 Show the current speed of the remote connection.
22212 @item set serial parity @var{parity}
22213 Set the parity for the remote serial I/O. Supported values of @var{parity} are:
22214 @code{even}, @code{none}, and @code{odd}. The default is @code{none}.
22216 @item show serial parity
22217 Show the current parity of the serial port.
22219 @item set remotebreak
22220 @cindex interrupt remote programs
22221 @cindex BREAK signal instead of Ctrl-C
22222 @anchor{set remotebreak}
22223 If set to on, @value{GDBN} sends a @code{BREAK} signal to the remote
22224 when you type @kbd{Ctrl-c} to interrupt the program running
22225 on the remote. If set to off, @value{GDBN} sends the @samp{Ctrl-C}
22226 character instead. The default is off, since most remote systems
22227 expect to see @samp{Ctrl-C} as the interrupt signal.
22229 @item show remotebreak
22230 Show whether @value{GDBN} sends @code{BREAK} or @samp{Ctrl-C} to
22231 interrupt the remote program.
22233 @item set remoteflow on
22234 @itemx set remoteflow off
22235 @kindex set remoteflow
22236 Enable or disable hardware flow control (@code{RTS}/@code{CTS})
22237 on the serial port used to communicate to the remote target.
22239 @item show remoteflow
22240 @kindex show remoteflow
22241 Show the current setting of hardware flow control.
22243 @item set remotelogbase @var{base}
22244 Set the base (a.k.a.@: radix) of logging serial protocol
22245 communications to @var{base}. Supported values of @var{base} are:
22246 @code{ascii}, @code{octal}, and @code{hex}. The default is
22249 @item show remotelogbase
22250 Show the current setting of the radix for logging remote serial
22253 @item set remotelogfile @var{file}
22254 @cindex record serial communications on file
22255 Record remote serial communications on the named @var{file}. The
22256 default is not to record at all.
22258 @item show remotelogfile
22259 Show the current setting of the file name on which to record the
22260 serial communications.
22262 @item set remotetimeout @var{num}
22263 @cindex timeout for serial communications
22264 @cindex remote timeout
22265 Set the timeout limit to wait for the remote target to respond to
22266 @var{num} seconds. The default is 2 seconds.
22268 @item show remotetimeout
22269 Show the current number of seconds to wait for the remote target
22272 @cindex limit hardware breakpoints and watchpoints
22273 @cindex remote target, limit break- and watchpoints
22274 @anchor{set remote hardware-watchpoint-limit}
22275 @anchor{set remote hardware-breakpoint-limit}
22276 @item set remote hardware-watchpoint-limit @var{limit}
22277 @itemx set remote hardware-breakpoint-limit @var{limit}
22278 Restrict @value{GDBN} to using @var{limit} remote hardware watchpoints
22279 or breakpoints. The @var{limit} can be set to 0 to disable hardware
22280 watchpoints or breakpoints, and @code{unlimited} for unlimited
22281 watchpoints or breakpoints.
22283 @item show remote hardware-watchpoint-limit
22284 @itemx show remote hardware-breakpoint-limit
22285 Show the current limit for the number of hardware watchpoints or
22286 breakpoints that @value{GDBN} can use.
22288 @cindex limit hardware watchpoints length
22289 @cindex remote target, limit watchpoints length
22290 @anchor{set remote hardware-watchpoint-length-limit}
22291 @item set remote hardware-watchpoint-length-limit @var{limit}
22292 Restrict @value{GDBN} to using @var{limit} bytes for the maximum
22293 length of a remote hardware watchpoint. A @var{limit} of 0 disables
22294 hardware watchpoints and @code{unlimited} allows watchpoints of any
22297 @item show remote hardware-watchpoint-length-limit
22298 Show the current limit (in bytes) of the maximum length of
22299 a remote hardware watchpoint.
22301 @item set remote exec-file @var{filename}
22302 @itemx show remote exec-file
22303 @anchor{set remote exec-file}
22304 @cindex executable file, for remote target
22305 Select the file used for @code{run} with @code{target
22306 extended-remote}. This should be set to a filename valid on the
22307 target system. If it is not set, the target will use a default
22308 filename (e.g.@: the last program run).
22310 @item set remote interrupt-sequence
22311 @cindex interrupt remote programs
22312 @cindex select Ctrl-C, BREAK or BREAK-g
22313 Allow the user to select one of @samp{Ctrl-C}, a @code{BREAK} or
22314 @samp{BREAK-g} as the
22315 sequence to the remote target in order to interrupt the execution.
22316 @samp{Ctrl-C} is a default. Some system prefers @code{BREAK} which
22317 is high level of serial line for some certain time.
22318 Linux kernel prefers @samp{BREAK-g}, a.k.a Magic SysRq g.
22319 It is @code{BREAK} signal followed by character @code{g}.
22321 @item show interrupt-sequence
22322 Show which of @samp{Ctrl-C}, @code{BREAK} or @code{BREAK-g}
22323 is sent by @value{GDBN} to interrupt the remote program.
22324 @code{BREAK-g} is BREAK signal followed by @code{g} and
22325 also known as Magic SysRq g.
22327 @item set remote interrupt-on-connect
22328 @cindex send interrupt-sequence on start
22329 Specify whether interrupt-sequence is sent to remote target when
22330 @value{GDBN} connects to it. This is mostly needed when you debug
22331 Linux kernel. Linux kernel expects @code{BREAK} followed by @code{g}
22332 which is known as Magic SysRq g in order to connect @value{GDBN}.
22334 @item show interrupt-on-connect
22335 Show whether interrupt-sequence is sent
22336 to remote target when @value{GDBN} connects to it.
22340 @item set tcp auto-retry on
22341 @cindex auto-retry, for remote TCP target
22342 Enable auto-retry for remote TCP connections. This is useful if the remote
22343 debugging agent is launched in parallel with @value{GDBN}; there is a race
22344 condition because the agent may not become ready to accept the connection
22345 before @value{GDBN} attempts to connect. When auto-retry is
22346 enabled, if the initial attempt to connect fails, @value{GDBN} reattempts
22347 to establish the connection using the timeout specified by
22348 @code{set tcp connect-timeout}.
22350 @item set tcp auto-retry off
22351 Do not auto-retry failed TCP connections.
22353 @item show tcp auto-retry
22354 Show the current auto-retry setting.
22356 @item set tcp connect-timeout @var{seconds}
22357 @itemx set tcp connect-timeout unlimited
22358 @cindex connection timeout, for remote TCP target
22359 @cindex timeout, for remote target connection
22360 Set the timeout for establishing a TCP connection to the remote target to
22361 @var{seconds}. The timeout affects both polling to retry failed connections
22362 (enabled by @code{set tcp auto-retry on}) and waiting for connections
22363 that are merely slow to complete, and represents an approximate cumulative
22364 value. If @var{seconds} is @code{unlimited}, there is no timeout and
22365 @value{GDBN} will keep attempting to establish a connection forever,
22366 unless interrupted with @kbd{Ctrl-c}. The default is 15 seconds.
22368 @item show tcp connect-timeout
22369 Show the current connection timeout setting.
22372 @cindex remote packets, enabling and disabling
22373 The @value{GDBN} remote protocol autodetects the packets supported by
22374 your debugging stub. If you need to override the autodetection, you
22375 can use these commands to enable or disable individual packets. Each
22376 packet can be set to @samp{on} (the remote target supports this
22377 packet), @samp{off} (the remote target does not support this packet),
22378 or @samp{auto} (detect remote target support for this packet). They
22379 all default to @samp{auto}. For more information about each packet,
22380 see @ref{Remote Protocol}.
22382 During normal use, you should not have to use any of these commands.
22383 If you do, that may be a bug in your remote debugging stub, or a bug
22384 in @value{GDBN}. You may want to report the problem to the
22385 @value{GDBN} developers.
22387 For each packet @var{name}, the command to enable or disable the
22388 packet is @code{set remote @var{name}-packet}. The available settings
22391 @multitable @columnfractions 0.28 0.32 0.25
22394 @tab Related Features
22396 @item @code{fetch-register}
22398 @tab @code{info registers}
22400 @item @code{set-register}
22404 @item @code{binary-download}
22406 @tab @code{load}, @code{set}
22408 @item @code{read-aux-vector}
22409 @tab @code{qXfer:auxv:read}
22410 @tab @code{info auxv}
22412 @item @code{symbol-lookup}
22413 @tab @code{qSymbol}
22414 @tab Detecting multiple threads
22416 @item @code{attach}
22417 @tab @code{vAttach}
22420 @item @code{verbose-resume}
22422 @tab Stepping or resuming multiple threads
22428 @item @code{software-breakpoint}
22432 @item @code{hardware-breakpoint}
22436 @item @code{write-watchpoint}
22440 @item @code{read-watchpoint}
22444 @item @code{access-watchpoint}
22448 @item @code{pid-to-exec-file}
22449 @tab @code{qXfer:exec-file:read}
22450 @tab @code{attach}, @code{run}
22452 @item @code{target-features}
22453 @tab @code{qXfer:features:read}
22454 @tab @code{set architecture}
22456 @item @code{library-info}
22457 @tab @code{qXfer:libraries:read}
22458 @tab @code{info sharedlibrary}
22460 @item @code{memory-map}
22461 @tab @code{qXfer:memory-map:read}
22462 @tab @code{info mem}
22464 @item @code{read-sdata-object}
22465 @tab @code{qXfer:sdata:read}
22466 @tab @code{print $_sdata}
22468 @item @code{read-spu-object}
22469 @tab @code{qXfer:spu:read}
22470 @tab @code{info spu}
22472 @item @code{write-spu-object}
22473 @tab @code{qXfer:spu:write}
22474 @tab @code{info spu}
22476 @item @code{read-siginfo-object}
22477 @tab @code{qXfer:siginfo:read}
22478 @tab @code{print $_siginfo}
22480 @item @code{write-siginfo-object}
22481 @tab @code{qXfer:siginfo:write}
22482 @tab @code{set $_siginfo}
22484 @item @code{threads}
22485 @tab @code{qXfer:threads:read}
22486 @tab @code{info threads}
22488 @item @code{get-thread-local-@*storage-address}
22489 @tab @code{qGetTLSAddr}
22490 @tab Displaying @code{__thread} variables
22492 @item @code{get-thread-information-block-address}
22493 @tab @code{qGetTIBAddr}
22494 @tab Display MS-Windows Thread Information Block.
22496 @item @code{search-memory}
22497 @tab @code{qSearch:memory}
22500 @item @code{supported-packets}
22501 @tab @code{qSupported}
22502 @tab Remote communications parameters
22504 @item @code{catch-syscalls}
22505 @tab @code{QCatchSyscalls}
22506 @tab @code{catch syscall}
22508 @item @code{pass-signals}
22509 @tab @code{QPassSignals}
22510 @tab @code{handle @var{signal}}
22512 @item @code{program-signals}
22513 @tab @code{QProgramSignals}
22514 @tab @code{handle @var{signal}}
22516 @item @code{hostio-close-packet}
22517 @tab @code{vFile:close}
22518 @tab @code{remote get}, @code{remote put}
22520 @item @code{hostio-open-packet}
22521 @tab @code{vFile:open}
22522 @tab @code{remote get}, @code{remote put}
22524 @item @code{hostio-pread-packet}
22525 @tab @code{vFile:pread}
22526 @tab @code{remote get}, @code{remote put}
22528 @item @code{hostio-pwrite-packet}
22529 @tab @code{vFile:pwrite}
22530 @tab @code{remote get}, @code{remote put}
22532 @item @code{hostio-unlink-packet}
22533 @tab @code{vFile:unlink}
22534 @tab @code{remote delete}
22536 @item @code{hostio-readlink-packet}
22537 @tab @code{vFile:readlink}
22540 @item @code{hostio-fstat-packet}
22541 @tab @code{vFile:fstat}
22544 @item @code{hostio-setfs-packet}
22545 @tab @code{vFile:setfs}
22548 @item @code{noack-packet}
22549 @tab @code{QStartNoAckMode}
22550 @tab Packet acknowledgment
22552 @item @code{osdata}
22553 @tab @code{qXfer:osdata:read}
22554 @tab @code{info os}
22556 @item @code{query-attached}
22557 @tab @code{qAttached}
22558 @tab Querying remote process attach state.
22560 @item @code{trace-buffer-size}
22561 @tab @code{QTBuffer:size}
22562 @tab @code{set trace-buffer-size}
22564 @item @code{trace-status}
22565 @tab @code{qTStatus}
22566 @tab @code{tstatus}
22568 @item @code{traceframe-info}
22569 @tab @code{qXfer:traceframe-info:read}
22570 @tab Traceframe info
22572 @item @code{install-in-trace}
22573 @tab @code{InstallInTrace}
22574 @tab Install tracepoint in tracing
22576 @item @code{disable-randomization}
22577 @tab @code{QDisableRandomization}
22578 @tab @code{set disable-randomization}
22580 @item @code{startup-with-shell}
22581 @tab @code{QStartupWithShell}
22582 @tab @code{set startup-with-shell}
22584 @item @code{environment-hex-encoded}
22585 @tab @code{QEnvironmentHexEncoded}
22586 @tab @code{set environment}
22588 @item @code{environment-unset}
22589 @tab @code{QEnvironmentUnset}
22590 @tab @code{unset environment}
22592 @item @code{environment-reset}
22593 @tab @code{QEnvironmentReset}
22594 @tab @code{Reset the inferior environment (i.e., unset user-set variables)}
22596 @item @code{set-working-dir}
22597 @tab @code{QSetWorkingDir}
22598 @tab @code{set cwd}
22600 @item @code{conditional-breakpoints-packet}
22601 @tab @code{Z0 and Z1}
22602 @tab @code{Support for target-side breakpoint condition evaluation}
22604 @item @code{multiprocess-extensions}
22605 @tab @code{multiprocess extensions}
22606 @tab Debug multiple processes and remote process PID awareness
22608 @item @code{swbreak-feature}
22609 @tab @code{swbreak stop reason}
22612 @item @code{hwbreak-feature}
22613 @tab @code{hwbreak stop reason}
22616 @item @code{fork-event-feature}
22617 @tab @code{fork stop reason}
22620 @item @code{vfork-event-feature}
22621 @tab @code{vfork stop reason}
22624 @item @code{exec-event-feature}
22625 @tab @code{exec stop reason}
22628 @item @code{thread-events}
22629 @tab @code{QThreadEvents}
22630 @tab Tracking thread lifetime.
22632 @item @code{no-resumed-stop-reply}
22633 @tab @code{no resumed thread left stop reply}
22634 @tab Tracking thread lifetime.
22639 @section Implementing a Remote Stub
22641 @cindex debugging stub, example
22642 @cindex remote stub, example
22643 @cindex stub example, remote debugging
22644 The stub files provided with @value{GDBN} implement the target side of the
22645 communication protocol, and the @value{GDBN} side is implemented in the
22646 @value{GDBN} source file @file{remote.c}. Normally, you can simply allow
22647 these subroutines to communicate, and ignore the details. (If you're
22648 implementing your own stub file, you can still ignore the details: start
22649 with one of the existing stub files. @file{sparc-stub.c} is the best
22650 organized, and therefore the easiest to read.)
22652 @cindex remote serial debugging, overview
22653 To debug a program running on another machine (the debugging
22654 @dfn{target} machine), you must first arrange for all the usual
22655 prerequisites for the program to run by itself. For example, for a C
22660 A startup routine to set up the C runtime environment; these usually
22661 have a name like @file{crt0}. The startup routine may be supplied by
22662 your hardware supplier, or you may have to write your own.
22665 A C subroutine library to support your program's
22666 subroutine calls, notably managing input and output.
22669 A way of getting your program to the other machine---for example, a
22670 download program. These are often supplied by the hardware
22671 manufacturer, but you may have to write your own from hardware
22675 The next step is to arrange for your program to use a serial port to
22676 communicate with the machine where @value{GDBN} is running (the @dfn{host}
22677 machine). In general terms, the scheme looks like this:
22681 @value{GDBN} already understands how to use this protocol; when everything
22682 else is set up, you can simply use the @samp{target remote} command
22683 (@pxref{Targets,,Specifying a Debugging Target}).
22685 @item On the target,
22686 you must link with your program a few special-purpose subroutines that
22687 implement the @value{GDBN} remote serial protocol. The file containing these
22688 subroutines is called a @dfn{debugging stub}.
22690 On certain remote targets, you can use an auxiliary program
22691 @code{gdbserver} instead of linking a stub into your program.
22692 @xref{Server,,Using the @code{gdbserver} Program}, for details.
22695 The debugging stub is specific to the architecture of the remote
22696 machine; for example, use @file{sparc-stub.c} to debug programs on
22699 @cindex remote serial stub list
22700 These working remote stubs are distributed with @value{GDBN}:
22705 @cindex @file{i386-stub.c}
22708 For Intel 386 and compatible architectures.
22711 @cindex @file{m68k-stub.c}
22712 @cindex Motorola 680x0
22714 For Motorola 680x0 architectures.
22717 @cindex @file{sh-stub.c}
22720 For Renesas SH architectures.
22723 @cindex @file{sparc-stub.c}
22725 For @sc{sparc} architectures.
22727 @item sparcl-stub.c
22728 @cindex @file{sparcl-stub.c}
22731 For Fujitsu @sc{sparclite} architectures.
22735 The @file{README} file in the @value{GDBN} distribution may list other
22736 recently added stubs.
22739 * Stub Contents:: What the stub can do for you
22740 * Bootstrapping:: What you must do for the stub
22741 * Debug Session:: Putting it all together
22744 @node Stub Contents
22745 @subsection What the Stub Can Do for You
22747 @cindex remote serial stub
22748 The debugging stub for your architecture supplies these three
22752 @item set_debug_traps
22753 @findex set_debug_traps
22754 @cindex remote serial stub, initialization
22755 This routine arranges for @code{handle_exception} to run when your
22756 program stops. You must call this subroutine explicitly in your
22757 program's startup code.
22759 @item handle_exception
22760 @findex handle_exception
22761 @cindex remote serial stub, main routine
22762 This is the central workhorse, but your program never calls it
22763 explicitly---the setup code arranges for @code{handle_exception} to
22764 run when a trap is triggered.
22766 @code{handle_exception} takes control when your program stops during
22767 execution (for example, on a breakpoint), and mediates communications
22768 with @value{GDBN} on the host machine. This is where the communications
22769 protocol is implemented; @code{handle_exception} acts as the @value{GDBN}
22770 representative on the target machine. It begins by sending summary
22771 information on the state of your program, then continues to execute,
22772 retrieving and transmitting any information @value{GDBN} needs, until you
22773 execute a @value{GDBN} command that makes your program resume; at that point,
22774 @code{handle_exception} returns control to your own code on the target
22778 @cindex @code{breakpoint} subroutine, remote
22779 Use this auxiliary subroutine to make your program contain a
22780 breakpoint. Depending on the particular situation, this may be the only
22781 way for @value{GDBN} to get control. For instance, if your target
22782 machine has some sort of interrupt button, you won't need to call this;
22783 pressing the interrupt button transfers control to
22784 @code{handle_exception}---in effect, to @value{GDBN}. On some machines,
22785 simply receiving characters on the serial port may also trigger a trap;
22786 again, in that situation, you don't need to call @code{breakpoint} from
22787 your own program---simply running @samp{target remote} from the host
22788 @value{GDBN} session gets control.
22790 Call @code{breakpoint} if none of these is true, or if you simply want
22791 to make certain your program stops at a predetermined point for the
22792 start of your debugging session.
22795 @node Bootstrapping
22796 @subsection What You Must Do for the Stub
22798 @cindex remote stub, support routines
22799 The debugging stubs that come with @value{GDBN} are set up for a particular
22800 chip architecture, but they have no information about the rest of your
22801 debugging target machine.
22803 First of all you need to tell the stub how to communicate with the
22807 @item int getDebugChar()
22808 @findex getDebugChar
22809 Write this subroutine to read a single character from the serial port.
22810 It may be identical to @code{getchar} for your target system; a
22811 different name is used to allow you to distinguish the two if you wish.
22813 @item void putDebugChar(int)
22814 @findex putDebugChar
22815 Write this subroutine to write a single character to the serial port.
22816 It may be identical to @code{putchar} for your target system; a
22817 different name is used to allow you to distinguish the two if you wish.
22820 @cindex control C, and remote debugging
22821 @cindex interrupting remote targets
22822 If you want @value{GDBN} to be able to stop your program while it is
22823 running, you need to use an interrupt-driven serial driver, and arrange
22824 for it to stop when it receives a @code{^C} (@samp{\003}, the control-C
22825 character). That is the character which @value{GDBN} uses to tell the
22826 remote system to stop.
22828 Getting the debugging target to return the proper status to @value{GDBN}
22829 probably requires changes to the standard stub; one quick and dirty way
22830 is to just execute a breakpoint instruction (the ``dirty'' part is that
22831 @value{GDBN} reports a @code{SIGTRAP} instead of a @code{SIGINT}).
22833 Other routines you need to supply are:
22836 @item void exceptionHandler (int @var{exception_number}, void *@var{exception_address})
22837 @findex exceptionHandler
22838 Write this function to install @var{exception_address} in the exception
22839 handling tables. You need to do this because the stub does not have any
22840 way of knowing what the exception handling tables on your target system
22841 are like (for example, the processor's table might be in @sc{rom},
22842 containing entries which point to a table in @sc{ram}).
22843 The @var{exception_number} specifies the exception which should be changed;
22844 its meaning is architecture-dependent (for example, different numbers
22845 might represent divide by zero, misaligned access, etc). When this
22846 exception occurs, control should be transferred directly to
22847 @var{exception_address}, and the processor state (stack, registers,
22848 and so on) should be just as it is when a processor exception occurs. So if
22849 you want to use a jump instruction to reach @var{exception_address}, it
22850 should be a simple jump, not a jump to subroutine.
22852 For the 386, @var{exception_address} should be installed as an interrupt
22853 gate so that interrupts are masked while the handler runs. The gate
22854 should be at privilege level 0 (the most privileged level). The
22855 @sc{sparc} and 68k stubs are able to mask interrupts themselves without
22856 help from @code{exceptionHandler}.
22858 @item void flush_i_cache()
22859 @findex flush_i_cache
22860 On @sc{sparc} and @sc{sparclite} only, write this subroutine to flush the
22861 instruction cache, if any, on your target machine. If there is no
22862 instruction cache, this subroutine may be a no-op.
22864 On target machines that have instruction caches, @value{GDBN} requires this
22865 function to make certain that the state of your program is stable.
22869 You must also make sure this library routine is available:
22872 @item void *memset(void *, int, int)
22874 This is the standard library function @code{memset} that sets an area of
22875 memory to a known value. If you have one of the free versions of
22876 @code{libc.a}, @code{memset} can be found there; otherwise, you must
22877 either obtain it from your hardware manufacturer, or write your own.
22880 If you do not use the GNU C compiler, you may need other standard
22881 library subroutines as well; this varies from one stub to another,
22882 but in general the stubs are likely to use any of the common library
22883 subroutines which @code{@value{NGCC}} generates as inline code.
22886 @node Debug Session
22887 @subsection Putting it All Together
22889 @cindex remote serial debugging summary
22890 In summary, when your program is ready to debug, you must follow these
22895 Make sure you have defined the supporting low-level routines
22896 (@pxref{Bootstrapping,,What You Must Do for the Stub}):
22898 @code{getDebugChar}, @code{putDebugChar},
22899 @code{flush_i_cache}, @code{memset}, @code{exceptionHandler}.
22903 Insert these lines in your program's startup code, before the main
22904 procedure is called:
22911 On some machines, when a breakpoint trap is raised, the hardware
22912 automatically makes the PC point to the instruction after the
22913 breakpoint. If your machine doesn't do that, you may need to adjust
22914 @code{handle_exception} to arrange for it to return to the instruction
22915 after the breakpoint on this first invocation, so that your program
22916 doesn't keep hitting the initial breakpoint instead of making
22920 For the 680x0 stub only, you need to provide a variable called
22921 @code{exceptionHook}. Normally you just use:
22924 void (*exceptionHook)() = 0;
22928 but if before calling @code{set_debug_traps}, you set it to point to a
22929 function in your program, that function is called when
22930 @code{@value{GDBN}} continues after stopping on a trap (for example, bus
22931 error). The function indicated by @code{exceptionHook} is called with
22932 one parameter: an @code{int} which is the exception number.
22935 Compile and link together: your program, the @value{GDBN} debugging stub for
22936 your target architecture, and the supporting subroutines.
22939 Make sure you have a serial connection between your target machine and
22940 the @value{GDBN} host, and identify the serial port on the host.
22943 @c The "remote" target now provides a `load' command, so we should
22944 @c document that. FIXME.
22945 Download your program to your target machine (or get it there by
22946 whatever means the manufacturer provides), and start it.
22949 Start @value{GDBN} on the host, and connect to the target
22950 (@pxref{Connecting,,Connecting to a Remote Target}).
22954 @node Configurations
22955 @chapter Configuration-Specific Information
22957 While nearly all @value{GDBN} commands are available for all native and
22958 cross versions of the debugger, there are some exceptions. This chapter
22959 describes things that are only available in certain configurations.
22961 There are three major categories of configurations: native
22962 configurations, where the host and target are the same, embedded
22963 operating system configurations, which are usually the same for several
22964 different processor architectures, and bare embedded processors, which
22965 are quite different from each other.
22970 * Embedded Processors::
22977 This section describes details specific to particular native
22981 * BSD libkvm Interface:: Debugging BSD kernel memory images
22982 * Process Information:: Process information
22983 * DJGPP Native:: Features specific to the DJGPP port
22984 * Cygwin Native:: Features specific to the Cygwin port
22985 * Hurd Native:: Features specific to @sc{gnu} Hurd
22986 * Darwin:: Features specific to Darwin
22987 * FreeBSD:: Features specific to FreeBSD
22990 @node BSD libkvm Interface
22991 @subsection BSD libkvm Interface
22994 @cindex kernel memory image
22995 @cindex kernel crash dump
22997 BSD-derived systems (FreeBSD/NetBSD/OpenBSD) have a kernel memory
22998 interface that provides a uniform interface for accessing kernel virtual
22999 memory images, including live systems and crash dumps. @value{GDBN}
23000 uses this interface to allow you to debug live kernels and kernel crash
23001 dumps on many native BSD configurations. This is implemented as a
23002 special @code{kvm} debugging target. For debugging a live system, load
23003 the currently running kernel into @value{GDBN} and connect to the
23007 (@value{GDBP}) @b{target kvm}
23010 For debugging crash dumps, provide the file name of the crash dump as an
23014 (@value{GDBP}) @b{target kvm /var/crash/bsd.0}
23017 Once connected to the @code{kvm} target, the following commands are
23023 Set current context from the @dfn{Process Control Block} (PCB) address.
23026 Set current context from proc address. This command isn't available on
23027 modern FreeBSD systems.
23030 @node Process Information
23031 @subsection Process Information
23033 @cindex examine process image
23034 @cindex process info via @file{/proc}
23036 Some operating systems provide interfaces to fetch additional
23037 information about running processes beyond memory and per-thread
23038 register state. If @value{GDBN} is configured for an operating system
23039 with a supported interface, the command @code{info proc} is available
23040 to report information about the process running your program, or about
23041 any process running on your system.
23043 One supported interface is a facility called @samp{/proc} that can be
23044 used to examine the image of a running process using file-system
23045 subroutines. This facility is supported on @sc{gnu}/Linux and Solaris
23048 On FreeBSD systems, system control nodes are used to query process
23051 In addition, some systems may provide additional process information
23052 in core files. Note that a core file may include a subset of the
23053 information available from a live process. Process information is
23054 currently avaiable from cores created on @sc{gnu}/Linux and FreeBSD
23061 @itemx info proc @var{process-id}
23062 Summarize available information about a process. If a
23063 process ID is specified by @var{process-id}, display information about
23064 that process; otherwise display information about the program being
23065 debugged. The summary includes the debugged process ID, the command
23066 line used to invoke it, its current working directory, and its
23067 executable file's absolute file name.
23069 On some systems, @var{process-id} can be of the form
23070 @samp{[@var{pid}]/@var{tid}} which specifies a certain thread ID
23071 within a process. If the optional @var{pid} part is missing, it means
23072 a thread from the process being debugged (the leading @samp{/} still
23073 needs to be present, or else @value{GDBN} will interpret the number as
23074 a process ID rather than a thread ID).
23076 @item info proc cmdline
23077 @cindex info proc cmdline
23078 Show the original command line of the process. This command is
23079 supported on @sc{gnu}/Linux and FreeBSD.
23081 @item info proc cwd
23082 @cindex info proc cwd
23083 Show the current working directory of the process. This command is
23084 supported on @sc{gnu}/Linux and FreeBSD.
23086 @item info proc exe
23087 @cindex info proc exe
23088 Show the name of executable of the process. This command is supported
23089 on @sc{gnu}/Linux and FreeBSD.
23091 @item info proc files
23092 @cindex info proc files
23093 Show the file descriptors open by the process. For each open file
23094 descriptor, @value{GDBN} shows its number, type (file, directory,
23095 character device, socket), file pointer offset, and the name of the
23096 resource open on the descriptor. The resource name can be a file name
23097 (for files, directories, and devices) or a protocol followed by socket
23098 address (for network connections). This command is supported on
23101 This example shows the open file descriptors for a process using a
23102 tty for standard input and output as well as two network sockets:
23105 (gdb) info proc files 22136
23109 FD Type Offset Flags Name
23110 text file - r-------- /usr/bin/ssh
23111 ctty chr - rw------- /dev/pts/20
23112 cwd dir - r-------- /usr/home/john
23113 root dir - r-------- /
23114 0 chr 0x32933a4 rw------- /dev/pts/20
23115 1 chr 0x32933a4 rw------- /dev/pts/20
23116 2 chr 0x32933a4 rw------- /dev/pts/20
23117 3 socket 0x0 rw----n-- tcp4 10.0.1.2:53014 -> 10.0.1.10:22
23118 4 socket 0x0 rw------- unix stream:/tmp/ssh-FIt89oAzOn5f/agent.2456
23121 @item info proc mappings
23122 @cindex memory address space mappings
23123 Report the memory address space ranges accessible in a process. On
23124 Solaris and FreeBSD systems, each memory range includes information on
23125 whether the process has read, write, or execute access rights to each
23126 range. On @sc{gnu}/Linux and FreeBSD systems, each memory range
23127 includes the object file which is mapped to that range.
23129 @item info proc stat
23130 @itemx info proc status
23131 @cindex process detailed status information
23132 Show additional process-related information, including the user ID and
23133 group ID; virtual memory usage; the signals that are pending, blocked,
23134 and ignored; its TTY; its consumption of system and user time; its
23135 stack size; its @samp{nice} value; etc. These commands are supported
23136 on @sc{gnu}/Linux and FreeBSD.
23138 For @sc{gnu}/Linux systems, see the @samp{proc} man page for more
23139 information (type @kbd{man 5 proc} from your shell prompt).
23141 For FreeBSD systems, @code{info proc stat} is an alias for @code{info
23144 @item info proc all
23145 Show all the information about the process described under all of the
23146 above @code{info proc} subcommands.
23149 @comment These sub-options of 'info proc' were not included when
23150 @comment procfs.c was re-written. Keep their descriptions around
23151 @comment against the day when someone finds the time to put them back in.
23152 @kindex info proc times
23153 @item info proc times
23154 Starting time, user CPU time, and system CPU time for your program and
23157 @kindex info proc id
23159 Report on the process IDs related to your program: its own process ID,
23160 the ID of its parent, the process group ID, and the session ID.
23163 @item set procfs-trace
23164 @kindex set procfs-trace
23165 @cindex @code{procfs} API calls
23166 This command enables and disables tracing of @code{procfs} API calls.
23168 @item show procfs-trace
23169 @kindex show procfs-trace
23170 Show the current state of @code{procfs} API call tracing.
23172 @item set procfs-file @var{file}
23173 @kindex set procfs-file
23174 Tell @value{GDBN} to write @code{procfs} API trace to the named
23175 @var{file}. @value{GDBN} appends the trace info to the previous
23176 contents of the file. The default is to display the trace on the
23179 @item show procfs-file
23180 @kindex show procfs-file
23181 Show the file to which @code{procfs} API trace is written.
23183 @item proc-trace-entry
23184 @itemx proc-trace-exit
23185 @itemx proc-untrace-entry
23186 @itemx proc-untrace-exit
23187 @kindex proc-trace-entry
23188 @kindex proc-trace-exit
23189 @kindex proc-untrace-entry
23190 @kindex proc-untrace-exit
23191 These commands enable and disable tracing of entries into and exits
23192 from the @code{syscall} interface.
23195 @kindex info pidlist
23196 @cindex process list, QNX Neutrino
23197 For QNX Neutrino only, this command displays the list of all the
23198 processes and all the threads within each process.
23201 @kindex info meminfo
23202 @cindex mapinfo list, QNX Neutrino
23203 For QNX Neutrino only, this command displays the list of all mapinfos.
23207 @subsection Features for Debugging @sc{djgpp} Programs
23208 @cindex @sc{djgpp} debugging
23209 @cindex native @sc{djgpp} debugging
23210 @cindex MS-DOS-specific commands
23213 @sc{djgpp} is a port of the @sc{gnu} development tools to MS-DOS and
23214 MS-Windows. @sc{djgpp} programs are 32-bit protected-mode programs
23215 that use the @dfn{DPMI} (DOS Protected-Mode Interface) API to run on
23216 top of real-mode DOS systems and their emulations.
23218 @value{GDBN} supports native debugging of @sc{djgpp} programs, and
23219 defines a few commands specific to the @sc{djgpp} port. This
23220 subsection describes those commands.
23225 This is a prefix of @sc{djgpp}-specific commands which print
23226 information about the target system and important OS structures.
23229 @cindex MS-DOS system info
23230 @cindex free memory information (MS-DOS)
23231 @item info dos sysinfo
23232 This command displays assorted information about the underlying
23233 platform: the CPU type and features, the OS version and flavor, the
23234 DPMI version, and the available conventional and DPMI memory.
23239 @cindex segment descriptor tables
23240 @cindex descriptor tables display
23242 @itemx info dos ldt
23243 @itemx info dos idt
23244 These 3 commands display entries from, respectively, Global, Local,
23245 and Interrupt Descriptor Tables (GDT, LDT, and IDT). The descriptor
23246 tables are data structures which store a descriptor for each segment
23247 that is currently in use. The segment's selector is an index into a
23248 descriptor table; the table entry for that index holds the
23249 descriptor's base address and limit, and its attributes and access
23252 A typical @sc{djgpp} program uses 3 segments: a code segment, a data
23253 segment (used for both data and the stack), and a DOS segment (which
23254 allows access to DOS/BIOS data structures and absolute addresses in
23255 conventional memory). However, the DPMI host will usually define
23256 additional segments in order to support the DPMI environment.
23258 @cindex garbled pointers
23259 These commands allow to display entries from the descriptor tables.
23260 Without an argument, all entries from the specified table are
23261 displayed. An argument, which should be an integer expression, means
23262 display a single entry whose index is given by the argument. For
23263 example, here's a convenient way to display information about the
23264 debugged program's data segment:
23267 @exdent @code{(@value{GDBP}) info dos ldt $ds}
23268 @exdent @code{0x13f: base=0x11970000 limit=0x0009ffff 32-Bit Data (Read/Write, Exp-up)}
23272 This comes in handy when you want to see whether a pointer is outside
23273 the data segment's limit (i.e.@: @dfn{garbled}).
23275 @cindex page tables display (MS-DOS)
23277 @itemx info dos pte
23278 These two commands display entries from, respectively, the Page
23279 Directory and the Page Tables. Page Directories and Page Tables are
23280 data structures which control how virtual memory addresses are mapped
23281 into physical addresses. A Page Table includes an entry for every
23282 page of memory that is mapped into the program's address space; there
23283 may be several Page Tables, each one holding up to 4096 entries. A
23284 Page Directory has up to 4096 entries, one each for every Page Table
23285 that is currently in use.
23287 Without an argument, @kbd{info dos pde} displays the entire Page
23288 Directory, and @kbd{info dos pte} displays all the entries in all of
23289 the Page Tables. An argument, an integer expression, given to the
23290 @kbd{info dos pde} command means display only that entry from the Page
23291 Directory table. An argument given to the @kbd{info dos pte} command
23292 means display entries from a single Page Table, the one pointed to by
23293 the specified entry in the Page Directory.
23295 @cindex direct memory access (DMA) on MS-DOS
23296 These commands are useful when your program uses @dfn{DMA} (Direct
23297 Memory Access), which needs physical addresses to program the DMA
23300 These commands are supported only with some DPMI servers.
23302 @cindex physical address from linear address
23303 @item info dos address-pte @var{addr}
23304 This command displays the Page Table entry for a specified linear
23305 address. The argument @var{addr} is a linear address which should
23306 already have the appropriate segment's base address added to it,
23307 because this command accepts addresses which may belong to @emph{any}
23308 segment. For example, here's how to display the Page Table entry for
23309 the page where a variable @code{i} is stored:
23312 @exdent @code{(@value{GDBP}) info dos address-pte __djgpp_base_address + (char *)&i}
23313 @exdent @code{Page Table entry for address 0x11a00d30:}
23314 @exdent @code{Base=0x02698000 Dirty Acc. Not-Cached Write-Back Usr Read-Write +0xd30}
23318 This says that @code{i} is stored at offset @code{0xd30} from the page
23319 whose physical base address is @code{0x02698000}, and shows all the
23320 attributes of that page.
23322 Note that you must cast the addresses of variables to a @code{char *},
23323 since otherwise the value of @code{__djgpp_base_address}, the base
23324 address of all variables and functions in a @sc{djgpp} program, will
23325 be added using the rules of C pointer arithmetics: if @code{i} is
23326 declared an @code{int}, @value{GDBN} will add 4 times the value of
23327 @code{__djgpp_base_address} to the address of @code{i}.
23329 Here's another example, it displays the Page Table entry for the
23333 @exdent @code{(@value{GDBP}) info dos address-pte *((unsigned *)&_go32_info_block + 3)}
23334 @exdent @code{Page Table entry for address 0x29110:}
23335 @exdent @code{Base=0x00029000 Dirty Acc. Not-Cached Write-Back Usr Read-Write +0x110}
23339 (The @code{+ 3} offset is because the transfer buffer's address is the
23340 3rd member of the @code{_go32_info_block} structure.) The output
23341 clearly shows that this DPMI server maps the addresses in conventional
23342 memory 1:1, i.e.@: the physical (@code{0x00029000} + @code{0x110}) and
23343 linear (@code{0x29110}) addresses are identical.
23345 This command is supported only with some DPMI servers.
23348 @cindex DOS serial data link, remote debugging
23349 In addition to native debugging, the DJGPP port supports remote
23350 debugging via a serial data link. The following commands are specific
23351 to remote serial debugging in the DJGPP port of @value{GDBN}.
23354 @kindex set com1base
23355 @kindex set com1irq
23356 @kindex set com2base
23357 @kindex set com2irq
23358 @kindex set com3base
23359 @kindex set com3irq
23360 @kindex set com4base
23361 @kindex set com4irq
23362 @item set com1base @var{addr}
23363 This command sets the base I/O port address of the @file{COM1} serial
23366 @item set com1irq @var{irq}
23367 This command sets the @dfn{Interrupt Request} (@code{IRQ}) line to use
23368 for the @file{COM1} serial port.
23370 There are similar commands @samp{set com2base}, @samp{set com3irq},
23371 etc.@: for setting the port address and the @code{IRQ} lines for the
23374 @kindex show com1base
23375 @kindex show com1irq
23376 @kindex show com2base
23377 @kindex show com2irq
23378 @kindex show com3base
23379 @kindex show com3irq
23380 @kindex show com4base
23381 @kindex show com4irq
23382 The related commands @samp{show com1base}, @samp{show com1irq} etc.@:
23383 display the current settings of the base address and the @code{IRQ}
23384 lines used by the COM ports.
23387 @kindex info serial
23388 @cindex DOS serial port status
23389 This command prints the status of the 4 DOS serial ports. For each
23390 port, it prints whether it's active or not, its I/O base address and
23391 IRQ number, whether it uses a 16550-style FIFO, its baudrate, and the
23392 counts of various errors encountered so far.
23396 @node Cygwin Native
23397 @subsection Features for Debugging MS Windows PE Executables
23398 @cindex MS Windows debugging
23399 @cindex native Cygwin debugging
23400 @cindex Cygwin-specific commands
23402 @value{GDBN} supports native debugging of MS Windows programs, including
23403 DLLs with and without symbolic debugging information.
23405 @cindex Ctrl-BREAK, MS-Windows
23406 @cindex interrupt debuggee on MS-Windows
23407 MS-Windows programs that call @code{SetConsoleMode} to switch off the
23408 special meaning of the @samp{Ctrl-C} keystroke cannot be interrupted
23409 by typing @kbd{C-c}. For this reason, @value{GDBN} on MS-Windows
23410 supports @kbd{C-@key{BREAK}} as an alternative interrupt key
23411 sequence, which can be used to interrupt the debuggee even if it
23414 There are various additional Cygwin-specific commands, described in
23415 this section. Working with DLLs that have no debugging symbols is
23416 described in @ref{Non-debug DLL Symbols}.
23421 This is a prefix of MS Windows-specific commands which print
23422 information about the target system and important OS structures.
23424 @item info w32 selector
23425 This command displays information returned by
23426 the Win32 API @code{GetThreadSelectorEntry} function.
23427 It takes an optional argument that is evaluated to
23428 a long value to give the information about this given selector.
23429 Without argument, this command displays information
23430 about the six segment registers.
23432 @item info w32 thread-information-block
23433 This command displays thread specific information stored in the
23434 Thread Information Block (readable on the X86 CPU family using @code{$fs}
23435 selector for 32-bit programs and @code{$gs} for 64-bit programs).
23437 @kindex signal-event
23438 @item signal-event @var{id}
23439 This command signals an event with user-provided @var{id}. Used to resume
23440 crashing process when attached to it using MS-Windows JIT debugging (AeDebug).
23442 To use it, create or edit the following keys in
23443 @code{HKLM\SOFTWARE\Microsoft\Windows NT\CurrentVersion\AeDebug} and/or
23444 @code{HKLM\SOFTWARE\Wow6432Node\Microsoft\Windows NT\CurrentVersion\AeDebug}
23445 (for x86_64 versions):
23449 @code{Debugger} (REG_SZ) --- a command to launch the debugger.
23450 Suggested command is: @code{@var{fully-qualified-path-to-gdb.exe} -ex
23451 "attach %ld" -ex "signal-event %ld" -ex "continue"}.
23453 The first @code{%ld} will be replaced by the process ID of the
23454 crashing process, the second @code{%ld} will be replaced by the ID of
23455 the event that blocks the crashing process, waiting for @value{GDBN}
23459 @code{Auto} (REG_SZ) --- either @code{1} or @code{0}. @code{1} will
23460 make the system run debugger specified by the Debugger key
23461 automatically, @code{0} will cause a dialog box with ``OK'' and
23462 ``Cancel'' buttons to appear, which allows the user to either
23463 terminate the crashing process (OK) or debug it (Cancel).
23466 @kindex set cygwin-exceptions
23467 @cindex debugging the Cygwin DLL
23468 @cindex Cygwin DLL, debugging
23469 @item set cygwin-exceptions @var{mode}
23470 If @var{mode} is @code{on}, @value{GDBN} will break on exceptions that
23471 happen inside the Cygwin DLL. If @var{mode} is @code{off},
23472 @value{GDBN} will delay recognition of exceptions, and may ignore some
23473 exceptions which seem to be caused by internal Cygwin DLL
23474 ``bookkeeping''. This option is meant primarily for debugging the
23475 Cygwin DLL itself; the default value is @code{off} to avoid annoying
23476 @value{GDBN} users with false @code{SIGSEGV} signals.
23478 @kindex show cygwin-exceptions
23479 @item show cygwin-exceptions
23480 Displays whether @value{GDBN} will break on exceptions that happen
23481 inside the Cygwin DLL itself.
23483 @kindex set new-console
23484 @item set new-console @var{mode}
23485 If @var{mode} is @code{on} the debuggee will
23486 be started in a new console on next start.
23487 If @var{mode} is @code{off}, the debuggee will
23488 be started in the same console as the debugger.
23490 @kindex show new-console
23491 @item show new-console
23492 Displays whether a new console is used
23493 when the debuggee is started.
23495 @kindex set new-group
23496 @item set new-group @var{mode}
23497 This boolean value controls whether the debuggee should
23498 start a new group or stay in the same group as the debugger.
23499 This affects the way the Windows OS handles
23502 @kindex show new-group
23503 @item show new-group
23504 Displays current value of new-group boolean.
23506 @kindex set debugevents
23507 @item set debugevents
23508 This boolean value adds debug output concerning kernel events related
23509 to the debuggee seen by the debugger. This includes events that
23510 signal thread and process creation and exit, DLL loading and
23511 unloading, console interrupts, and debugging messages produced by the
23512 Windows @code{OutputDebugString} API call.
23514 @kindex set debugexec
23515 @item set debugexec
23516 This boolean value adds debug output concerning execute events
23517 (such as resume thread) seen by the debugger.
23519 @kindex set debugexceptions
23520 @item set debugexceptions
23521 This boolean value adds debug output concerning exceptions in the
23522 debuggee seen by the debugger.
23524 @kindex set debugmemory
23525 @item set debugmemory
23526 This boolean value adds debug output concerning debuggee memory reads
23527 and writes by the debugger.
23531 This boolean values specifies whether the debuggee is called
23532 via a shell or directly (default value is on).
23536 Displays if the debuggee will be started with a shell.
23541 * Non-debug DLL Symbols:: Support for DLLs without debugging symbols
23544 @node Non-debug DLL Symbols
23545 @subsubsection Support for DLLs without Debugging Symbols
23546 @cindex DLLs with no debugging symbols
23547 @cindex Minimal symbols and DLLs
23549 Very often on windows, some of the DLLs that your program relies on do
23550 not include symbolic debugging information (for example,
23551 @file{kernel32.dll}). When @value{GDBN} doesn't recognize any debugging
23552 symbols in a DLL, it relies on the minimal amount of symbolic
23553 information contained in the DLL's export table. This section
23554 describes working with such symbols, known internally to @value{GDBN} as
23555 ``minimal symbols''.
23557 Note that before the debugged program has started execution, no DLLs
23558 will have been loaded. The easiest way around this problem is simply to
23559 start the program --- either by setting a breakpoint or letting the
23560 program run once to completion.
23562 @subsubsection DLL Name Prefixes
23564 In keeping with the naming conventions used by the Microsoft debugging
23565 tools, DLL export symbols are made available with a prefix based on the
23566 DLL name, for instance @code{KERNEL32!CreateFileA}. The plain name is
23567 also entered into the symbol table, so @code{CreateFileA} is often
23568 sufficient. In some cases there will be name clashes within a program
23569 (particularly if the executable itself includes full debugging symbols)
23570 necessitating the use of the fully qualified name when referring to the
23571 contents of the DLL. Use single-quotes around the name to avoid the
23572 exclamation mark (``!'') being interpreted as a language operator.
23574 Note that the internal name of the DLL may be all upper-case, even
23575 though the file name of the DLL is lower-case, or vice-versa. Since
23576 symbols within @value{GDBN} are @emph{case-sensitive} this may cause
23577 some confusion. If in doubt, try the @code{info functions} and
23578 @code{info variables} commands or even @code{maint print msymbols}
23579 (@pxref{Symbols}). Here's an example:
23582 (@value{GDBP}) info function CreateFileA
23583 All functions matching regular expression "CreateFileA":
23585 Non-debugging symbols:
23586 0x77e885f4 CreateFileA
23587 0x77e885f4 KERNEL32!CreateFileA
23591 (@value{GDBP}) info function !
23592 All functions matching regular expression "!":
23594 Non-debugging symbols:
23595 0x6100114c cygwin1!__assert
23596 0x61004034 cygwin1!_dll_crt0@@0
23597 0x61004240 cygwin1!dll_crt0(per_process *)
23601 @subsubsection Working with Minimal Symbols
23603 Symbols extracted from a DLL's export table do not contain very much
23604 type information. All that @value{GDBN} can do is guess whether a symbol
23605 refers to a function or variable depending on the linker section that
23606 contains the symbol. Also note that the actual contents of the memory
23607 contained in a DLL are not available unless the program is running. This
23608 means that you cannot examine the contents of a variable or disassemble
23609 a function within a DLL without a running program.
23611 Variables are generally treated as pointers and dereferenced
23612 automatically. For this reason, it is often necessary to prefix a
23613 variable name with the address-of operator (``&'') and provide explicit
23614 type information in the command. Here's an example of the type of
23618 (@value{GDBP}) print 'cygwin1!__argv'
23619 'cygwin1!__argv' has unknown type; cast it to its declared type
23623 (@value{GDBP}) x 'cygwin1!__argv'
23624 'cygwin1!__argv' has unknown type; cast it to its declared type
23627 And two possible solutions:
23630 (@value{GDBP}) print ((char **)'cygwin1!__argv')[0]
23631 $2 = 0x22fd98 "/cygdrive/c/mydirectory/myprogram"
23635 (@value{GDBP}) x/2x &'cygwin1!__argv'
23636 0x610c0aa8 <cygwin1!__argv>: 0x10021608 0x00000000
23637 (@value{GDBP}) x/x 0x10021608
23638 0x10021608: 0x0022fd98
23639 (@value{GDBP}) x/s 0x0022fd98
23640 0x22fd98: "/cygdrive/c/mydirectory/myprogram"
23643 Setting a break point within a DLL is possible even before the program
23644 starts execution. However, under these circumstances, @value{GDBN} can't
23645 examine the initial instructions of the function in order to skip the
23646 function's frame set-up code. You can work around this by using ``*&''
23647 to set the breakpoint at a raw memory address:
23650 (@value{GDBP}) break *&'python22!PyOS_Readline'
23651 Breakpoint 1 at 0x1e04eff0
23654 The author of these extensions is not entirely convinced that setting a
23655 break point within a shared DLL like @file{kernel32.dll} is completely
23659 @subsection Commands Specific to @sc{gnu} Hurd Systems
23660 @cindex @sc{gnu} Hurd debugging
23662 This subsection describes @value{GDBN} commands specific to the
23663 @sc{gnu} Hurd native debugging.
23668 @kindex set signals@r{, Hurd command}
23669 @kindex set sigs@r{, Hurd command}
23670 This command toggles the state of inferior signal interception by
23671 @value{GDBN}. Mach exceptions, such as breakpoint traps, are not
23672 affected by this command. @code{sigs} is a shorthand alias for
23677 @kindex show signals@r{, Hurd command}
23678 @kindex show sigs@r{, Hurd command}
23679 Show the current state of intercepting inferior's signals.
23681 @item set signal-thread
23682 @itemx set sigthread
23683 @kindex set signal-thread
23684 @kindex set sigthread
23685 This command tells @value{GDBN} which thread is the @code{libc} signal
23686 thread. That thread is run when a signal is delivered to a running
23687 process. @code{set sigthread} is the shorthand alias of @code{set
23690 @item show signal-thread
23691 @itemx show sigthread
23692 @kindex show signal-thread
23693 @kindex show sigthread
23694 These two commands show which thread will run when the inferior is
23695 delivered a signal.
23698 @kindex set stopped@r{, Hurd command}
23699 This commands tells @value{GDBN} that the inferior process is stopped,
23700 as with the @code{SIGSTOP} signal. The stopped process can be
23701 continued by delivering a signal to it.
23704 @kindex show stopped@r{, Hurd command}
23705 This command shows whether @value{GDBN} thinks the debuggee is
23708 @item set exceptions
23709 @kindex set exceptions@r{, Hurd command}
23710 Use this command to turn off trapping of exceptions in the inferior.
23711 When exception trapping is off, neither breakpoints nor
23712 single-stepping will work. To restore the default, set exception
23715 @item show exceptions
23716 @kindex show exceptions@r{, Hurd command}
23717 Show the current state of trapping exceptions in the inferior.
23719 @item set task pause
23720 @kindex set task@r{, Hurd commands}
23721 @cindex task attributes (@sc{gnu} Hurd)
23722 @cindex pause current task (@sc{gnu} Hurd)
23723 This command toggles task suspension when @value{GDBN} has control.
23724 Setting it to on takes effect immediately, and the task is suspended
23725 whenever @value{GDBN} gets control. Setting it to off will take
23726 effect the next time the inferior is continued. If this option is set
23727 to off, you can use @code{set thread default pause on} or @code{set
23728 thread pause on} (see below) to pause individual threads.
23730 @item show task pause
23731 @kindex show task@r{, Hurd commands}
23732 Show the current state of task suspension.
23734 @item set task detach-suspend-count
23735 @cindex task suspend count
23736 @cindex detach from task, @sc{gnu} Hurd
23737 This command sets the suspend count the task will be left with when
23738 @value{GDBN} detaches from it.
23740 @item show task detach-suspend-count
23741 Show the suspend count the task will be left with when detaching.
23743 @item set task exception-port
23744 @itemx set task excp
23745 @cindex task exception port, @sc{gnu} Hurd
23746 This command sets the task exception port to which @value{GDBN} will
23747 forward exceptions. The argument should be the value of the @dfn{send
23748 rights} of the task. @code{set task excp} is a shorthand alias.
23750 @item set noninvasive
23751 @cindex noninvasive task options
23752 This command switches @value{GDBN} to a mode that is the least
23753 invasive as far as interfering with the inferior is concerned. This
23754 is the same as using @code{set task pause}, @code{set exceptions}, and
23755 @code{set signals} to values opposite to the defaults.
23757 @item info send-rights
23758 @itemx info receive-rights
23759 @itemx info port-rights
23760 @itemx info port-sets
23761 @itemx info dead-names
23764 @cindex send rights, @sc{gnu} Hurd
23765 @cindex receive rights, @sc{gnu} Hurd
23766 @cindex port rights, @sc{gnu} Hurd
23767 @cindex port sets, @sc{gnu} Hurd
23768 @cindex dead names, @sc{gnu} Hurd
23769 These commands display information about, respectively, send rights,
23770 receive rights, port rights, port sets, and dead names of a task.
23771 There are also shorthand aliases: @code{info ports} for @code{info
23772 port-rights} and @code{info psets} for @code{info port-sets}.
23774 @item set thread pause
23775 @kindex set thread@r{, Hurd command}
23776 @cindex thread properties, @sc{gnu} Hurd
23777 @cindex pause current thread (@sc{gnu} Hurd)
23778 This command toggles current thread suspension when @value{GDBN} has
23779 control. Setting it to on takes effect immediately, and the current
23780 thread is suspended whenever @value{GDBN} gets control. Setting it to
23781 off will take effect the next time the inferior is continued.
23782 Normally, this command has no effect, since when @value{GDBN} has
23783 control, the whole task is suspended. However, if you used @code{set
23784 task pause off} (see above), this command comes in handy to suspend
23785 only the current thread.
23787 @item show thread pause
23788 @kindex show thread@r{, Hurd command}
23789 This command shows the state of current thread suspension.
23791 @item set thread run
23792 This command sets whether the current thread is allowed to run.
23794 @item show thread run
23795 Show whether the current thread is allowed to run.
23797 @item set thread detach-suspend-count
23798 @cindex thread suspend count, @sc{gnu} Hurd
23799 @cindex detach from thread, @sc{gnu} Hurd
23800 This command sets the suspend count @value{GDBN} will leave on a
23801 thread when detaching. This number is relative to the suspend count
23802 found by @value{GDBN} when it notices the thread; use @code{set thread
23803 takeover-suspend-count} to force it to an absolute value.
23805 @item show thread detach-suspend-count
23806 Show the suspend count @value{GDBN} will leave on the thread when
23809 @item set thread exception-port
23810 @itemx set thread excp
23811 Set the thread exception port to which to forward exceptions. This
23812 overrides the port set by @code{set task exception-port} (see above).
23813 @code{set thread excp} is the shorthand alias.
23815 @item set thread takeover-suspend-count
23816 Normally, @value{GDBN}'s thread suspend counts are relative to the
23817 value @value{GDBN} finds when it notices each thread. This command
23818 changes the suspend counts to be absolute instead.
23820 @item set thread default
23821 @itemx show thread default
23822 @cindex thread default settings, @sc{gnu} Hurd
23823 Each of the above @code{set thread} commands has a @code{set thread
23824 default} counterpart (e.g., @code{set thread default pause}, @code{set
23825 thread default exception-port}, etc.). The @code{thread default}
23826 variety of commands sets the default thread properties for all
23827 threads; you can then change the properties of individual threads with
23828 the non-default commands.
23835 @value{GDBN} provides the following commands specific to the Darwin target:
23838 @item set debug darwin @var{num}
23839 @kindex set debug darwin
23840 When set to a non zero value, enables debugging messages specific to
23841 the Darwin support. Higher values produce more verbose output.
23843 @item show debug darwin
23844 @kindex show debug darwin
23845 Show the current state of Darwin messages.
23847 @item set debug mach-o @var{num}
23848 @kindex set debug mach-o
23849 When set to a non zero value, enables debugging messages while
23850 @value{GDBN} is reading Darwin object files. (@dfn{Mach-O} is the
23851 file format used on Darwin for object and executable files.) Higher
23852 values produce more verbose output. This is a command to diagnose
23853 problems internal to @value{GDBN} and should not be needed in normal
23856 @item show debug mach-o
23857 @kindex show debug mach-o
23858 Show the current state of Mach-O file messages.
23860 @item set mach-exceptions on
23861 @itemx set mach-exceptions off
23862 @kindex set mach-exceptions
23863 On Darwin, faults are first reported as a Mach exception and are then
23864 mapped to a Posix signal. Use this command to turn on trapping of
23865 Mach exceptions in the inferior. This might be sometimes useful to
23866 better understand the cause of a fault. The default is off.
23868 @item show mach-exceptions
23869 @kindex show mach-exceptions
23870 Show the current state of exceptions trapping.
23874 @subsection FreeBSD
23877 When the ABI of a system call is changed in the FreeBSD kernel, this
23878 is implemented by leaving a compatibility system call using the old
23879 ABI at the existing number and allocating a new system call number for
23880 the version using the new ABI. As a convenience, when a system call
23881 is caught by name (@pxref{catch syscall}), compatibility system calls
23884 For example, FreeBSD 12 introduced a new variant of the @code{kevent}
23885 system call and catching the @code{kevent} system call by name catches
23889 (@value{GDBP}) catch syscall kevent
23890 Catchpoint 1 (syscalls 'freebsd11_kevent' [363] 'kevent' [560])
23896 @section Embedded Operating Systems
23898 This section describes configurations involving the debugging of
23899 embedded operating systems that are available for several different
23902 @value{GDBN} includes the ability to debug programs running on
23903 various real-time operating systems.
23905 @node Embedded Processors
23906 @section Embedded Processors
23908 This section goes into details specific to particular embedded
23911 @cindex send command to simulator
23912 Whenever a specific embedded processor has a simulator, @value{GDBN}
23913 allows to send an arbitrary command to the simulator.
23916 @item sim @var{command}
23917 @kindex sim@r{, a command}
23918 Send an arbitrary @var{command} string to the simulator. Consult the
23919 documentation for the specific simulator in use for information about
23920 acceptable commands.
23925 * ARC:: Synopsys ARC
23927 * M68K:: Motorola M68K
23928 * MicroBlaze:: Xilinx MicroBlaze
23929 * MIPS Embedded:: MIPS Embedded
23930 * OpenRISC 1000:: OpenRISC 1000 (or1k)
23931 * PowerPC Embedded:: PowerPC Embedded
23934 * Super-H:: Renesas Super-H
23938 @subsection Synopsys ARC
23939 @cindex Synopsys ARC
23940 @cindex ARC specific commands
23946 @value{GDBN} provides the following ARC-specific commands:
23949 @item set debug arc
23950 @kindex set debug arc
23951 Control the level of ARC specific debug messages. Use 0 for no messages (the
23952 default), 1 for debug messages, and 2 for even more debug messages.
23954 @item show debug arc
23955 @kindex show debug arc
23956 Show the level of ARC specific debugging in operation.
23958 @item maint print arc arc-instruction @var{address}
23959 @kindex maint print arc arc-instruction
23960 Print internal disassembler information about instruction at a given address.
23967 @value{GDBN} provides the following ARM-specific commands:
23970 @item set arm disassembler
23972 This commands selects from a list of disassembly styles. The
23973 @code{"std"} style is the standard style.
23975 @item show arm disassembler
23977 Show the current disassembly style.
23979 @item set arm apcs32
23980 @cindex ARM 32-bit mode
23981 This command toggles ARM operation mode between 32-bit and 26-bit.
23983 @item show arm apcs32
23984 Display the current usage of the ARM 32-bit mode.
23986 @item set arm fpu @var{fputype}
23987 This command sets the ARM floating-point unit (FPU) type. The
23988 argument @var{fputype} can be one of these:
23992 Determine the FPU type by querying the OS ABI.
23994 Software FPU, with mixed-endian doubles on little-endian ARM
23997 GCC-compiled FPA co-processor.
23999 Software FPU with pure-endian doubles.
24005 Show the current type of the FPU.
24008 This command forces @value{GDBN} to use the specified ABI.
24011 Show the currently used ABI.
24013 @item set arm fallback-mode (arm|thumb|auto)
24014 @value{GDBN} uses the symbol table, when available, to determine
24015 whether instructions are ARM or Thumb. This command controls
24016 @value{GDBN}'s default behavior when the symbol table is not
24017 available. The default is @samp{auto}, which causes @value{GDBN} to
24018 use the current execution mode (from the @code{T} bit in the @code{CPSR}
24021 @item show arm fallback-mode
24022 Show the current fallback instruction mode.
24024 @item set arm force-mode (arm|thumb|auto)
24025 This command overrides use of the symbol table to determine whether
24026 instructions are ARM or Thumb. The default is @samp{auto}, which
24027 causes @value{GDBN} to use the symbol table and then the setting
24028 of @samp{set arm fallback-mode}.
24030 @item show arm force-mode
24031 Show the current forced instruction mode.
24033 @item set debug arm
24034 Toggle whether to display ARM-specific debugging messages from the ARM
24035 target support subsystem.
24037 @item show debug arm
24038 Show whether ARM-specific debugging messages are enabled.
24042 @item target sim @r{[}@var{simargs}@r{]} @dots{}
24043 The @value{GDBN} ARM simulator accepts the following optional arguments.
24046 @item --swi-support=@var{type}
24047 Tell the simulator which SWI interfaces to support. The argument
24048 @var{type} may be a comma separated list of the following values.
24049 The default value is @code{all}.
24064 The Motorola m68k configuration includes ColdFire support.
24067 @subsection MicroBlaze
24068 @cindex Xilinx MicroBlaze
24069 @cindex XMD, Xilinx Microprocessor Debugger
24071 The MicroBlaze is a soft-core processor supported on various Xilinx
24072 FPGAs, such as Spartan or Virtex series. Boards with these processors
24073 usually have JTAG ports which connect to a host system running the Xilinx
24074 Embedded Development Kit (EDK) or Software Development Kit (SDK).
24075 This host system is used to download the configuration bitstream to
24076 the target FPGA. The Xilinx Microprocessor Debugger (XMD) program
24077 communicates with the target board using the JTAG interface and
24078 presents a @code{gdbserver} interface to the board. By default
24079 @code{xmd} uses port @code{1234}. (While it is possible to change
24080 this default port, it requires the use of undocumented @code{xmd}
24081 commands. Contact Xilinx support if you need to do this.)
24083 Use these GDB commands to connect to the MicroBlaze target processor.
24086 @item target remote :1234
24087 Use this command to connect to the target if you are running @value{GDBN}
24088 on the same system as @code{xmd}.
24090 @item target remote @var{xmd-host}:1234
24091 Use this command to connect to the target if it is connected to @code{xmd}
24092 running on a different system named @var{xmd-host}.
24095 Use this command to download a program to the MicroBlaze target.
24097 @item set debug microblaze @var{n}
24098 Enable MicroBlaze-specific debugging messages if non-zero.
24100 @item show debug microblaze @var{n}
24101 Show MicroBlaze-specific debugging level.
24104 @node MIPS Embedded
24105 @subsection @acronym{MIPS} Embedded
24108 @value{GDBN} supports these special commands for @acronym{MIPS} targets:
24111 @item set mipsfpu double
24112 @itemx set mipsfpu single
24113 @itemx set mipsfpu none
24114 @itemx set mipsfpu auto
24115 @itemx show mipsfpu
24116 @kindex set mipsfpu
24117 @kindex show mipsfpu
24118 @cindex @acronym{MIPS} remote floating point
24119 @cindex floating point, @acronym{MIPS} remote
24120 If your target board does not support the @acronym{MIPS} floating point
24121 coprocessor, you should use the command @samp{set mipsfpu none} (if you
24122 need this, you may wish to put the command in your @value{GDBN} init
24123 file). This tells @value{GDBN} how to find the return value of
24124 functions which return floating point values. It also allows
24125 @value{GDBN} to avoid saving the floating point registers when calling
24126 functions on the board. If you are using a floating point coprocessor
24127 with only single precision floating point support, as on the @sc{r4650}
24128 processor, use the command @samp{set mipsfpu single}. The default
24129 double precision floating point coprocessor may be selected using
24130 @samp{set mipsfpu double}.
24132 In previous versions the only choices were double precision or no
24133 floating point, so @samp{set mipsfpu on} will select double precision
24134 and @samp{set mipsfpu off} will select no floating point.
24136 As usual, you can inquire about the @code{mipsfpu} variable with
24137 @samp{show mipsfpu}.
24140 @node OpenRISC 1000
24141 @subsection OpenRISC 1000
24142 @cindex OpenRISC 1000
24145 The OpenRISC 1000 provides a free RISC instruction set architecture. It is
24146 mainly provided as a soft-core which can run on Xilinx, Altera and other
24149 @value{GDBN} for OpenRISC supports the below commands when connecting to
24157 Runs the builtin CPU simulator which can run very basic
24158 programs but does not support most hardware functions like MMU.
24159 For more complex use cases the user is advised to run an external
24160 target, and connect using @samp{target remote}.
24162 Example: @code{target sim}
24164 @item set debug or1k
24165 Toggle whether to display OpenRISC-specific debugging messages from the
24166 OpenRISC target support subsystem.
24168 @item show debug or1k
24169 Show whether OpenRISC-specific debugging messages are enabled.
24172 @node PowerPC Embedded
24173 @subsection PowerPC Embedded
24175 @cindex DVC register
24176 @value{GDBN} supports using the DVC (Data Value Compare) register to
24177 implement in hardware simple hardware watchpoint conditions of the form:
24180 (@value{GDBP}) watch @var{ADDRESS|VARIABLE} \
24181 if @var{ADDRESS|VARIABLE} == @var{CONSTANT EXPRESSION}
24184 The DVC register will be automatically used when @value{GDBN} detects
24185 such pattern in a condition expression, and the created watchpoint uses one
24186 debug register (either the @code{exact-watchpoints} option is on and the
24187 variable is scalar, or the variable has a length of one byte). This feature
24188 is available in native @value{GDBN} running on a Linux kernel version 2.6.34
24191 When running on PowerPC embedded processors, @value{GDBN} automatically uses
24192 ranged hardware watchpoints, unless the @code{exact-watchpoints} option is on,
24193 in which case watchpoints using only one debug register are created when
24194 watching variables of scalar types.
24196 You can create an artificial array to watch an arbitrary memory
24197 region using one of the following commands (@pxref{Expressions}):
24200 (@value{GDBP}) watch *((char *) @var{address})@@@var{length}
24201 (@value{GDBP}) watch @{char[@var{length}]@} @var{address}
24204 PowerPC embedded processors support masked watchpoints. See the discussion
24205 about the @code{mask} argument in @ref{Set Watchpoints}.
24207 @cindex ranged breakpoint
24208 PowerPC embedded processors support hardware accelerated
24209 @dfn{ranged breakpoints}. A ranged breakpoint stops execution of
24210 the inferior whenever it executes an instruction at any address within
24211 the range it specifies. To set a ranged breakpoint in @value{GDBN},
24212 use the @code{break-range} command.
24214 @value{GDBN} provides the following PowerPC-specific commands:
24217 @kindex break-range
24218 @item break-range @var{start-location}, @var{end-location}
24219 Set a breakpoint for an address range given by
24220 @var{start-location} and @var{end-location}, which can specify a function name,
24221 a line number, an offset of lines from the current line or from the start
24222 location, or an address of an instruction (see @ref{Specify Location},
24223 for a list of all the possible ways to specify a @var{location}.)
24224 The breakpoint will stop execution of the inferior whenever it
24225 executes an instruction at any address within the specified range,
24226 (including @var{start-location} and @var{end-location}.)
24228 @kindex set powerpc
24229 @item set powerpc soft-float
24230 @itemx show powerpc soft-float
24231 Force @value{GDBN} to use (or not use) a software floating point calling
24232 convention. By default, @value{GDBN} selects the calling convention based
24233 on the selected architecture and the provided executable file.
24235 @item set powerpc vector-abi
24236 @itemx show powerpc vector-abi
24237 Force @value{GDBN} to use the specified calling convention for vector
24238 arguments and return values. The valid options are @samp{auto};
24239 @samp{generic}, to avoid vector registers even if they are present;
24240 @samp{altivec}, to use AltiVec registers; and @samp{spe} to use SPE
24241 registers. By default, @value{GDBN} selects the calling convention
24242 based on the selected architecture and the provided executable file.
24244 @item set powerpc exact-watchpoints
24245 @itemx show powerpc exact-watchpoints
24246 Allow @value{GDBN} to use only one debug register when watching a variable
24247 of scalar type, thus assuming that the variable is accessed through the
24248 address of its first byte.
24253 @subsection Atmel AVR
24256 When configured for debugging the Atmel AVR, @value{GDBN} supports the
24257 following AVR-specific commands:
24260 @item info io_registers
24261 @kindex info io_registers@r{, AVR}
24262 @cindex I/O registers (Atmel AVR)
24263 This command displays information about the AVR I/O registers. For
24264 each register, @value{GDBN} prints its number and value.
24271 When configured for debugging CRIS, @value{GDBN} provides the
24272 following CRIS-specific commands:
24275 @item set cris-version @var{ver}
24276 @cindex CRIS version
24277 Set the current CRIS version to @var{ver}, either @samp{10} or @samp{32}.
24278 The CRIS version affects register names and sizes. This command is useful in
24279 case autodetection of the CRIS version fails.
24281 @item show cris-version
24282 Show the current CRIS version.
24284 @item set cris-dwarf2-cfi
24285 @cindex DWARF-2 CFI and CRIS
24286 Set the usage of DWARF-2 CFI for CRIS debugging. The default is @samp{on}.
24287 Change to @samp{off} when using @code{gcc-cris} whose version is below
24290 @item show cris-dwarf2-cfi
24291 Show the current state of using DWARF-2 CFI.
24293 @item set cris-mode @var{mode}
24295 Set the current CRIS mode to @var{mode}. It should only be changed when
24296 debugging in guru mode, in which case it should be set to
24297 @samp{guru} (the default is @samp{normal}).
24299 @item show cris-mode
24300 Show the current CRIS mode.
24304 @subsection Renesas Super-H
24307 For the Renesas Super-H processor, @value{GDBN} provides these
24311 @item set sh calling-convention @var{convention}
24312 @kindex set sh calling-convention
24313 Set the calling-convention used when calling functions from @value{GDBN}.
24314 Allowed values are @samp{gcc}, which is the default setting, and @samp{renesas}.
24315 With the @samp{gcc} setting, functions are called using the @value{NGCC} calling
24316 convention. If the DWARF-2 information of the called function specifies
24317 that the function follows the Renesas calling convention, the function
24318 is called using the Renesas calling convention. If the calling convention
24319 is set to @samp{renesas}, the Renesas calling convention is always used,
24320 regardless of the DWARF-2 information. This can be used to override the
24321 default of @samp{gcc} if debug information is missing, or the compiler
24322 does not emit the DWARF-2 calling convention entry for a function.
24324 @item show sh calling-convention
24325 @kindex show sh calling-convention
24326 Show the current calling convention setting.
24331 @node Architectures
24332 @section Architectures
24334 This section describes characteristics of architectures that affect
24335 all uses of @value{GDBN} with the architecture, both native and cross.
24342 * HPPA:: HP PA architecture
24343 * SPU:: Cell Broadband Engine SPU architecture
24351 @subsection AArch64
24352 @cindex AArch64 support
24354 When @value{GDBN} is debugging the AArch64 architecture, it provides the
24355 following special commands:
24358 @item set debug aarch64
24359 @kindex set debug aarch64
24360 This command determines whether AArch64 architecture-specific debugging
24361 messages are to be displayed.
24363 @item show debug aarch64
24364 Show whether AArch64 debugging messages are displayed.
24368 @subsubsection AArch64 SVE.
24369 @cindex AArch64 SVE.
24371 When @value{GDBN} is debugging the AArch64 architecture, if the Scalable Vector
24372 Extension (SVE) is present, then @value{GDBN} will provide the vector registers
24373 @code{$z0} through @code{$z31}, vector predicate registers @code{$p0} through
24374 @code{$p15}, and the @code{$ffr} register. In addition, the pseudo register
24375 @code{$vg} will be provided. This is the vector granule for the current thread
24376 and represents the number of 64-bit chunks in an SVE @code{z} register.
24378 If the vector length changes, then the @code{$vg} register will be updated,
24379 but the lengths of the @code{z} and @code{p} registers will not change. This
24380 is a known limitation of @value{GDBN} and does not affect the execution of the
24385 @subsection x86 Architecture-specific Issues
24388 @item set struct-convention @var{mode}
24389 @kindex set struct-convention
24390 @cindex struct return convention
24391 @cindex struct/union returned in registers
24392 Set the convention used by the inferior to return @code{struct}s and
24393 @code{union}s from functions to @var{mode}. Possible values of
24394 @var{mode} are @code{"pcc"}, @code{"reg"}, and @code{"default"} (the
24395 default). @code{"default"} or @code{"pcc"} means that @code{struct}s
24396 are returned on the stack, while @code{"reg"} means that a
24397 @code{struct} or a @code{union} whose size is 1, 2, 4, or 8 bytes will
24398 be returned in a register.
24400 @item show struct-convention
24401 @kindex show struct-convention
24402 Show the current setting of the convention to return @code{struct}s
24407 @subsubsection Intel @dfn{Memory Protection Extensions} (MPX).
24408 @cindex Intel Memory Protection Extensions (MPX).
24410 Memory Protection Extension (MPX) adds the bound registers @samp{BND0}
24411 @footnote{The register named with capital letters represent the architecture
24412 registers.} through @samp{BND3}. Bound registers store a pair of 64-bit values
24413 which are the lower bound and upper bound. Bounds are effective addresses or
24414 memory locations. The upper bounds are architecturally represented in 1's
24415 complement form. A bound having lower bound = 0, and upper bound = 0
24416 (1's complement of all bits set) will allow access to the entire address space.
24418 @samp{BND0} through @samp{BND3} are represented in @value{GDBN} as @samp{bnd0raw}
24419 through @samp{bnd3raw}. Pseudo registers @samp{bnd0} through @samp{bnd3}
24420 display the upper bound performing the complement of one operation on the
24421 upper bound value, i.e.@ when upper bound in @samp{bnd0raw} is 0 in the
24422 @value{GDBN} @samp{bnd0} it will be @code{0xfff@dots{}}. In this sense it
24423 can also be noted that the upper bounds are inclusive.
24425 As an example, assume that the register BND0 holds bounds for a pointer having
24426 access allowed for the range between 0x32 and 0x71. The values present on
24427 bnd0raw and bnd registers are presented as follows:
24430 bnd0raw = @{0x32, 0xffffffff8e@}
24431 bnd0 = @{lbound = 0x32, ubound = 0x71@} : size 64
24434 This way the raw value can be accessed via bnd0raw@dots{}bnd3raw. Any
24435 change on bnd0@dots{}bnd3 or bnd0raw@dots{}bnd3raw is reflect on its
24436 counterpart. When the bnd0@dots{}bnd3 registers are displayed via
24437 Python, the display includes the memory size, in bits, accessible to
24440 Bounds can also be stored in bounds tables, which are stored in
24441 application memory. These tables store bounds for pointers by specifying
24442 the bounds pointer's value along with its bounds. Evaluating and changing
24443 bounds located in bound tables is therefore interesting while investigating
24444 bugs on MPX context. @value{GDBN} provides commands for this purpose:
24447 @item show mpx bound @var{pointer}
24448 @kindex show mpx bound
24449 Display bounds of the given @var{pointer}.
24451 @item set mpx bound @var{pointer}, @var{lbound}, @var{ubound}
24452 @kindex set mpx bound
24453 Set the bounds of a pointer in the bound table.
24454 This command takes three parameters: @var{pointer} is the pointers
24455 whose bounds are to be changed, @var{lbound} and @var{ubound} are new values
24456 for lower and upper bounds respectively.
24459 When you call an inferior function on an Intel MPX enabled program,
24460 GDB sets the inferior's bound registers to the init (disabled) state
24461 before calling the function. As a consequence, bounds checks for the
24462 pointer arguments passed to the function will always pass.
24464 This is necessary because when you call an inferior function, the
24465 program is usually in the middle of the execution of other function.
24466 Since at that point bound registers are in an arbitrary state, not
24467 clearing them would lead to random bound violations in the called
24470 You can still examine the influence of the bound registers on the
24471 execution of the called function by stopping the execution of the
24472 called function at its prologue, setting bound registers, and
24473 continuing the execution. For example:
24477 Breakpoint 2 at 0x4009de: file i386-mpx-call.c, line 47.
24478 $ print upper (a, b, c, d, 1)
24479 Breakpoint 2, upper (a=0x0, b=0x6e0000005b, c=0x0, d=0x0, len=48)....
24481 @{lbound = 0x0, ubound = ffffffff@} : size -1
24484 At this last step the value of bnd0 can be changed for investigation of bound
24485 violations caused along the execution of the call. In order to know how to
24486 set the bound registers or bound table for the call consult the ABI.
24491 See the following section.
24494 @subsection @acronym{MIPS}
24496 @cindex stack on Alpha
24497 @cindex stack on @acronym{MIPS}
24498 @cindex Alpha stack
24499 @cindex @acronym{MIPS} stack
24500 Alpha- and @acronym{MIPS}-based computers use an unusual stack frame, which
24501 sometimes requires @value{GDBN} to search backward in the object code to
24502 find the beginning of a function.
24504 @cindex response time, @acronym{MIPS} debugging
24505 To improve response time (especially for embedded applications, where
24506 @value{GDBN} may be restricted to a slow serial line for this search)
24507 you may want to limit the size of this search, using one of these
24511 @cindex @code{heuristic-fence-post} (Alpha, @acronym{MIPS})
24512 @item set heuristic-fence-post @var{limit}
24513 Restrict @value{GDBN} to examining at most @var{limit} bytes in its
24514 search for the beginning of a function. A value of @var{0} (the
24515 default) means there is no limit. However, except for @var{0}, the
24516 larger the limit the more bytes @code{heuristic-fence-post} must search
24517 and therefore the longer it takes to run. You should only need to use
24518 this command when debugging a stripped executable.
24520 @item show heuristic-fence-post
24521 Display the current limit.
24525 These commands are available @emph{only} when @value{GDBN} is configured
24526 for debugging programs on Alpha or @acronym{MIPS} processors.
24528 Several @acronym{MIPS}-specific commands are available when debugging @acronym{MIPS}
24532 @item set mips abi @var{arg}
24533 @kindex set mips abi
24534 @cindex set ABI for @acronym{MIPS}
24535 Tell @value{GDBN} which @acronym{MIPS} ABI is used by the inferior. Possible
24536 values of @var{arg} are:
24540 The default ABI associated with the current binary (this is the
24550 @item show mips abi
24551 @kindex show mips abi
24552 Show the @acronym{MIPS} ABI used by @value{GDBN} to debug the inferior.
24554 @item set mips compression @var{arg}
24555 @kindex set mips compression
24556 @cindex code compression, @acronym{MIPS}
24557 Tell @value{GDBN} which @acronym{MIPS} compressed
24558 @acronym{ISA, Instruction Set Architecture} encoding is used by the
24559 inferior. @value{GDBN} uses this for code disassembly and other
24560 internal interpretation purposes. This setting is only referred to
24561 when no executable has been associated with the debugging session or
24562 the executable does not provide information about the encoding it uses.
24563 Otherwise this setting is automatically updated from information
24564 provided by the executable.
24566 Possible values of @var{arg} are @samp{mips16} and @samp{micromips}.
24567 The default compressed @acronym{ISA} encoding is @samp{mips16}, as
24568 executables containing @acronym{MIPS16} code frequently are not
24569 identified as such.
24571 This setting is ``sticky''; that is, it retains its value across
24572 debugging sessions until reset either explicitly with this command or
24573 implicitly from an executable.
24575 The compiler and/or assembler typically add symbol table annotations to
24576 identify functions compiled for the @acronym{MIPS16} or
24577 @acronym{microMIPS} @acronym{ISA}s. If these function-scope annotations
24578 are present, @value{GDBN} uses them in preference to the global
24579 compressed @acronym{ISA} encoding setting.
24581 @item show mips compression
24582 @kindex show mips compression
24583 Show the @acronym{MIPS} compressed @acronym{ISA} encoding used by
24584 @value{GDBN} to debug the inferior.
24587 @itemx show mipsfpu
24588 @xref{MIPS Embedded, set mipsfpu}.
24590 @item set mips mask-address @var{arg}
24591 @kindex set mips mask-address
24592 @cindex @acronym{MIPS} addresses, masking
24593 This command determines whether the most-significant 32 bits of 64-bit
24594 @acronym{MIPS} addresses are masked off. The argument @var{arg} can be
24595 @samp{on}, @samp{off}, or @samp{auto}. The latter is the default
24596 setting, which lets @value{GDBN} determine the correct value.
24598 @item show mips mask-address
24599 @kindex show mips mask-address
24600 Show whether the upper 32 bits of @acronym{MIPS} addresses are masked off or
24603 @item set remote-mips64-transfers-32bit-regs
24604 @kindex set remote-mips64-transfers-32bit-regs
24605 This command controls compatibility with 64-bit @acronym{MIPS} targets that
24606 transfer data in 32-bit quantities. If you have an old @acronym{MIPS} 64 target
24607 that transfers 32 bits for some registers, like @sc{sr} and @sc{fsr},
24608 and 64 bits for other registers, set this option to @samp{on}.
24610 @item show remote-mips64-transfers-32bit-regs
24611 @kindex show remote-mips64-transfers-32bit-regs
24612 Show the current setting of compatibility with older @acronym{MIPS} 64 targets.
24614 @item set debug mips
24615 @kindex set debug mips
24616 This command turns on and off debugging messages for the @acronym{MIPS}-specific
24617 target code in @value{GDBN}.
24619 @item show debug mips
24620 @kindex show debug mips
24621 Show the current setting of @acronym{MIPS} debugging messages.
24627 @cindex HPPA support
24629 When @value{GDBN} is debugging the HP PA architecture, it provides the
24630 following special commands:
24633 @item set debug hppa
24634 @kindex set debug hppa
24635 This command determines whether HPPA architecture-specific debugging
24636 messages are to be displayed.
24638 @item show debug hppa
24639 Show whether HPPA debugging messages are displayed.
24641 @item maint print unwind @var{address}
24642 @kindex maint print unwind@r{, HPPA}
24643 This command displays the contents of the unwind table entry at the
24644 given @var{address}.
24650 @subsection Cell Broadband Engine SPU architecture
24651 @cindex Cell Broadband Engine
24654 When @value{GDBN} is debugging the Cell Broadband Engine SPU architecture,
24655 it provides the following special commands:
24658 @item info spu event
24660 Display SPU event facility status. Shows current event mask
24661 and pending event status.
24663 @item info spu signal
24664 Display SPU signal notification facility status. Shows pending
24665 signal-control word and signal notification mode of both signal
24666 notification channels.
24668 @item info spu mailbox
24669 Display SPU mailbox facility status. Shows all pending entries,
24670 in order of processing, in each of the SPU Write Outbound,
24671 SPU Write Outbound Interrupt, and SPU Read Inbound mailboxes.
24674 Display MFC DMA status. Shows all pending commands in the MFC
24675 DMA queue. For each entry, opcode, tag, class IDs, effective
24676 and local store addresses and transfer size are shown.
24678 @item info spu proxydma
24679 Display MFC Proxy-DMA status. Shows all pending commands in the MFC
24680 Proxy-DMA queue. For each entry, opcode, tag, class IDs, effective
24681 and local store addresses and transfer size are shown.
24685 When @value{GDBN} is debugging a combined PowerPC/SPU application
24686 on the Cell Broadband Engine, it provides in addition the following
24690 @item set spu stop-on-load @var{arg}
24692 Set whether to stop for new SPE threads. When set to @code{on}, @value{GDBN}
24693 will give control to the user when a new SPE thread enters its @code{main}
24694 function. The default is @code{off}.
24696 @item show spu stop-on-load
24698 Show whether to stop for new SPE threads.
24700 @item set spu auto-flush-cache @var{arg}
24701 Set whether to automatically flush the software-managed cache. When set to
24702 @code{on}, @value{GDBN} will automatically cause the SPE software-managed
24703 cache to be flushed whenever SPE execution stops. This provides a consistent
24704 view of PowerPC memory that is accessed via the cache. If an application
24705 does not use the software-managed cache, this option has no effect.
24707 @item show spu auto-flush-cache
24708 Show whether to automatically flush the software-managed cache.
24713 @subsection PowerPC
24714 @cindex PowerPC architecture
24716 When @value{GDBN} is debugging the PowerPC architecture, it provides a set of
24717 pseudo-registers to enable inspection of 128-bit wide Decimal Floating Point
24718 numbers stored in the floating point registers. These values must be stored
24719 in two consecutive registers, always starting at an even register like
24720 @code{f0} or @code{f2}.
24722 The pseudo-registers go from @code{$dl0} through @code{$dl15}, and are formed
24723 by joining the even/odd register pairs @code{f0} and @code{f1} for @code{$dl0},
24724 @code{f2} and @code{f3} for @code{$dl1} and so on.
24726 For POWER7 processors, @value{GDBN} provides a set of pseudo-registers, the 64-bit
24727 wide Extended Floating Point Registers (@samp{f32} through @samp{f63}).
24730 @subsection Nios II
24731 @cindex Nios II architecture
24733 When @value{GDBN} is debugging the Nios II architecture,
24734 it provides the following special commands:
24738 @item set debug nios2
24739 @kindex set debug nios2
24740 This command turns on and off debugging messages for the Nios II
24741 target code in @value{GDBN}.
24743 @item show debug nios2
24744 @kindex show debug nios2
24745 Show the current setting of Nios II debugging messages.
24749 @subsection Sparc64
24750 @cindex Sparc64 support
24751 @cindex Application Data Integrity
24752 @subsubsection ADI Support
24754 The M7 processor supports an Application Data Integrity (ADI) feature that
24755 detects invalid data accesses. When software allocates memory and enables
24756 ADI on the allocated memory, it chooses a 4-bit version number, sets the
24757 version in the upper 4 bits of the 64-bit pointer to that data, and stores
24758 the 4-bit version in every cacheline of that data. Hardware saves the latter
24759 in spare bits in the cache and memory hierarchy. On each load and store,
24760 the processor compares the upper 4 VA (virtual address) bits to the
24761 cacheline's version. If there is a mismatch, the processor generates a
24762 version mismatch trap which can be either precise or disrupting. The trap
24763 is an error condition which the kernel delivers to the process as a SIGSEGV
24766 Note that only 64-bit applications can use ADI and need to be built with
24769 Values of the ADI version tags, which are in granularity of a
24770 cacheline (64 bytes), can be viewed or modified.
24774 @kindex adi examine
24775 @item adi (examine | x) [ / @var{n} ] @var{addr}
24777 The @code{adi examine} command displays the value of one ADI version tag per
24780 @var{n} is a decimal integer specifying the number in bytes; the default
24781 is 1. It specifies how much ADI version information, at the ratio of 1:ADI
24782 block size, to display.
24784 @var{addr} is the address in user address space where you want @value{GDBN}
24785 to begin displaying the ADI version tags.
24787 Below is an example of displaying ADI versions of variable "shmaddr".
24790 (@value{GDBP}) adi x/100 shmaddr
24791 0xfff800010002c000: 0 0
24795 @item adi (assign | a) [ / @var{n} ] @var{addr} = @var{tag}
24797 The @code{adi assign} command is used to assign new ADI version tag
24800 @var{n} is a decimal integer specifying the number in bytes;
24801 the default is 1. It specifies how much ADI version information, at the
24802 ratio of 1:ADI block size, to modify.
24804 @var{addr} is the address in user address space where you want @value{GDBN}
24805 to begin modifying the ADI version tags.
24807 @var{tag} is the new ADI version tag.
24809 For example, do the following to modify then verify ADI versions of
24810 variable "shmaddr":
24813 (@value{GDBP}) adi a/100 shmaddr = 7
24814 (@value{GDBP}) adi x/100 shmaddr
24815 0xfff800010002c000: 7 7
24822 @cindex S12Z support
24824 When @value{GDBN} is debugging the S12Z architecture,
24825 it provides the following special command:
24828 @item maint info bdccsr
24829 @kindex maint info bdccsr@r{, S12Z}
24830 This command displays the current value of the microprocessor's
24835 @node Controlling GDB
24836 @chapter Controlling @value{GDBN}
24838 You can alter the way @value{GDBN} interacts with you by using the
24839 @code{set} command. For commands controlling how @value{GDBN} displays
24840 data, see @ref{Print Settings, ,Print Settings}. Other settings are
24845 * Editing:: Command editing
24846 * Command History:: Command history
24847 * Screen Size:: Screen size
24848 * Output Styling:: Output styling
24849 * Numbers:: Numbers
24850 * ABI:: Configuring the current ABI
24851 * Auto-loading:: Automatically loading associated files
24852 * Messages/Warnings:: Optional warnings and messages
24853 * Debugging Output:: Optional messages about internal happenings
24854 * Other Misc Settings:: Other Miscellaneous Settings
24862 @value{GDBN} indicates its readiness to read a command by printing a string
24863 called the @dfn{prompt}. This string is normally @samp{(@value{GDBP})}. You
24864 can change the prompt string with the @code{set prompt} command. For
24865 instance, when debugging @value{GDBN} with @value{GDBN}, it is useful to change
24866 the prompt in one of the @value{GDBN} sessions so that you can always tell
24867 which one you are talking to.
24869 @emph{Note:} @code{set prompt} does not add a space for you after the
24870 prompt you set. This allows you to set a prompt which ends in a space
24871 or a prompt that does not.
24875 @item set prompt @var{newprompt}
24876 Directs @value{GDBN} to use @var{newprompt} as its prompt string henceforth.
24878 @kindex show prompt
24880 Prints a line of the form: @samp{Gdb's prompt is: @var{your-prompt}}
24883 Versions of @value{GDBN} that ship with Python scripting enabled have
24884 prompt extensions. The commands for interacting with these extensions
24888 @kindex set extended-prompt
24889 @item set extended-prompt @var{prompt}
24890 Set an extended prompt that allows for substitutions.
24891 @xref{gdb.prompt}, for a list of escape sequences that can be used for
24892 substitution. Any escape sequences specified as part of the prompt
24893 string are replaced with the corresponding strings each time the prompt
24899 set extended-prompt Current working directory: \w (gdb)
24902 Note that when an extended-prompt is set, it takes control of the
24903 @var{prompt_hook} hook. @xref{prompt_hook}, for further information.
24905 @kindex show extended-prompt
24906 @item show extended-prompt
24907 Prints the extended prompt. Any escape sequences specified as part of
24908 the prompt string with @code{set extended-prompt}, are replaced with the
24909 corresponding strings each time the prompt is displayed.
24913 @section Command Editing
24915 @cindex command line editing
24917 @value{GDBN} reads its input commands via the @dfn{Readline} interface. This
24918 @sc{gnu} library provides consistent behavior for programs which provide a
24919 command line interface to the user. Advantages are @sc{gnu} Emacs-style
24920 or @dfn{vi}-style inline editing of commands, @code{csh}-like history
24921 substitution, and a storage and recall of command history across
24922 debugging sessions.
24924 You may control the behavior of command line editing in @value{GDBN} with the
24925 command @code{set}.
24928 @kindex set editing
24931 @itemx set editing on
24932 Enable command line editing (enabled by default).
24934 @item set editing off
24935 Disable command line editing.
24937 @kindex show editing
24939 Show whether command line editing is enabled.
24942 @ifset SYSTEM_READLINE
24943 @xref{Command Line Editing, , , rluserman, GNU Readline Library},
24945 @ifclear SYSTEM_READLINE
24946 @xref{Command Line Editing},
24948 for more details about the Readline
24949 interface. Users unfamiliar with @sc{gnu} Emacs or @code{vi} are
24950 encouraged to read that chapter.
24952 @node Command History
24953 @section Command History
24954 @cindex command history
24956 @value{GDBN} can keep track of the commands you type during your
24957 debugging sessions, so that you can be certain of precisely what
24958 happened. Use these commands to manage the @value{GDBN} command
24961 @value{GDBN} uses the @sc{gnu} History library, a part of the Readline
24962 package, to provide the history facility.
24963 @ifset SYSTEM_READLINE
24964 @xref{Using History Interactively, , , history, GNU History Library},
24966 @ifclear SYSTEM_READLINE
24967 @xref{Using History Interactively},
24969 for the detailed description of the History library.
24971 To issue a command to @value{GDBN} without affecting certain aspects of
24972 the state which is seen by users, prefix it with @samp{server }
24973 (@pxref{Server Prefix}). This
24974 means that this command will not affect the command history, nor will it
24975 affect @value{GDBN}'s notion of which command to repeat if @key{RET} is
24976 pressed on a line by itself.
24978 @cindex @code{server}, command prefix
24979 The server prefix does not affect the recording of values into the value
24980 history; to print a value without recording it into the value history,
24981 use the @code{output} command instead of the @code{print} command.
24983 Here is the description of @value{GDBN} commands related to command
24987 @cindex history substitution
24988 @cindex history file
24989 @kindex set history filename
24990 @cindex @env{GDBHISTFILE}, environment variable
24991 @item set history filename @var{fname}
24992 Set the name of the @value{GDBN} command history file to @var{fname}.
24993 This is the file where @value{GDBN} reads an initial command history
24994 list, and where it writes the command history from this session when it
24995 exits. You can access this list through history expansion or through
24996 the history command editing characters listed below. This file defaults
24997 to the value of the environment variable @code{GDBHISTFILE}, or to
24998 @file{./.gdb_history} (@file{./_gdb_history} on MS-DOS) if this variable
25001 @cindex save command history
25002 @kindex set history save
25003 @item set history save
25004 @itemx set history save on
25005 Record command history in a file, whose name may be specified with the
25006 @code{set history filename} command. By default, this option is disabled.
25008 @item set history save off
25009 Stop recording command history in a file.
25011 @cindex history size
25012 @kindex set history size
25013 @cindex @env{GDBHISTSIZE}, environment variable
25014 @item set history size @var{size}
25015 @itemx set history size unlimited
25016 Set the number of commands which @value{GDBN} keeps in its history list.
25017 This defaults to the value of the environment variable @env{GDBHISTSIZE}, or
25018 to 256 if this variable is not set. Non-numeric values of @env{GDBHISTSIZE}
25019 are ignored. If @var{size} is @code{unlimited} or if @env{GDBHISTSIZE} is
25020 either a negative number or the empty string, then the number of commands
25021 @value{GDBN} keeps in the history list is unlimited.
25023 @cindex remove duplicate history
25024 @kindex set history remove-duplicates
25025 @item set history remove-duplicates @var{count}
25026 @itemx set history remove-duplicates unlimited
25027 Control the removal of duplicate history entries in the command history list.
25028 If @var{count} is non-zero, @value{GDBN} will look back at the last @var{count}
25029 history entries and remove the first entry that is a duplicate of the current
25030 entry being added to the command history list. If @var{count} is
25031 @code{unlimited} then this lookbehind is unbounded. If @var{count} is 0, then
25032 removal of duplicate history entries is disabled.
25034 Only history entries added during the current session are considered for
25035 removal. This option is set to 0 by default.
25039 History expansion assigns special meaning to the character @kbd{!}.
25040 @ifset SYSTEM_READLINE
25041 @xref{Event Designators, , , history, GNU History Library},
25043 @ifclear SYSTEM_READLINE
25044 @xref{Event Designators},
25048 @cindex history expansion, turn on/off
25049 Since @kbd{!} is also the logical not operator in C, history expansion
25050 is off by default. If you decide to enable history expansion with the
25051 @code{set history expansion on} command, you may sometimes need to
25052 follow @kbd{!} (when it is used as logical not, in an expression) with
25053 a space or a tab to prevent it from being expanded. The readline
25054 history facilities do not attempt substitution on the strings
25055 @kbd{!=} and @kbd{!(}, even when history expansion is enabled.
25057 The commands to control history expansion are:
25060 @item set history expansion on
25061 @itemx set history expansion
25062 @kindex set history expansion
25063 Enable history expansion. History expansion is off by default.
25065 @item set history expansion off
25066 Disable history expansion.
25069 @kindex show history
25071 @itemx show history filename
25072 @itemx show history save
25073 @itemx show history size
25074 @itemx show history expansion
25075 These commands display the state of the @value{GDBN} history parameters.
25076 @code{show history} by itself displays all four states.
25081 @kindex show commands
25082 @cindex show last commands
25083 @cindex display command history
25084 @item show commands
25085 Display the last ten commands in the command history.
25087 @item show commands @var{n}
25088 Print ten commands centered on command number @var{n}.
25090 @item show commands +
25091 Print ten commands just after the commands last printed.
25095 @section Screen Size
25096 @cindex size of screen
25097 @cindex screen size
25100 @cindex pauses in output
25102 Certain commands to @value{GDBN} may produce large amounts of
25103 information output to the screen. To help you read all of it,
25104 @value{GDBN} pauses and asks you for input at the end of each page of
25105 output. Type @key{RET} when you want to see one more page of output,
25106 @kbd{q} to discard the remaining output, or @kbd{c} to continue
25107 without paging for the rest of the current command. Also, the screen
25108 width setting determines when to wrap lines of output. Depending on
25109 what is being printed, @value{GDBN} tries to break the line at a
25110 readable place, rather than simply letting it overflow onto the
25113 Normally @value{GDBN} knows the size of the screen from the terminal
25114 driver software. For example, on Unix @value{GDBN} uses the termcap data base
25115 together with the value of the @code{TERM} environment variable and the
25116 @code{stty rows} and @code{stty cols} settings. If this is not correct,
25117 you can override it with the @code{set height} and @code{set
25124 @kindex show height
25125 @item set height @var{lpp}
25126 @itemx set height unlimited
25128 @itemx set width @var{cpl}
25129 @itemx set width unlimited
25131 These @code{set} commands specify a screen height of @var{lpp} lines and
25132 a screen width of @var{cpl} characters. The associated @code{show}
25133 commands display the current settings.
25135 If you specify a height of either @code{unlimited} or zero lines,
25136 @value{GDBN} does not pause during output no matter how long the
25137 output is. This is useful if output is to a file or to an editor
25140 Likewise, you can specify @samp{set width unlimited} or @samp{set
25141 width 0} to prevent @value{GDBN} from wrapping its output.
25143 @item set pagination on
25144 @itemx set pagination off
25145 @kindex set pagination
25146 Turn the output pagination on or off; the default is on. Turning
25147 pagination off is the alternative to @code{set height unlimited}. Note that
25148 running @value{GDBN} with the @option{--batch} option (@pxref{Mode
25149 Options, -batch}) also automatically disables pagination.
25151 @item show pagination
25152 @kindex show pagination
25153 Show the current pagination mode.
25156 @node Output Styling
25157 @section Output Styling
25163 @value{GDBN} can style its output on a capable terminal. This is
25164 enabled by default on most systems, but disabled by default when in
25165 batch mode (@pxref{Mode Options}). Various style settings are available;
25166 and styles can also be disabled entirely.
25169 @item set style enabled @samp{on|off}
25170 Enable or disable all styling. The default is host-dependent, with
25171 most hosts defaulting to @samp{on}.
25173 @item show style enabled
25174 Show the current state of styling.
25176 @item set style sources @samp{on|off}
25177 Enable or disable source code styling. This affects whether source
25178 code, such as the output of the @code{list} command, is styled. Note
25179 that source styling only works if styling in general is enabled, and
25180 if @value{GDBN} was linked with the GNU Source Highlight library. The
25181 default is @samp{on}.
25183 @item show style sources
25184 Show the current state of source code styling.
25187 Subcommands of @code{set style} control specific forms of styling.
25188 These subcommands all follow the same pattern: each style-able object
25189 can be styled with a foreground color, a background color, and an
25192 For example, the style of file names can be controlled using the
25193 @code{set style filename} group of commands:
25196 @item set style filename background @var{color}
25197 Set the background to @var{color}. Valid colors are @samp{none}
25198 (meaning the terminal's default color), @samp{black}, @samp{red},
25199 @samp{green}, @samp{yellow}, @samp{blue}, @samp{magenta}, @samp{cyan},
25202 @item set style filename foreground @var{color}
25203 Set the foreground to @var{color}. Valid colors are @samp{none}
25204 (meaning the terminal's default color), @samp{black}, @samp{red},
25205 @samp{green}, @samp{yellow}, @samp{blue}, @samp{magenta}, @samp{cyan},
25208 @item set style filename intensity @var{value}
25209 Set the intensity to @var{value}. Valid intensities are @samp{normal}
25210 (the default), @samp{bold}, and @samp{dim}.
25213 The @code{show style} command and its subcommands are styling
25214 a style name in their output using its own style.
25215 So, use @command{show style} to see the complete list of styles,
25216 their characteristics and the visual aspect of each style.
25218 The style-able objects are:
25221 Control the styling of file names. By default, this style's
25222 foreground color is green.
25225 Control the styling of function names. These are managed with the
25226 @code{set style function} family of commands. By default, this
25227 style's foreground color is yellow.
25230 Control the styling of variable names. These are managed with the
25231 @code{set style variable} family of commands. By default, this style's
25232 foreground color is cyan.
25235 Control the styling of addresses. These are managed with the
25236 @code{set style address} family of commands. By default, this style's
25237 foreground color is blue.
25240 Control the styling of titles. These are managed with the
25241 @code{set style title} family of commands. By default, this style's
25242 intensity is bold. Commands are using the title style to improve
25243 the readibility of large output. For example, the commands
25244 @command{apropos} and @command{help} are using the title style
25245 for the command names.
25248 Control the styling of highlightings. These are managed with the
25249 @code{set style highlight} family of commands. By default, this style's
25250 foreground color is red. Commands are using the highlight style to draw
25251 the user attention to some specific parts of their output. For example,
25252 the command @command{apropos -v REGEXP} uses the highlight style to
25253 mark the documentation parts matching @var{regexp}.
25259 @cindex number representation
25260 @cindex entering numbers
25262 You can always enter numbers in octal, decimal, or hexadecimal in
25263 @value{GDBN} by the usual conventions: octal numbers begin with
25264 @samp{0}, decimal numbers end with @samp{.}, and hexadecimal numbers
25265 begin with @samp{0x}. Numbers that neither begin with @samp{0} or
25266 @samp{0x}, nor end with a @samp{.} are, by default, entered in base
25267 10; likewise, the default display for numbers---when no particular
25268 format is specified---is base 10. You can change the default base for
25269 both input and output with the commands described below.
25272 @kindex set input-radix
25273 @item set input-radix @var{base}
25274 Set the default base for numeric input. Supported choices
25275 for @var{base} are decimal 8, 10, or 16. The base must itself be
25276 specified either unambiguously or using the current input radix; for
25280 set input-radix 012
25281 set input-radix 10.
25282 set input-radix 0xa
25286 sets the input base to decimal. On the other hand, @samp{set input-radix 10}
25287 leaves the input radix unchanged, no matter what it was, since
25288 @samp{10}, being without any leading or trailing signs of its base, is
25289 interpreted in the current radix. Thus, if the current radix is 16,
25290 @samp{10} is interpreted in hex, i.e.@: as 16 decimal, which doesn't
25293 @kindex set output-radix
25294 @item set output-radix @var{base}
25295 Set the default base for numeric display. Supported choices
25296 for @var{base} are decimal 8, 10, or 16. The base must itself be
25297 specified either unambiguously or using the current input radix.
25299 @kindex show input-radix
25300 @item show input-radix
25301 Display the current default base for numeric input.
25303 @kindex show output-radix
25304 @item show output-radix
25305 Display the current default base for numeric display.
25307 @item set radix @r{[}@var{base}@r{]}
25311 These commands set and show the default base for both input and output
25312 of numbers. @code{set radix} sets the radix of input and output to
25313 the same base; without an argument, it resets the radix back to its
25314 default value of 10.
25319 @section Configuring the Current ABI
25321 @value{GDBN} can determine the @dfn{ABI} (Application Binary Interface) of your
25322 application automatically. However, sometimes you need to override its
25323 conclusions. Use these commands to manage @value{GDBN}'s view of the
25329 @cindex Newlib OS ABI and its influence on the longjmp handling
25331 One @value{GDBN} configuration can debug binaries for multiple operating
25332 system targets, either via remote debugging or native emulation.
25333 @value{GDBN} will autodetect the @dfn{OS ABI} (Operating System ABI) in use,
25334 but you can override its conclusion using the @code{set osabi} command.
25335 One example where this is useful is in debugging of binaries which use
25336 an alternate C library (e.g.@: @sc{uClibc} for @sc{gnu}/Linux) which does
25337 not have the same identifying marks that the standard C library for your
25340 When @value{GDBN} is debugging the AArch64 architecture, it provides a
25341 ``Newlib'' OS ABI. This is useful for handling @code{setjmp} and
25342 @code{longjmp} when debugging binaries that use the @sc{newlib} C library.
25343 The ``Newlib'' OS ABI can be selected by @code{set osabi Newlib}.
25347 Show the OS ABI currently in use.
25350 With no argument, show the list of registered available OS ABI's.
25352 @item set osabi @var{abi}
25353 Set the current OS ABI to @var{abi}.
25356 @cindex float promotion
25358 Generally, the way that an argument of type @code{float} is passed to a
25359 function depends on whether the function is prototyped. For a prototyped
25360 (i.e.@: ANSI/ISO style) function, @code{float} arguments are passed unchanged,
25361 according to the architecture's convention for @code{float}. For unprototyped
25362 (i.e.@: K&R style) functions, @code{float} arguments are first promoted to type
25363 @code{double} and then passed.
25365 Unfortunately, some forms of debug information do not reliably indicate whether
25366 a function is prototyped. If @value{GDBN} calls a function that is not marked
25367 as prototyped, it consults @kbd{set coerce-float-to-double}.
25370 @kindex set coerce-float-to-double
25371 @item set coerce-float-to-double
25372 @itemx set coerce-float-to-double on
25373 Arguments of type @code{float} will be promoted to @code{double} when passed
25374 to an unprototyped function. This is the default setting.
25376 @item set coerce-float-to-double off
25377 Arguments of type @code{float} will be passed directly to unprototyped
25380 @kindex show coerce-float-to-double
25381 @item show coerce-float-to-double
25382 Show the current setting of promoting @code{float} to @code{double}.
25386 @kindex show cp-abi
25387 @value{GDBN} needs to know the ABI used for your program's C@t{++}
25388 objects. The correct C@t{++} ABI depends on which C@t{++} compiler was
25389 used to build your application. @value{GDBN} only fully supports
25390 programs with a single C@t{++} ABI; if your program contains code using
25391 multiple C@t{++} ABI's or if @value{GDBN} can not identify your
25392 program's ABI correctly, you can tell @value{GDBN} which ABI to use.
25393 Currently supported ABI's include ``gnu-v2'', for @code{g++} versions
25394 before 3.0, ``gnu-v3'', for @code{g++} versions 3.0 and later, and
25395 ``hpaCC'' for the HP ANSI C@t{++} compiler. Other C@t{++} compilers may
25396 use the ``gnu-v2'' or ``gnu-v3'' ABI's as well. The default setting is
25401 Show the C@t{++} ABI currently in use.
25404 With no argument, show the list of supported C@t{++} ABI's.
25406 @item set cp-abi @var{abi}
25407 @itemx set cp-abi auto
25408 Set the current C@t{++} ABI to @var{abi}, or return to automatic detection.
25412 @section Automatically loading associated files
25413 @cindex auto-loading
25415 @value{GDBN} sometimes reads files with commands and settings automatically,
25416 without being explicitly told so by the user. We call this feature
25417 @dfn{auto-loading}. While auto-loading is useful for automatically adapting
25418 @value{GDBN} to the needs of your project, it can sometimes produce unexpected
25419 results or introduce security risks (e.g., if the file comes from untrusted
25423 * Init File in the Current Directory:: @samp{set/show/info auto-load local-gdbinit}
25424 * libthread_db.so.1 file:: @samp{set/show/info auto-load libthread-db}
25426 * Auto-loading safe path:: @samp{set/show/info auto-load safe-path}
25427 * Auto-loading verbose mode:: @samp{set/show debug auto-load}
25430 There are various kinds of files @value{GDBN} can automatically load.
25431 In addition to these files, @value{GDBN} supports auto-loading code written
25432 in various extension languages. @xref{Auto-loading extensions}.
25434 Note that loading of these associated files (including the local @file{.gdbinit}
25435 file) requires accordingly configured @code{auto-load safe-path}
25436 (@pxref{Auto-loading safe path}).
25438 For these reasons, @value{GDBN} includes commands and options to let you
25439 control when to auto-load files and which files should be auto-loaded.
25442 @anchor{set auto-load off}
25443 @kindex set auto-load off
25444 @item set auto-load off
25445 Globally disable loading of all auto-loaded files.
25446 You may want to use this command with the @samp{-iex} option
25447 (@pxref{Option -init-eval-command}) such as:
25449 $ @kbd{gdb -iex "set auto-load off" untrusted-executable corefile}
25452 Be aware that system init file (@pxref{System-wide configuration})
25453 and init files from your home directory (@pxref{Home Directory Init File})
25454 still get read (as they come from generally trusted directories).
25455 To prevent @value{GDBN} from auto-loading even those init files, use the
25456 @option{-nx} option (@pxref{Mode Options}), in addition to
25457 @code{set auto-load no}.
25459 @anchor{show auto-load}
25460 @kindex show auto-load
25461 @item show auto-load
25462 Show whether auto-loading of each specific @samp{auto-load} file(s) is enabled
25466 (gdb) show auto-load
25467 gdb-scripts: Auto-loading of canned sequences of commands scripts is on.
25468 libthread-db: Auto-loading of inferior specific libthread_db is on.
25469 local-gdbinit: Auto-loading of .gdbinit script from current directory
25471 python-scripts: Auto-loading of Python scripts is on.
25472 safe-path: List of directories from which it is safe to auto-load files
25473 is $debugdir:$datadir/auto-load.
25474 scripts-directory: List of directories from which to load auto-loaded scripts
25475 is $debugdir:$datadir/auto-load.
25478 @anchor{info auto-load}
25479 @kindex info auto-load
25480 @item info auto-load
25481 Print whether each specific @samp{auto-load} file(s) have been auto-loaded or
25485 (gdb) info auto-load
25488 Yes /home/user/gdb/gdb-gdb.gdb
25489 libthread-db: No auto-loaded libthread-db.
25490 local-gdbinit: Local .gdbinit file "/home/user/gdb/.gdbinit" has been
25494 Yes /home/user/gdb/gdb-gdb.py
25498 These are @value{GDBN} control commands for the auto-loading:
25500 @multitable @columnfractions .5 .5
25501 @item @xref{set auto-load off}.
25502 @tab Disable auto-loading globally.
25503 @item @xref{show auto-load}.
25504 @tab Show setting of all kinds of files.
25505 @item @xref{info auto-load}.
25506 @tab Show state of all kinds of files.
25507 @item @xref{set auto-load gdb-scripts}.
25508 @tab Control for @value{GDBN} command scripts.
25509 @item @xref{show auto-load gdb-scripts}.
25510 @tab Show setting of @value{GDBN} command scripts.
25511 @item @xref{info auto-load gdb-scripts}.
25512 @tab Show state of @value{GDBN} command scripts.
25513 @item @xref{set auto-load python-scripts}.
25514 @tab Control for @value{GDBN} Python scripts.
25515 @item @xref{show auto-load python-scripts}.
25516 @tab Show setting of @value{GDBN} Python scripts.
25517 @item @xref{info auto-load python-scripts}.
25518 @tab Show state of @value{GDBN} Python scripts.
25519 @item @xref{set auto-load guile-scripts}.
25520 @tab Control for @value{GDBN} Guile scripts.
25521 @item @xref{show auto-load guile-scripts}.
25522 @tab Show setting of @value{GDBN} Guile scripts.
25523 @item @xref{info auto-load guile-scripts}.
25524 @tab Show state of @value{GDBN} Guile scripts.
25525 @item @xref{set auto-load scripts-directory}.
25526 @tab Control for @value{GDBN} auto-loaded scripts location.
25527 @item @xref{show auto-load scripts-directory}.
25528 @tab Show @value{GDBN} auto-loaded scripts location.
25529 @item @xref{add-auto-load-scripts-directory}.
25530 @tab Add directory for auto-loaded scripts location list.
25531 @item @xref{set auto-load local-gdbinit}.
25532 @tab Control for init file in the current directory.
25533 @item @xref{show auto-load local-gdbinit}.
25534 @tab Show setting of init file in the current directory.
25535 @item @xref{info auto-load local-gdbinit}.
25536 @tab Show state of init file in the current directory.
25537 @item @xref{set auto-load libthread-db}.
25538 @tab Control for thread debugging library.
25539 @item @xref{show auto-load libthread-db}.
25540 @tab Show setting of thread debugging library.
25541 @item @xref{info auto-load libthread-db}.
25542 @tab Show state of thread debugging library.
25543 @item @xref{set auto-load safe-path}.
25544 @tab Control directories trusted for automatic loading.
25545 @item @xref{show auto-load safe-path}.
25546 @tab Show directories trusted for automatic loading.
25547 @item @xref{add-auto-load-safe-path}.
25548 @tab Add directory trusted for automatic loading.
25551 @node Init File in the Current Directory
25552 @subsection Automatically loading init file in the current directory
25553 @cindex auto-loading init file in the current directory
25555 By default, @value{GDBN} reads and executes the canned sequences of commands
25556 from init file (if any) in the current working directory,
25557 see @ref{Init File in the Current Directory during Startup}.
25559 Note that loading of this local @file{.gdbinit} file also requires accordingly
25560 configured @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
25563 @anchor{set auto-load local-gdbinit}
25564 @kindex set auto-load local-gdbinit
25565 @item set auto-load local-gdbinit [on|off]
25566 Enable or disable the auto-loading of canned sequences of commands
25567 (@pxref{Sequences}) found in init file in the current directory.
25569 @anchor{show auto-load local-gdbinit}
25570 @kindex show auto-load local-gdbinit
25571 @item show auto-load local-gdbinit
25572 Show whether auto-loading of canned sequences of commands from init file in the
25573 current directory is enabled or disabled.
25575 @anchor{info auto-load local-gdbinit}
25576 @kindex info auto-load local-gdbinit
25577 @item info auto-load local-gdbinit
25578 Print whether canned sequences of commands from init file in the
25579 current directory have been auto-loaded.
25582 @node libthread_db.so.1 file
25583 @subsection Automatically loading thread debugging library
25584 @cindex auto-loading libthread_db.so.1
25586 This feature is currently present only on @sc{gnu}/Linux native hosts.
25588 @value{GDBN} reads in some cases thread debugging library from places specific
25589 to the inferior (@pxref{set libthread-db-search-path}).
25591 The special @samp{libthread-db-search-path} entry @samp{$sdir} is processed
25592 without checking this @samp{set auto-load libthread-db} switch as system
25593 libraries have to be trusted in general. In all other cases of
25594 @samp{libthread-db-search-path} entries @value{GDBN} checks first if @samp{set
25595 auto-load libthread-db} is enabled before trying to open such thread debugging
25598 Note that loading of this debugging library also requires accordingly configured
25599 @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
25602 @anchor{set auto-load libthread-db}
25603 @kindex set auto-load libthread-db
25604 @item set auto-load libthread-db [on|off]
25605 Enable or disable the auto-loading of inferior specific thread debugging library.
25607 @anchor{show auto-load libthread-db}
25608 @kindex show auto-load libthread-db
25609 @item show auto-load libthread-db
25610 Show whether auto-loading of inferior specific thread debugging library is
25611 enabled or disabled.
25613 @anchor{info auto-load libthread-db}
25614 @kindex info auto-load libthread-db
25615 @item info auto-load libthread-db
25616 Print the list of all loaded inferior specific thread debugging libraries and
25617 for each such library print list of inferior @var{pid}s using it.
25620 @node Auto-loading safe path
25621 @subsection Security restriction for auto-loading
25622 @cindex auto-loading safe-path
25624 As the files of inferior can come from untrusted source (such as submitted by
25625 an application user) @value{GDBN} does not always load any files automatically.
25626 @value{GDBN} provides the @samp{set auto-load safe-path} setting to list
25627 directories trusted for loading files not explicitly requested by user.
25628 Each directory can also be a shell wildcard pattern.
25630 If the path is not set properly you will see a warning and the file will not
25635 Reading symbols from /home/user/gdb/gdb...done.
25636 warning: File "/home/user/gdb/gdb-gdb.gdb" auto-loading has been
25637 declined by your `auto-load safe-path' set
25638 to "$debugdir:$datadir/auto-load".
25639 warning: File "/home/user/gdb/gdb-gdb.py" auto-loading has been
25640 declined by your `auto-load safe-path' set
25641 to "$debugdir:$datadir/auto-load".
25645 To instruct @value{GDBN} to go ahead and use the init files anyway,
25646 invoke @value{GDBN} like this:
25649 $ gdb -q -iex "set auto-load safe-path /home/user/gdb" ./gdb
25652 The list of trusted directories is controlled by the following commands:
25655 @anchor{set auto-load safe-path}
25656 @kindex set auto-load safe-path
25657 @item set auto-load safe-path @r{[}@var{directories}@r{]}
25658 Set the list of directories (and their subdirectories) trusted for automatic
25659 loading and execution of scripts. You can also enter a specific trusted file.
25660 Each directory can also be a shell wildcard pattern; wildcards do not match
25661 directory separator - see @code{FNM_PATHNAME} for system function @code{fnmatch}
25662 (@pxref{Wildcard Matching, fnmatch, , libc, GNU C Library Reference Manual}).
25663 If you omit @var{directories}, @samp{auto-load safe-path} will be reset to
25664 its default value as specified during @value{GDBN} compilation.
25666 The list of directories uses path separator (@samp{:} on GNU and Unix
25667 systems, @samp{;} on MS-Windows and MS-DOS) to separate directories, similarly
25668 to the @env{PATH} environment variable.
25670 @anchor{show auto-load safe-path}
25671 @kindex show auto-load safe-path
25672 @item show auto-load safe-path
25673 Show the list of directories trusted for automatic loading and execution of
25676 @anchor{add-auto-load-safe-path}
25677 @kindex add-auto-load-safe-path
25678 @item add-auto-load-safe-path
25679 Add an entry (or list of entries) to the list of directories trusted for
25680 automatic loading and execution of scripts. Multiple entries may be delimited
25681 by the host platform path separator in use.
25684 This variable defaults to what @code{--with-auto-load-dir} has been configured
25685 to (@pxref{with-auto-load-dir}). @file{$debugdir} and @file{$datadir}
25686 substitution applies the same as for @ref{set auto-load scripts-directory}.
25687 The default @code{set auto-load safe-path} value can be also overriden by
25688 @value{GDBN} configuration option @option{--with-auto-load-safe-path}.
25690 Setting this variable to @file{/} disables this security protection,
25691 corresponding @value{GDBN} configuration option is
25692 @option{--without-auto-load-safe-path}.
25693 This variable is supposed to be set to the system directories writable by the
25694 system superuser only. Users can add their source directories in init files in
25695 their home directories (@pxref{Home Directory Init File}). See also deprecated
25696 init file in the current directory
25697 (@pxref{Init File in the Current Directory during Startup}).
25699 To force @value{GDBN} to load the files it declined to load in the previous
25700 example, you could use one of the following ways:
25703 @item @file{~/.gdbinit}: @samp{add-auto-load-safe-path ~/src/gdb}
25704 Specify this trusted directory (or a file) as additional component of the list.
25705 You have to specify also any existing directories displayed by
25706 by @samp{show auto-load safe-path} (such as @samp{/usr:/bin} in this example).
25708 @item @kbd{gdb -iex "set auto-load safe-path /usr:/bin:~/src/gdb" @dots{}}
25709 Specify this directory as in the previous case but just for a single
25710 @value{GDBN} session.
25712 @item @kbd{gdb -iex "set auto-load safe-path /" @dots{}}
25713 Disable auto-loading safety for a single @value{GDBN} session.
25714 This assumes all the files you debug during this @value{GDBN} session will come
25715 from trusted sources.
25717 @item @kbd{./configure --without-auto-load-safe-path}
25718 During compilation of @value{GDBN} you may disable any auto-loading safety.
25719 This assumes all the files you will ever debug with this @value{GDBN} come from
25723 On the other hand you can also explicitly forbid automatic files loading which
25724 also suppresses any such warning messages:
25727 @item @kbd{gdb -iex "set auto-load no" @dots{}}
25728 You can use @value{GDBN} command-line option for a single @value{GDBN} session.
25730 @item @file{~/.gdbinit}: @samp{set auto-load no}
25731 Disable auto-loading globally for the user
25732 (@pxref{Home Directory Init File}). While it is improbable, you could also
25733 use system init file instead (@pxref{System-wide configuration}).
25736 This setting applies to the file names as entered by user. If no entry matches
25737 @value{GDBN} tries as a last resort to also resolve all the file names into
25738 their canonical form (typically resolving symbolic links) and compare the
25739 entries again. @value{GDBN} already canonicalizes most of the filenames on its
25740 own before starting the comparison so a canonical form of directories is
25741 recommended to be entered.
25743 @node Auto-loading verbose mode
25744 @subsection Displaying files tried for auto-load
25745 @cindex auto-loading verbose mode
25747 For better visibility of all the file locations where you can place scripts to
25748 be auto-loaded with inferior --- or to protect yourself against accidental
25749 execution of untrusted scripts --- @value{GDBN} provides a feature for printing
25750 all the files attempted to be loaded. Both existing and non-existing files may
25753 For example the list of directories from which it is safe to auto-load files
25754 (@pxref{Auto-loading safe path}) applies also to canonicalized filenames which
25755 may not be too obvious while setting it up.
25758 (gdb) set debug auto-load on
25759 (gdb) file ~/src/t/true
25760 auto-load: Loading canned sequences of commands script "/tmp/true-gdb.gdb"
25761 for objfile "/tmp/true".
25762 auto-load: Updating directories of "/usr:/opt".
25763 auto-load: Using directory "/usr".
25764 auto-load: Using directory "/opt".
25765 warning: File "/tmp/true-gdb.gdb" auto-loading has been declined
25766 by your `auto-load safe-path' set to "/usr:/opt".
25770 @anchor{set debug auto-load}
25771 @kindex set debug auto-load
25772 @item set debug auto-load [on|off]
25773 Set whether to print the filenames attempted to be auto-loaded.
25775 @anchor{show debug auto-load}
25776 @kindex show debug auto-load
25777 @item show debug auto-load
25778 Show whether printing of the filenames attempted to be auto-loaded is turned
25782 @node Messages/Warnings
25783 @section Optional Warnings and Messages
25785 @cindex verbose operation
25786 @cindex optional warnings
25787 By default, @value{GDBN} is silent about its inner workings. If you are
25788 running on a slow machine, you may want to use the @code{set verbose}
25789 command. This makes @value{GDBN} tell you when it does a lengthy
25790 internal operation, so you will not think it has crashed.
25792 Currently, the messages controlled by @code{set verbose} are those
25793 which announce that the symbol table for a source file is being read;
25794 see @code{symbol-file} in @ref{Files, ,Commands to Specify Files}.
25797 @kindex set verbose
25798 @item set verbose on
25799 Enables @value{GDBN} output of certain informational messages.
25801 @item set verbose off
25802 Disables @value{GDBN} output of certain informational messages.
25804 @kindex show verbose
25806 Displays whether @code{set verbose} is on or off.
25809 By default, if @value{GDBN} encounters bugs in the symbol table of an
25810 object file, it is silent; but if you are debugging a compiler, you may
25811 find this information useful (@pxref{Symbol Errors, ,Errors Reading
25816 @kindex set complaints
25817 @item set complaints @var{limit}
25818 Permits @value{GDBN} to output @var{limit} complaints about each type of
25819 unusual symbols before becoming silent about the problem. Set
25820 @var{limit} to zero to suppress all complaints; set it to a large number
25821 to prevent complaints from being suppressed.
25823 @kindex show complaints
25824 @item show complaints
25825 Displays how many symbol complaints @value{GDBN} is permitted to produce.
25829 @anchor{confirmation requests}
25830 By default, @value{GDBN} is cautious, and asks what sometimes seems to be a
25831 lot of stupid questions to confirm certain commands. For example, if
25832 you try to run a program which is already running:
25836 The program being debugged has been started already.
25837 Start it from the beginning? (y or n)
25840 If you are willing to unflinchingly face the consequences of your own
25841 commands, you can disable this ``feature'':
25845 @kindex set confirm
25847 @cindex confirmation
25848 @cindex stupid questions
25849 @item set confirm off
25850 Disables confirmation requests. Note that running @value{GDBN} with
25851 the @option{--batch} option (@pxref{Mode Options, -batch}) also
25852 automatically disables confirmation requests.
25854 @item set confirm on
25855 Enables confirmation requests (the default).
25857 @kindex show confirm
25859 Displays state of confirmation requests.
25863 @cindex command tracing
25864 If you need to debug user-defined commands or sourced files you may find it
25865 useful to enable @dfn{command tracing}. In this mode each command will be
25866 printed as it is executed, prefixed with one or more @samp{+} symbols, the
25867 quantity denoting the call depth of each command.
25870 @kindex set trace-commands
25871 @cindex command scripts, debugging
25872 @item set trace-commands on
25873 Enable command tracing.
25874 @item set trace-commands off
25875 Disable command tracing.
25876 @item show trace-commands
25877 Display the current state of command tracing.
25880 @node Debugging Output
25881 @section Optional Messages about Internal Happenings
25882 @cindex optional debugging messages
25884 @value{GDBN} has commands that enable optional debugging messages from
25885 various @value{GDBN} subsystems; normally these commands are of
25886 interest to @value{GDBN} maintainers, or when reporting a bug. This
25887 section documents those commands.
25890 @kindex set exec-done-display
25891 @item set exec-done-display
25892 Turns on or off the notification of asynchronous commands'
25893 completion. When on, @value{GDBN} will print a message when an
25894 asynchronous command finishes its execution. The default is off.
25895 @kindex show exec-done-display
25896 @item show exec-done-display
25897 Displays the current setting of asynchronous command completion
25900 @cindex ARM AArch64
25901 @item set debug aarch64
25902 Turns on or off display of debugging messages related to ARM AArch64.
25903 The default is off.
25905 @item show debug aarch64
25906 Displays the current state of displaying debugging messages related to
25908 @cindex gdbarch debugging info
25909 @cindex architecture debugging info
25910 @item set debug arch
25911 Turns on or off display of gdbarch debugging info. The default is off
25912 @item show debug arch
25913 Displays the current state of displaying gdbarch debugging info.
25914 @item set debug aix-solib
25915 @cindex AIX shared library debugging
25916 Control display of debugging messages from the AIX shared library
25917 support module. The default is off.
25918 @item show debug aix-thread
25919 Show the current state of displaying AIX shared library debugging messages.
25920 @item set debug aix-thread
25921 @cindex AIX threads
25922 Display debugging messages about inner workings of the AIX thread
25924 @item show debug aix-thread
25925 Show the current state of AIX thread debugging info display.
25926 @item set debug check-physname
25928 Check the results of the ``physname'' computation. When reading DWARF
25929 debugging information for C@t{++}, @value{GDBN} attempts to compute
25930 each entity's name. @value{GDBN} can do this computation in two
25931 different ways, depending on exactly what information is present.
25932 When enabled, this setting causes @value{GDBN} to compute the names
25933 both ways and display any discrepancies.
25934 @item show debug check-physname
25935 Show the current state of ``physname'' checking.
25936 @item set debug coff-pe-read
25937 @cindex COFF/PE exported symbols
25938 Control display of debugging messages related to reading of COFF/PE
25939 exported symbols. The default is off.
25940 @item show debug coff-pe-read
25941 Displays the current state of displaying debugging messages related to
25942 reading of COFF/PE exported symbols.
25943 @item set debug dwarf-die
25945 Dump DWARF DIEs after they are read in.
25946 The value is the number of nesting levels to print.
25947 A value of zero turns off the display.
25948 @item show debug dwarf-die
25949 Show the current state of DWARF DIE debugging.
25950 @item set debug dwarf-line
25951 @cindex DWARF Line Tables
25952 Turns on or off display of debugging messages related to reading
25953 DWARF line tables. The default is 0 (off).
25954 A value of 1 provides basic information.
25955 A value greater than 1 provides more verbose information.
25956 @item show debug dwarf-line
25957 Show the current state of DWARF line table debugging.
25958 @item set debug dwarf-read
25959 @cindex DWARF Reading
25960 Turns on or off display of debugging messages related to reading
25961 DWARF debug info. The default is 0 (off).
25962 A value of 1 provides basic information.
25963 A value greater than 1 provides more verbose information.
25964 @item show debug dwarf-read
25965 Show the current state of DWARF reader debugging.
25966 @item set debug displaced
25967 @cindex displaced stepping debugging info
25968 Turns on or off display of @value{GDBN} debugging info for the
25969 displaced stepping support. The default is off.
25970 @item show debug displaced
25971 Displays the current state of displaying @value{GDBN} debugging info
25972 related to displaced stepping.
25973 @item set debug event
25974 @cindex event debugging info
25975 Turns on or off display of @value{GDBN} event debugging info. The
25977 @item show debug event
25978 Displays the current state of displaying @value{GDBN} event debugging
25980 @item set debug expression
25981 @cindex expression debugging info
25982 Turns on or off display of debugging info about @value{GDBN}
25983 expression parsing. The default is off.
25984 @item show debug expression
25985 Displays the current state of displaying debugging info about
25986 @value{GDBN} expression parsing.
25987 @item set debug fbsd-lwp
25988 @cindex FreeBSD LWP debug messages
25989 Turns on or off debugging messages from the FreeBSD LWP debug support.
25990 @item show debug fbsd-lwp
25991 Show the current state of FreeBSD LWP debugging messages.
25992 @item set debug fbsd-nat
25993 @cindex FreeBSD native target debug messages
25994 Turns on or off debugging messages from the FreeBSD native target.
25995 @item show debug fbsd-nat
25996 Show the current state of FreeBSD native target debugging messages.
25997 @item set debug frame
25998 @cindex frame debugging info
25999 Turns on or off display of @value{GDBN} frame debugging info. The
26001 @item show debug frame
26002 Displays the current state of displaying @value{GDBN} frame debugging
26004 @item set debug gnu-nat
26005 @cindex @sc{gnu}/Hurd debug messages
26006 Turn on or off debugging messages from the @sc{gnu}/Hurd debug support.
26007 @item show debug gnu-nat
26008 Show the current state of @sc{gnu}/Hurd debugging messages.
26009 @item set debug infrun
26010 @cindex inferior debugging info
26011 Turns on or off display of @value{GDBN} debugging info for running the inferior.
26012 The default is off. @file{infrun.c} contains GDB's runtime state machine used
26013 for implementing operations such as single-stepping the inferior.
26014 @item show debug infrun
26015 Displays the current state of @value{GDBN} inferior debugging.
26016 @item set debug jit
26017 @cindex just-in-time compilation, debugging messages
26018 Turn on or off debugging messages from JIT debug support.
26019 @item show debug jit
26020 Displays the current state of @value{GDBN} JIT debugging.
26021 @item set debug lin-lwp
26022 @cindex @sc{gnu}/Linux LWP debug messages
26023 @cindex Linux lightweight processes
26024 Turn on or off debugging messages from the Linux LWP debug support.
26025 @item show debug lin-lwp
26026 Show the current state of Linux LWP debugging messages.
26027 @item set debug linux-namespaces
26028 @cindex @sc{gnu}/Linux namespaces debug messages
26029 Turn on or off debugging messages from the Linux namespaces debug support.
26030 @item show debug linux-namespaces
26031 Show the current state of Linux namespaces debugging messages.
26032 @item set debug mach-o
26033 @cindex Mach-O symbols processing
26034 Control display of debugging messages related to Mach-O symbols
26035 processing. The default is off.
26036 @item show debug mach-o
26037 Displays the current state of displaying debugging messages related to
26038 reading of COFF/PE exported symbols.
26039 @item set debug notification
26040 @cindex remote async notification debugging info
26041 Turn on or off debugging messages about remote async notification.
26042 The default is off.
26043 @item show debug notification
26044 Displays the current state of remote async notification debugging messages.
26045 @item set debug observer
26046 @cindex observer debugging info
26047 Turns on or off display of @value{GDBN} observer debugging. This
26048 includes info such as the notification of observable events.
26049 @item show debug observer
26050 Displays the current state of observer debugging.
26051 @item set debug overload
26052 @cindex C@t{++} overload debugging info
26053 Turns on or off display of @value{GDBN} C@t{++} overload debugging
26054 info. This includes info such as ranking of functions, etc. The default
26056 @item show debug overload
26057 Displays the current state of displaying @value{GDBN} C@t{++} overload
26059 @cindex expression parser, debugging info
26060 @cindex debug expression parser
26061 @item set debug parser
26062 Turns on or off the display of expression parser debugging output.
26063 Internally, this sets the @code{yydebug} variable in the expression
26064 parser. @xref{Tracing, , Tracing Your Parser, bison, Bison}, for
26065 details. The default is off.
26066 @item show debug parser
26067 Show the current state of expression parser debugging.
26068 @cindex packets, reporting on stdout
26069 @cindex serial connections, debugging
26070 @cindex debug remote protocol
26071 @cindex remote protocol debugging
26072 @cindex display remote packets
26073 @item set debug remote
26074 Turns on or off display of reports on all packets sent back and forth across
26075 the serial line to the remote machine. The info is printed on the
26076 @value{GDBN} standard output stream. The default is off.
26077 @item show debug remote
26078 Displays the state of display of remote packets.
26080 @item set debug separate-debug-file
26081 Turns on or off display of debug output about separate debug file search.
26082 @item show debug separate-debug-file
26083 Displays the state of separate debug file search debug output.
26085 @item set debug serial
26086 Turns on or off display of @value{GDBN} serial debugging info. The
26088 @item show debug serial
26089 Displays the current state of displaying @value{GDBN} serial debugging
26091 @item set debug solib-frv
26092 @cindex FR-V shared-library debugging
26093 Turn on or off debugging messages for FR-V shared-library code.
26094 @item show debug solib-frv
26095 Display the current state of FR-V shared-library code debugging
26097 @item set debug symbol-lookup
26098 @cindex symbol lookup
26099 Turns on or off display of debugging messages related to symbol lookup.
26100 The default is 0 (off).
26101 A value of 1 provides basic information.
26102 A value greater than 1 provides more verbose information.
26103 @item show debug symbol-lookup
26104 Show the current state of symbol lookup debugging messages.
26105 @item set debug symfile
26106 @cindex symbol file functions
26107 Turns on or off display of debugging messages related to symbol file functions.
26108 The default is off. @xref{Files}.
26109 @item show debug symfile
26110 Show the current state of symbol file debugging messages.
26111 @item set debug symtab-create
26112 @cindex symbol table creation
26113 Turns on or off display of debugging messages related to symbol table creation.
26114 The default is 0 (off).
26115 A value of 1 provides basic information.
26116 A value greater than 1 provides more verbose information.
26117 @item show debug symtab-create
26118 Show the current state of symbol table creation debugging.
26119 @item set debug target
26120 @cindex target debugging info
26121 Turns on or off display of @value{GDBN} target debugging info. This info
26122 includes what is going on at the target level of GDB, as it happens. The
26123 default is 0. Set it to 1 to track events, and to 2 to also track the
26124 value of large memory transfers.
26125 @item show debug target
26126 Displays the current state of displaying @value{GDBN} target debugging
26128 @item set debug timestamp
26129 @cindex timestampping debugging info
26130 Turns on or off display of timestamps with @value{GDBN} debugging info.
26131 When enabled, seconds and microseconds are displayed before each debugging
26133 @item show debug timestamp
26134 Displays the current state of displaying timestamps with @value{GDBN}
26136 @item set debug varobj
26137 @cindex variable object debugging info
26138 Turns on or off display of @value{GDBN} variable object debugging
26139 info. The default is off.
26140 @item show debug varobj
26141 Displays the current state of displaying @value{GDBN} variable object
26143 @item set debug xml
26144 @cindex XML parser debugging
26145 Turn on or off debugging messages for built-in XML parsers.
26146 @item show debug xml
26147 Displays the current state of XML debugging messages.
26150 @node Other Misc Settings
26151 @section Other Miscellaneous Settings
26152 @cindex miscellaneous settings
26155 @kindex set interactive-mode
26156 @item set interactive-mode
26157 If @code{on}, forces @value{GDBN} to assume that GDB was started
26158 in a terminal. In practice, this means that @value{GDBN} should wait
26159 for the user to answer queries generated by commands entered at
26160 the command prompt. If @code{off}, forces @value{GDBN} to operate
26161 in the opposite mode, and it uses the default answers to all queries.
26162 If @code{auto} (the default), @value{GDBN} tries to determine whether
26163 its standard input is a terminal, and works in interactive-mode if it
26164 is, non-interactively otherwise.
26166 In the vast majority of cases, the debugger should be able to guess
26167 correctly which mode should be used. But this setting can be useful
26168 in certain specific cases, such as running a MinGW @value{GDBN}
26169 inside a cygwin window.
26171 @kindex show interactive-mode
26172 @item show interactive-mode
26173 Displays whether the debugger is operating in interactive mode or not.
26176 @node Extending GDB
26177 @chapter Extending @value{GDBN}
26178 @cindex extending GDB
26180 @value{GDBN} provides several mechanisms for extension.
26181 @value{GDBN} also provides the ability to automatically load
26182 extensions when it reads a file for debugging. This allows the
26183 user to automatically customize @value{GDBN} for the program
26187 * Sequences:: Canned Sequences of @value{GDBN} Commands
26188 * Python:: Extending @value{GDBN} using Python
26189 * Guile:: Extending @value{GDBN} using Guile
26190 * Auto-loading extensions:: Automatically loading extensions
26191 * Multiple Extension Languages:: Working with multiple extension languages
26192 * Aliases:: Creating new spellings of existing commands
26195 To facilitate the use of extension languages, @value{GDBN} is capable
26196 of evaluating the contents of a file. When doing so, @value{GDBN}
26197 can recognize which extension language is being used by looking at
26198 the filename extension. Files with an unrecognized filename extension
26199 are always treated as a @value{GDBN} Command Files.
26200 @xref{Command Files,, Command files}.
26202 You can control how @value{GDBN} evaluates these files with the following
26206 @kindex set script-extension
26207 @kindex show script-extension
26208 @item set script-extension off
26209 All scripts are always evaluated as @value{GDBN} Command Files.
26211 @item set script-extension soft
26212 The debugger determines the scripting language based on filename
26213 extension. If this scripting language is supported, @value{GDBN}
26214 evaluates the script using that language. Otherwise, it evaluates
26215 the file as a @value{GDBN} Command File.
26217 @item set script-extension strict
26218 The debugger determines the scripting language based on filename
26219 extension, and evaluates the script using that language. If the
26220 language is not supported, then the evaluation fails.
26222 @item show script-extension
26223 Display the current value of the @code{script-extension} option.
26228 @section Canned Sequences of Commands
26230 Aside from breakpoint commands (@pxref{Break Commands, ,Breakpoint
26231 Command Lists}), @value{GDBN} provides two ways to store sequences of
26232 commands for execution as a unit: user-defined commands and command
26236 * Define:: How to define your own commands
26237 * Hooks:: Hooks for user-defined commands
26238 * Command Files:: How to write scripts of commands to be stored in a file
26239 * Output:: Commands for controlled output
26240 * Auto-loading sequences:: Controlling auto-loaded command files
26244 @subsection User-defined Commands
26246 @cindex user-defined command
26247 @cindex arguments, to user-defined commands
26248 A @dfn{user-defined command} is a sequence of @value{GDBN} commands to
26249 which you assign a new name as a command. This is done with the
26250 @code{define} command. User commands may accept an unlimited number of arguments
26251 separated by whitespace. Arguments are accessed within the user command
26252 via @code{$arg0@dots{}$argN}. A trivial example:
26256 print $arg0 + $arg1 + $arg2
26261 To execute the command use:
26268 This defines the command @code{adder}, which prints the sum of
26269 its three arguments. Note the arguments are text substitutions, so they may
26270 reference variables, use complex expressions, or even perform inferior
26273 @cindex argument count in user-defined commands
26274 @cindex how many arguments (user-defined commands)
26275 In addition, @code{$argc} may be used to find out how many arguments have
26281 print $arg0 + $arg1
26284 print $arg0 + $arg1 + $arg2
26289 Combining with the @code{eval} command (@pxref{eval}) makes it easier
26290 to process a variable number of arguments:
26297 eval "set $sum = $sum + $arg%d", $i
26307 @item define @var{commandname}
26308 Define a command named @var{commandname}. If there is already a command
26309 by that name, you are asked to confirm that you want to redefine it.
26310 The argument @var{commandname} may be a bare command name consisting of letters,
26311 numbers, dashes, and underscores. It may also start with any predefined
26312 prefix command. For example, @samp{define target my-target} creates
26313 a user-defined @samp{target my-target} command.
26315 The definition of the command is made up of other @value{GDBN} command lines,
26316 which are given following the @code{define} command. The end of these
26317 commands is marked by a line containing @code{end}.
26320 @kindex end@r{ (user-defined commands)}
26321 @item document @var{commandname}
26322 Document the user-defined command @var{commandname}, so that it can be
26323 accessed by @code{help}. The command @var{commandname} must already be
26324 defined. This command reads lines of documentation just as @code{define}
26325 reads the lines of the command definition, ending with @code{end}.
26326 After the @code{document} command is finished, @code{help} on command
26327 @var{commandname} displays the documentation you have written.
26329 You may use the @code{document} command again to change the
26330 documentation of a command. Redefining the command with @code{define}
26331 does not change the documentation.
26333 @kindex dont-repeat
26334 @cindex don't repeat command
26336 Used inside a user-defined command, this tells @value{GDBN} that this
26337 command should not be repeated when the user hits @key{RET}
26338 (@pxref{Command Syntax, repeat last command}).
26340 @kindex help user-defined
26341 @item help user-defined
26342 List all user-defined commands and all python commands defined in class
26343 COMAND_USER. The first line of the documentation or docstring is
26348 @itemx show user @var{commandname}
26349 Display the @value{GDBN} commands used to define @var{commandname} (but
26350 not its documentation). If no @var{commandname} is given, display the
26351 definitions for all user-defined commands.
26352 This does not work for user-defined python commands.
26354 @cindex infinite recursion in user-defined commands
26355 @kindex show max-user-call-depth
26356 @kindex set max-user-call-depth
26357 @item show max-user-call-depth
26358 @itemx set max-user-call-depth
26359 The value of @code{max-user-call-depth} controls how many recursion
26360 levels are allowed in user-defined commands before @value{GDBN} suspects an
26361 infinite recursion and aborts the command.
26362 This does not apply to user-defined python commands.
26365 In addition to the above commands, user-defined commands frequently
26366 use control flow commands, described in @ref{Command Files}.
26368 When user-defined commands are executed, the
26369 commands of the definition are not printed. An error in any command
26370 stops execution of the user-defined command.
26372 If used interactively, commands that would ask for confirmation proceed
26373 without asking when used inside a user-defined command. Many @value{GDBN}
26374 commands that normally print messages to say what they are doing omit the
26375 messages when used in a user-defined command.
26378 @subsection User-defined Command Hooks
26379 @cindex command hooks
26380 @cindex hooks, for commands
26381 @cindex hooks, pre-command
26384 You may define @dfn{hooks}, which are a special kind of user-defined
26385 command. Whenever you run the command @samp{foo}, if the user-defined
26386 command @samp{hook-foo} exists, it is executed (with no arguments)
26387 before that command.
26389 @cindex hooks, post-command
26391 A hook may also be defined which is run after the command you executed.
26392 Whenever you run the command @samp{foo}, if the user-defined command
26393 @samp{hookpost-foo} exists, it is executed (with no arguments) after
26394 that command. Post-execution hooks may exist simultaneously with
26395 pre-execution hooks, for the same command.
26397 It is valid for a hook to call the command which it hooks. If this
26398 occurs, the hook is not re-executed, thereby avoiding infinite recursion.
26400 @c It would be nice if hookpost could be passed a parameter indicating
26401 @c if the command it hooks executed properly or not. FIXME!
26403 @kindex stop@r{, a pseudo-command}
26404 In addition, a pseudo-command, @samp{stop} exists. Defining
26405 (@samp{hook-stop}) makes the associated commands execute every time
26406 execution stops in your program: before breakpoint commands are run,
26407 displays are printed, or the stack frame is printed.
26409 For example, to ignore @code{SIGALRM} signals while
26410 single-stepping, but treat them normally during normal execution,
26415 handle SIGALRM nopass
26419 handle SIGALRM pass
26422 define hook-continue
26423 handle SIGALRM pass
26427 As a further example, to hook at the beginning and end of the @code{echo}
26428 command, and to add extra text to the beginning and end of the message,
26436 define hookpost-echo
26440 (@value{GDBP}) echo Hello World
26441 <<<---Hello World--->>>
26446 You can define a hook for any single-word command in @value{GDBN}, but
26447 not for command aliases; you should define a hook for the basic command
26448 name, e.g.@: @code{backtrace} rather than @code{bt}.
26449 @c FIXME! So how does Joe User discover whether a command is an alias
26451 You can hook a multi-word command by adding @code{hook-} or
26452 @code{hookpost-} to the last word of the command, e.g.@:
26453 @samp{define target hook-remote} to add a hook to @samp{target remote}.
26455 If an error occurs during the execution of your hook, execution of
26456 @value{GDBN} commands stops and @value{GDBN} issues a prompt
26457 (before the command that you actually typed had a chance to run).
26459 If you try to define a hook which does not match any known command, you
26460 get a warning from the @code{define} command.
26462 @node Command Files
26463 @subsection Command Files
26465 @cindex command files
26466 @cindex scripting commands
26467 A command file for @value{GDBN} is a text file made of lines that are
26468 @value{GDBN} commands. Comments (lines starting with @kbd{#}) may
26469 also be included. An empty line in a command file does nothing; it
26470 does not mean to repeat the last command, as it would from the
26473 You can request the execution of a command file with the @code{source}
26474 command. Note that the @code{source} command is also used to evaluate
26475 scripts that are not Command Files. The exact behavior can be configured
26476 using the @code{script-extension} setting.
26477 @xref{Extending GDB,, Extending GDB}.
26481 @cindex execute commands from a file
26482 @item source [-s] [-v] @var{filename}
26483 Execute the command file @var{filename}.
26486 The lines in a command file are generally executed sequentially,
26487 unless the order of execution is changed by one of the
26488 @emph{flow-control commands} described below. The commands are not
26489 printed as they are executed. An error in any command terminates
26490 execution of the command file and control is returned to the console.
26492 @value{GDBN} first searches for @var{filename} in the current directory.
26493 If the file is not found there, and @var{filename} does not specify a
26494 directory, then @value{GDBN} also looks for the file on the source search path
26495 (specified with the @samp{directory} command);
26496 except that @file{$cdir} is not searched because the compilation directory
26497 is not relevant to scripts.
26499 If @code{-s} is specified, then @value{GDBN} searches for @var{filename}
26500 on the search path even if @var{filename} specifies a directory.
26501 The search is done by appending @var{filename} to each element of the
26502 search path. So, for example, if @var{filename} is @file{mylib/myscript}
26503 and the search path contains @file{/home/user} then @value{GDBN} will
26504 look for the script @file{/home/user/mylib/myscript}.
26505 The search is also done if @var{filename} is an absolute path.
26506 For example, if @var{filename} is @file{/tmp/myscript} and
26507 the search path contains @file{/home/user} then @value{GDBN} will
26508 look for the script @file{/home/user/tmp/myscript}.
26509 For DOS-like systems, if @var{filename} contains a drive specification,
26510 it is stripped before concatenation. For example, if @var{filename} is
26511 @file{d:myscript} and the search path contains @file{c:/tmp} then @value{GDBN}
26512 will look for the script @file{c:/tmp/myscript}.
26514 If @code{-v}, for verbose mode, is given then @value{GDBN} displays
26515 each command as it is executed. The option must be given before
26516 @var{filename}, and is interpreted as part of the filename anywhere else.
26518 Commands that would ask for confirmation if used interactively proceed
26519 without asking when used in a command file. Many @value{GDBN} commands that
26520 normally print messages to say what they are doing omit the messages
26521 when called from command files.
26523 @value{GDBN} also accepts command input from standard input. In this
26524 mode, normal output goes to standard output and error output goes to
26525 standard error. Errors in a command file supplied on standard input do
26526 not terminate execution of the command file---execution continues with
26530 gdb < cmds > log 2>&1
26533 (The syntax above will vary depending on the shell used.) This example
26534 will execute commands from the file @file{cmds}. All output and errors
26535 would be directed to @file{log}.
26537 Since commands stored on command files tend to be more general than
26538 commands typed interactively, they frequently need to deal with
26539 complicated situations, such as different or unexpected values of
26540 variables and symbols, changes in how the program being debugged is
26541 built, etc. @value{GDBN} provides a set of flow-control commands to
26542 deal with these complexities. Using these commands, you can write
26543 complex scripts that loop over data structures, execute commands
26544 conditionally, etc.
26551 This command allows to include in your script conditionally executed
26552 commands. The @code{if} command takes a single argument, which is an
26553 expression to evaluate. It is followed by a series of commands that
26554 are executed only if the expression is true (its value is nonzero).
26555 There can then optionally be an @code{else} line, followed by a series
26556 of commands that are only executed if the expression was false. The
26557 end of the list is marked by a line containing @code{end}.
26561 This command allows to write loops. Its syntax is similar to
26562 @code{if}: the command takes a single argument, which is an expression
26563 to evaluate, and must be followed by the commands to execute, one per
26564 line, terminated by an @code{end}. These commands are called the
26565 @dfn{body} of the loop. The commands in the body of @code{while} are
26566 executed repeatedly as long as the expression evaluates to true.
26570 This command exits the @code{while} loop in whose body it is included.
26571 Execution of the script continues after that @code{while}s @code{end}
26574 @kindex loop_continue
26575 @item loop_continue
26576 This command skips the execution of the rest of the body of commands
26577 in the @code{while} loop in whose body it is included. Execution
26578 branches to the beginning of the @code{while} loop, where it evaluates
26579 the controlling expression.
26581 @kindex end@r{ (if/else/while commands)}
26583 Terminate the block of commands that are the body of @code{if},
26584 @code{else}, or @code{while} flow-control commands.
26589 @subsection Commands for Controlled Output
26591 During the execution of a command file or a user-defined command, normal
26592 @value{GDBN} output is suppressed; the only output that appears is what is
26593 explicitly printed by the commands in the definition. This section
26594 describes three commands useful for generating exactly the output you
26599 @item echo @var{text}
26600 @c I do not consider backslash-space a standard C escape sequence
26601 @c because it is not in ANSI.
26602 Print @var{text}. Nonprinting characters can be included in
26603 @var{text} using C escape sequences, such as @samp{\n} to print a
26604 newline. @strong{No newline is printed unless you specify one.}
26605 In addition to the standard C escape sequences, a backslash followed
26606 by a space stands for a space. This is useful for displaying a
26607 string with spaces at the beginning or the end, since leading and
26608 trailing spaces are otherwise trimmed from all arguments.
26609 To print @samp{@w{ }and foo =@w{ }}, use the command
26610 @samp{echo \@w{ }and foo = \@w{ }}.
26612 A backslash at the end of @var{text} can be used, as in C, to continue
26613 the command onto subsequent lines. For example,
26616 echo This is some text\n\
26617 which is continued\n\
26618 onto several lines.\n
26621 produces the same output as
26624 echo This is some text\n
26625 echo which is continued\n
26626 echo onto several lines.\n
26630 @item output @var{expression}
26631 Print the value of @var{expression} and nothing but that value: no
26632 newlines, no @samp{$@var{nn} = }. The value is not entered in the
26633 value history either. @xref{Expressions, ,Expressions}, for more information
26636 @item output/@var{fmt} @var{expression}
26637 Print the value of @var{expression} in format @var{fmt}. You can use
26638 the same formats as for @code{print}. @xref{Output Formats,,Output
26639 Formats}, for more information.
26642 @item printf @var{template}, @var{expressions}@dots{}
26643 Print the values of one or more @var{expressions} under the control of
26644 the string @var{template}. To print several values, make
26645 @var{expressions} be a comma-separated list of individual expressions,
26646 which may be either numbers or pointers. Their values are printed as
26647 specified by @var{template}, exactly as a C program would do by
26648 executing the code below:
26651 printf (@var{template}, @var{expressions}@dots{});
26654 As in @code{C} @code{printf}, ordinary characters in @var{template}
26655 are printed verbatim, while @dfn{conversion specification} introduced
26656 by the @samp{%} character cause subsequent @var{expressions} to be
26657 evaluated, their values converted and formatted according to type and
26658 style information encoded in the conversion specifications, and then
26661 For example, you can print two values in hex like this:
26664 printf "foo, bar-foo = 0x%x, 0x%x\n", foo, bar-foo
26667 @code{printf} supports all the standard @code{C} conversion
26668 specifications, including the flags and modifiers between the @samp{%}
26669 character and the conversion letter, with the following exceptions:
26673 The argument-ordering modifiers, such as @samp{2$}, are not supported.
26676 The modifier @samp{*} is not supported for specifying precision or
26680 The @samp{'} flag (for separation of digits into groups according to
26681 @code{LC_NUMERIC'}) is not supported.
26684 The type modifiers @samp{hh}, @samp{j}, @samp{t}, and @samp{z} are not
26688 The conversion letter @samp{n} (as in @samp{%n}) is not supported.
26691 The conversion letters @samp{a} and @samp{A} are not supported.
26695 Note that the @samp{ll} type modifier is supported only if the
26696 underlying @code{C} implementation used to build @value{GDBN} supports
26697 the @code{long long int} type, and the @samp{L} type modifier is
26698 supported only if @code{long double} type is available.
26700 As in @code{C}, @code{printf} supports simple backslash-escape
26701 sequences, such as @code{\n}, @samp{\t}, @samp{\\}, @samp{\"},
26702 @samp{\a}, and @samp{\f}, that consist of backslash followed by a
26703 single character. Octal and hexadecimal escape sequences are not
26706 Additionally, @code{printf} supports conversion specifications for DFP
26707 (@dfn{Decimal Floating Point}) types using the following length modifiers
26708 together with a floating point specifier.
26713 @samp{H} for printing @code{Decimal32} types.
26716 @samp{D} for printing @code{Decimal64} types.
26719 @samp{DD} for printing @code{Decimal128} types.
26722 If the underlying @code{C} implementation used to build @value{GDBN} has
26723 support for the three length modifiers for DFP types, other modifiers
26724 such as width and precision will also be available for @value{GDBN} to use.
26726 In case there is no such @code{C} support, no additional modifiers will be
26727 available and the value will be printed in the standard way.
26729 Here's an example of printing DFP types using the above conversion letters:
26731 printf "D32: %Hf - D64: %Df - D128: %DDf\n",1.2345df,1.2E10dd,1.2E1dl
26736 @item eval @var{template}, @var{expressions}@dots{}
26737 Convert the values of one or more @var{expressions} under the control of
26738 the string @var{template} to a command line, and call it.
26742 @node Auto-loading sequences
26743 @subsection Controlling auto-loading native @value{GDBN} scripts
26744 @cindex native script auto-loading
26746 When a new object file is read (for example, due to the @code{file}
26747 command, or because the inferior has loaded a shared library),
26748 @value{GDBN} will look for the command file @file{@var{objfile}-gdb.gdb}.
26749 @xref{Auto-loading extensions}.
26751 Auto-loading can be enabled or disabled,
26752 and the list of auto-loaded scripts can be printed.
26755 @anchor{set auto-load gdb-scripts}
26756 @kindex set auto-load gdb-scripts
26757 @item set auto-load gdb-scripts [on|off]
26758 Enable or disable the auto-loading of canned sequences of commands scripts.
26760 @anchor{show auto-load gdb-scripts}
26761 @kindex show auto-load gdb-scripts
26762 @item show auto-load gdb-scripts
26763 Show whether auto-loading of canned sequences of commands scripts is enabled or
26766 @anchor{info auto-load gdb-scripts}
26767 @kindex info auto-load gdb-scripts
26768 @cindex print list of auto-loaded canned sequences of commands scripts
26769 @item info auto-load gdb-scripts [@var{regexp}]
26770 Print the list of all canned sequences of commands scripts that @value{GDBN}
26774 If @var{regexp} is supplied only canned sequences of commands scripts with
26775 matching names are printed.
26777 @c Python docs live in a separate file.
26778 @include python.texi
26780 @c Guile docs live in a separate file.
26781 @include guile.texi
26783 @node Auto-loading extensions
26784 @section Auto-loading extensions
26785 @cindex auto-loading extensions
26787 @value{GDBN} provides two mechanisms for automatically loading extensions
26788 when a new object file is read (for example, due to the @code{file}
26789 command, or because the inferior has loaded a shared library):
26790 @file{@var{objfile}-gdb.@var{ext}} and the @code{.debug_gdb_scripts}
26791 section of modern file formats like ELF.
26794 * objfile-gdb.ext file: objfile-gdbdotext file. The @file{@var{objfile}-gdb.@var{ext}} file
26795 * .debug_gdb_scripts section: dotdebug_gdb_scripts section. The @code{.debug_gdb_scripts} section
26796 * Which flavor to choose?::
26799 The auto-loading feature is useful for supplying application-specific
26800 debugging commands and features.
26802 Auto-loading can be enabled or disabled,
26803 and the list of auto-loaded scripts can be printed.
26804 See the @samp{auto-loading} section of each extension language
26805 for more information.
26806 For @value{GDBN} command files see @ref{Auto-loading sequences}.
26807 For Python files see @ref{Python Auto-loading}.
26809 Note that loading of this script file also requires accordingly configured
26810 @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
26812 @node objfile-gdbdotext file
26813 @subsection The @file{@var{objfile}-gdb.@var{ext}} file
26814 @cindex @file{@var{objfile}-gdb.gdb}
26815 @cindex @file{@var{objfile}-gdb.py}
26816 @cindex @file{@var{objfile}-gdb.scm}
26818 When a new object file is read, @value{GDBN} looks for a file named
26819 @file{@var{objfile}-gdb.@var{ext}} (we call it @var{script-name} below),
26820 where @var{objfile} is the object file's name and
26821 where @var{ext} is the file extension for the extension language:
26824 @item @file{@var{objfile}-gdb.gdb}
26825 GDB's own command language
26826 @item @file{@var{objfile}-gdb.py}
26828 @item @file{@var{objfile}-gdb.scm}
26832 @var{script-name} is formed by ensuring that the file name of @var{objfile}
26833 is absolute, following all symlinks, and resolving @code{.} and @code{..}
26834 components, and appending the @file{-gdb.@var{ext}} suffix.
26835 If this file exists and is readable, @value{GDBN} will evaluate it as a
26836 script in the specified extension language.
26838 If this file does not exist, then @value{GDBN} will look for
26839 @var{script-name} file in all of the directories as specified below.
26841 Note that loading of these files requires an accordingly configured
26842 @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
26844 For object files using @file{.exe} suffix @value{GDBN} tries to load first the
26845 scripts normally according to its @file{.exe} filename. But if no scripts are
26846 found @value{GDBN} also tries script filenames matching the object file without
26847 its @file{.exe} suffix. This @file{.exe} stripping is case insensitive and it
26848 is attempted on any platform. This makes the script filenames compatible
26849 between Unix and MS-Windows hosts.
26852 @anchor{set auto-load scripts-directory}
26853 @kindex set auto-load scripts-directory
26854 @item set auto-load scripts-directory @r{[}@var{directories}@r{]}
26855 Control @value{GDBN} auto-loaded scripts location. Multiple directory entries
26856 may be delimited by the host platform path separator in use
26857 (@samp{:} on Unix, @samp{;} on MS-Windows and MS-DOS).
26859 Each entry here needs to be covered also by the security setting
26860 @code{set auto-load safe-path} (@pxref{set auto-load safe-path}).
26862 @anchor{with-auto-load-dir}
26863 This variable defaults to @file{$debugdir:$datadir/auto-load}. The default
26864 @code{set auto-load safe-path} value can be also overriden by @value{GDBN}
26865 configuration option @option{--with-auto-load-dir}.
26867 Any reference to @file{$debugdir} will get replaced by
26868 @var{debug-file-directory} value (@pxref{Separate Debug Files}) and any
26869 reference to @file{$datadir} will get replaced by @var{data-directory} which is
26870 determined at @value{GDBN} startup (@pxref{Data Files}). @file{$debugdir} and
26871 @file{$datadir} must be placed as a directory component --- either alone or
26872 delimited by @file{/} or @file{\} directory separators, depending on the host
26875 The list of directories uses path separator (@samp{:} on GNU and Unix
26876 systems, @samp{;} on MS-Windows and MS-DOS) to separate directories, similarly
26877 to the @env{PATH} environment variable.
26879 @anchor{show auto-load scripts-directory}
26880 @kindex show auto-load scripts-directory
26881 @item show auto-load scripts-directory
26882 Show @value{GDBN} auto-loaded scripts location.
26884 @anchor{add-auto-load-scripts-directory}
26885 @kindex add-auto-load-scripts-directory
26886 @item add-auto-load-scripts-directory @r{[}@var{directories}@dots{}@r{]}
26887 Add an entry (or list of entries) to the list of auto-loaded scripts locations.
26888 Multiple entries may be delimited by the host platform path separator in use.
26891 @value{GDBN} does not track which files it has already auto-loaded this way.
26892 @value{GDBN} will load the associated script every time the corresponding
26893 @var{objfile} is opened.
26894 So your @file{-gdb.@var{ext}} file should be careful to avoid errors if it
26895 is evaluated more than once.
26897 @node dotdebug_gdb_scripts section
26898 @subsection The @code{.debug_gdb_scripts} section
26899 @cindex @code{.debug_gdb_scripts} section
26901 For systems using file formats like ELF and COFF,
26902 when @value{GDBN} loads a new object file
26903 it will look for a special section named @code{.debug_gdb_scripts}.
26904 If this section exists, its contents is a list of null-terminated entries
26905 specifying scripts to load. Each entry begins with a non-null prefix byte that
26906 specifies the kind of entry, typically the extension language and whether the
26907 script is in a file or inlined in @code{.debug_gdb_scripts}.
26909 The following entries are supported:
26912 @item SECTION_SCRIPT_ID_PYTHON_FILE = 1
26913 @item SECTION_SCRIPT_ID_SCHEME_FILE = 3
26914 @item SECTION_SCRIPT_ID_PYTHON_TEXT = 4
26915 @item SECTION_SCRIPT_ID_SCHEME_TEXT = 6
26918 @subsubsection Script File Entries
26920 If the entry specifies a file, @value{GDBN} will look for the file first
26921 in the current directory and then along the source search path
26922 (@pxref{Source Path, ,Specifying Source Directories}),
26923 except that @file{$cdir} is not searched, since the compilation
26924 directory is not relevant to scripts.
26926 File entries can be placed in section @code{.debug_gdb_scripts} with,
26927 for example, this GCC macro for Python scripts.
26930 /* Note: The "MS" section flags are to remove duplicates. */
26931 #define DEFINE_GDB_PY_SCRIPT(script_name) \
26933 .pushsection \".debug_gdb_scripts\", \"MS\",@@progbits,1\n\
26934 .byte 1 /* Python */\n\
26935 .asciz \"" script_name "\"\n\
26941 For Guile scripts, replace @code{.byte 1} with @code{.byte 3}.
26942 Then one can reference the macro in a header or source file like this:
26945 DEFINE_GDB_PY_SCRIPT ("my-app-scripts.py")
26948 The script name may include directories if desired.
26950 Note that loading of this script file also requires accordingly configured
26951 @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
26953 If the macro invocation is put in a header, any application or library
26954 using this header will get a reference to the specified script,
26955 and with the use of @code{"MS"} attributes on the section, the linker
26956 will remove duplicates.
26958 @subsubsection Script Text Entries
26960 Script text entries allow to put the executable script in the entry
26961 itself instead of loading it from a file.
26962 The first line of the entry, everything after the prefix byte and up to
26963 the first newline (@code{0xa}) character, is the script name, and must not
26964 contain any kind of space character, e.g., spaces or tabs.
26965 The rest of the entry, up to the trailing null byte, is the script to
26966 execute in the specified language. The name needs to be unique among
26967 all script names, as @value{GDBN} executes each script only once based
26970 Here is an example from file @file{py-section-script.c} in the @value{GDBN}
26974 #include "symcat.h"
26975 #include "gdb/section-scripts.h"
26977 ".pushsection \".debug_gdb_scripts\", \"MS\",@@progbits,1\n"
26978 ".byte " XSTRING (SECTION_SCRIPT_ID_PYTHON_TEXT) "\n"
26979 ".ascii \"gdb.inlined-script\\n\"\n"
26980 ".ascii \"class test_cmd (gdb.Command):\\n\"\n"
26981 ".ascii \" def __init__ (self):\\n\"\n"
26982 ".ascii \" super (test_cmd, self).__init__ ("
26983 "\\\"test-cmd\\\", gdb.COMMAND_OBSCURE)\\n\"\n"
26984 ".ascii \" def invoke (self, arg, from_tty):\\n\"\n"
26985 ".ascii \" print (\\\"test-cmd output, arg = %s\\\" % arg)\\n\"\n"
26986 ".ascii \"test_cmd ()\\n\"\n"
26992 Loading of inlined scripts requires a properly configured
26993 @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
26994 The path to specify in @code{auto-load safe-path} is the path of the file
26995 containing the @code{.debug_gdb_scripts} section.
26997 @node Which flavor to choose?
26998 @subsection Which flavor to choose?
27000 Given the multiple ways of auto-loading extensions, it might not always
27001 be clear which one to choose. This section provides some guidance.
27004 Benefits of the @file{-gdb.@var{ext}} way:
27008 Can be used with file formats that don't support multiple sections.
27011 Ease of finding scripts for public libraries.
27013 Scripts specified in the @code{.debug_gdb_scripts} section are searched for
27014 in the source search path.
27015 For publicly installed libraries, e.g., @file{libstdc++}, there typically
27016 isn't a source directory in which to find the script.
27019 Doesn't require source code additions.
27023 Benefits of the @code{.debug_gdb_scripts} way:
27027 Works with static linking.
27029 Scripts for libraries done the @file{-gdb.@var{ext}} way require an objfile to
27030 trigger their loading. When an application is statically linked the only
27031 objfile available is the executable, and it is cumbersome to attach all the
27032 scripts from all the input libraries to the executable's
27033 @file{-gdb.@var{ext}} script.
27036 Works with classes that are entirely inlined.
27038 Some classes can be entirely inlined, and thus there may not be an associated
27039 shared library to attach a @file{-gdb.@var{ext}} script to.
27042 Scripts needn't be copied out of the source tree.
27044 In some circumstances, apps can be built out of large collections of internal
27045 libraries, and the build infrastructure necessary to install the
27046 @file{-gdb.@var{ext}} scripts in a place where @value{GDBN} can find them is
27047 cumbersome. It may be easier to specify the scripts in the
27048 @code{.debug_gdb_scripts} section as relative paths, and add a path to the
27049 top of the source tree to the source search path.
27052 @node Multiple Extension Languages
27053 @section Multiple Extension Languages
27055 The Guile and Python extension languages do not share any state,
27056 and generally do not interfere with each other.
27057 There are some things to be aware of, however.
27059 @subsection Python comes first
27061 Python was @value{GDBN}'s first extension language, and to avoid breaking
27062 existing behaviour Python comes first. This is generally solved by the
27063 ``first one wins'' principle. @value{GDBN} maintains a list of enabled
27064 extension languages, and when it makes a call to an extension language,
27065 (say to pretty-print a value), it tries each in turn until an extension
27066 language indicates it has performed the request (e.g., has returned the
27067 pretty-printed form of a value).
27068 This extends to errors while performing such requests: If an error happens
27069 while, for example, trying to pretty-print an object then the error is
27070 reported and any following extension languages are not tried.
27073 @section Creating new spellings of existing commands
27074 @cindex aliases for commands
27076 It is often useful to define alternate spellings of existing commands.
27077 For example, if a new @value{GDBN} command defined in Python has
27078 a long name to type, it is handy to have an abbreviated version of it
27079 that involves less typing.
27081 @value{GDBN} itself uses aliases. For example @samp{s} is an alias
27082 of the @samp{step} command even though it is otherwise an ambiguous
27083 abbreviation of other commands like @samp{set} and @samp{show}.
27085 Aliases are also used to provide shortened or more common versions
27086 of multi-word commands. For example, @value{GDBN} provides the
27087 @samp{tty} alias of the @samp{set inferior-tty} command.
27089 You can define a new alias with the @samp{alias} command.
27094 @item alias [-a] [--] @var{ALIAS} = @var{COMMAND}
27098 @var{ALIAS} specifies the name of the new alias.
27099 Each word of @var{ALIAS} must consist of letters, numbers, dashes and
27102 @var{COMMAND} specifies the name of an existing command
27103 that is being aliased.
27105 The @samp{-a} option specifies that the new alias is an abbreviation
27106 of the command. Abbreviations are not shown in command
27107 lists displayed by the @samp{help} command.
27109 The @samp{--} option specifies the end of options,
27110 and is useful when @var{ALIAS} begins with a dash.
27112 Here is a simple example showing how to make an abbreviation
27113 of a command so that there is less to type.
27114 Suppose you were tired of typing @samp{disas}, the current
27115 shortest unambiguous abbreviation of the @samp{disassemble} command
27116 and you wanted an even shorter version named @samp{di}.
27117 The following will accomplish this.
27120 (gdb) alias -a di = disas
27123 Note that aliases are different from user-defined commands.
27124 With a user-defined command, you also need to write documentation
27125 for it with the @samp{document} command.
27126 An alias automatically picks up the documentation of the existing command.
27128 Here is an example where we make @samp{elms} an abbreviation of
27129 @samp{elements} in the @samp{set print elements} command.
27130 This is to show that you can make an abbreviation of any part
27134 (gdb) alias -a set print elms = set print elements
27135 (gdb) alias -a show print elms = show print elements
27136 (gdb) set p elms 20
27138 Limit on string chars or array elements to print is 200.
27141 Note that if you are defining an alias of a @samp{set} command,
27142 and you want to have an alias for the corresponding @samp{show}
27143 command, then you need to define the latter separately.
27145 Unambiguously abbreviated commands are allowed in @var{COMMAND} and
27146 @var{ALIAS}, just as they are normally.
27149 (gdb) alias -a set pr elms = set p ele
27152 Finally, here is an example showing the creation of a one word
27153 alias for a more complex command.
27154 This creates alias @samp{spe} of the command @samp{set print elements}.
27157 (gdb) alias spe = set print elements
27162 @chapter Command Interpreters
27163 @cindex command interpreters
27165 @value{GDBN} supports multiple command interpreters, and some command
27166 infrastructure to allow users or user interface writers to switch
27167 between interpreters or run commands in other interpreters.
27169 @value{GDBN} currently supports two command interpreters, the console
27170 interpreter (sometimes called the command-line interpreter or @sc{cli})
27171 and the machine interface interpreter (or @sc{gdb/mi}). This manual
27172 describes both of these interfaces in great detail.
27174 By default, @value{GDBN} will start with the console interpreter.
27175 However, the user may choose to start @value{GDBN} with another
27176 interpreter by specifying the @option{-i} or @option{--interpreter}
27177 startup options. Defined interpreters include:
27181 @cindex console interpreter
27182 The traditional console or command-line interpreter. This is the most often
27183 used interpreter with @value{GDBN}. With no interpreter specified at runtime,
27184 @value{GDBN} will use this interpreter.
27187 @cindex mi interpreter
27188 The newest @sc{gdb/mi} interface (currently @code{mi3}). Used primarily
27189 by programs wishing to use @value{GDBN} as a backend for a debugger GUI
27190 or an IDE. For more information, see @ref{GDB/MI, ,The @sc{gdb/mi}
27194 @cindex mi3 interpreter
27195 The @sc{gdb/mi} interface introduced in @value{GDBN} 9.1.
27198 @cindex mi2 interpreter
27199 The @sc{gdb/mi} interface introduced in @value{GDBN} 6.0.
27202 @cindex mi1 interpreter
27203 The @sc{gdb/mi} interface introduced in @value{GDBN} 5.1.
27207 @cindex invoke another interpreter
27209 @kindex interpreter-exec
27210 You may execute commands in any interpreter from the current
27211 interpreter using the appropriate command. If you are running the
27212 console interpreter, simply use the @code{interpreter-exec} command:
27215 interpreter-exec mi "-data-list-register-names"
27218 @sc{gdb/mi} has a similar command, although it is only available in versions of
27219 @value{GDBN} which support @sc{gdb/mi} version 2 (or greater).
27221 Note that @code{interpreter-exec} only changes the interpreter for the
27222 duration of the specified command. It does not change the interpreter
27225 @cindex start a new independent interpreter
27227 Although you may only choose a single interpreter at startup, it is
27228 possible to run an independent interpreter on a specified input/output
27229 device (usually a tty).
27231 For example, consider a debugger GUI or IDE that wants to provide a
27232 @value{GDBN} console view. It may do so by embedding a terminal
27233 emulator widget in its GUI, starting @value{GDBN} in the traditional
27234 command-line mode with stdin/stdout/stderr redirected to that
27235 terminal, and then creating an MI interpreter running on a specified
27236 input/output device. The console interpreter created by @value{GDBN}
27237 at startup handles commands the user types in the terminal widget,
27238 while the GUI controls and synchronizes state with @value{GDBN} using
27239 the separate MI interpreter.
27241 To start a new secondary @dfn{user interface} running MI, use the
27242 @code{new-ui} command:
27245 @cindex new user interface
27247 new-ui @var{interpreter} @var{tty}
27250 The @var{interpreter} parameter specifies the interpreter to run.
27251 This accepts the same values as the @code{interpreter-exec} command.
27252 For example, @samp{console}, @samp{mi}, @samp{mi2}, etc. The
27253 @var{tty} parameter specifies the name of the bidirectional file the
27254 interpreter uses for input/output, usually the name of a
27255 pseudoterminal slave on Unix systems. For example:
27258 (@value{GDBP}) new-ui mi /dev/pts/9
27262 runs an MI interpreter on @file{/dev/pts/9}.
27265 @chapter @value{GDBN} Text User Interface
27267 @cindex Text User Interface
27270 * TUI Overview:: TUI overview
27271 * TUI Keys:: TUI key bindings
27272 * TUI Single Key Mode:: TUI single key mode
27273 * TUI Commands:: TUI-specific commands
27274 * TUI Configuration:: TUI configuration variables
27277 The @value{GDBN} Text User Interface (TUI) is a terminal
27278 interface which uses the @code{curses} library to show the source
27279 file, the assembly output, the program registers and @value{GDBN}
27280 commands in separate text windows. The TUI mode is supported only
27281 on platforms where a suitable version of the @code{curses} library
27284 The TUI mode is enabled by default when you invoke @value{GDBN} as
27285 @samp{@value{GDBP} -tui}.
27286 You can also switch in and out of TUI mode while @value{GDBN} runs by
27287 using various TUI commands and key bindings, such as @command{tui
27288 enable} or @kbd{C-x C-a}. @xref{TUI Commands, ,TUI Commands}, and
27289 @ref{TUI Keys, ,TUI Key Bindings}.
27292 @section TUI Overview
27294 In TUI mode, @value{GDBN} can display several text windows:
27298 This window is the @value{GDBN} command window with the @value{GDBN}
27299 prompt and the @value{GDBN} output. The @value{GDBN} input is still
27300 managed using readline.
27303 The source window shows the source file of the program. The current
27304 line and active breakpoints are displayed in this window.
27307 The assembly window shows the disassembly output of the program.
27310 This window shows the processor registers. Registers are highlighted
27311 when their values change.
27314 The source and assembly windows show the current program position
27315 by highlighting the current line and marking it with a @samp{>} marker.
27316 Breakpoints are indicated with two markers. The first marker
27317 indicates the breakpoint type:
27321 Breakpoint which was hit at least once.
27324 Breakpoint which was never hit.
27327 Hardware breakpoint which was hit at least once.
27330 Hardware breakpoint which was never hit.
27333 The second marker indicates whether the breakpoint is enabled or not:
27337 Breakpoint is enabled.
27340 Breakpoint is disabled.
27343 The source, assembly and register windows are updated when the current
27344 thread changes, when the frame changes, or when the program counter
27347 These windows are not all visible at the same time. The command
27348 window is always visible. The others can be arranged in several
27359 source and assembly,
27362 source and registers, or
27365 assembly and registers.
27368 A status line above the command window shows the following information:
27372 Indicates the current @value{GDBN} target.
27373 (@pxref{Targets, ,Specifying a Debugging Target}).
27376 Gives the current process or thread number.
27377 When no process is being debugged, this field is set to @code{No process}.
27380 Gives the current function name for the selected frame.
27381 The name is demangled if demangling is turned on (@pxref{Print Settings}).
27382 When there is no symbol corresponding to the current program counter,
27383 the string @code{??} is displayed.
27386 Indicates the current line number for the selected frame.
27387 When the current line number is not known, the string @code{??} is displayed.
27390 Indicates the current program counter address.
27394 @section TUI Key Bindings
27395 @cindex TUI key bindings
27397 The TUI installs several key bindings in the readline keymaps
27398 @ifset SYSTEM_READLINE
27399 (@pxref{Command Line Editing, , , rluserman, GNU Readline Library}).
27401 @ifclear SYSTEM_READLINE
27402 (@pxref{Command Line Editing}).
27404 The following key bindings are installed for both TUI mode and the
27405 @value{GDBN} standard mode.
27414 Enter or leave the TUI mode. When leaving the TUI mode,
27415 the curses window management stops and @value{GDBN} operates using
27416 its standard mode, writing on the terminal directly. When reentering
27417 the TUI mode, control is given back to the curses windows.
27418 The screen is then refreshed.
27422 Use a TUI layout with only one window. The layout will
27423 either be @samp{source} or @samp{assembly}. When the TUI mode
27424 is not active, it will switch to the TUI mode.
27426 Think of this key binding as the Emacs @kbd{C-x 1} binding.
27430 Use a TUI layout with at least two windows. When the current
27431 layout already has two windows, the next layout with two windows is used.
27432 When a new layout is chosen, one window will always be common to the
27433 previous layout and the new one.
27435 Think of it as the Emacs @kbd{C-x 2} binding.
27439 Change the active window. The TUI associates several key bindings
27440 (like scrolling and arrow keys) with the active window. This command
27441 gives the focus to the next TUI window.
27443 Think of it as the Emacs @kbd{C-x o} binding.
27447 Switch in and out of the TUI SingleKey mode that binds single
27448 keys to @value{GDBN} commands (@pxref{TUI Single Key Mode}).
27451 The following key bindings only work in the TUI mode:
27456 Scroll the active window one page up.
27460 Scroll the active window one page down.
27464 Scroll the active window one line up.
27468 Scroll the active window one line down.
27472 Scroll the active window one column left.
27476 Scroll the active window one column right.
27480 Refresh the screen.
27483 Because the arrow keys scroll the active window in the TUI mode, they
27484 are not available for their normal use by readline unless the command
27485 window has the focus. When another window is active, you must use
27486 other readline key bindings such as @kbd{C-p}, @kbd{C-n}, @kbd{C-b}
27487 and @kbd{C-f} to control the command window.
27489 @node TUI Single Key Mode
27490 @section TUI Single Key Mode
27491 @cindex TUI single key mode
27493 The TUI also provides a @dfn{SingleKey} mode, which binds several
27494 frequently used @value{GDBN} commands to single keys. Type @kbd{C-x s} to
27495 switch into this mode, where the following key bindings are used:
27498 @kindex c @r{(SingleKey TUI key)}
27502 @kindex d @r{(SingleKey TUI key)}
27506 @kindex f @r{(SingleKey TUI key)}
27510 @kindex n @r{(SingleKey TUI key)}
27514 @kindex o @r{(SingleKey TUI key)}
27516 nexti. The shortcut letter @samp{o} stands for ``step Over''.
27518 @kindex q @r{(SingleKey TUI key)}
27520 exit the SingleKey mode.
27522 @kindex r @r{(SingleKey TUI key)}
27526 @kindex s @r{(SingleKey TUI key)}
27530 @kindex i @r{(SingleKey TUI key)}
27532 stepi. The shortcut letter @samp{i} stands for ``step Into''.
27534 @kindex u @r{(SingleKey TUI key)}
27538 @kindex v @r{(SingleKey TUI key)}
27542 @kindex w @r{(SingleKey TUI key)}
27547 Other keys temporarily switch to the @value{GDBN} command prompt.
27548 The key that was pressed is inserted in the editing buffer so that
27549 it is possible to type most @value{GDBN} commands without interaction
27550 with the TUI SingleKey mode. Once the command is entered the TUI
27551 SingleKey mode is restored. The only way to permanently leave
27552 this mode is by typing @kbd{q} or @kbd{C-x s}.
27556 @section TUI-specific Commands
27557 @cindex TUI commands
27559 The TUI has specific commands to control the text windows.
27560 These commands are always available, even when @value{GDBN} is not in
27561 the TUI mode. When @value{GDBN} is in the standard mode, most
27562 of these commands will automatically switch to the TUI mode.
27564 Note that if @value{GDBN}'s @code{stdout} is not connected to a
27565 terminal, or @value{GDBN} has been started with the machine interface
27566 interpreter (@pxref{GDB/MI, ,The @sc{gdb/mi} Interface}), most of
27567 these commands will fail with an error, because it would not be
27568 possible or desirable to enable curses window management.
27573 Activate TUI mode. The last active TUI window layout will be used if
27574 TUI mode has prevsiouly been used in the current debugging session,
27575 otherwise a default layout is used.
27578 @kindex tui disable
27579 Disable TUI mode, returning to the console interpreter.
27583 List and give the size of all displayed windows.
27585 @item layout @var{name}
27587 Changes which TUI windows are displayed. In each layout the command
27588 window is always displayed, the @var{name} parameter controls which
27589 additional windows are displayed, and can be any of the following:
27593 Display the next layout.
27596 Display the previous layout.
27599 Display the source and command windows.
27602 Display the assembly and command windows.
27605 Display the source, assembly, and command windows.
27608 When in @code{src} layout display the register, source, and command
27609 windows. When in @code{asm} or @code{split} layout display the
27610 register, assembler, and command windows.
27613 @item focus @var{name}
27615 Changes which TUI window is currently active for scrolling. The
27616 @var{name} parameter can be any of the following:
27620 Make the next window active for scrolling.
27623 Make the previous window active for scrolling.
27626 Make the source window active for scrolling.
27629 Make the assembly window active for scrolling.
27632 Make the register window active for scrolling.
27635 Make the command window active for scrolling.
27640 Refresh the screen. This is similar to typing @kbd{C-L}.
27642 @item tui reg @var{group}
27644 Changes the register group displayed in the tui register window to
27645 @var{group}. If the register window is not currently displayed this
27646 command will cause the register window to be displayed. The list of
27647 register groups, as well as their order is target specific. The
27648 following groups are available on most targets:
27651 Repeatedly selecting this group will cause the display to cycle
27652 through all of the available register groups.
27655 Repeatedly selecting this group will cause the display to cycle
27656 through all of the available register groups in the reverse order to
27660 Display the general registers.
27662 Display the floating point registers.
27664 Display the system registers.
27666 Display the vector registers.
27668 Display all registers.
27673 Update the source window and the current execution point.
27675 @item winheight @var{name} +@var{count}
27676 @itemx winheight @var{name} -@var{count}
27678 Change the height of the window @var{name} by @var{count}
27679 lines. Positive counts increase the height, while negative counts
27680 decrease it. The @var{name} parameter can be one of @code{src} (the
27681 source window), @code{cmd} (the command window), @code{asm} (the
27682 disassembly window), or @code{regs} (the register display window).
27685 @node TUI Configuration
27686 @section TUI Configuration Variables
27687 @cindex TUI configuration variables
27689 Several configuration variables control the appearance of TUI windows.
27692 @item set tui border-kind @var{kind}
27693 @kindex set tui border-kind
27694 Select the border appearance for the source, assembly and register windows.
27695 The possible values are the following:
27698 Use a space character to draw the border.
27701 Use @sc{ascii} characters @samp{+}, @samp{-} and @samp{|} to draw the border.
27704 Use the Alternate Character Set to draw the border. The border is
27705 drawn using character line graphics if the terminal supports them.
27708 @item set tui border-mode @var{mode}
27709 @kindex set tui border-mode
27710 @itemx set tui active-border-mode @var{mode}
27711 @kindex set tui active-border-mode
27712 Select the display attributes for the borders of the inactive windows
27713 or the active window. The @var{mode} can be one of the following:
27716 Use normal attributes to display the border.
27722 Use reverse video mode.
27725 Use half bright mode.
27727 @item half-standout
27728 Use half bright and standout mode.
27731 Use extra bright or bold mode.
27733 @item bold-standout
27734 Use extra bright or bold and standout mode.
27737 @item set tui tab-width @var{nchars}
27738 @kindex set tui tab-width
27740 Set the width of tab stops to be @var{nchars} characters. This
27741 setting affects the display of TAB characters in the source and
27746 @chapter Using @value{GDBN} under @sc{gnu} Emacs
27749 @cindex @sc{gnu} Emacs
27750 A special interface allows you to use @sc{gnu} Emacs to view (and
27751 edit) the source files for the program you are debugging with
27754 To use this interface, use the command @kbd{M-x gdb} in Emacs. Give the
27755 executable file you want to debug as an argument. This command starts
27756 @value{GDBN} as a subprocess of Emacs, with input and output through a newly
27757 created Emacs buffer.
27758 @c (Do not use the @code{-tui} option to run @value{GDBN} from Emacs.)
27760 Running @value{GDBN} under Emacs can be just like running @value{GDBN} normally except for two
27765 All ``terminal'' input and output goes through an Emacs buffer, called
27768 This applies both to @value{GDBN} commands and their output, and to the input
27769 and output done by the program you are debugging.
27771 This is useful because it means that you can copy the text of previous
27772 commands and input them again; you can even use parts of the output
27775 All the facilities of Emacs' Shell mode are available for interacting
27776 with your program. In particular, you can send signals the usual
27777 way---for example, @kbd{C-c C-c} for an interrupt, @kbd{C-c C-z} for a
27781 @value{GDBN} displays source code through Emacs.
27783 Each time @value{GDBN} displays a stack frame, Emacs automatically finds the
27784 source file for that frame and puts an arrow (@samp{=>}) at the
27785 left margin of the current line. Emacs uses a separate buffer for
27786 source display, and splits the screen to show both your @value{GDBN} session
27789 Explicit @value{GDBN} @code{list} or search commands still produce output as
27790 usual, but you probably have no reason to use them from Emacs.
27793 We call this @dfn{text command mode}. Emacs 22.1, and later, also uses
27794 a graphical mode, enabled by default, which provides further buffers
27795 that can control the execution and describe the state of your program.
27796 @xref{GDB Graphical Interface,,, Emacs, The @sc{gnu} Emacs Manual}.
27798 If you specify an absolute file name when prompted for the @kbd{M-x
27799 gdb} argument, then Emacs sets your current working directory to where
27800 your program resides. If you only specify the file name, then Emacs
27801 sets your current working directory to the directory associated
27802 with the previous buffer. In this case, @value{GDBN} may find your
27803 program by searching your environment's @code{PATH} variable, but on
27804 some operating systems it might not find the source. So, although the
27805 @value{GDBN} input and output session proceeds normally, the auxiliary
27806 buffer does not display the current source and line of execution.
27808 The initial working directory of @value{GDBN} is printed on the top
27809 line of the GUD buffer and this serves as a default for the commands
27810 that specify files for @value{GDBN} to operate on. @xref{Files,
27811 ,Commands to Specify Files}.
27813 By default, @kbd{M-x gdb} calls the program called @file{gdb}. If you
27814 need to call @value{GDBN} by a different name (for example, if you
27815 keep several configurations around, with different names) you can
27816 customize the Emacs variable @code{gud-gdb-command-name} to run the
27819 In the GUD buffer, you can use these special Emacs commands in
27820 addition to the standard Shell mode commands:
27824 Describe the features of Emacs' GUD Mode.
27827 Execute to another source line, like the @value{GDBN} @code{step} command; also
27828 update the display window to show the current file and location.
27831 Execute to next source line in this function, skipping all function
27832 calls, like the @value{GDBN} @code{next} command. Then update the display window
27833 to show the current file and location.
27836 Execute one instruction, like the @value{GDBN} @code{stepi} command; update
27837 display window accordingly.
27840 Execute until exit from the selected stack frame, like the @value{GDBN}
27841 @code{finish} command.
27844 Continue execution of your program, like the @value{GDBN} @code{continue}
27848 Go up the number of frames indicated by the numeric argument
27849 (@pxref{Arguments, , Numeric Arguments, Emacs, The @sc{gnu} Emacs Manual}),
27850 like the @value{GDBN} @code{up} command.
27853 Go down the number of frames indicated by the numeric argument, like the
27854 @value{GDBN} @code{down} command.
27857 In any source file, the Emacs command @kbd{C-x @key{SPC}} (@code{gud-break})
27858 tells @value{GDBN} to set a breakpoint on the source line point is on.
27860 In text command mode, if you type @kbd{M-x speedbar}, Emacs displays a
27861 separate frame which shows a backtrace when the GUD buffer is current.
27862 Move point to any frame in the stack and type @key{RET} to make it
27863 become the current frame and display the associated source in the
27864 source buffer. Alternatively, click @kbd{Mouse-2} to make the
27865 selected frame become the current one. In graphical mode, the
27866 speedbar displays watch expressions.
27868 If you accidentally delete the source-display buffer, an easy way to get
27869 it back is to type the command @code{f} in the @value{GDBN} buffer, to
27870 request a frame display; when you run under Emacs, this recreates
27871 the source buffer if necessary to show you the context of the current
27874 The source files displayed in Emacs are in ordinary Emacs buffers
27875 which are visiting the source files in the usual way. You can edit
27876 the files with these buffers if you wish; but keep in mind that @value{GDBN}
27877 communicates with Emacs in terms of line numbers. If you add or
27878 delete lines from the text, the line numbers that @value{GDBN} knows cease
27879 to correspond properly with the code.
27881 A more detailed description of Emacs' interaction with @value{GDBN} is
27882 given in the Emacs manual (@pxref{Debuggers,,, Emacs, The @sc{gnu}
27886 @chapter The @sc{gdb/mi} Interface
27888 @unnumberedsec Function and Purpose
27890 @cindex @sc{gdb/mi}, its purpose
27891 @sc{gdb/mi} is a line based machine oriented text interface to
27892 @value{GDBN} and is activated by specifying using the
27893 @option{--interpreter} command line option (@pxref{Mode Options}). It
27894 is specifically intended to support the development of systems which
27895 use the debugger as just one small component of a larger system.
27897 This chapter is a specification of the @sc{gdb/mi} interface. It is written
27898 in the form of a reference manual.
27900 Note that @sc{gdb/mi} is still under construction, so some of the
27901 features described below are incomplete and subject to change
27902 (@pxref{GDB/MI Development and Front Ends, , @sc{gdb/mi} Development and Front Ends}).
27904 @unnumberedsec Notation and Terminology
27906 @cindex notational conventions, for @sc{gdb/mi}
27907 This chapter uses the following notation:
27911 @code{|} separates two alternatives.
27914 @code{[ @var{something} ]} indicates that @var{something} is optional:
27915 it may or may not be given.
27918 @code{( @var{group} )*} means that @var{group} inside the parentheses
27919 may repeat zero or more times.
27922 @code{( @var{group} )+} means that @var{group} inside the parentheses
27923 may repeat one or more times.
27926 @code{"@var{string}"} means a literal @var{string}.
27930 @heading Dependencies
27934 * GDB/MI General Design::
27935 * GDB/MI Command Syntax::
27936 * GDB/MI Compatibility with CLI::
27937 * GDB/MI Development and Front Ends::
27938 * GDB/MI Output Records::
27939 * GDB/MI Simple Examples::
27940 * GDB/MI Command Description Format::
27941 * GDB/MI Breakpoint Commands::
27942 * GDB/MI Catchpoint Commands::
27943 * GDB/MI Program Context::
27944 * GDB/MI Thread Commands::
27945 * GDB/MI Ada Tasking Commands::
27946 * GDB/MI Program Execution::
27947 * GDB/MI Stack Manipulation::
27948 * GDB/MI Variable Objects::
27949 * GDB/MI Data Manipulation::
27950 * GDB/MI Tracepoint Commands::
27951 * GDB/MI Symbol Query::
27952 * GDB/MI File Commands::
27954 * GDB/MI Kod Commands::
27955 * GDB/MI Memory Overlay Commands::
27956 * GDB/MI Signal Handling Commands::
27958 * GDB/MI Target Manipulation::
27959 * GDB/MI File Transfer Commands::
27960 * GDB/MI Ada Exceptions Commands::
27961 * GDB/MI Support Commands::
27962 * GDB/MI Miscellaneous Commands::
27965 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
27966 @node GDB/MI General Design
27967 @section @sc{gdb/mi} General Design
27968 @cindex GDB/MI General Design
27970 Interaction of a @sc{GDB/MI} frontend with @value{GDBN} involves three
27971 parts---commands sent to @value{GDBN}, responses to those commands
27972 and notifications. Each command results in exactly one response,
27973 indicating either successful completion of the command, or an error.
27974 For the commands that do not resume the target, the response contains the
27975 requested information. For the commands that resume the target, the
27976 response only indicates whether the target was successfully resumed.
27977 Notifications is the mechanism for reporting changes in the state of the
27978 target, or in @value{GDBN} state, that cannot conveniently be associated with
27979 a command and reported as part of that command response.
27981 The important examples of notifications are:
27985 Exec notifications. These are used to report changes in
27986 target state---when a target is resumed, or stopped. It would not
27987 be feasible to include this information in response of resuming
27988 commands, because one resume commands can result in multiple events in
27989 different threads. Also, quite some time may pass before any event
27990 happens in the target, while a frontend needs to know whether the resuming
27991 command itself was successfully executed.
27994 Console output, and status notifications. Console output
27995 notifications are used to report output of CLI commands, as well as
27996 diagnostics for other commands. Status notifications are used to
27997 report the progress of a long-running operation. Naturally, including
27998 this information in command response would mean no output is produced
27999 until the command is finished, which is undesirable.
28002 General notifications. Commands may have various side effects on
28003 the @value{GDBN} or target state beyond their official purpose. For example,
28004 a command may change the selected thread. Although such changes can
28005 be included in command response, using notification allows for more
28006 orthogonal frontend design.
28010 There's no guarantee that whenever an MI command reports an error,
28011 @value{GDBN} or the target are in any specific state, and especially,
28012 the state is not reverted to the state before the MI command was
28013 processed. Therefore, whenever an MI command results in an error,
28014 we recommend that the frontend refreshes all the information shown in
28015 the user interface.
28019 * Context management::
28020 * Asynchronous and non-stop modes::
28024 @node Context management
28025 @subsection Context management
28027 @subsubsection Threads and Frames
28029 In most cases when @value{GDBN} accesses the target, this access is
28030 done in context of a specific thread and frame (@pxref{Frames}).
28031 Often, even when accessing global data, the target requires that a thread
28032 be specified. The CLI interface maintains the selected thread and frame,
28033 and supplies them to target on each command. This is convenient,
28034 because a command line user would not want to specify that information
28035 explicitly on each command, and because user interacts with
28036 @value{GDBN} via a single terminal, so no confusion is possible as
28037 to what thread and frame are the current ones.
28039 In the case of MI, the concept of selected thread and frame is less
28040 useful. First, a frontend can easily remember this information
28041 itself. Second, a graphical frontend can have more than one window,
28042 each one used for debugging a different thread, and the frontend might
28043 want to access additional threads for internal purposes. This
28044 increases the risk that by relying on implicitly selected thread, the
28045 frontend may be operating on a wrong one. Therefore, each MI command
28046 should explicitly specify which thread and frame to operate on. To
28047 make it possible, each MI command accepts the @samp{--thread} and
28048 @samp{--frame} options, the value to each is @value{GDBN} global
28049 identifier for thread and frame to operate on.
28051 Usually, each top-level window in a frontend allows the user to select
28052 a thread and a frame, and remembers the user selection for further
28053 operations. However, in some cases @value{GDBN} may suggest that the
28054 current thread or frame be changed. For example, when stopping on a
28055 breakpoint it is reasonable to switch to the thread where breakpoint is
28056 hit. For another example, if the user issues the CLI @samp{thread} or
28057 @samp{frame} commands via the frontend, it is desirable to change the
28058 frontend's selection to the one specified by user. @value{GDBN}
28059 communicates the suggestion to change current thread and frame using the
28060 @samp{=thread-selected} notification.
28062 Note that historically, MI shares the selected thread with CLI, so
28063 frontends used the @code{-thread-select} to execute commands in the
28064 right context. However, getting this to work right is cumbersome. The
28065 simplest way is for frontend to emit @code{-thread-select} command
28066 before every command. This doubles the number of commands that need
28067 to be sent. The alternative approach is to suppress @code{-thread-select}
28068 if the selected thread in @value{GDBN} is supposed to be identical to the
28069 thread the frontend wants to operate on. However, getting this
28070 optimization right can be tricky. In particular, if the frontend
28071 sends several commands to @value{GDBN}, and one of the commands changes the
28072 selected thread, then the behaviour of subsequent commands will
28073 change. So, a frontend should either wait for response from such
28074 problematic commands, or explicitly add @code{-thread-select} for
28075 all subsequent commands. No frontend is known to do this exactly
28076 right, so it is suggested to just always pass the @samp{--thread} and
28077 @samp{--frame} options.
28079 @subsubsection Language
28081 The execution of several commands depends on which language is selected.
28082 By default, the current language (@pxref{show language}) is used.
28083 But for commands known to be language-sensitive, it is recommended
28084 to use the @samp{--language} option. This option takes one argument,
28085 which is the name of the language to use while executing the command.
28089 -data-evaluate-expression --language c "sizeof (void*)"
28094 The valid language names are the same names accepted by the
28095 @samp{set language} command (@pxref{Manually}), excluding @samp{auto},
28096 @samp{local} or @samp{unknown}.
28098 @node Asynchronous and non-stop modes
28099 @subsection Asynchronous command execution and non-stop mode
28101 On some targets, @value{GDBN} is capable of processing MI commands
28102 even while the target is running. This is called @dfn{asynchronous
28103 command execution} (@pxref{Background Execution}). The frontend may
28104 specify a preferrence for asynchronous execution using the
28105 @code{-gdb-set mi-async 1} command, which should be emitted before
28106 either running the executable or attaching to the target. After the
28107 frontend has started the executable or attached to the target, it can
28108 find if asynchronous execution is enabled using the
28109 @code{-list-target-features} command.
28112 @item -gdb-set mi-async on
28113 @item -gdb-set mi-async off
28114 Set whether MI is in asynchronous mode.
28116 When @code{off}, which is the default, MI execution commands (e.g.,
28117 @code{-exec-continue}) are foreground commands, and @value{GDBN} waits
28118 for the program to stop before processing further commands.
28120 When @code{on}, MI execution commands are background execution
28121 commands (e.g., @code{-exec-continue} becomes the equivalent of the
28122 @code{c&} CLI command), and so @value{GDBN} is capable of processing
28123 MI commands even while the target is running.
28125 @item -gdb-show mi-async
28126 Show whether MI asynchronous mode is enabled.
28129 Note: In @value{GDBN} version 7.7 and earlier, this option was called
28130 @code{target-async} instead of @code{mi-async}, and it had the effect
28131 of both putting MI in asynchronous mode and making CLI background
28132 commands possible. CLI background commands are now always possible
28133 ``out of the box'' if the target supports them. The old spelling is
28134 kept as a deprecated alias for backwards compatibility.
28136 Even if @value{GDBN} can accept a command while target is running,
28137 many commands that access the target do not work when the target is
28138 running. Therefore, asynchronous command execution is most useful
28139 when combined with non-stop mode (@pxref{Non-Stop Mode}). Then,
28140 it is possible to examine the state of one thread, while other threads
28143 When a given thread is running, MI commands that try to access the
28144 target in the context of that thread may not work, or may work only on
28145 some targets. In particular, commands that try to operate on thread's
28146 stack will not work, on any target. Commands that read memory, or
28147 modify breakpoints, may work or not work, depending on the target. Note
28148 that even commands that operate on global state, such as @code{print},
28149 @code{set}, and breakpoint commands, still access the target in the
28150 context of a specific thread, so frontend should try to find a
28151 stopped thread and perform the operation on that thread (using the
28152 @samp{--thread} option).
28154 Which commands will work in the context of a running thread is
28155 highly target dependent. However, the two commands
28156 @code{-exec-interrupt}, to stop a thread, and @code{-thread-info},
28157 to find the state of a thread, will always work.
28159 @node Thread groups
28160 @subsection Thread groups
28161 @value{GDBN} may be used to debug several processes at the same time.
28162 On some platfroms, @value{GDBN} may support debugging of several
28163 hardware systems, each one having several cores with several different
28164 processes running on each core. This section describes the MI
28165 mechanism to support such debugging scenarios.
28167 The key observation is that regardless of the structure of the
28168 target, MI can have a global list of threads, because most commands that
28169 accept the @samp{--thread} option do not need to know what process that
28170 thread belongs to. Therefore, it is not necessary to introduce
28171 neither additional @samp{--process} option, nor an notion of the
28172 current process in the MI interface. The only strictly new feature
28173 that is required is the ability to find how the threads are grouped
28176 To allow the user to discover such grouping, and to support arbitrary
28177 hierarchy of machines/cores/processes, MI introduces the concept of a
28178 @dfn{thread group}. Thread group is a collection of threads and other
28179 thread groups. A thread group always has a string identifier, a type,
28180 and may have additional attributes specific to the type. A new
28181 command, @code{-list-thread-groups}, returns the list of top-level
28182 thread groups, which correspond to processes that @value{GDBN} is
28183 debugging at the moment. By passing an identifier of a thread group
28184 to the @code{-list-thread-groups} command, it is possible to obtain
28185 the members of specific thread group.
28187 To allow the user to easily discover processes, and other objects, he
28188 wishes to debug, a concept of @dfn{available thread group} is
28189 introduced. Available thread group is an thread group that
28190 @value{GDBN} is not debugging, but that can be attached to, using the
28191 @code{-target-attach} command. The list of available top-level thread
28192 groups can be obtained using @samp{-list-thread-groups --available}.
28193 In general, the content of a thread group may be only retrieved only
28194 after attaching to that thread group.
28196 Thread groups are related to inferiors (@pxref{Inferiors and
28197 Programs}). Each inferior corresponds to a thread group of a special
28198 type @samp{process}, and some additional operations are permitted on
28199 such thread groups.
28201 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
28202 @node GDB/MI Command Syntax
28203 @section @sc{gdb/mi} Command Syntax
28206 * GDB/MI Input Syntax::
28207 * GDB/MI Output Syntax::
28210 @node GDB/MI Input Syntax
28211 @subsection @sc{gdb/mi} Input Syntax
28213 @cindex input syntax for @sc{gdb/mi}
28214 @cindex @sc{gdb/mi}, input syntax
28216 @item @var{command} @expansion{}
28217 @code{@var{cli-command} | @var{mi-command}}
28219 @item @var{cli-command} @expansion{}
28220 @code{[ @var{token} ] @var{cli-command} @var{nl}}, where
28221 @var{cli-command} is any existing @value{GDBN} CLI command.
28223 @item @var{mi-command} @expansion{}
28224 @code{[ @var{token} ] "-" @var{operation} ( " " @var{option} )*
28225 @code{[} " --" @code{]} ( " " @var{parameter} )* @var{nl}}
28227 @item @var{token} @expansion{}
28228 "any sequence of digits"
28230 @item @var{option} @expansion{}
28231 @code{"-" @var{parameter} [ " " @var{parameter} ]}
28233 @item @var{parameter} @expansion{}
28234 @code{@var{non-blank-sequence} | @var{c-string}}
28236 @item @var{operation} @expansion{}
28237 @emph{any of the operations described in this chapter}
28239 @item @var{non-blank-sequence} @expansion{}
28240 @emph{anything, provided it doesn't contain special characters such as
28241 "-", @var{nl}, """ and of course " "}
28243 @item @var{c-string} @expansion{}
28244 @code{""" @var{seven-bit-iso-c-string-content} """}
28246 @item @var{nl} @expansion{}
28255 The CLI commands are still handled by the @sc{mi} interpreter; their
28256 output is described below.
28259 The @code{@var{token}}, when present, is passed back when the command
28263 Some @sc{mi} commands accept optional arguments as part of the parameter
28264 list. Each option is identified by a leading @samp{-} (dash) and may be
28265 followed by an optional argument parameter. Options occur first in the
28266 parameter list and can be delimited from normal parameters using
28267 @samp{--} (this is useful when some parameters begin with a dash).
28274 We want easy access to the existing CLI syntax (for debugging).
28277 We want it to be easy to spot a @sc{mi} operation.
28280 @node GDB/MI Output Syntax
28281 @subsection @sc{gdb/mi} Output Syntax
28283 @cindex output syntax of @sc{gdb/mi}
28284 @cindex @sc{gdb/mi}, output syntax
28285 The output from @sc{gdb/mi} consists of zero or more out-of-band records
28286 followed, optionally, by a single result record. This result record
28287 is for the most recent command. The sequence of output records is
28288 terminated by @samp{(gdb)}.
28290 If an input command was prefixed with a @code{@var{token}} then the
28291 corresponding output for that command will also be prefixed by that same
28295 @item @var{output} @expansion{}
28296 @code{( @var{out-of-band-record} )* [ @var{result-record} ] "(gdb)" @var{nl}}
28298 @item @var{result-record} @expansion{}
28299 @code{ [ @var{token} ] "^" @var{result-class} ( "," @var{result} )* @var{nl}}
28301 @item @var{out-of-band-record} @expansion{}
28302 @code{@var{async-record} | @var{stream-record}}
28304 @item @var{async-record} @expansion{}
28305 @code{@var{exec-async-output} | @var{status-async-output} | @var{notify-async-output}}
28307 @item @var{exec-async-output} @expansion{}
28308 @code{[ @var{token} ] "*" @var{async-output nl}}
28310 @item @var{status-async-output} @expansion{}
28311 @code{[ @var{token} ] "+" @var{async-output nl}}
28313 @item @var{notify-async-output} @expansion{}
28314 @code{[ @var{token} ] "=" @var{async-output nl}}
28316 @item @var{async-output} @expansion{}
28317 @code{@var{async-class} ( "," @var{result} )*}
28319 @item @var{result-class} @expansion{}
28320 @code{"done" | "running" | "connected" | "error" | "exit"}
28322 @item @var{async-class} @expansion{}
28323 @code{"stopped" | @var{others}} (where @var{others} will be added
28324 depending on the needs---this is still in development).
28326 @item @var{result} @expansion{}
28327 @code{ @var{variable} "=" @var{value}}
28329 @item @var{variable} @expansion{}
28330 @code{ @var{string} }
28332 @item @var{value} @expansion{}
28333 @code{ @var{const} | @var{tuple} | @var{list} }
28335 @item @var{const} @expansion{}
28336 @code{@var{c-string}}
28338 @item @var{tuple} @expansion{}
28339 @code{ "@{@}" | "@{" @var{result} ( "," @var{result} )* "@}" }
28341 @item @var{list} @expansion{}
28342 @code{ "[]" | "[" @var{value} ( "," @var{value} )* "]" | "["
28343 @var{result} ( "," @var{result} )* "]" }
28345 @item @var{stream-record} @expansion{}
28346 @code{@var{console-stream-output} | @var{target-stream-output} | @var{log-stream-output}}
28348 @item @var{console-stream-output} @expansion{}
28349 @code{"~" @var{c-string nl}}
28351 @item @var{target-stream-output} @expansion{}
28352 @code{"@@" @var{c-string nl}}
28354 @item @var{log-stream-output} @expansion{}
28355 @code{"&" @var{c-string nl}}
28357 @item @var{nl} @expansion{}
28360 @item @var{token} @expansion{}
28361 @emph{any sequence of digits}.
28369 All output sequences end in a single line containing a period.
28372 The @code{@var{token}} is from the corresponding request. Note that
28373 for all async output, while the token is allowed by the grammar and
28374 may be output by future versions of @value{GDBN} for select async
28375 output messages, it is generally omitted. Frontends should treat
28376 all async output as reporting general changes in the state of the
28377 target and there should be no need to associate async output to any
28381 @cindex status output in @sc{gdb/mi}
28382 @var{status-async-output} contains on-going status information about the
28383 progress of a slow operation. It can be discarded. All status output is
28384 prefixed by @samp{+}.
28387 @cindex async output in @sc{gdb/mi}
28388 @var{exec-async-output} contains asynchronous state change on the target
28389 (stopped, started, disappeared). All async output is prefixed by
28393 @cindex notify output in @sc{gdb/mi}
28394 @var{notify-async-output} contains supplementary information that the
28395 client should handle (e.g., a new breakpoint information). All notify
28396 output is prefixed by @samp{=}.
28399 @cindex console output in @sc{gdb/mi}
28400 @var{console-stream-output} is output that should be displayed as is in the
28401 console. It is the textual response to a CLI command. All the console
28402 output is prefixed by @samp{~}.
28405 @cindex target output in @sc{gdb/mi}
28406 @var{target-stream-output} is the output produced by the target program.
28407 All the target output is prefixed by @samp{@@}.
28410 @cindex log output in @sc{gdb/mi}
28411 @var{log-stream-output} is output text coming from @value{GDBN}'s internals, for
28412 instance messages that should be displayed as part of an error log. All
28413 the log output is prefixed by @samp{&}.
28416 @cindex list output in @sc{gdb/mi}
28417 New @sc{gdb/mi} commands should only output @var{lists} containing
28423 @xref{GDB/MI Stream Records, , @sc{gdb/mi} Stream Records}, for more
28424 details about the various output records.
28426 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
28427 @node GDB/MI Compatibility with CLI
28428 @section @sc{gdb/mi} Compatibility with CLI
28430 @cindex compatibility, @sc{gdb/mi} and CLI
28431 @cindex @sc{gdb/mi}, compatibility with CLI
28433 For the developers convenience CLI commands can be entered directly,
28434 but there may be some unexpected behaviour. For example, commands
28435 that query the user will behave as if the user replied yes, breakpoint
28436 command lists are not executed and some CLI commands, such as
28437 @code{if}, @code{when} and @code{define}, prompt for further input with
28438 @samp{>}, which is not valid MI output.
28440 This feature may be removed at some stage in the future and it is
28441 recommended that front ends use the @code{-interpreter-exec} command
28442 (@pxref{-interpreter-exec}).
28444 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
28445 @node GDB/MI Development and Front Ends
28446 @section @sc{gdb/mi} Development and Front Ends
28447 @cindex @sc{gdb/mi} development
28449 The application which takes the MI output and presents the state of the
28450 program being debugged to the user is called a @dfn{front end}.
28452 Since @sc{gdb/mi} is used by a variety of front ends to @value{GDBN}, changes
28453 to the MI interface may break existing usage. This section describes how the
28454 protocol changes and how to request previous version of the protocol when it
28457 Some changes in MI need not break a carefully designed front end, and
28458 for these the MI version will remain unchanged. The following is a
28459 list of changes that may occur within one level, so front ends should
28460 parse MI output in a way that can handle them:
28464 New MI commands may be added.
28467 New fields may be added to the output of any MI command.
28470 The range of values for fields with specified values, e.g.,
28471 @code{in_scope} (@pxref{-var-update}) may be extended.
28473 @c The format of field's content e.g type prefix, may change so parse it
28474 @c at your own risk. Yes, in general?
28476 @c The order of fields may change? Shouldn't really matter but it might
28477 @c resolve inconsistencies.
28480 If the changes are likely to break front ends, the MI version level
28481 will be increased by one. The new versions of the MI protocol are not compatible
28482 with the old versions. Old versions of MI remain available, allowing front ends
28483 to keep using them until they are modified to use the latest MI version.
28485 Since @code{--interpreter=mi} always points to the latest MI version, it is
28486 recommended that front ends request a specific version of MI when launching
28487 @value{GDBN} (e.g. @code{--interpreter=mi2}) to make sure they get an
28488 interpreter with the MI version they expect.
28490 The following table gives a summary of the the released versions of the MI
28491 interface: the version number, the version of GDB in which it first appeared
28492 and the breaking changes compared to the previous version.
28494 @multitable @columnfractions .05 .05 .9
28495 @headitem MI version @tab GDB version @tab Breaking changes
28512 The @code{-environment-pwd}, @code{-environment-directory} and
28513 @code{-environment-path} commands now returns values using the MI output
28514 syntax, rather than CLI output syntax.
28517 @code{-var-list-children}'s @code{children} result field is now a list, rather
28521 @code{-var-update}'s @code{changelist} result field is now a list, rather than
28533 The output of information about multi-location breakpoints has changed in the
28534 responses to the @code{-break-insert} and @code{-break-info} commands, as well
28535 as in the @code{=breakpoint-created} and @code{=breakpoint-modified} events.
28536 The multiple locations are now placed in a @code{locations} field, whose value
28542 If your front end cannot yet migrate to a more recent version of the
28543 MI protocol, you can nevertheless selectively enable specific features
28544 available in those recent MI versions, using the following commands:
28548 @item -fix-multi-location-breakpoint-output
28549 Use the output for multi-location breakpoints which was introduced by
28550 MI 3, even when using MI versions 2 or 1. This command has no
28551 effect when using MI version 3 or later.
28555 The best way to avoid unexpected changes in MI that might break your front
28556 end is to make your project known to @value{GDBN} developers and
28557 follow development on @email{gdb@@sourceware.org} and
28558 @email{gdb-patches@@sourceware.org}.
28559 @cindex mailing lists
28561 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
28562 @node GDB/MI Output Records
28563 @section @sc{gdb/mi} Output Records
28566 * GDB/MI Result Records::
28567 * GDB/MI Stream Records::
28568 * GDB/MI Async Records::
28569 * GDB/MI Breakpoint Information::
28570 * GDB/MI Frame Information::
28571 * GDB/MI Thread Information::
28572 * GDB/MI Ada Exception Information::
28575 @node GDB/MI Result Records
28576 @subsection @sc{gdb/mi} Result Records
28578 @cindex result records in @sc{gdb/mi}
28579 @cindex @sc{gdb/mi}, result records
28580 In addition to a number of out-of-band notifications, the response to a
28581 @sc{gdb/mi} command includes one of the following result indications:
28585 @item "^done" [ "," @var{results} ]
28586 The synchronous operation was successful, @code{@var{results}} are the return
28591 This result record is equivalent to @samp{^done}. Historically, it
28592 was output instead of @samp{^done} if the command has resumed the
28593 target. This behaviour is maintained for backward compatibility, but
28594 all frontends should treat @samp{^done} and @samp{^running}
28595 identically and rely on the @samp{*running} output record to determine
28596 which threads are resumed.
28600 @value{GDBN} has connected to a remote target.
28602 @item "^error" "," "msg=" @var{c-string} [ "," "code=" @var{c-string} ]
28604 The operation failed. The @code{msg=@var{c-string}} variable contains
28605 the corresponding error message.
28607 If present, the @code{code=@var{c-string}} variable provides an error
28608 code on which consumers can rely on to detect the corresponding
28609 error condition. At present, only one error code is defined:
28612 @item "undefined-command"
28613 Indicates that the command causing the error does not exist.
28618 @value{GDBN} has terminated.
28622 @node GDB/MI Stream Records
28623 @subsection @sc{gdb/mi} Stream Records
28625 @cindex @sc{gdb/mi}, stream records
28626 @cindex stream records in @sc{gdb/mi}
28627 @value{GDBN} internally maintains a number of output streams: the console, the
28628 target, and the log. The output intended for each of these streams is
28629 funneled through the @sc{gdb/mi} interface using @dfn{stream records}.
28631 Each stream record begins with a unique @dfn{prefix character} which
28632 identifies its stream (@pxref{GDB/MI Output Syntax, , @sc{gdb/mi} Output
28633 Syntax}). In addition to the prefix, each stream record contains a
28634 @code{@var{string-output}}. This is either raw text (with an implicit new
28635 line) or a quoted C string (which does not contain an implicit newline).
28638 @item "~" @var{string-output}
28639 The console output stream contains text that should be displayed in the
28640 CLI console window. It contains the textual responses to CLI commands.
28642 @item "@@" @var{string-output}
28643 The target output stream contains any textual output from the running
28644 target. This is only present when GDB's event loop is truly
28645 asynchronous, which is currently only the case for remote targets.
28647 @item "&" @var{string-output}
28648 The log stream contains debugging messages being produced by @value{GDBN}'s
28652 @node GDB/MI Async Records
28653 @subsection @sc{gdb/mi} Async Records
28655 @cindex async records in @sc{gdb/mi}
28656 @cindex @sc{gdb/mi}, async records
28657 @dfn{Async} records are used to notify the @sc{gdb/mi} client of
28658 additional changes that have occurred. Those changes can either be a
28659 consequence of @sc{gdb/mi} commands (e.g., a breakpoint modified) or a result of
28660 target activity (e.g., target stopped).
28662 The following is the list of possible async records:
28666 @item *running,thread-id="@var{thread}"
28667 The target is now running. The @var{thread} field can be the global
28668 thread ID of the the thread that is now running, and it can be
28669 @samp{all} if all threads are running. The frontend should assume
28670 that no interaction with a running thread is possible after this
28671 notification is produced. The frontend should not assume that this
28672 notification is output only once for any command. @value{GDBN} may
28673 emit this notification several times, either for different threads,
28674 because it cannot resume all threads together, or even for a single
28675 thread, if the thread must be stepped though some code before letting
28678 @item *stopped,reason="@var{reason}",thread-id="@var{id}",stopped-threads="@var{stopped}",core="@var{core}"
28679 The target has stopped. The @var{reason} field can have one of the
28683 @item breakpoint-hit
28684 A breakpoint was reached.
28685 @item watchpoint-trigger
28686 A watchpoint was triggered.
28687 @item read-watchpoint-trigger
28688 A read watchpoint was triggered.
28689 @item access-watchpoint-trigger
28690 An access watchpoint was triggered.
28691 @item function-finished
28692 An -exec-finish or similar CLI command was accomplished.
28693 @item location-reached
28694 An -exec-until or similar CLI command was accomplished.
28695 @item watchpoint-scope
28696 A watchpoint has gone out of scope.
28697 @item end-stepping-range
28698 An -exec-next, -exec-next-instruction, -exec-step, -exec-step-instruction or
28699 similar CLI command was accomplished.
28700 @item exited-signalled
28701 The inferior exited because of a signal.
28703 The inferior exited.
28704 @item exited-normally
28705 The inferior exited normally.
28706 @item signal-received
28707 A signal was received by the inferior.
28709 The inferior has stopped due to a library being loaded or unloaded.
28710 This can happen when @code{stop-on-solib-events} (@pxref{Files}) is
28711 set or when a @code{catch load} or @code{catch unload} catchpoint is
28712 in use (@pxref{Set Catchpoints}).
28714 The inferior has forked. This is reported when @code{catch fork}
28715 (@pxref{Set Catchpoints}) has been used.
28717 The inferior has vforked. This is reported in when @code{catch vfork}
28718 (@pxref{Set Catchpoints}) has been used.
28719 @item syscall-entry
28720 The inferior entered a system call. This is reported when @code{catch
28721 syscall} (@pxref{Set Catchpoints}) has been used.
28722 @item syscall-return
28723 The inferior returned from a system call. This is reported when
28724 @code{catch syscall} (@pxref{Set Catchpoints}) has been used.
28726 The inferior called @code{exec}. This is reported when @code{catch exec}
28727 (@pxref{Set Catchpoints}) has been used.
28730 The @var{id} field identifies the global thread ID of the thread
28731 that directly caused the stop -- for example by hitting a breakpoint.
28732 Depending on whether all-stop
28733 mode is in effect (@pxref{All-Stop Mode}), @value{GDBN} may either
28734 stop all threads, or only the thread that directly triggered the stop.
28735 If all threads are stopped, the @var{stopped} field will have the
28736 value of @code{"all"}. Otherwise, the value of the @var{stopped}
28737 field will be a list of thread identifiers. Presently, this list will
28738 always include a single thread, but frontend should be prepared to see
28739 several threads in the list. The @var{core} field reports the
28740 processor core on which the stop event has happened. This field may be absent
28741 if such information is not available.
28743 @item =thread-group-added,id="@var{id}"
28744 @itemx =thread-group-removed,id="@var{id}"
28745 A thread group was either added or removed. The @var{id} field
28746 contains the @value{GDBN} identifier of the thread group. When a thread
28747 group is added, it generally might not be associated with a running
28748 process. When a thread group is removed, its id becomes invalid and
28749 cannot be used in any way.
28751 @item =thread-group-started,id="@var{id}",pid="@var{pid}"
28752 A thread group became associated with a running program,
28753 either because the program was just started or the thread group
28754 was attached to a program. The @var{id} field contains the
28755 @value{GDBN} identifier of the thread group. The @var{pid} field
28756 contains process identifier, specific to the operating system.
28758 @item =thread-group-exited,id="@var{id}"[,exit-code="@var{code}"]
28759 A thread group is no longer associated with a running program,
28760 either because the program has exited, or because it was detached
28761 from. The @var{id} field contains the @value{GDBN} identifier of the
28762 thread group. The @var{code} field is the exit code of the inferior; it exists
28763 only when the inferior exited with some code.
28765 @item =thread-created,id="@var{id}",group-id="@var{gid}"
28766 @itemx =thread-exited,id="@var{id}",group-id="@var{gid}"
28767 A thread either was created, or has exited. The @var{id} field
28768 contains the global @value{GDBN} identifier of the thread. The @var{gid}
28769 field identifies the thread group this thread belongs to.
28771 @item =thread-selected,id="@var{id}"[,frame="@var{frame}"]
28772 Informs that the selected thread or frame were changed. This notification
28773 is not emitted as result of the @code{-thread-select} or
28774 @code{-stack-select-frame} commands, but is emitted whenever an MI command
28775 that is not documented to change the selected thread and frame actually
28776 changes them. In particular, invoking, directly or indirectly
28777 (via user-defined command), the CLI @code{thread} or @code{frame} commands,
28778 will generate this notification. Changing the thread or frame from another
28779 user interface (see @ref{Interpreters}) will also generate this notification.
28781 The @var{frame} field is only present if the newly selected thread is
28782 stopped. See @ref{GDB/MI Frame Information} for the format of its value.
28784 We suggest that in response to this notification, front ends
28785 highlight the selected thread and cause subsequent commands to apply to
28788 @item =library-loaded,...
28789 Reports that a new library file was loaded by the program. This
28790 notification has 5 fields---@var{id}, @var{target-name},
28791 @var{host-name}, @var{symbols-loaded} and @var{ranges}. The @var{id} field is an
28792 opaque identifier of the library. For remote debugging case,
28793 @var{target-name} and @var{host-name} fields give the name of the
28794 library file on the target, and on the host respectively. For native
28795 debugging, both those fields have the same value. The
28796 @var{symbols-loaded} field is emitted only for backward compatibility
28797 and should not be relied on to convey any useful information. The
28798 @var{thread-group} field, if present, specifies the id of the thread
28799 group in whose context the library was loaded. If the field is
28800 absent, it means the library was loaded in the context of all present
28801 thread groups. The @var{ranges} field specifies the ranges of addresses belonging
28804 @item =library-unloaded,...
28805 Reports that a library was unloaded by the program. This notification
28806 has 3 fields---@var{id}, @var{target-name} and @var{host-name} with
28807 the same meaning as for the @code{=library-loaded} notification.
28808 The @var{thread-group} field, if present, specifies the id of the
28809 thread group in whose context the library was unloaded. If the field is
28810 absent, it means the library was unloaded in the context of all present
28813 @item =traceframe-changed,num=@var{tfnum},tracepoint=@var{tpnum}
28814 @itemx =traceframe-changed,end
28815 Reports that the trace frame was changed and its new number is
28816 @var{tfnum}. The number of the tracepoint associated with this trace
28817 frame is @var{tpnum}.
28819 @item =tsv-created,name=@var{name},initial=@var{initial}
28820 Reports that the new trace state variable @var{name} is created with
28821 initial value @var{initial}.
28823 @item =tsv-deleted,name=@var{name}
28824 @itemx =tsv-deleted
28825 Reports that the trace state variable @var{name} is deleted or all
28826 trace state variables are deleted.
28828 @item =tsv-modified,name=@var{name},initial=@var{initial}[,current=@var{current}]
28829 Reports that the trace state variable @var{name} is modified with
28830 the initial value @var{initial}. The current value @var{current} of
28831 trace state variable is optional and is reported if the current
28832 value of trace state variable is known.
28834 @item =breakpoint-created,bkpt=@{...@}
28835 @itemx =breakpoint-modified,bkpt=@{...@}
28836 @itemx =breakpoint-deleted,id=@var{number}
28837 Reports that a breakpoint was created, modified, or deleted,
28838 respectively. Only user-visible breakpoints are reported to the MI
28841 The @var{bkpt} argument is of the same form as returned by the various
28842 breakpoint commands; @xref{GDB/MI Breakpoint Commands}. The
28843 @var{number} is the ordinal number of the breakpoint.
28845 Note that if a breakpoint is emitted in the result record of a
28846 command, then it will not also be emitted in an async record.
28848 @item =record-started,thread-group="@var{id}",method="@var{method}"[,format="@var{format}"]
28849 @itemx =record-stopped,thread-group="@var{id}"
28850 Execution log recording was either started or stopped on an
28851 inferior. The @var{id} is the @value{GDBN} identifier of the thread
28852 group corresponding to the affected inferior.
28854 The @var{method} field indicates the method used to record execution. If the
28855 method in use supports multiple recording formats, @var{format} will be present
28856 and contain the currently used format. @xref{Process Record and Replay},
28857 for existing method and format values.
28859 @item =cmd-param-changed,param=@var{param},value=@var{value}
28860 Reports that a parameter of the command @code{set @var{param}} is
28861 changed to @var{value}. In the multi-word @code{set} command,
28862 the @var{param} is the whole parameter list to @code{set} command.
28863 For example, In command @code{set check type on}, @var{param}
28864 is @code{check type} and @var{value} is @code{on}.
28866 @item =memory-changed,thread-group=@var{id},addr=@var{addr},len=@var{len}[,type="code"]
28867 Reports that bytes from @var{addr} to @var{data} + @var{len} were
28868 written in an inferior. The @var{id} is the identifier of the
28869 thread group corresponding to the affected inferior. The optional
28870 @code{type="code"} part is reported if the memory written to holds
28874 @node GDB/MI Breakpoint Information
28875 @subsection @sc{gdb/mi} Breakpoint Information
28877 When @value{GDBN} reports information about a breakpoint, a
28878 tracepoint, a watchpoint, or a catchpoint, it uses a tuple with the
28883 The breakpoint number.
28886 The type of the breakpoint. For ordinary breakpoints this will be
28887 @samp{breakpoint}, but many values are possible.
28890 If the type of the breakpoint is @samp{catchpoint}, then this
28891 indicates the exact type of catchpoint.
28894 This is the breakpoint disposition---either @samp{del}, meaning that
28895 the breakpoint will be deleted at the next stop, or @samp{keep},
28896 meaning that the breakpoint will not be deleted.
28899 This indicates whether the breakpoint is enabled, in which case the
28900 value is @samp{y}, or disabled, in which case the value is @samp{n}.
28901 Note that this is not the same as the field @code{enable}.
28904 The address of the breakpoint. This may be a hexidecimal number,
28905 giving the address; or the string @samp{<PENDING>}, for a pending
28906 breakpoint; or the string @samp{<MULTIPLE>}, for a breakpoint with
28907 multiple locations. This field will not be present if no address can
28908 be determined. For example, a watchpoint does not have an address.
28911 If known, the function in which the breakpoint appears.
28912 If not known, this field is not present.
28915 The name of the source file which contains this function, if known.
28916 If not known, this field is not present.
28919 The full file name of the source file which contains this function, if
28920 known. If not known, this field is not present.
28923 The line number at which this breakpoint appears, if known.
28924 If not known, this field is not present.
28927 If the source file is not known, this field may be provided. If
28928 provided, this holds the address of the breakpoint, possibly followed
28932 If this breakpoint is pending, this field is present and holds the
28933 text used to set the breakpoint, as entered by the user.
28936 Where this breakpoint's condition is evaluated, either @samp{host} or
28940 If this is a thread-specific breakpoint, then this identifies the
28941 thread in which the breakpoint can trigger.
28944 If this breakpoint is restricted to a particular Ada task, then this
28945 field will hold the task identifier.
28948 If the breakpoint is conditional, this is the condition expression.
28951 The ignore count of the breakpoint.
28954 The enable count of the breakpoint.
28956 @item traceframe-usage
28959 @item static-tracepoint-marker-string-id
28960 For a static tracepoint, the name of the static tracepoint marker.
28963 For a masked watchpoint, this is the mask.
28966 A tracepoint's pass count.
28968 @item original-location
28969 The location of the breakpoint as originally specified by the user.
28970 This field is optional.
28973 The number of times the breakpoint has been hit.
28976 This field is only given for tracepoints. This is either @samp{y},
28977 meaning that the tracepoint is installed, or @samp{n}, meaning that it
28981 Some extra data, the exact contents of which are type-dependent.
28984 This field is present if the breakpoint has multiple locations. It is also
28985 exceptionally present if the breakpoint is enabled and has a single, disabled
28988 The value is a list of locations. The format of a location is decribed below.
28992 A location in a multi-location breakpoint is represented as a tuple with the
28998 The location number as a dotted pair, like @samp{1.2}. The first digit is the
28999 number of the parent breakpoint. The second digit is the number of the
29000 location within that breakpoint.
29003 This indicates whether the location is enabled, in which case the
29004 value is @samp{y}, or disabled, in which case the value is @samp{n}.
29005 Note that this is not the same as the field @code{enable}.
29008 The address of this location as an hexidecimal number.
29011 If known, the function in which the location appears.
29012 If not known, this field is not present.
29015 The name of the source file which contains this location, if known.
29016 If not known, this field is not present.
29019 The full file name of the source file which contains this location, if
29020 known. If not known, this field is not present.
29023 The line number at which this location appears, if known.
29024 If not known, this field is not present.
29026 @item thread-groups
29027 The thread groups this location is in.
29031 For example, here is what the output of @code{-break-insert}
29032 (@pxref{GDB/MI Breakpoint Commands}) might be:
29035 -> -break-insert main
29036 <- ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
29037 enabled="y",addr="0x08048564",func="main",file="myprog.c",
29038 fullname="/home/nickrob/myprog.c",line="68",thread-groups=["i1"],
29043 @node GDB/MI Frame Information
29044 @subsection @sc{gdb/mi} Frame Information
29046 Response from many MI commands includes an information about stack
29047 frame. This information is a tuple that may have the following
29052 The level of the stack frame. The innermost frame has the level of
29053 zero. This field is always present.
29056 The name of the function corresponding to the frame. This field may
29057 be absent if @value{GDBN} is unable to determine the function name.
29060 The code address for the frame. This field is always present.
29063 The name of the source files that correspond to the frame's code
29064 address. This field may be absent.
29067 The source line corresponding to the frames' code address. This field
29071 The name of the binary file (either executable or shared library) the
29072 corresponds to the frame's code address. This field may be absent.
29076 @node GDB/MI Thread Information
29077 @subsection @sc{gdb/mi} Thread Information
29079 Whenever @value{GDBN} has to report an information about a thread, it
29080 uses a tuple with the following fields. The fields are always present unless
29085 The global numeric id assigned to the thread by @value{GDBN}.
29088 The target-specific string identifying the thread.
29091 Additional information about the thread provided by the target.
29092 It is supposed to be human-readable and not interpreted by the
29093 frontend. This field is optional.
29096 The name of the thread. If the user specified a name using the
29097 @code{thread name} command, then this name is given. Otherwise, if
29098 @value{GDBN} can extract the thread name from the target, then that
29099 name is given. If @value{GDBN} cannot find the thread name, then this
29103 The execution state of the thread, either @samp{stopped} or @samp{running},
29104 depending on whether the thread is presently running.
29107 The stack frame currently executing in the thread. This field is only present
29108 if the thread is stopped. Its format is documented in
29109 @ref{GDB/MI Frame Information}.
29112 The value of this field is an integer number of the processor core the
29113 thread was last seen on. This field is optional.
29116 @node GDB/MI Ada Exception Information
29117 @subsection @sc{gdb/mi} Ada Exception Information
29119 Whenever a @code{*stopped} record is emitted because the program
29120 stopped after hitting an exception catchpoint (@pxref{Set Catchpoints}),
29121 @value{GDBN} provides the name of the exception that was raised via
29122 the @code{exception-name} field. Also, for exceptions that were raised
29123 with an exception message, @value{GDBN} provides that message via
29124 the @code{exception-message} field.
29126 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
29127 @node GDB/MI Simple Examples
29128 @section Simple Examples of @sc{gdb/mi} Interaction
29129 @cindex @sc{gdb/mi}, simple examples
29131 This subsection presents several simple examples of interaction using
29132 the @sc{gdb/mi} interface. In these examples, @samp{->} means that the
29133 following line is passed to @sc{gdb/mi} as input, while @samp{<-} means
29134 the output received from @sc{gdb/mi}.
29136 Note the line breaks shown in the examples are here only for
29137 readability, they don't appear in the real output.
29139 @subheading Setting a Breakpoint
29141 Setting a breakpoint generates synchronous output which contains detailed
29142 information of the breakpoint.
29145 -> -break-insert main
29146 <- ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
29147 enabled="y",addr="0x08048564",func="main",file="myprog.c",
29148 fullname="/home/nickrob/myprog.c",line="68",thread-groups=["i1"],
29153 @subheading Program Execution
29155 Program execution generates asynchronous records and MI gives the
29156 reason that execution stopped.
29162 <- *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",thread-id="0",
29163 frame=@{addr="0x08048564",func="main",
29164 args=[@{name="argc",value="1"@},@{name="argv",value="0xbfc4d4d4"@}],
29165 file="myprog.c",fullname="/home/nickrob/myprog.c",line="68",
29166 arch="i386:x86_64"@}
29171 <- *stopped,reason="exited-normally"
29175 @subheading Quitting @value{GDBN}
29177 Quitting @value{GDBN} just prints the result class @samp{^exit}.
29185 Please note that @samp{^exit} is printed immediately, but it might
29186 take some time for @value{GDBN} to actually exit. During that time, @value{GDBN}
29187 performs necessary cleanups, including killing programs being debugged
29188 or disconnecting from debug hardware, so the frontend should wait till
29189 @value{GDBN} exits and should only forcibly kill @value{GDBN} if it
29190 fails to exit in reasonable time.
29192 @subheading A Bad Command
29194 Here's what happens if you pass a non-existent command:
29198 <- ^error,msg="Undefined MI command: rubbish"
29203 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
29204 @node GDB/MI Command Description Format
29205 @section @sc{gdb/mi} Command Description Format
29207 The remaining sections describe blocks of commands. Each block of
29208 commands is laid out in a fashion similar to this section.
29210 @subheading Motivation
29212 The motivation for this collection of commands.
29214 @subheading Introduction
29216 A brief introduction to this collection of commands as a whole.
29218 @subheading Commands
29220 For each command in the block, the following is described:
29222 @subsubheading Synopsis
29225 -command @var{args}@dots{}
29228 @subsubheading Result
29230 @subsubheading @value{GDBN} Command
29232 The corresponding @value{GDBN} CLI command(s), if any.
29234 @subsubheading Example
29236 Example(s) formatted for readability. Some of the described commands have
29237 not been implemented yet and these are labeled N.A.@: (not available).
29240 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
29241 @node GDB/MI Breakpoint Commands
29242 @section @sc{gdb/mi} Breakpoint Commands
29244 @cindex breakpoint commands for @sc{gdb/mi}
29245 @cindex @sc{gdb/mi}, breakpoint commands
29246 This section documents @sc{gdb/mi} commands for manipulating
29249 @subheading The @code{-break-after} Command
29250 @findex -break-after
29252 @subsubheading Synopsis
29255 -break-after @var{number} @var{count}
29258 The breakpoint number @var{number} is not in effect until it has been
29259 hit @var{count} times. To see how this is reflected in the output of
29260 the @samp{-break-list} command, see the description of the
29261 @samp{-break-list} command below.
29263 @subsubheading @value{GDBN} Command
29265 The corresponding @value{GDBN} command is @samp{ignore}.
29267 @subsubheading Example
29272 ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
29273 enabled="y",addr="0x000100d0",func="main",file="hello.c",
29274 fullname="/home/foo/hello.c",line="5",thread-groups=["i1"],
29282 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
29283 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
29284 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
29285 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
29286 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
29287 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
29288 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
29289 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
29290 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
29291 line="5",thread-groups=["i1"],times="0",ignore="3"@}]@}
29296 @subheading The @code{-break-catch} Command
29297 @findex -break-catch
29300 @subheading The @code{-break-commands} Command
29301 @findex -break-commands
29303 @subsubheading Synopsis
29306 -break-commands @var{number} [ @var{command1} ... @var{commandN} ]
29309 Specifies the CLI commands that should be executed when breakpoint
29310 @var{number} is hit. The parameters @var{command1} to @var{commandN}
29311 are the commands. If no command is specified, any previously-set
29312 commands are cleared. @xref{Break Commands}. Typical use of this
29313 functionality is tracing a program, that is, printing of values of
29314 some variables whenever breakpoint is hit and then continuing.
29316 @subsubheading @value{GDBN} Command
29318 The corresponding @value{GDBN} command is @samp{commands}.
29320 @subsubheading Example
29325 ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
29326 enabled="y",addr="0x000100d0",func="main",file="hello.c",
29327 fullname="/home/foo/hello.c",line="5",thread-groups=["i1"],
29330 -break-commands 1 "print v" "continue"
29335 @subheading The @code{-break-condition} Command
29336 @findex -break-condition
29338 @subsubheading Synopsis
29341 -break-condition @var{number} @var{expr}
29344 Breakpoint @var{number} will stop the program only if the condition in
29345 @var{expr} is true. The condition becomes part of the
29346 @samp{-break-list} output (see the description of the @samp{-break-list}
29349 @subsubheading @value{GDBN} Command
29351 The corresponding @value{GDBN} command is @samp{condition}.
29353 @subsubheading Example
29357 -break-condition 1 1
29361 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
29362 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
29363 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
29364 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
29365 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
29366 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
29367 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
29368 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
29369 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
29370 line="5",cond="1",thread-groups=["i1"],times="0",ignore="3"@}]@}
29374 @subheading The @code{-break-delete} Command
29375 @findex -break-delete
29377 @subsubheading Synopsis
29380 -break-delete ( @var{breakpoint} )+
29383 Delete the breakpoint(s) whose number(s) are specified in the argument
29384 list. This is obviously reflected in the breakpoint list.
29386 @subsubheading @value{GDBN} Command
29388 The corresponding @value{GDBN} command is @samp{delete}.
29390 @subsubheading Example
29398 ^done,BreakpointTable=@{nr_rows="0",nr_cols="6",
29399 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
29400 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
29401 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
29402 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
29403 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
29404 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
29409 @subheading The @code{-break-disable} Command
29410 @findex -break-disable
29412 @subsubheading Synopsis
29415 -break-disable ( @var{breakpoint} )+
29418 Disable the named @var{breakpoint}(s). The field @samp{enabled} in the
29419 break list is now set to @samp{n} for the named @var{breakpoint}(s).
29421 @subsubheading @value{GDBN} Command
29423 The corresponding @value{GDBN} command is @samp{disable}.
29425 @subsubheading Example
29433 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
29434 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
29435 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
29436 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
29437 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
29438 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
29439 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
29440 body=[bkpt=@{number="2",type="breakpoint",disp="keep",enabled="n",
29441 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
29442 line="5",thread-groups=["i1"],times="0"@}]@}
29446 @subheading The @code{-break-enable} Command
29447 @findex -break-enable
29449 @subsubheading Synopsis
29452 -break-enable ( @var{breakpoint} )+
29455 Enable (previously disabled) @var{breakpoint}(s).
29457 @subsubheading @value{GDBN} Command
29459 The corresponding @value{GDBN} command is @samp{enable}.
29461 @subsubheading Example
29469 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
29470 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
29471 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
29472 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
29473 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
29474 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
29475 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
29476 body=[bkpt=@{number="2",type="breakpoint",disp="keep",enabled="y",
29477 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
29478 line="5",thread-groups=["i1"],times="0"@}]@}
29482 @subheading The @code{-break-info} Command
29483 @findex -break-info
29485 @subsubheading Synopsis
29488 -break-info @var{breakpoint}
29492 Get information about a single breakpoint.
29494 The result is a table of breakpoints. @xref{GDB/MI Breakpoint
29495 Information}, for details on the format of each breakpoint in the
29498 @subsubheading @value{GDBN} Command
29500 The corresponding @value{GDBN} command is @samp{info break @var{breakpoint}}.
29502 @subsubheading Example
29505 @subheading The @code{-break-insert} Command
29506 @findex -break-insert
29507 @anchor{-break-insert}
29509 @subsubheading Synopsis
29512 -break-insert [ -t ] [ -h ] [ -f ] [ -d ] [ -a ]
29513 [ -c @var{condition} ] [ -i @var{ignore-count} ]
29514 [ -p @var{thread-id} ] [ @var{location} ]
29518 If specified, @var{location}, can be one of:
29521 @item linespec location
29522 A linespec location. @xref{Linespec Locations}.
29524 @item explicit location
29525 An explicit location. @sc{gdb/mi} explicit locations are
29526 analogous to the CLI's explicit locations using the option names
29527 listed below. @xref{Explicit Locations}.
29530 @item --source @var{filename}
29531 The source file name of the location. This option requires the use
29532 of either @samp{--function} or @samp{--line}.
29534 @item --function @var{function}
29535 The name of a function or method.
29537 @item --label @var{label}
29538 The name of a label.
29540 @item --line @var{lineoffset}
29541 An absolute or relative line offset from the start of the location.
29544 @item address location
29545 An address location, *@var{address}. @xref{Address Locations}.
29549 The possible optional parameters of this command are:
29553 Insert a temporary breakpoint.
29555 Insert a hardware breakpoint.
29557 If @var{location} cannot be parsed (for example if it
29558 refers to unknown files or functions), create a pending
29559 breakpoint. Without this flag, @value{GDBN} will report
29560 an error, and won't create a breakpoint, if @var{location}
29563 Create a disabled breakpoint.
29565 Create a tracepoint. @xref{Tracepoints}. When this parameter
29566 is used together with @samp{-h}, a fast tracepoint is created.
29567 @item -c @var{condition}
29568 Make the breakpoint conditional on @var{condition}.
29569 @item -i @var{ignore-count}
29570 Initialize the @var{ignore-count}.
29571 @item -p @var{thread-id}
29572 Restrict the breakpoint to the thread with the specified global
29576 @subsubheading Result
29578 @xref{GDB/MI Breakpoint Information}, for details on the format of the
29579 resulting breakpoint.
29581 Note: this format is open to change.
29582 @c An out-of-band breakpoint instead of part of the result?
29584 @subsubheading @value{GDBN} Command
29586 The corresponding @value{GDBN} commands are @samp{break}, @samp{tbreak},
29587 @samp{hbreak}, and @samp{thbreak}. @c and @samp{rbreak}.
29589 @subsubheading Example
29594 ^done,bkpt=@{number="1",addr="0x0001072c",file="recursive2.c",
29595 fullname="/home/foo/recursive2.c,line="4",thread-groups=["i1"],
29598 -break-insert -t foo
29599 ^done,bkpt=@{number="2",addr="0x00010774",file="recursive2.c",
29600 fullname="/home/foo/recursive2.c,line="11",thread-groups=["i1"],
29604 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
29605 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
29606 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
29607 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
29608 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
29609 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
29610 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
29611 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
29612 addr="0x0001072c", func="main",file="recursive2.c",
29613 fullname="/home/foo/recursive2.c,"line="4",thread-groups=["i1"],
29615 bkpt=@{number="2",type="breakpoint",disp="del",enabled="y",
29616 addr="0x00010774",func="foo",file="recursive2.c",
29617 fullname="/home/foo/recursive2.c",line="11",thread-groups=["i1"],
29620 @c -break-insert -r foo.*
29621 @c ~int foo(int, int);
29622 @c ^done,bkpt=@{number="3",addr="0x00010774",file="recursive2.c,
29623 @c "fullname="/home/foo/recursive2.c",line="11",thread-groups=["i1"],
29628 @subheading The @code{-dprintf-insert} Command
29629 @findex -dprintf-insert
29631 @subsubheading Synopsis
29634 -dprintf-insert [ -t ] [ -f ] [ -d ]
29635 [ -c @var{condition} ] [ -i @var{ignore-count} ]
29636 [ -p @var{thread-id} ] [ @var{location} ] [ @var{format} ]
29641 If supplied, @var{location} may be specified the same way as for
29642 the @code{-break-insert} command. @xref{-break-insert}.
29644 The possible optional parameters of this command are:
29648 Insert a temporary breakpoint.
29650 If @var{location} cannot be parsed (for example, if it
29651 refers to unknown files or functions), create a pending
29652 breakpoint. Without this flag, @value{GDBN} will report
29653 an error, and won't create a breakpoint, if @var{location}
29656 Create a disabled breakpoint.
29657 @item -c @var{condition}
29658 Make the breakpoint conditional on @var{condition}.
29659 @item -i @var{ignore-count}
29660 Set the ignore count of the breakpoint (@pxref{Conditions, ignore count})
29661 to @var{ignore-count}.
29662 @item -p @var{thread-id}
29663 Restrict the breakpoint to the thread with the specified global
29667 @subsubheading Result
29669 @xref{GDB/MI Breakpoint Information}, for details on the format of the
29670 resulting breakpoint.
29672 @c An out-of-band breakpoint instead of part of the result?
29674 @subsubheading @value{GDBN} Command
29676 The corresponding @value{GDBN} command is @samp{dprintf}.
29678 @subsubheading Example
29682 4-dprintf-insert foo "At foo entry\n"
29683 4^done,bkpt=@{number="1",type="dprintf",disp="keep",enabled="y",
29684 addr="0x000000000040061b",func="foo",file="mi-dprintf.c",
29685 fullname="mi-dprintf.c",line="25",thread-groups=["i1"],
29686 times="0",script=@{"printf \"At foo entry\\n\"","continue"@},
29687 original-location="foo"@}
29689 5-dprintf-insert 26 "arg=%d, g=%d\n" arg g
29690 5^done,bkpt=@{number="2",type="dprintf",disp="keep",enabled="y",
29691 addr="0x000000000040062a",func="foo",file="mi-dprintf.c",
29692 fullname="mi-dprintf.c",line="26",thread-groups=["i1"],
29693 times="0",script=@{"printf \"arg=%d, g=%d\\n\", arg, g","continue"@},
29694 original-location="mi-dprintf.c:26"@}
29698 @subheading The @code{-break-list} Command
29699 @findex -break-list
29701 @subsubheading Synopsis
29707 Displays the list of inserted breakpoints, showing the following fields:
29711 number of the breakpoint
29713 type of the breakpoint: @samp{breakpoint} or @samp{watchpoint}
29715 should the breakpoint be deleted or disabled when it is hit: @samp{keep}
29718 is the breakpoint enabled or no: @samp{y} or @samp{n}
29720 memory location at which the breakpoint is set
29722 logical location of the breakpoint, expressed by function name, file
29724 @item Thread-groups
29725 list of thread groups to which this breakpoint applies
29727 number of times the breakpoint has been hit
29730 If there are no breakpoints or watchpoints, the @code{BreakpointTable}
29731 @code{body} field is an empty list.
29733 @subsubheading @value{GDBN} Command
29735 The corresponding @value{GDBN} command is @samp{info break}.
29737 @subsubheading Example
29742 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
29743 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
29744 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
29745 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
29746 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
29747 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
29748 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
29749 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
29750 addr="0x000100d0",func="main",file="hello.c",line="5",thread-groups=["i1"],
29752 bkpt=@{number="2",type="breakpoint",disp="keep",enabled="y",
29753 addr="0x00010114",func="foo",file="hello.c",fullname="/home/foo/hello.c",
29754 line="13",thread-groups=["i1"],times="0"@}]@}
29758 Here's an example of the result when there are no breakpoints:
29763 ^done,BreakpointTable=@{nr_rows="0",nr_cols="6",
29764 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
29765 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
29766 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
29767 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
29768 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
29769 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
29774 @subheading The @code{-break-passcount} Command
29775 @findex -break-passcount
29777 @subsubheading Synopsis
29780 -break-passcount @var{tracepoint-number} @var{passcount}
29783 Set the passcount for tracepoint @var{tracepoint-number} to
29784 @var{passcount}. If the breakpoint referred to by @var{tracepoint-number}
29785 is not a tracepoint, error is emitted. This corresponds to CLI
29786 command @samp{passcount}.
29788 @subheading The @code{-break-watch} Command
29789 @findex -break-watch
29791 @subsubheading Synopsis
29794 -break-watch [ -a | -r ]
29797 Create a watchpoint. With the @samp{-a} option it will create an
29798 @dfn{access} watchpoint, i.e., a watchpoint that triggers either on a
29799 read from or on a write to the memory location. With the @samp{-r}
29800 option, the watchpoint created is a @dfn{read} watchpoint, i.e., it will
29801 trigger only when the memory location is accessed for reading. Without
29802 either of the options, the watchpoint created is a regular watchpoint,
29803 i.e., it will trigger when the memory location is accessed for writing.
29804 @xref{Set Watchpoints, , Setting Watchpoints}.
29806 Note that @samp{-break-list} will report a single list of watchpoints and
29807 breakpoints inserted.
29809 @subsubheading @value{GDBN} Command
29811 The corresponding @value{GDBN} commands are @samp{watch}, @samp{awatch}, and
29814 @subsubheading Example
29816 Setting a watchpoint on a variable in the @code{main} function:
29821 ^done,wpt=@{number="2",exp="x"@}
29826 *stopped,reason="watchpoint-trigger",wpt=@{number="2",exp="x"@},
29827 value=@{old="-268439212",new="55"@},
29828 frame=@{func="main",args=[],file="recursive2.c",
29829 fullname="/home/foo/bar/recursive2.c",line="5",arch="i386:x86_64"@}
29833 Setting a watchpoint on a variable local to a function. @value{GDBN} will stop
29834 the program execution twice: first for the variable changing value, then
29835 for the watchpoint going out of scope.
29840 ^done,wpt=@{number="5",exp="C"@}
29845 *stopped,reason="watchpoint-trigger",
29846 wpt=@{number="5",exp="C"@},value=@{old="-276895068",new="3"@},
29847 frame=@{func="callee4",args=[],
29848 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
29849 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="13",
29850 arch="i386:x86_64"@}
29855 *stopped,reason="watchpoint-scope",wpnum="5",
29856 frame=@{func="callee3",args=[@{name="strarg",
29857 value="0x11940 \"A string argument.\""@}],
29858 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
29859 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="18",
29860 arch="i386:x86_64"@}
29864 Listing breakpoints and watchpoints, at different points in the program
29865 execution. Note that once the watchpoint goes out of scope, it is
29871 ^done,wpt=@{number="2",exp="C"@}
29874 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
29875 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
29876 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
29877 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
29878 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
29879 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
29880 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
29881 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
29882 addr="0x00010734",func="callee4",
29883 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
29884 fullname="/home/foo/devo/gdb/testsuite/gdb.mi/basics.c"line="8",thread-groups=["i1"],
29886 bkpt=@{number="2",type="watchpoint",disp="keep",
29887 enabled="y",addr="",what="C",thread-groups=["i1"],times="0"@}]@}
29892 *stopped,reason="watchpoint-trigger",wpt=@{number="2",exp="C"@},
29893 value=@{old="-276895068",new="3"@},
29894 frame=@{func="callee4",args=[],
29895 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
29896 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="13",
29897 arch="i386:x86_64"@}
29900 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
29901 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
29902 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
29903 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
29904 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
29905 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
29906 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
29907 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
29908 addr="0x00010734",func="callee4",
29909 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
29910 fullname="/home/foo/devo/gdb/testsuite/gdb.mi/basics.c",line="8",thread-groups=["i1"],
29912 bkpt=@{number="2",type="watchpoint",disp="keep",
29913 enabled="y",addr="",what="C",thread-groups=["i1"],times="-5"@}]@}
29917 ^done,reason="watchpoint-scope",wpnum="2",
29918 frame=@{func="callee3",args=[@{name="strarg",
29919 value="0x11940 \"A string argument.\""@}],
29920 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
29921 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="18",
29922 arch="i386:x86_64"@}
29925 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
29926 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
29927 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
29928 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
29929 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
29930 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
29931 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
29932 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
29933 addr="0x00010734",func="callee4",
29934 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
29935 fullname="/home/foo/devo/gdb/testsuite/gdb.mi/basics.c",line="8",
29936 thread-groups=["i1"],times="1"@}]@}
29941 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
29942 @node GDB/MI Catchpoint Commands
29943 @section @sc{gdb/mi} Catchpoint Commands
29945 This section documents @sc{gdb/mi} commands for manipulating
29949 * Shared Library GDB/MI Catchpoint Commands::
29950 * Ada Exception GDB/MI Catchpoint Commands::
29951 * C++ Exception GDB/MI Catchpoint Commands::
29954 @node Shared Library GDB/MI Catchpoint Commands
29955 @subsection Shared Library @sc{gdb/mi} Catchpoints
29957 @subheading The @code{-catch-load} Command
29958 @findex -catch-load
29960 @subsubheading Synopsis
29963 -catch-load [ -t ] [ -d ] @var{regexp}
29966 Add a catchpoint for library load events. If the @samp{-t} option is used,
29967 the catchpoint is a temporary one (@pxref{Set Breaks, ,Setting
29968 Breakpoints}). If the @samp{-d} option is used, the catchpoint is created
29969 in a disabled state. The @samp{regexp} argument is a regular
29970 expression used to match the name of the loaded library.
29973 @subsubheading @value{GDBN} Command
29975 The corresponding @value{GDBN} command is @samp{catch load}.
29977 @subsubheading Example
29980 -catch-load -t foo.so
29981 ^done,bkpt=@{number="1",type="catchpoint",disp="del",enabled="y",
29982 what="load of library matching foo.so",catch-type="load",times="0"@}
29987 @subheading The @code{-catch-unload} Command
29988 @findex -catch-unload
29990 @subsubheading Synopsis
29993 -catch-unload [ -t ] [ -d ] @var{regexp}
29996 Add a catchpoint for library unload events. If the @samp{-t} option is
29997 used, the catchpoint is a temporary one (@pxref{Set Breaks, ,Setting
29998 Breakpoints}). If the @samp{-d} option is used, the catchpoint is
29999 created in a disabled state. The @samp{regexp} argument is a regular
30000 expression used to match the name of the unloaded library.
30002 @subsubheading @value{GDBN} Command
30004 The corresponding @value{GDBN} command is @samp{catch unload}.
30006 @subsubheading Example
30009 -catch-unload -d bar.so
30010 ^done,bkpt=@{number="2",type="catchpoint",disp="keep",enabled="n",
30011 what="load of library matching bar.so",catch-type="unload",times="0"@}
30015 @node Ada Exception GDB/MI Catchpoint Commands
30016 @subsection Ada Exception @sc{gdb/mi} Catchpoints
30018 The following @sc{gdb/mi} commands can be used to create catchpoints
30019 that stop the execution when Ada exceptions are being raised.
30021 @subheading The @code{-catch-assert} Command
30022 @findex -catch-assert
30024 @subsubheading Synopsis
30027 -catch-assert [ -c @var{condition}] [ -d ] [ -t ]
30030 Add a catchpoint for failed Ada assertions.
30032 The possible optional parameters for this command are:
30035 @item -c @var{condition}
30036 Make the catchpoint conditional on @var{condition}.
30038 Create a disabled catchpoint.
30040 Create a temporary catchpoint.
30043 @subsubheading @value{GDBN} Command
30045 The corresponding @value{GDBN} command is @samp{catch assert}.
30047 @subsubheading Example
30051 ^done,bkptno="5",bkpt=@{number="5",type="breakpoint",disp="keep",
30052 enabled="y",addr="0x0000000000404888",what="failed Ada assertions",
30053 thread-groups=["i1"],times="0",
30054 original-location="__gnat_debug_raise_assert_failure"@}
30058 @subheading The @code{-catch-exception} Command
30059 @findex -catch-exception
30061 @subsubheading Synopsis
30064 -catch-exception [ -c @var{condition}] [ -d ] [ -e @var{exception-name} ]
30068 Add a catchpoint stopping when Ada exceptions are raised.
30069 By default, the command stops the program when any Ada exception
30070 gets raised. But it is also possible, by using some of the
30071 optional parameters described below, to create more selective
30074 The possible optional parameters for this command are:
30077 @item -c @var{condition}
30078 Make the catchpoint conditional on @var{condition}.
30080 Create a disabled catchpoint.
30081 @item -e @var{exception-name}
30082 Only stop when @var{exception-name} is raised. This option cannot
30083 be used combined with @samp{-u}.
30085 Create a temporary catchpoint.
30087 Stop only when an unhandled exception gets raised. This option
30088 cannot be used combined with @samp{-e}.
30091 @subsubheading @value{GDBN} Command
30093 The corresponding @value{GDBN} commands are @samp{catch exception}
30094 and @samp{catch exception unhandled}.
30096 @subsubheading Example
30099 -catch-exception -e Program_Error
30100 ^done,bkptno="4",bkpt=@{number="4",type="breakpoint",disp="keep",
30101 enabled="y",addr="0x0000000000404874",
30102 what="`Program_Error' Ada exception", thread-groups=["i1"],
30103 times="0",original-location="__gnat_debug_raise_exception"@}
30107 @subheading The @code{-catch-handlers} Command
30108 @findex -catch-handlers
30110 @subsubheading Synopsis
30113 -catch-handlers [ -c @var{condition}] [ -d ] [ -e @var{exception-name} ]
30117 Add a catchpoint stopping when Ada exceptions are handled.
30118 By default, the command stops the program when any Ada exception
30119 gets handled. But it is also possible, by using some of the
30120 optional parameters described below, to create more selective
30123 The possible optional parameters for this command are:
30126 @item -c @var{condition}
30127 Make the catchpoint conditional on @var{condition}.
30129 Create a disabled catchpoint.
30130 @item -e @var{exception-name}
30131 Only stop when @var{exception-name} is handled.
30133 Create a temporary catchpoint.
30136 @subsubheading @value{GDBN} Command
30138 The corresponding @value{GDBN} command is @samp{catch handlers}.
30140 @subsubheading Example
30143 -catch-handlers -e Constraint_Error
30144 ^done,bkptno="4",bkpt=@{number="4",type="breakpoint",disp="keep",
30145 enabled="y",addr="0x0000000000402f68",
30146 what="`Constraint_Error' Ada exception handlers",thread-groups=["i1"],
30147 times="0",original-location="__gnat_begin_handler"@}
30151 @node C++ Exception GDB/MI Catchpoint Commands
30152 @subsection C@t{++} Exception @sc{gdb/mi} Catchpoints
30154 The following @sc{gdb/mi} commands can be used to create catchpoints
30155 that stop the execution when C@t{++} exceptions are being throw, rethrown,
30158 @subheading The @code{-catch-throw} Command
30159 @findex -catch-throw
30161 @subsubheading Synopsis
30164 -catch-throw [ -t ] [ -r @var{regexp}]
30167 Stop when the debuggee throws a C@t{++} exception. If @var{regexp} is
30168 given, then only exceptions whose type matches the regular expression
30171 If @samp{-t} is given, then the catchpoint is enabled only for one
30172 stop, the catchpoint is automatically deleted after stopping once for
30175 @subsubheading @value{GDBN} Command
30177 The corresponding @value{GDBN} commands are @samp{catch throw}
30178 and @samp{tcatch throw} (@pxref{Set Catchpoints}).
30180 @subsubheading Example
30183 -catch-throw -r exception_type
30184 ^done,bkpt=@{number="1",type="catchpoint",disp="keep",enabled="y",
30185 what="exception throw",catch-type="throw",
30186 thread-groups=["i1"],
30187 regexp="exception_type",times="0"@}
30193 ~"Catchpoint 1 (exception thrown), 0x00007ffff7ae00ed
30194 in __cxa_throw () from /lib64/libstdc++.so.6\n"
30195 *stopped,bkptno="1",reason="breakpoint-hit",disp="keep",
30196 frame=@{addr="0x00007ffff7ae00ed",func="__cxa_throw",
30197 args=[],from="/lib64/libstdc++.so.6",arch="i386:x86-64"@},
30198 thread-id="1",stopped-threads="all",core="6"
30202 @subheading The @code{-catch-rethrow} Command
30203 @findex -catch-rethrow
30205 @subsubheading Synopsis
30208 -catch-rethrow [ -t ] [ -r @var{regexp}]
30211 Stop when a C@t{++} exception is re-thrown. If @var{regexp} is given,
30212 then only exceptions whose type matches the regular expression will be
30215 If @samp{-t} is given, then the catchpoint is enabled only for one
30216 stop, the catchpoint is automatically deleted after the first event is
30219 @subsubheading @value{GDBN} Command
30221 The corresponding @value{GDBN} commands are @samp{catch rethrow}
30222 and @samp{tcatch rethrow} (@pxref{Set Catchpoints}).
30224 @subsubheading Example
30227 -catch-rethrow -r exception_type
30228 ^done,bkpt=@{number="1",type="catchpoint",disp="keep",enabled="y",
30229 what="exception rethrow",catch-type="rethrow",
30230 thread-groups=["i1"],
30231 regexp="exception_type",times="0"@}
30237 ~"Catchpoint 1 (exception rethrown), 0x00007ffff7ae00ed
30238 in __cxa_rethrow () from /lib64/libstdc++.so.6\n"
30239 *stopped,bkptno="1",reason="breakpoint-hit",disp="keep",
30240 frame=@{addr="0x00007ffff7ae00ed",func="__cxa_rethrow",
30241 args=[],from="/lib64/libstdc++.so.6",arch="i386:x86-64"@},
30242 thread-id="1",stopped-threads="all",core="6"
30246 @subheading The @code{-catch-catch} Command
30247 @findex -catch-catch
30249 @subsubheading Synopsis
30252 -catch-catch [ -t ] [ -r @var{regexp}]
30255 Stop when the debuggee catches a C@t{++} exception. If @var{regexp}
30256 is given, then only exceptions whose type matches the regular
30257 expression will be caught.
30259 If @samp{-t} is given, then the catchpoint is enabled only for one
30260 stop, the catchpoint is automatically deleted after the first event is
30263 @subsubheading @value{GDBN} Command
30265 The corresponding @value{GDBN} commands are @samp{catch catch}
30266 and @samp{tcatch catch} (@pxref{Set Catchpoints}).
30268 @subsubheading Example
30271 -catch-catch -r exception_type
30272 ^done,bkpt=@{number="1",type="catchpoint",disp="keep",enabled="y",
30273 what="exception catch",catch-type="catch",
30274 thread-groups=["i1"],
30275 regexp="exception_type",times="0"@}
30281 ~"Catchpoint 1 (exception caught), 0x00007ffff7ae00ed
30282 in __cxa_begin_catch () from /lib64/libstdc++.so.6\n"
30283 *stopped,bkptno="1",reason="breakpoint-hit",disp="keep",
30284 frame=@{addr="0x00007ffff7ae00ed",func="__cxa_begin_catch",
30285 args=[],from="/lib64/libstdc++.so.6",arch="i386:x86-64"@},
30286 thread-id="1",stopped-threads="all",core="6"
30290 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
30291 @node GDB/MI Program Context
30292 @section @sc{gdb/mi} Program Context
30294 @subheading The @code{-exec-arguments} Command
30295 @findex -exec-arguments
30298 @subsubheading Synopsis
30301 -exec-arguments @var{args}
30304 Set the inferior program arguments, to be used in the next
30307 @subsubheading @value{GDBN} Command
30309 The corresponding @value{GDBN} command is @samp{set args}.
30311 @subsubheading Example
30315 -exec-arguments -v word
30322 @subheading The @code{-exec-show-arguments} Command
30323 @findex -exec-show-arguments
30325 @subsubheading Synopsis
30328 -exec-show-arguments
30331 Print the arguments of the program.
30333 @subsubheading @value{GDBN} Command
30335 The corresponding @value{GDBN} command is @samp{show args}.
30337 @subsubheading Example
30342 @subheading The @code{-environment-cd} Command
30343 @findex -environment-cd
30345 @subsubheading Synopsis
30348 -environment-cd @var{pathdir}
30351 Set @value{GDBN}'s working directory.
30353 @subsubheading @value{GDBN} Command
30355 The corresponding @value{GDBN} command is @samp{cd}.
30357 @subsubheading Example
30361 -environment-cd /kwikemart/marge/ezannoni/flathead-dev/devo/gdb
30367 @subheading The @code{-environment-directory} Command
30368 @findex -environment-directory
30370 @subsubheading Synopsis
30373 -environment-directory [ -r ] [ @var{pathdir} ]+
30376 Add directories @var{pathdir} to beginning of search path for source files.
30377 If the @samp{-r} option is used, the search path is reset to the default
30378 search path. If directories @var{pathdir} are supplied in addition to the
30379 @samp{-r} option, the search path is first reset and then addition
30381 Multiple directories may be specified, separated by blanks. Specifying
30382 multiple directories in a single command
30383 results in the directories added to the beginning of the
30384 search path in the same order they were presented in the command.
30385 If blanks are needed as
30386 part of a directory name, double-quotes should be used around
30387 the name. In the command output, the path will show up separated
30388 by the system directory-separator character. The directory-separator
30389 character must not be used
30390 in any directory name.
30391 If no directories are specified, the current search path is displayed.
30393 @subsubheading @value{GDBN} Command
30395 The corresponding @value{GDBN} command is @samp{dir}.
30397 @subsubheading Example
30401 -environment-directory /kwikemart/marge/ezannoni/flathead-dev/devo/gdb
30402 ^done,source-path="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb:$cdir:$cwd"
30404 -environment-directory ""
30405 ^done,source-path="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb:$cdir:$cwd"
30407 -environment-directory -r /home/jjohnstn/src/gdb /usr/src
30408 ^done,source-path="/home/jjohnstn/src/gdb:/usr/src:$cdir:$cwd"
30410 -environment-directory -r
30411 ^done,source-path="$cdir:$cwd"
30416 @subheading The @code{-environment-path} Command
30417 @findex -environment-path
30419 @subsubheading Synopsis
30422 -environment-path [ -r ] [ @var{pathdir} ]+
30425 Add directories @var{pathdir} to beginning of search path for object files.
30426 If the @samp{-r} option is used, the search path is reset to the original
30427 search path that existed at gdb start-up. If directories @var{pathdir} are
30428 supplied in addition to the
30429 @samp{-r} option, the search path is first reset and then addition
30431 Multiple directories may be specified, separated by blanks. Specifying
30432 multiple directories in a single command
30433 results in the directories added to the beginning of the
30434 search path in the same order they were presented in the command.
30435 If blanks are needed as
30436 part of a directory name, double-quotes should be used around
30437 the name. In the command output, the path will show up separated
30438 by the system directory-separator character. The directory-separator
30439 character must not be used
30440 in any directory name.
30441 If no directories are specified, the current path is displayed.
30444 @subsubheading @value{GDBN} Command
30446 The corresponding @value{GDBN} command is @samp{path}.
30448 @subsubheading Example
30453 ^done,path="/usr/bin"
30455 -environment-path /kwikemart/marge/ezannoni/flathead-dev/ppc-eabi/gdb /bin
30456 ^done,path="/kwikemart/marge/ezannoni/flathead-dev/ppc-eabi/gdb:/bin:/usr/bin"
30458 -environment-path -r /usr/local/bin
30459 ^done,path="/usr/local/bin:/usr/bin"
30464 @subheading The @code{-environment-pwd} Command
30465 @findex -environment-pwd
30467 @subsubheading Synopsis
30473 Show the current working directory.
30475 @subsubheading @value{GDBN} Command
30477 The corresponding @value{GDBN} command is @samp{pwd}.
30479 @subsubheading Example
30484 ^done,cwd="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb"
30488 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
30489 @node GDB/MI Thread Commands
30490 @section @sc{gdb/mi} Thread Commands
30493 @subheading The @code{-thread-info} Command
30494 @findex -thread-info
30496 @subsubheading Synopsis
30499 -thread-info [ @var{thread-id} ]
30502 Reports information about either a specific thread, if the
30503 @var{thread-id} parameter is present, or about all threads.
30504 @var{thread-id} is the thread's global thread ID. When printing
30505 information about all threads, also reports the global ID of the
30508 @subsubheading @value{GDBN} Command
30510 The @samp{info thread} command prints the same information
30513 @subsubheading Result
30515 The result contains the following attributes:
30519 A list of threads. The format of the elements of the list is described in
30520 @ref{GDB/MI Thread Information}.
30522 @item current-thread-id
30523 The global id of the currently selected thread. This field is omitted if there
30524 is no selected thread (for example, when the selected inferior is not running,
30525 and therefore has no threads) or if a @var{thread-id} argument was passed to
30530 @subsubheading Example
30535 @{id="2",target-id="Thread 0xb7e14b90 (LWP 21257)",
30536 frame=@{level="0",addr="0xffffe410",func="__kernel_vsyscall",
30537 args=[]@},state="running"@},
30538 @{id="1",target-id="Thread 0xb7e156b0 (LWP 21254)",
30539 frame=@{level="0",addr="0x0804891f",func="foo",
30540 args=[@{name="i",value="10"@}],
30541 file="/tmp/a.c",fullname="/tmp/a.c",line="158",arch="i386:x86_64"@},
30542 state="running"@}],
30543 current-thread-id="1"
30547 @subheading The @code{-thread-list-ids} Command
30548 @findex -thread-list-ids
30550 @subsubheading Synopsis
30556 Produces a list of the currently known global @value{GDBN} thread ids.
30557 At the end of the list it also prints the total number of such
30560 This command is retained for historical reasons, the
30561 @code{-thread-info} command should be used instead.
30563 @subsubheading @value{GDBN} Command
30565 Part of @samp{info threads} supplies the same information.
30567 @subsubheading Example
30572 ^done,thread-ids=@{thread-id="3",thread-id="2",thread-id="1"@},
30573 current-thread-id="1",number-of-threads="3"
30578 @subheading The @code{-thread-select} Command
30579 @findex -thread-select
30581 @subsubheading Synopsis
30584 -thread-select @var{thread-id}
30587 Make thread with global thread number @var{thread-id} the current
30588 thread. It prints the number of the new current thread, and the
30589 topmost frame for that thread.
30591 This command is deprecated in favor of explicitly using the
30592 @samp{--thread} option to each command.
30594 @subsubheading @value{GDBN} Command
30596 The corresponding @value{GDBN} command is @samp{thread}.
30598 @subsubheading Example
30605 *stopped,reason="end-stepping-range",thread-id="2",line="187",
30606 file="../../../devo/gdb/testsuite/gdb.threads/linux-dp.c"
30610 thread-ids=@{thread-id="3",thread-id="2",thread-id="1"@},
30611 number-of-threads="3"
30614 ^done,new-thread-id="3",
30615 frame=@{level="0",func="vprintf",
30616 args=[@{name="format",value="0x8048e9c \"%*s%c %d %c\\n\""@},
30617 @{name="arg",value="0x2"@}],file="vprintf.c",line="31",arch="i386:x86_64"@}
30621 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
30622 @node GDB/MI Ada Tasking Commands
30623 @section @sc{gdb/mi} Ada Tasking Commands
30625 @subheading The @code{-ada-task-info} Command
30626 @findex -ada-task-info
30628 @subsubheading Synopsis
30631 -ada-task-info [ @var{task-id} ]
30634 Reports information about either a specific Ada task, if the
30635 @var{task-id} parameter is present, or about all Ada tasks.
30637 @subsubheading @value{GDBN} Command
30639 The @samp{info tasks} command prints the same information
30640 about all Ada tasks (@pxref{Ada Tasks}).
30642 @subsubheading Result
30644 The result is a table of Ada tasks. The following columns are
30645 defined for each Ada task:
30649 This field exists only for the current thread. It has the value @samp{*}.
30652 The identifier that @value{GDBN} uses to refer to the Ada task.
30655 The identifier that the target uses to refer to the Ada task.
30658 The global thread identifier of the thread corresponding to the Ada
30661 This field should always exist, as Ada tasks are always implemented
30662 on top of a thread. But if @value{GDBN} cannot find this corresponding
30663 thread for any reason, the field is omitted.
30666 This field exists only when the task was created by another task.
30667 In this case, it provides the ID of the parent task.
30670 The base priority of the task.
30673 The current state of the task. For a detailed description of the
30674 possible states, see @ref{Ada Tasks}.
30677 The name of the task.
30681 @subsubheading Example
30685 ^done,tasks=@{nr_rows="3",nr_cols="8",
30686 hdr=[@{width="1",alignment="-1",col_name="current",colhdr=""@},
30687 @{width="3",alignment="1",col_name="id",colhdr="ID"@},
30688 @{width="9",alignment="1",col_name="task-id",colhdr="TID"@},
30689 @{width="4",alignment="1",col_name="thread-id",colhdr=""@},
30690 @{width="4",alignment="1",col_name="parent-id",colhdr="P-ID"@},
30691 @{width="3",alignment="1",col_name="priority",colhdr="Pri"@},
30692 @{width="22",alignment="-1",col_name="state",colhdr="State"@},
30693 @{width="1",alignment="2",col_name="name",colhdr="Name"@}],
30694 body=[@{current="*",id="1",task-id=" 644010",thread-id="1",priority="48",
30695 state="Child Termination Wait",name="main_task"@}]@}
30699 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
30700 @node GDB/MI Program Execution
30701 @section @sc{gdb/mi} Program Execution
30703 These are the asynchronous commands which generate the out-of-band
30704 record @samp{*stopped}. Currently @value{GDBN} only really executes
30705 asynchronously with remote targets and this interaction is mimicked in
30708 @subheading The @code{-exec-continue} Command
30709 @findex -exec-continue
30711 @subsubheading Synopsis
30714 -exec-continue [--reverse] [--all|--thread-group N]
30717 Resumes the execution of the inferior program, which will continue
30718 to execute until it reaches a debugger stop event. If the
30719 @samp{--reverse} option is specified, execution resumes in reverse until
30720 it reaches a stop event. Stop events may include
30723 breakpoints or watchpoints
30725 signals or exceptions
30727 the end of the process (or its beginning under @samp{--reverse})
30729 the end or beginning of a replay log if one is being used.
30731 In all-stop mode (@pxref{All-Stop
30732 Mode}), may resume only one thread, or all threads, depending on the
30733 value of the @samp{scheduler-locking} variable. If @samp{--all} is
30734 specified, all threads (in all inferiors) will be resumed. The @samp{--all} option is
30735 ignored in all-stop mode. If the @samp{--thread-group} options is
30736 specified, then all threads in that thread group are resumed.
30738 @subsubheading @value{GDBN} Command
30740 The corresponding @value{GDBN} corresponding is @samp{continue}.
30742 @subsubheading Example
30749 *stopped,reason="breakpoint-hit",disp="keep",bkptno="2",frame=@{
30750 func="foo",args=[],file="hello.c",fullname="/home/foo/bar/hello.c",
30751 line="13",arch="i386:x86_64"@}
30756 @subheading The @code{-exec-finish} Command
30757 @findex -exec-finish
30759 @subsubheading Synopsis
30762 -exec-finish [--reverse]
30765 Resumes the execution of the inferior program until the current
30766 function is exited. Displays the results returned by the function.
30767 If the @samp{--reverse} option is specified, resumes the reverse
30768 execution of the inferior program until the point where current
30769 function was called.
30771 @subsubheading @value{GDBN} Command
30773 The corresponding @value{GDBN} command is @samp{finish}.
30775 @subsubheading Example
30777 Function returning @code{void}.
30784 *stopped,reason="function-finished",frame=@{func="main",args=[],
30785 file="hello.c",fullname="/home/foo/bar/hello.c",line="7",arch="i386:x86_64"@}
30789 Function returning other than @code{void}. The name of the internal
30790 @value{GDBN} variable storing the result is printed, together with the
30797 *stopped,reason="function-finished",frame=@{addr="0x000107b0",func="foo",
30798 args=[@{name="a",value="1"],@{name="b",value="9"@}@},
30799 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
30800 arch="i386:x86_64"@},
30801 gdb-result-var="$1",return-value="0"
30806 @subheading The @code{-exec-interrupt} Command
30807 @findex -exec-interrupt
30809 @subsubheading Synopsis
30812 -exec-interrupt [--all|--thread-group N]
30815 Interrupts the background execution of the target. Note how the token
30816 associated with the stop message is the one for the execution command
30817 that has been interrupted. The token for the interrupt itself only
30818 appears in the @samp{^done} output. If the user is trying to
30819 interrupt a non-running program, an error message will be printed.
30821 Note that when asynchronous execution is enabled, this command is
30822 asynchronous just like other execution commands. That is, first the
30823 @samp{^done} response will be printed, and the target stop will be
30824 reported after that using the @samp{*stopped} notification.
30826 In non-stop mode, only the context thread is interrupted by default.
30827 All threads (in all inferiors) will be interrupted if the
30828 @samp{--all} option is specified. If the @samp{--thread-group}
30829 option is specified, all threads in that group will be interrupted.
30831 @subsubheading @value{GDBN} Command
30833 The corresponding @value{GDBN} command is @samp{interrupt}.
30835 @subsubheading Example
30846 111*stopped,signal-name="SIGINT",signal-meaning="Interrupt",
30847 frame=@{addr="0x00010140",func="foo",args=[],file="try.c",
30848 fullname="/home/foo/bar/try.c",line="13",arch="i386:x86_64"@}
30853 ^error,msg="mi_cmd_exec_interrupt: Inferior not executing."
30857 @subheading The @code{-exec-jump} Command
30860 @subsubheading Synopsis
30863 -exec-jump @var{location}
30866 Resumes execution of the inferior program at the location specified by
30867 parameter. @xref{Specify Location}, for a description of the
30868 different forms of @var{location}.
30870 @subsubheading @value{GDBN} Command
30872 The corresponding @value{GDBN} command is @samp{jump}.
30874 @subsubheading Example
30877 -exec-jump foo.c:10
30878 *running,thread-id="all"
30883 @subheading The @code{-exec-next} Command
30886 @subsubheading Synopsis
30889 -exec-next [--reverse]
30892 Resumes execution of the inferior program, stopping when the beginning
30893 of the next source line is reached.
30895 If the @samp{--reverse} option is specified, resumes reverse execution
30896 of the inferior program, stopping at the beginning of the previous
30897 source line. If you issue this command on the first line of a
30898 function, it will take you back to the caller of that function, to the
30899 source line where the function was called.
30902 @subsubheading @value{GDBN} Command
30904 The corresponding @value{GDBN} command is @samp{next}.
30906 @subsubheading Example
30912 *stopped,reason="end-stepping-range",line="8",file="hello.c"
30917 @subheading The @code{-exec-next-instruction} Command
30918 @findex -exec-next-instruction
30920 @subsubheading Synopsis
30923 -exec-next-instruction [--reverse]
30926 Executes one machine instruction. If the instruction is a function
30927 call, continues until the function returns. If the program stops at an
30928 instruction in the middle of a source line, the address will be
30931 If the @samp{--reverse} option is specified, resumes reverse execution
30932 of the inferior program, stopping at the previous instruction. If the
30933 previously executed instruction was a return from another function,
30934 it will continue to execute in reverse until the call to that function
30935 (from the current stack frame) is reached.
30937 @subsubheading @value{GDBN} Command
30939 The corresponding @value{GDBN} command is @samp{nexti}.
30941 @subsubheading Example
30945 -exec-next-instruction
30949 *stopped,reason="end-stepping-range",
30950 addr="0x000100d4",line="5",file="hello.c"
30955 @subheading The @code{-exec-return} Command
30956 @findex -exec-return
30958 @subsubheading Synopsis
30964 Makes current function return immediately. Doesn't execute the inferior.
30965 Displays the new current frame.
30967 @subsubheading @value{GDBN} Command
30969 The corresponding @value{GDBN} command is @samp{return}.
30971 @subsubheading Example
30975 200-break-insert callee4
30976 200^done,bkpt=@{number="1",addr="0x00010734",
30977 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",line="8"@}
30982 000*stopped,reason="breakpoint-hit",disp="keep",bkptno="1",
30983 frame=@{func="callee4",args=[],
30984 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
30985 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="8",
30986 arch="i386:x86_64"@}
30992 111^done,frame=@{level="0",func="callee3",
30993 args=[@{name="strarg",
30994 value="0x11940 \"A string argument.\""@}],
30995 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
30996 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="18",
30997 arch="i386:x86_64"@}
31002 @subheading The @code{-exec-run} Command
31005 @subsubheading Synopsis
31008 -exec-run [ --all | --thread-group N ] [ --start ]
31011 Starts execution of the inferior from the beginning. The inferior
31012 executes until either a breakpoint is encountered or the program
31013 exits. In the latter case the output will include an exit code, if
31014 the program has exited exceptionally.
31016 When neither the @samp{--all} nor the @samp{--thread-group} option
31017 is specified, the current inferior is started. If the
31018 @samp{--thread-group} option is specified, it should refer to a thread
31019 group of type @samp{process}, and that thread group will be started.
31020 If the @samp{--all} option is specified, then all inferiors will be started.
31022 Using the @samp{--start} option instructs the debugger to stop
31023 the execution at the start of the inferior's main subprogram,
31024 following the same behavior as the @code{start} command
31025 (@pxref{Starting}).
31027 @subsubheading @value{GDBN} Command
31029 The corresponding @value{GDBN} command is @samp{run}.
31031 @subsubheading Examples
31036 ^done,bkpt=@{number="1",addr="0x0001072c",file="recursive2.c",line="4"@}
31041 *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",
31042 frame=@{func="main",args=[],file="recursive2.c",
31043 fullname="/home/foo/bar/recursive2.c",line="4",arch="i386:x86_64"@}
31048 Program exited normally:
31056 *stopped,reason="exited-normally"
31061 Program exited exceptionally:
31069 *stopped,reason="exited",exit-code="01"
31073 Another way the program can terminate is if it receives a signal such as
31074 @code{SIGINT}. In this case, @sc{gdb/mi} displays this:
31078 *stopped,reason="exited-signalled",signal-name="SIGINT",
31079 signal-meaning="Interrupt"
31083 @c @subheading -exec-signal
31086 @subheading The @code{-exec-step} Command
31089 @subsubheading Synopsis
31092 -exec-step [--reverse]
31095 Resumes execution of the inferior program, stopping when the beginning
31096 of the next source line is reached, if the next source line is not a
31097 function call. If it is, stop at the first instruction of the called
31098 function. If the @samp{--reverse} option is specified, resumes reverse
31099 execution of the inferior program, stopping at the beginning of the
31100 previously executed source line.
31102 @subsubheading @value{GDBN} Command
31104 The corresponding @value{GDBN} command is @samp{step}.
31106 @subsubheading Example
31108 Stepping into a function:
31114 *stopped,reason="end-stepping-range",
31115 frame=@{func="foo",args=[@{name="a",value="10"@},
31116 @{name="b",value="0"@}],file="recursive2.c",
31117 fullname="/home/foo/bar/recursive2.c",line="11",arch="i386:x86_64"@}
31127 *stopped,reason="end-stepping-range",line="14",file="recursive2.c"
31132 @subheading The @code{-exec-step-instruction} Command
31133 @findex -exec-step-instruction
31135 @subsubheading Synopsis
31138 -exec-step-instruction [--reverse]
31141 Resumes the inferior which executes one machine instruction. If the
31142 @samp{--reverse} option is specified, resumes reverse execution of the
31143 inferior program, stopping at the previously executed instruction.
31144 The output, once @value{GDBN} has stopped, will vary depending on
31145 whether we have stopped in the middle of a source line or not. In the
31146 former case, the address at which the program stopped will be printed
31149 @subsubheading @value{GDBN} Command
31151 The corresponding @value{GDBN} command is @samp{stepi}.
31153 @subsubheading Example
31157 -exec-step-instruction
31161 *stopped,reason="end-stepping-range",
31162 frame=@{func="foo",args=[],file="try.c",
31163 fullname="/home/foo/bar/try.c",line="10",arch="i386:x86_64"@}
31165 -exec-step-instruction
31169 *stopped,reason="end-stepping-range",
31170 frame=@{addr="0x000100f4",func="foo",args=[],file="try.c",
31171 fullname="/home/foo/bar/try.c",line="10",arch="i386:x86_64"@}
31176 @subheading The @code{-exec-until} Command
31177 @findex -exec-until
31179 @subsubheading Synopsis
31182 -exec-until [ @var{location} ]
31185 Executes the inferior until the @var{location} specified in the
31186 argument is reached. If there is no argument, the inferior executes
31187 until a source line greater than the current one is reached. The
31188 reason for stopping in this case will be @samp{location-reached}.
31190 @subsubheading @value{GDBN} Command
31192 The corresponding @value{GDBN} command is @samp{until}.
31194 @subsubheading Example
31198 -exec-until recursive2.c:6
31202 *stopped,reason="location-reached",frame=@{func="main",args=[],
31203 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="6",
31204 arch="i386:x86_64"@}
31209 @subheading -file-clear
31210 Is this going away????
31213 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
31214 @node GDB/MI Stack Manipulation
31215 @section @sc{gdb/mi} Stack Manipulation Commands
31217 @subheading The @code{-enable-frame-filters} Command
31218 @findex -enable-frame-filters
31221 -enable-frame-filters
31224 @value{GDBN} allows Python-based frame filters to affect the output of
31225 the MI commands relating to stack traces. As there is no way to
31226 implement this in a fully backward-compatible way, a front end must
31227 request that this functionality be enabled.
31229 Once enabled, this feature cannot be disabled.
31231 Note that if Python support has not been compiled into @value{GDBN},
31232 this command will still succeed (and do nothing).
31234 @subheading The @code{-stack-info-frame} Command
31235 @findex -stack-info-frame
31237 @subsubheading Synopsis
31243 Get info on the selected frame.
31245 @subsubheading @value{GDBN} Command
31247 The corresponding @value{GDBN} command is @samp{info frame} or @samp{frame}
31248 (without arguments).
31250 @subsubheading Example
31255 ^done,frame=@{level="1",addr="0x0001076c",func="callee3",
31256 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
31257 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="17",
31258 arch="i386:x86_64"@}
31262 @subheading The @code{-stack-info-depth} Command
31263 @findex -stack-info-depth
31265 @subsubheading Synopsis
31268 -stack-info-depth [ @var{max-depth} ]
31271 Return the depth of the stack. If the integer argument @var{max-depth}
31272 is specified, do not count beyond @var{max-depth} frames.
31274 @subsubheading @value{GDBN} Command
31276 There's no equivalent @value{GDBN} command.
31278 @subsubheading Example
31280 For a stack with frame levels 0 through 11:
31287 -stack-info-depth 4
31290 -stack-info-depth 12
31293 -stack-info-depth 11
31296 -stack-info-depth 13
31301 @anchor{-stack-list-arguments}
31302 @subheading The @code{-stack-list-arguments} Command
31303 @findex -stack-list-arguments
31305 @subsubheading Synopsis
31308 -stack-list-arguments [ --no-frame-filters ] [ --skip-unavailable ] @var{print-values}
31309 [ @var{low-frame} @var{high-frame} ]
31312 Display a list of the arguments for the frames between @var{low-frame}
31313 and @var{high-frame} (inclusive). If @var{low-frame} and
31314 @var{high-frame} are not provided, list the arguments for the whole
31315 call stack. If the two arguments are equal, show the single frame
31316 at the corresponding level. It is an error if @var{low-frame} is
31317 larger than the actual number of frames. On the other hand,
31318 @var{high-frame} may be larger than the actual number of frames, in
31319 which case only existing frames will be returned.
31321 If @var{print-values} is 0 or @code{--no-values}, print only the names of
31322 the variables; if it is 1 or @code{--all-values}, print also their
31323 values; and if it is 2 or @code{--simple-values}, print the name,
31324 type and value for simple data types, and the name and type for arrays,
31325 structures and unions. If the option @code{--no-frame-filters} is
31326 supplied, then Python frame filters will not be executed.
31328 If the @code{--skip-unavailable} option is specified, arguments that
31329 are not available are not listed. Partially available arguments
31330 are still displayed, however.
31332 Use of this command to obtain arguments in a single frame is
31333 deprecated in favor of the @samp{-stack-list-variables} command.
31335 @subsubheading @value{GDBN} Command
31337 @value{GDBN} does not have an equivalent command. @code{gdbtk} has a
31338 @samp{gdb_get_args} command which partially overlaps with the
31339 functionality of @samp{-stack-list-arguments}.
31341 @subsubheading Example
31348 frame=@{level="0",addr="0x00010734",func="callee4",
31349 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
31350 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="8",
31351 arch="i386:x86_64"@},
31352 frame=@{level="1",addr="0x0001076c",func="callee3",
31353 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
31354 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="17",
31355 arch="i386:x86_64"@},
31356 frame=@{level="2",addr="0x0001078c",func="callee2",
31357 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
31358 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="22",
31359 arch="i386:x86_64"@},
31360 frame=@{level="3",addr="0x000107b4",func="callee1",
31361 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
31362 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="27",
31363 arch="i386:x86_64"@},
31364 frame=@{level="4",addr="0x000107e0",func="main",
31365 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
31366 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="32",
31367 arch="i386:x86_64"@}]
31369 -stack-list-arguments 0
31372 frame=@{level="0",args=[]@},
31373 frame=@{level="1",args=[name="strarg"]@},
31374 frame=@{level="2",args=[name="intarg",name="strarg"]@},
31375 frame=@{level="3",args=[name="intarg",name="strarg",name="fltarg"]@},
31376 frame=@{level="4",args=[]@}]
31378 -stack-list-arguments 1
31381 frame=@{level="0",args=[]@},
31383 args=[@{name="strarg",value="0x11940 \"A string argument.\""@}]@},
31384 frame=@{level="2",args=[
31385 @{name="intarg",value="2"@},
31386 @{name="strarg",value="0x11940 \"A string argument.\""@}]@},
31387 @{frame=@{level="3",args=[
31388 @{name="intarg",value="2"@},
31389 @{name="strarg",value="0x11940 \"A string argument.\""@},
31390 @{name="fltarg",value="3.5"@}]@},
31391 frame=@{level="4",args=[]@}]
31393 -stack-list-arguments 0 2 2
31394 ^done,stack-args=[frame=@{level="2",args=[name="intarg",name="strarg"]@}]
31396 -stack-list-arguments 1 2 2
31397 ^done,stack-args=[frame=@{level="2",
31398 args=[@{name="intarg",value="2"@},
31399 @{name="strarg",value="0x11940 \"A string argument.\""@}]@}]
31403 @c @subheading -stack-list-exception-handlers
31406 @anchor{-stack-list-frames}
31407 @subheading The @code{-stack-list-frames} Command
31408 @findex -stack-list-frames
31410 @subsubheading Synopsis
31413 -stack-list-frames [ --no-frame-filters @var{low-frame} @var{high-frame} ]
31416 List the frames currently on the stack. For each frame it displays the
31421 The frame number, 0 being the topmost frame, i.e., the innermost function.
31423 The @code{$pc} value for that frame.
31427 File name of the source file where the function lives.
31428 @item @var{fullname}
31429 The full file name of the source file where the function lives.
31431 Line number corresponding to the @code{$pc}.
31433 The shared library where this function is defined. This is only given
31434 if the frame's function is not known.
31436 Frame's architecture.
31439 If invoked without arguments, this command prints a backtrace for the
31440 whole stack. If given two integer arguments, it shows the frames whose
31441 levels are between the two arguments (inclusive). If the two arguments
31442 are equal, it shows the single frame at the corresponding level. It is
31443 an error if @var{low-frame} is larger than the actual number of
31444 frames. On the other hand, @var{high-frame} may be larger than the
31445 actual number of frames, in which case only existing frames will be
31446 returned. If the option @code{--no-frame-filters} is supplied, then
31447 Python frame filters will not be executed.
31449 @subsubheading @value{GDBN} Command
31451 The corresponding @value{GDBN} commands are @samp{backtrace} and @samp{where}.
31453 @subsubheading Example
31455 Full stack backtrace:
31461 [frame=@{level="0",addr="0x0001076c",func="foo",
31462 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="11",
31463 arch="i386:x86_64"@},
31464 frame=@{level="1",addr="0x000107a4",func="foo",
31465 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
31466 arch="i386:x86_64"@},
31467 frame=@{level="2",addr="0x000107a4",func="foo",
31468 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
31469 arch="i386:x86_64"@},
31470 frame=@{level="3",addr="0x000107a4",func="foo",
31471 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
31472 arch="i386:x86_64"@},
31473 frame=@{level="4",addr="0x000107a4",func="foo",
31474 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
31475 arch="i386:x86_64"@},
31476 frame=@{level="5",addr="0x000107a4",func="foo",
31477 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
31478 arch="i386:x86_64"@},
31479 frame=@{level="6",addr="0x000107a4",func="foo",
31480 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
31481 arch="i386:x86_64"@},
31482 frame=@{level="7",addr="0x000107a4",func="foo",
31483 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
31484 arch="i386:x86_64"@},
31485 frame=@{level="8",addr="0x000107a4",func="foo",
31486 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
31487 arch="i386:x86_64"@},
31488 frame=@{level="9",addr="0x000107a4",func="foo",
31489 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
31490 arch="i386:x86_64"@},
31491 frame=@{level="10",addr="0x000107a4",func="foo",
31492 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
31493 arch="i386:x86_64"@},
31494 frame=@{level="11",addr="0x00010738",func="main",
31495 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="4",
31496 arch="i386:x86_64"@}]
31500 Show frames between @var{low_frame} and @var{high_frame}:
31504 -stack-list-frames 3 5
31506 [frame=@{level="3",addr="0x000107a4",func="foo",
31507 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
31508 arch="i386:x86_64"@},
31509 frame=@{level="4",addr="0x000107a4",func="foo",
31510 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
31511 arch="i386:x86_64"@},
31512 frame=@{level="5",addr="0x000107a4",func="foo",
31513 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
31514 arch="i386:x86_64"@}]
31518 Show a single frame:
31522 -stack-list-frames 3 3
31524 [frame=@{level="3",addr="0x000107a4",func="foo",
31525 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
31526 arch="i386:x86_64"@}]
31531 @subheading The @code{-stack-list-locals} Command
31532 @findex -stack-list-locals
31533 @anchor{-stack-list-locals}
31535 @subsubheading Synopsis
31538 -stack-list-locals [ --no-frame-filters ] [ --skip-unavailable ] @var{print-values}
31541 Display the local variable names for the selected frame. If
31542 @var{print-values} is 0 or @code{--no-values}, print only the names of
31543 the variables; if it is 1 or @code{--all-values}, print also their
31544 values; and if it is 2 or @code{--simple-values}, print the name,
31545 type and value for simple data types, and the name and type for arrays,
31546 structures and unions. In this last case, a frontend can immediately
31547 display the value of simple data types and create variable objects for
31548 other data types when the user wishes to explore their values in
31549 more detail. If the option @code{--no-frame-filters} is supplied, then
31550 Python frame filters will not be executed.
31552 If the @code{--skip-unavailable} option is specified, local variables
31553 that are not available are not listed. Partially available local
31554 variables are still displayed, however.
31556 This command is deprecated in favor of the
31557 @samp{-stack-list-variables} command.
31559 @subsubheading @value{GDBN} Command
31561 @samp{info locals} in @value{GDBN}, @samp{gdb_get_locals} in @code{gdbtk}.
31563 @subsubheading Example
31567 -stack-list-locals 0
31568 ^done,locals=[name="A",name="B",name="C"]
31570 -stack-list-locals --all-values
31571 ^done,locals=[@{name="A",value="1"@},@{name="B",value="2"@},
31572 @{name="C",value="@{1, 2, 3@}"@}]
31573 -stack-list-locals --simple-values
31574 ^done,locals=[@{name="A",type="int",value="1"@},
31575 @{name="B",type="int",value="2"@},@{name="C",type="int [3]"@}]
31579 @anchor{-stack-list-variables}
31580 @subheading The @code{-stack-list-variables} Command
31581 @findex -stack-list-variables
31583 @subsubheading Synopsis
31586 -stack-list-variables [ --no-frame-filters ] [ --skip-unavailable ] @var{print-values}
31589 Display the names of local variables and function arguments for the selected frame. If
31590 @var{print-values} is 0 or @code{--no-values}, print only the names of
31591 the variables; if it is 1 or @code{--all-values}, print also their
31592 values; and if it is 2 or @code{--simple-values}, print the name,
31593 type and value for simple data types, and the name and type for arrays,
31594 structures and unions. If the option @code{--no-frame-filters} is
31595 supplied, then Python frame filters will not be executed.
31597 If the @code{--skip-unavailable} option is specified, local variables
31598 and arguments that are not available are not listed. Partially
31599 available arguments and local variables are still displayed, however.
31601 @subsubheading Example
31605 -stack-list-variables --thread 1 --frame 0 --all-values
31606 ^done,variables=[@{name="x",value="11"@},@{name="s",value="@{a = 1, b = 2@}"@}]
31611 @subheading The @code{-stack-select-frame} Command
31612 @findex -stack-select-frame
31614 @subsubheading Synopsis
31617 -stack-select-frame @var{framenum}
31620 Change the selected frame. Select a different frame @var{framenum} on
31623 This command in deprecated in favor of passing the @samp{--frame}
31624 option to every command.
31626 @subsubheading @value{GDBN} Command
31628 The corresponding @value{GDBN} commands are @samp{frame}, @samp{up},
31629 @samp{down}, @samp{select-frame}, @samp{up-silent}, and @samp{down-silent}.
31631 @subsubheading Example
31635 -stack-select-frame 2
31640 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
31641 @node GDB/MI Variable Objects
31642 @section @sc{gdb/mi} Variable Objects
31646 @subheading Motivation for Variable Objects in @sc{gdb/mi}
31648 For the implementation of a variable debugger window (locals, watched
31649 expressions, etc.), we are proposing the adaptation of the existing code
31650 used by @code{Insight}.
31652 The two main reasons for that are:
31656 It has been proven in practice (it is already on its second generation).
31659 It will shorten development time (needless to say how important it is
31663 The original interface was designed to be used by Tcl code, so it was
31664 slightly changed so it could be used through @sc{gdb/mi}. This section
31665 describes the @sc{gdb/mi} operations that will be available and gives some
31666 hints about their use.
31668 @emph{Note}: In addition to the set of operations described here, we
31669 expect the @sc{gui} implementation of a variable window to require, at
31670 least, the following operations:
31673 @item @code{-gdb-show} @code{output-radix}
31674 @item @code{-stack-list-arguments}
31675 @item @code{-stack-list-locals}
31676 @item @code{-stack-select-frame}
31681 @subheading Introduction to Variable Objects
31683 @cindex variable objects in @sc{gdb/mi}
31685 Variable objects are "object-oriented" MI interface for examining and
31686 changing values of expressions. Unlike some other MI interfaces that
31687 work with expressions, variable objects are specifically designed for
31688 simple and efficient presentation in the frontend. A variable object
31689 is identified by string name. When a variable object is created, the
31690 frontend specifies the expression for that variable object. The
31691 expression can be a simple variable, or it can be an arbitrary complex
31692 expression, and can even involve CPU registers. After creating a
31693 variable object, the frontend can invoke other variable object
31694 operations---for example to obtain or change the value of a variable
31695 object, or to change display format.
31697 Variable objects have hierarchical tree structure. Any variable object
31698 that corresponds to a composite type, such as structure in C, has
31699 a number of child variable objects, for example corresponding to each
31700 element of a structure. A child variable object can itself have
31701 children, recursively. Recursion ends when we reach
31702 leaf variable objects, which always have built-in types. Child variable
31703 objects are created only by explicit request, so if a frontend
31704 is not interested in the children of a particular variable object, no
31705 child will be created.
31707 For a leaf variable object it is possible to obtain its value as a
31708 string, or set the value from a string. String value can be also
31709 obtained for a non-leaf variable object, but it's generally a string
31710 that only indicates the type of the object, and does not list its
31711 contents. Assignment to a non-leaf variable object is not allowed.
31713 A frontend does not need to read the values of all variable objects each time
31714 the program stops. Instead, MI provides an update command that lists all
31715 variable objects whose values has changed since the last update
31716 operation. This considerably reduces the amount of data that must
31717 be transferred to the frontend. As noted above, children variable
31718 objects are created on demand, and only leaf variable objects have a
31719 real value. As result, gdb will read target memory only for leaf
31720 variables that frontend has created.
31722 The automatic update is not always desirable. For example, a frontend
31723 might want to keep a value of some expression for future reference,
31724 and never update it. For another example, fetching memory is
31725 relatively slow for embedded targets, so a frontend might want
31726 to disable automatic update for the variables that are either not
31727 visible on the screen, or ``closed''. This is possible using so
31728 called ``frozen variable objects''. Such variable objects are never
31729 implicitly updated.
31731 Variable objects can be either @dfn{fixed} or @dfn{floating}. For the
31732 fixed variable object, the expression is parsed when the variable
31733 object is created, including associating identifiers to specific
31734 variables. The meaning of expression never changes. For a floating
31735 variable object the values of variables whose names appear in the
31736 expressions are re-evaluated every time in the context of the current
31737 frame. Consider this example:
31742 struct work_state state;
31749 If a fixed variable object for the @code{state} variable is created in
31750 this function, and we enter the recursive call, the variable
31751 object will report the value of @code{state} in the top-level
31752 @code{do_work} invocation. On the other hand, a floating variable
31753 object will report the value of @code{state} in the current frame.
31755 If an expression specified when creating a fixed variable object
31756 refers to a local variable, the variable object becomes bound to the
31757 thread and frame in which the variable object is created. When such
31758 variable object is updated, @value{GDBN} makes sure that the
31759 thread/frame combination the variable object is bound to still exists,
31760 and re-evaluates the variable object in context of that thread/frame.
31762 The following is the complete set of @sc{gdb/mi} operations defined to
31763 access this functionality:
31765 @multitable @columnfractions .4 .6
31766 @item @strong{Operation}
31767 @tab @strong{Description}
31769 @item @code{-enable-pretty-printing}
31770 @tab enable Python-based pretty-printing
31771 @item @code{-var-create}
31772 @tab create a variable object
31773 @item @code{-var-delete}
31774 @tab delete the variable object and/or its children
31775 @item @code{-var-set-format}
31776 @tab set the display format of this variable
31777 @item @code{-var-show-format}
31778 @tab show the display format of this variable
31779 @item @code{-var-info-num-children}
31780 @tab tells how many children this object has
31781 @item @code{-var-list-children}
31782 @tab return a list of the object's children
31783 @item @code{-var-info-type}
31784 @tab show the type of this variable object
31785 @item @code{-var-info-expression}
31786 @tab print parent-relative expression that this variable object represents
31787 @item @code{-var-info-path-expression}
31788 @tab print full expression that this variable object represents
31789 @item @code{-var-show-attributes}
31790 @tab is this variable editable? does it exist here?
31791 @item @code{-var-evaluate-expression}
31792 @tab get the value of this variable
31793 @item @code{-var-assign}
31794 @tab set the value of this variable
31795 @item @code{-var-update}
31796 @tab update the variable and its children
31797 @item @code{-var-set-frozen}
31798 @tab set frozeness attribute
31799 @item @code{-var-set-update-range}
31800 @tab set range of children to display on update
31803 In the next subsection we describe each operation in detail and suggest
31804 how it can be used.
31806 @subheading Description And Use of Operations on Variable Objects
31808 @subheading The @code{-enable-pretty-printing} Command
31809 @findex -enable-pretty-printing
31812 -enable-pretty-printing
31815 @value{GDBN} allows Python-based visualizers to affect the output of the
31816 MI variable object commands. However, because there was no way to
31817 implement this in a fully backward-compatible way, a front end must
31818 request that this functionality be enabled.
31820 Once enabled, this feature cannot be disabled.
31822 Note that if Python support has not been compiled into @value{GDBN},
31823 this command will still succeed (and do nothing).
31825 This feature is currently (as of @value{GDBN} 7.0) experimental, and
31826 may work differently in future versions of @value{GDBN}.
31828 @subheading The @code{-var-create} Command
31829 @findex -var-create
31831 @subsubheading Synopsis
31834 -var-create @{@var{name} | "-"@}
31835 @{@var{frame-addr} | "*" | "@@"@} @var{expression}
31838 This operation creates a variable object, which allows the monitoring of
31839 a variable, the result of an expression, a memory cell or a CPU
31842 The @var{name} parameter is the string by which the object can be
31843 referenced. It must be unique. If @samp{-} is specified, the varobj
31844 system will generate a string ``varNNNNNN'' automatically. It will be
31845 unique provided that one does not specify @var{name} of that format.
31846 The command fails if a duplicate name is found.
31848 The frame under which the expression should be evaluated can be
31849 specified by @var{frame-addr}. A @samp{*} indicates that the current
31850 frame should be used. A @samp{@@} indicates that a floating variable
31851 object must be created.
31853 @var{expression} is any expression valid on the current language set (must not
31854 begin with a @samp{*}), or one of the following:
31858 @samp{*@var{addr}}, where @var{addr} is the address of a memory cell
31861 @samp{*@var{addr}-@var{addr}} --- a memory address range (TBD)
31864 @samp{$@var{regname}} --- a CPU register name
31867 @cindex dynamic varobj
31868 A varobj's contents may be provided by a Python-based pretty-printer. In this
31869 case the varobj is known as a @dfn{dynamic varobj}. Dynamic varobjs
31870 have slightly different semantics in some cases. If the
31871 @code{-enable-pretty-printing} command is not sent, then @value{GDBN}
31872 will never create a dynamic varobj. This ensures backward
31873 compatibility for existing clients.
31875 @subsubheading Result
31877 This operation returns attributes of the newly-created varobj. These
31882 The name of the varobj.
31885 The number of children of the varobj. This number is not necessarily
31886 reliable for a dynamic varobj. Instead, you must examine the
31887 @samp{has_more} attribute.
31890 The varobj's scalar value. For a varobj whose type is some sort of
31891 aggregate (e.g., a @code{struct}), or for a dynamic varobj, this value
31892 will not be interesting.
31895 The varobj's type. This is a string representation of the type, as
31896 would be printed by the @value{GDBN} CLI. If @samp{print object}
31897 (@pxref{Print Settings, set print object}) is set to @code{on}, the
31898 @emph{actual} (derived) type of the object is shown rather than the
31899 @emph{declared} one.
31902 If a variable object is bound to a specific thread, then this is the
31903 thread's global identifier.
31906 For a dynamic varobj, this indicates whether there appear to be any
31907 children available. For a non-dynamic varobj, this will be 0.
31910 This attribute will be present and have the value @samp{1} if the
31911 varobj is a dynamic varobj. If the varobj is not a dynamic varobj,
31912 then this attribute will not be present.
31915 A dynamic varobj can supply a display hint to the front end. The
31916 value comes directly from the Python pretty-printer object's
31917 @code{display_hint} method. @xref{Pretty Printing API}.
31920 Typical output will look like this:
31923 name="@var{name}",numchild="@var{N}",type="@var{type}",thread-id="@var{M}",
31924 has_more="@var{has_more}"
31928 @subheading The @code{-var-delete} Command
31929 @findex -var-delete
31931 @subsubheading Synopsis
31934 -var-delete [ -c ] @var{name}
31937 Deletes a previously created variable object and all of its children.
31938 With the @samp{-c} option, just deletes the children.
31940 Returns an error if the object @var{name} is not found.
31943 @subheading The @code{-var-set-format} Command
31944 @findex -var-set-format
31946 @subsubheading Synopsis
31949 -var-set-format @var{name} @var{format-spec}
31952 Sets the output format for the value of the object @var{name} to be
31955 @anchor{-var-set-format}
31956 The syntax for the @var{format-spec} is as follows:
31959 @var{format-spec} @expansion{}
31960 @{binary | decimal | hexadecimal | octal | natural | zero-hexadecimal@}
31963 The natural format is the default format choosen automatically
31964 based on the variable type (like decimal for an @code{int}, hex
31965 for pointers, etc.).
31967 The zero-hexadecimal format has a representation similar to hexadecimal
31968 but with padding zeroes to the left of the value. For example, a 32-bit
31969 hexadecimal value of 0x1234 would be represented as 0x00001234 in the
31970 zero-hexadecimal format.
31972 For a variable with children, the format is set only on the
31973 variable itself, and the children are not affected.
31975 @subheading The @code{-var-show-format} Command
31976 @findex -var-show-format
31978 @subsubheading Synopsis
31981 -var-show-format @var{name}
31984 Returns the format used to display the value of the object @var{name}.
31987 @var{format} @expansion{}
31992 @subheading The @code{-var-info-num-children} Command
31993 @findex -var-info-num-children
31995 @subsubheading Synopsis
31998 -var-info-num-children @var{name}
32001 Returns the number of children of a variable object @var{name}:
32007 Note that this number is not completely reliable for a dynamic varobj.
32008 It will return the current number of children, but more children may
32012 @subheading The @code{-var-list-children} Command
32013 @findex -var-list-children
32015 @subsubheading Synopsis
32018 -var-list-children [@var{print-values}] @var{name} [@var{from} @var{to}]
32020 @anchor{-var-list-children}
32022 Return a list of the children of the specified variable object and
32023 create variable objects for them, if they do not already exist. With
32024 a single argument or if @var{print-values} has a value of 0 or
32025 @code{--no-values}, print only the names of the variables; if
32026 @var{print-values} is 1 or @code{--all-values}, also print their
32027 values; and if it is 2 or @code{--simple-values} print the name and
32028 value for simple data types and just the name for arrays, structures
32031 @var{from} and @var{to}, if specified, indicate the range of children
32032 to report. If @var{from} or @var{to} is less than zero, the range is
32033 reset and all children will be reported. Otherwise, children starting
32034 at @var{from} (zero-based) and up to and excluding @var{to} will be
32037 If a child range is requested, it will only affect the current call to
32038 @code{-var-list-children}, but not future calls to @code{-var-update}.
32039 For this, you must instead use @code{-var-set-update-range}. The
32040 intent of this approach is to enable a front end to implement any
32041 update approach it likes; for example, scrolling a view may cause the
32042 front end to request more children with @code{-var-list-children}, and
32043 then the front end could call @code{-var-set-update-range} with a
32044 different range to ensure that future updates are restricted to just
32047 For each child the following results are returned:
32052 Name of the variable object created for this child.
32055 The expression to be shown to the user by the front end to designate this child.
32056 For example this may be the name of a structure member.
32058 For a dynamic varobj, this value cannot be used to form an
32059 expression. There is no way to do this at all with a dynamic varobj.
32061 For C/C@t{++} structures there are several pseudo children returned to
32062 designate access qualifiers. For these pseudo children @var{exp} is
32063 @samp{public}, @samp{private}, or @samp{protected}. In this case the
32064 type and value are not present.
32066 A dynamic varobj will not report the access qualifying
32067 pseudo-children, regardless of the language. This information is not
32068 available at all with a dynamic varobj.
32071 Number of children this child has. For a dynamic varobj, this will be
32075 The type of the child. If @samp{print object}
32076 (@pxref{Print Settings, set print object}) is set to @code{on}, the
32077 @emph{actual} (derived) type of the object is shown rather than the
32078 @emph{declared} one.
32081 If values were requested, this is the value.
32084 If this variable object is associated with a thread, this is the
32085 thread's global thread id. Otherwise this result is not present.
32088 If the variable object is frozen, this variable will be present with a value of 1.
32091 A dynamic varobj can supply a display hint to the front end. The
32092 value comes directly from the Python pretty-printer object's
32093 @code{display_hint} method. @xref{Pretty Printing API}.
32096 This attribute will be present and have the value @samp{1} if the
32097 varobj is a dynamic varobj. If the varobj is not a dynamic varobj,
32098 then this attribute will not be present.
32102 The result may have its own attributes:
32106 A dynamic varobj can supply a display hint to the front end. The
32107 value comes directly from the Python pretty-printer object's
32108 @code{display_hint} method. @xref{Pretty Printing API}.
32111 This is an integer attribute which is nonzero if there are children
32112 remaining after the end of the selected range.
32115 @subsubheading Example
32119 -var-list-children n
32120 ^done,numchild=@var{n},children=[child=@{name=@var{name},exp=@var{exp},
32121 numchild=@var{n},type=@var{type}@},@r{(repeats N times)}]
32123 -var-list-children --all-values n
32124 ^done,numchild=@var{n},children=[child=@{name=@var{name},exp=@var{exp},
32125 numchild=@var{n},value=@var{value},type=@var{type}@},@r{(repeats N times)}]
32129 @subheading The @code{-var-info-type} Command
32130 @findex -var-info-type
32132 @subsubheading Synopsis
32135 -var-info-type @var{name}
32138 Returns the type of the specified variable @var{name}. The type is
32139 returned as a string in the same format as it is output by the
32143 type=@var{typename}
32147 @subheading The @code{-var-info-expression} Command
32148 @findex -var-info-expression
32150 @subsubheading Synopsis
32153 -var-info-expression @var{name}
32156 Returns a string that is suitable for presenting this
32157 variable object in user interface. The string is generally
32158 not valid expression in the current language, and cannot be evaluated.
32160 For example, if @code{a} is an array, and variable object
32161 @code{A} was created for @code{a}, then we'll get this output:
32164 (gdb) -var-info-expression A.1
32165 ^done,lang="C",exp="1"
32169 Here, the value of @code{lang} is the language name, which can be
32170 found in @ref{Supported Languages}.
32172 Note that the output of the @code{-var-list-children} command also
32173 includes those expressions, so the @code{-var-info-expression} command
32176 @subheading The @code{-var-info-path-expression} Command
32177 @findex -var-info-path-expression
32179 @subsubheading Synopsis
32182 -var-info-path-expression @var{name}
32185 Returns an expression that can be evaluated in the current
32186 context and will yield the same value that a variable object has.
32187 Compare this with the @code{-var-info-expression} command, which
32188 result can be used only for UI presentation. Typical use of
32189 the @code{-var-info-path-expression} command is creating a
32190 watchpoint from a variable object.
32192 This command is currently not valid for children of a dynamic varobj,
32193 and will give an error when invoked on one.
32195 For example, suppose @code{C} is a C@t{++} class, derived from class
32196 @code{Base}, and that the @code{Base} class has a member called
32197 @code{m_size}. Assume a variable @code{c} is has the type of
32198 @code{C} and a variable object @code{C} was created for variable
32199 @code{c}. Then, we'll get this output:
32201 (gdb) -var-info-path-expression C.Base.public.m_size
32202 ^done,path_expr=((Base)c).m_size)
32205 @subheading The @code{-var-show-attributes} Command
32206 @findex -var-show-attributes
32208 @subsubheading Synopsis
32211 -var-show-attributes @var{name}
32214 List attributes of the specified variable object @var{name}:
32217 status=@var{attr} [ ( ,@var{attr} )* ]
32221 where @var{attr} is @code{@{ @{ editable | noneditable @} | TBD @}}.
32223 @subheading The @code{-var-evaluate-expression} Command
32224 @findex -var-evaluate-expression
32226 @subsubheading Synopsis
32229 -var-evaluate-expression [-f @var{format-spec}] @var{name}
32232 Evaluates the expression that is represented by the specified variable
32233 object and returns its value as a string. The format of the string
32234 can be specified with the @samp{-f} option. The possible values of
32235 this option are the same as for @code{-var-set-format}
32236 (@pxref{-var-set-format}). If the @samp{-f} option is not specified,
32237 the current display format will be used. The current display format
32238 can be changed using the @code{-var-set-format} command.
32244 Note that one must invoke @code{-var-list-children} for a variable
32245 before the value of a child variable can be evaluated.
32247 @subheading The @code{-var-assign} Command
32248 @findex -var-assign
32250 @subsubheading Synopsis
32253 -var-assign @var{name} @var{expression}
32256 Assigns the value of @var{expression} to the variable object specified
32257 by @var{name}. The object must be @samp{editable}. If the variable's
32258 value is altered by the assign, the variable will show up in any
32259 subsequent @code{-var-update} list.
32261 @subsubheading Example
32269 ^done,changelist=[@{name="var1",in_scope="true",type_changed="false"@}]
32273 @subheading The @code{-var-update} Command
32274 @findex -var-update
32276 @subsubheading Synopsis
32279 -var-update [@var{print-values}] @{@var{name} | "*"@}
32282 Reevaluate the expressions corresponding to the variable object
32283 @var{name} and all its direct and indirect children, and return the
32284 list of variable objects whose values have changed; @var{name} must
32285 be a root variable object. Here, ``changed'' means that the result of
32286 @code{-var-evaluate-expression} before and after the
32287 @code{-var-update} is different. If @samp{*} is used as the variable
32288 object names, all existing variable objects are updated, except
32289 for frozen ones (@pxref{-var-set-frozen}). The option
32290 @var{print-values} determines whether both names and values, or just
32291 names are printed. The possible values of this option are the same
32292 as for @code{-var-list-children} (@pxref{-var-list-children}). It is
32293 recommended to use the @samp{--all-values} option, to reduce the
32294 number of MI commands needed on each program stop.
32296 With the @samp{*} parameter, if a variable object is bound to a
32297 currently running thread, it will not be updated, without any
32300 If @code{-var-set-update-range} was previously used on a varobj, then
32301 only the selected range of children will be reported.
32303 @code{-var-update} reports all the changed varobjs in a tuple named
32306 Each item in the change list is itself a tuple holding:
32310 The name of the varobj.
32313 If values were requested for this update, then this field will be
32314 present and will hold the value of the varobj.
32317 @anchor{-var-update}
32318 This field is a string which may take one of three values:
32322 The variable object's current value is valid.
32325 The variable object does not currently hold a valid value but it may
32326 hold one in the future if its associated expression comes back into
32330 The variable object no longer holds a valid value.
32331 This can occur when the executable file being debugged has changed,
32332 either through recompilation or by using the @value{GDBN} @code{file}
32333 command. The front end should normally choose to delete these variable
32337 In the future new values may be added to this list so the front should
32338 be prepared for this possibility. @xref{GDB/MI Development and Front Ends, ,@sc{GDB/MI} Development and Front Ends}.
32341 This is only present if the varobj is still valid. If the type
32342 changed, then this will be the string @samp{true}; otherwise it will
32345 When a varobj's type changes, its children are also likely to have
32346 become incorrect. Therefore, the varobj's children are automatically
32347 deleted when this attribute is @samp{true}. Also, the varobj's update
32348 range, when set using the @code{-var-set-update-range} command, is
32352 If the varobj's type changed, then this field will be present and will
32355 @item new_num_children
32356 For a dynamic varobj, if the number of children changed, or if the
32357 type changed, this will be the new number of children.
32359 The @samp{numchild} field in other varobj responses is generally not
32360 valid for a dynamic varobj -- it will show the number of children that
32361 @value{GDBN} knows about, but because dynamic varobjs lazily
32362 instantiate their children, this will not reflect the number of
32363 children which may be available.
32365 The @samp{new_num_children} attribute only reports changes to the
32366 number of children known by @value{GDBN}. This is the only way to
32367 detect whether an update has removed children (which necessarily can
32368 only happen at the end of the update range).
32371 The display hint, if any.
32374 This is an integer value, which will be 1 if there are more children
32375 available outside the varobj's update range.
32378 This attribute will be present and have the value @samp{1} if the
32379 varobj is a dynamic varobj. If the varobj is not a dynamic varobj,
32380 then this attribute will not be present.
32383 If new children were added to a dynamic varobj within the selected
32384 update range (as set by @code{-var-set-update-range}), then they will
32385 be listed in this attribute.
32388 @subsubheading Example
32395 -var-update --all-values var1
32396 ^done,changelist=[@{name="var1",value="3",in_scope="true",
32397 type_changed="false"@}]
32401 @subheading The @code{-var-set-frozen} Command
32402 @findex -var-set-frozen
32403 @anchor{-var-set-frozen}
32405 @subsubheading Synopsis
32408 -var-set-frozen @var{name} @var{flag}
32411 Set the frozenness flag on the variable object @var{name}. The
32412 @var{flag} parameter should be either @samp{1} to make the variable
32413 frozen or @samp{0} to make it unfrozen. If a variable object is
32414 frozen, then neither itself, nor any of its children, are
32415 implicitly updated by @code{-var-update} of
32416 a parent variable or by @code{-var-update *}. Only
32417 @code{-var-update} of the variable itself will update its value and
32418 values of its children. After a variable object is unfrozen, it is
32419 implicitly updated by all subsequent @code{-var-update} operations.
32420 Unfreezing a variable does not update it, only subsequent
32421 @code{-var-update} does.
32423 @subsubheading Example
32427 -var-set-frozen V 1
32432 @subheading The @code{-var-set-update-range} command
32433 @findex -var-set-update-range
32434 @anchor{-var-set-update-range}
32436 @subsubheading Synopsis
32439 -var-set-update-range @var{name} @var{from} @var{to}
32442 Set the range of children to be returned by future invocations of
32443 @code{-var-update}.
32445 @var{from} and @var{to} indicate the range of children to report. If
32446 @var{from} or @var{to} is less than zero, the range is reset and all
32447 children will be reported. Otherwise, children starting at @var{from}
32448 (zero-based) and up to and excluding @var{to} will be reported.
32450 @subsubheading Example
32454 -var-set-update-range V 1 2
32458 @subheading The @code{-var-set-visualizer} command
32459 @findex -var-set-visualizer
32460 @anchor{-var-set-visualizer}
32462 @subsubheading Synopsis
32465 -var-set-visualizer @var{name} @var{visualizer}
32468 Set a visualizer for the variable object @var{name}.
32470 @var{visualizer} is the visualizer to use. The special value
32471 @samp{None} means to disable any visualizer in use.
32473 If not @samp{None}, @var{visualizer} must be a Python expression.
32474 This expression must evaluate to a callable object which accepts a
32475 single argument. @value{GDBN} will call this object with the value of
32476 the varobj @var{name} as an argument (this is done so that the same
32477 Python pretty-printing code can be used for both the CLI and MI).
32478 When called, this object must return an object which conforms to the
32479 pretty-printing interface (@pxref{Pretty Printing API}).
32481 The pre-defined function @code{gdb.default_visualizer} may be used to
32482 select a visualizer by following the built-in process
32483 (@pxref{Selecting Pretty-Printers}). This is done automatically when
32484 a varobj is created, and so ordinarily is not needed.
32486 This feature is only available if Python support is enabled. The MI
32487 command @code{-list-features} (@pxref{GDB/MI Support Commands})
32488 can be used to check this.
32490 @subsubheading Example
32492 Resetting the visualizer:
32496 -var-set-visualizer V None
32500 Reselecting the default (type-based) visualizer:
32504 -var-set-visualizer V gdb.default_visualizer
32508 Suppose @code{SomeClass} is a visualizer class. A lambda expression
32509 can be used to instantiate this class for a varobj:
32513 -var-set-visualizer V "lambda val: SomeClass()"
32517 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
32518 @node GDB/MI Data Manipulation
32519 @section @sc{gdb/mi} Data Manipulation
32521 @cindex data manipulation, in @sc{gdb/mi}
32522 @cindex @sc{gdb/mi}, data manipulation
32523 This section describes the @sc{gdb/mi} commands that manipulate data:
32524 examine memory and registers, evaluate expressions, etc.
32526 For details about what an addressable memory unit is,
32527 @pxref{addressable memory unit}.
32529 @c REMOVED FROM THE INTERFACE.
32530 @c @subheading -data-assign
32531 @c Change the value of a program variable. Plenty of side effects.
32532 @c @subsubheading GDB Command
32534 @c @subsubheading Example
32537 @subheading The @code{-data-disassemble} Command
32538 @findex -data-disassemble
32540 @subsubheading Synopsis
32544 [ -s @var{start-addr} -e @var{end-addr} ]
32545 | [ -a @var{addr} ]
32546 | [ -f @var{filename} -l @var{linenum} [ -n @var{lines} ] ]
32554 @item @var{start-addr}
32555 is the beginning address (or @code{$pc})
32556 @item @var{end-addr}
32559 is an address anywhere within (or the name of) the function to
32560 disassemble. If an address is specified, the whole function
32561 surrounding that address will be disassembled. If a name is
32562 specified, the whole function with that name will be disassembled.
32563 @item @var{filename}
32564 is the name of the file to disassemble
32565 @item @var{linenum}
32566 is the line number to disassemble around
32568 is the number of disassembly lines to be produced. If it is -1,
32569 the whole function will be disassembled, in case no @var{end-addr} is
32570 specified. If @var{end-addr} is specified as a non-zero value, and
32571 @var{lines} is lower than the number of disassembly lines between
32572 @var{start-addr} and @var{end-addr}, only @var{lines} lines are
32573 displayed; if @var{lines} is higher than the number of lines between
32574 @var{start-addr} and @var{end-addr}, only the lines up to @var{end-addr}
32579 @item 0 disassembly only
32580 @item 1 mixed source and disassembly (deprecated)
32581 @item 2 disassembly with raw opcodes
32582 @item 3 mixed source and disassembly with raw opcodes (deprecated)
32583 @item 4 mixed source and disassembly
32584 @item 5 mixed source and disassembly with raw opcodes
32587 Modes 1 and 3 are deprecated. The output is ``source centric''
32588 which hasn't proved useful in practice.
32589 @xref{Machine Code}, for a discussion of the difference between
32590 @code{/m} and @code{/s} output of the @code{disassemble} command.
32593 @subsubheading Result
32595 The result of the @code{-data-disassemble} command will be a list named
32596 @samp{asm_insns}, the contents of this list depend on the @var{mode}
32597 used with the @code{-data-disassemble} command.
32599 For modes 0 and 2 the @samp{asm_insns} list contains tuples with the
32604 The address at which this instruction was disassembled.
32607 The name of the function this instruction is within.
32610 The decimal offset in bytes from the start of @samp{func-name}.
32613 The text disassembly for this @samp{address}.
32616 This field is only present for modes 2, 3 and 5. This contains the raw opcode
32617 bytes for the @samp{inst} field.
32621 For modes 1, 3, 4 and 5 the @samp{asm_insns} list contains tuples named
32622 @samp{src_and_asm_line}, each of which has the following fields:
32626 The line number within @samp{file}.
32629 The file name from the compilation unit. This might be an absolute
32630 file name or a relative file name depending on the compile command
32634 Absolute file name of @samp{file}. It is converted to a canonical form
32635 using the source file search path
32636 (@pxref{Source Path, ,Specifying Source Directories})
32637 and after resolving all the symbolic links.
32639 If the source file is not found this field will contain the path as
32640 present in the debug information.
32642 @item line_asm_insn
32643 This is a list of tuples containing the disassembly for @samp{line} in
32644 @samp{file}. The fields of each tuple are the same as for
32645 @code{-data-disassemble} in @var{mode} 0 and 2, so @samp{address},
32646 @samp{func-name}, @samp{offset}, @samp{inst}, and optionally
32651 Note that whatever included in the @samp{inst} field, is not
32652 manipulated directly by @sc{gdb/mi}, i.e., it is not possible to
32655 @subsubheading @value{GDBN} Command
32657 The corresponding @value{GDBN} command is @samp{disassemble}.
32659 @subsubheading Example
32661 Disassemble from the current value of @code{$pc} to @code{$pc + 20}:
32665 -data-disassemble -s $pc -e "$pc + 20" -- 0
32668 @{address="0x000107c0",func-name="main",offset="4",
32669 inst="mov 2, %o0"@},
32670 @{address="0x000107c4",func-name="main",offset="8",
32671 inst="sethi %hi(0x11800), %o2"@},
32672 @{address="0x000107c8",func-name="main",offset="12",
32673 inst="or %o2, 0x140, %o1\t! 0x11940 <_lib_version+8>"@},
32674 @{address="0x000107cc",func-name="main",offset="16",
32675 inst="sethi %hi(0x11800), %o2"@},
32676 @{address="0x000107d0",func-name="main",offset="20",
32677 inst="or %o2, 0x168, %o4\t! 0x11968 <_lib_version+48>"@}]
32681 Disassemble the whole @code{main} function. Line 32 is part of
32685 -data-disassemble -f basics.c -l 32 -- 0
32687 @{address="0x000107bc",func-name="main",offset="0",
32688 inst="save %sp, -112, %sp"@},
32689 @{address="0x000107c0",func-name="main",offset="4",
32690 inst="mov 2, %o0"@},
32691 @{address="0x000107c4",func-name="main",offset="8",
32692 inst="sethi %hi(0x11800), %o2"@},
32694 @{address="0x0001081c",func-name="main",offset="96",inst="ret "@},
32695 @{address="0x00010820",func-name="main",offset="100",inst="restore "@}]
32699 Disassemble 3 instructions from the start of @code{main}:
32703 -data-disassemble -f basics.c -l 32 -n 3 -- 0
32705 @{address="0x000107bc",func-name="main",offset="0",
32706 inst="save %sp, -112, %sp"@},
32707 @{address="0x000107c0",func-name="main",offset="4",
32708 inst="mov 2, %o0"@},
32709 @{address="0x000107c4",func-name="main",offset="8",
32710 inst="sethi %hi(0x11800), %o2"@}]
32714 Disassemble 3 instructions from the start of @code{main} in mixed mode:
32718 -data-disassemble -f basics.c -l 32 -n 3 -- 1
32720 src_and_asm_line=@{line="31",
32721 file="../../../src/gdb/testsuite/gdb.mi/basics.c",
32722 fullname="/absolute/path/to/src/gdb/testsuite/gdb.mi/basics.c",
32723 line_asm_insn=[@{address="0x000107bc",
32724 func-name="main",offset="0",inst="save %sp, -112, %sp"@}]@},
32725 src_and_asm_line=@{line="32",
32726 file="../../../src/gdb/testsuite/gdb.mi/basics.c",
32727 fullname="/absolute/path/to/src/gdb/testsuite/gdb.mi/basics.c",
32728 line_asm_insn=[@{address="0x000107c0",
32729 func-name="main",offset="4",inst="mov 2, %o0"@},
32730 @{address="0x000107c4",func-name="main",offset="8",
32731 inst="sethi %hi(0x11800), %o2"@}]@}]
32736 @subheading The @code{-data-evaluate-expression} Command
32737 @findex -data-evaluate-expression
32739 @subsubheading Synopsis
32742 -data-evaluate-expression @var{expr}
32745 Evaluate @var{expr} as an expression. The expression could contain an
32746 inferior function call. The function call will execute synchronously.
32747 If the expression contains spaces, it must be enclosed in double quotes.
32749 @subsubheading @value{GDBN} Command
32751 The corresponding @value{GDBN} commands are @samp{print}, @samp{output}, and
32752 @samp{call}. In @code{gdbtk} only, there's a corresponding
32753 @samp{gdb_eval} command.
32755 @subsubheading Example
32757 In the following example, the numbers that precede the commands are the
32758 @dfn{tokens} described in @ref{GDB/MI Command Syntax, ,@sc{gdb/mi}
32759 Command Syntax}. Notice how @sc{gdb/mi} returns the same tokens in its
32763 211-data-evaluate-expression A
32766 311-data-evaluate-expression &A
32767 311^done,value="0xefffeb7c"
32769 411-data-evaluate-expression A+3
32772 511-data-evaluate-expression "A + 3"
32778 @subheading The @code{-data-list-changed-registers} Command
32779 @findex -data-list-changed-registers
32781 @subsubheading Synopsis
32784 -data-list-changed-registers
32787 Display a list of the registers that have changed.
32789 @subsubheading @value{GDBN} Command
32791 @value{GDBN} doesn't have a direct analog for this command; @code{gdbtk}
32792 has the corresponding command @samp{gdb_changed_register_list}.
32794 @subsubheading Example
32796 On a PPC MBX board:
32804 *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",frame=@{
32805 func="main",args=[],file="try.c",fullname="/home/foo/bar/try.c",
32806 line="5",arch="powerpc"@}
32808 -data-list-changed-registers
32809 ^done,changed-registers=["0","1","2","4","5","6","7","8","9",
32810 "10","11","13","14","15","16","17","18","19","20","21","22","23",
32811 "24","25","26","27","28","30","31","64","65","66","67","69"]
32816 @subheading The @code{-data-list-register-names} Command
32817 @findex -data-list-register-names
32819 @subsubheading Synopsis
32822 -data-list-register-names [ ( @var{regno} )+ ]
32825 Show a list of register names for the current target. If no arguments
32826 are given, it shows a list of the names of all the registers. If
32827 integer numbers are given as arguments, it will print a list of the
32828 names of the registers corresponding to the arguments. To ensure
32829 consistency between a register name and its number, the output list may
32830 include empty register names.
32832 @subsubheading @value{GDBN} Command
32834 @value{GDBN} does not have a command which corresponds to
32835 @samp{-data-list-register-names}. In @code{gdbtk} there is a
32836 corresponding command @samp{gdb_regnames}.
32838 @subsubheading Example
32840 For the PPC MBX board:
32843 -data-list-register-names
32844 ^done,register-names=["r0","r1","r2","r3","r4","r5","r6","r7",
32845 "r8","r9","r10","r11","r12","r13","r14","r15","r16","r17","r18",
32846 "r19","r20","r21","r22","r23","r24","r25","r26","r27","r28","r29",
32847 "r30","r31","f0","f1","f2","f3","f4","f5","f6","f7","f8","f9",
32848 "f10","f11","f12","f13","f14","f15","f16","f17","f18","f19","f20",
32849 "f21","f22","f23","f24","f25","f26","f27","f28","f29","f30","f31",
32850 "", "pc","ps","cr","lr","ctr","xer"]
32852 -data-list-register-names 1 2 3
32853 ^done,register-names=["r1","r2","r3"]
32857 @subheading The @code{-data-list-register-values} Command
32858 @findex -data-list-register-values
32860 @subsubheading Synopsis
32863 -data-list-register-values
32864 [ @code{--skip-unavailable} ] @var{fmt} [ ( @var{regno} )*]
32867 Display the registers' contents. The format according to which the
32868 registers' contents are to be returned is given by @var{fmt}, followed
32869 by an optional list of numbers specifying the registers to display. A
32870 missing list of numbers indicates that the contents of all the
32871 registers must be returned. The @code{--skip-unavailable} option
32872 indicates that only the available registers are to be returned.
32874 Allowed formats for @var{fmt} are:
32891 @subsubheading @value{GDBN} Command
32893 The corresponding @value{GDBN} commands are @samp{info reg}, @samp{info
32894 all-reg}, and (in @code{gdbtk}) @samp{gdb_fetch_registers}.
32896 @subsubheading Example
32898 For a PPC MBX board (note: line breaks are for readability only, they
32899 don't appear in the actual output):
32903 -data-list-register-values r 64 65
32904 ^done,register-values=[@{number="64",value="0xfe00a300"@},
32905 @{number="65",value="0x00029002"@}]
32907 -data-list-register-values x
32908 ^done,register-values=[@{number="0",value="0xfe0043c8"@},
32909 @{number="1",value="0x3fff88"@},@{number="2",value="0xfffffffe"@},
32910 @{number="3",value="0x0"@},@{number="4",value="0xa"@},
32911 @{number="5",value="0x3fff68"@},@{number="6",value="0x3fff58"@},
32912 @{number="7",value="0xfe011e98"@},@{number="8",value="0x2"@},
32913 @{number="9",value="0xfa202820"@},@{number="10",value="0xfa202808"@},
32914 @{number="11",value="0x1"@},@{number="12",value="0x0"@},
32915 @{number="13",value="0x4544"@},@{number="14",value="0xffdfffff"@},
32916 @{number="15",value="0xffffffff"@},@{number="16",value="0xfffffeff"@},
32917 @{number="17",value="0xefffffed"@},@{number="18",value="0xfffffffe"@},
32918 @{number="19",value="0xffffffff"@},@{number="20",value="0xffffffff"@},
32919 @{number="21",value="0xffffffff"@},@{number="22",value="0xfffffff7"@},
32920 @{number="23",value="0xffffffff"@},@{number="24",value="0xffffffff"@},
32921 @{number="25",value="0xffffffff"@},@{number="26",value="0xfffffffb"@},
32922 @{number="27",value="0xffffffff"@},@{number="28",value="0xf7bfffff"@},
32923 @{number="29",value="0x0"@},@{number="30",value="0xfe010000"@},
32924 @{number="31",value="0x0"@},@{number="32",value="0x0"@},
32925 @{number="33",value="0x0"@},@{number="34",value="0x0"@},
32926 @{number="35",value="0x0"@},@{number="36",value="0x0"@},
32927 @{number="37",value="0x0"@},@{number="38",value="0x0"@},
32928 @{number="39",value="0x0"@},@{number="40",value="0x0"@},
32929 @{number="41",value="0x0"@},@{number="42",value="0x0"@},
32930 @{number="43",value="0x0"@},@{number="44",value="0x0"@},
32931 @{number="45",value="0x0"@},@{number="46",value="0x0"@},
32932 @{number="47",value="0x0"@},@{number="48",value="0x0"@},
32933 @{number="49",value="0x0"@},@{number="50",value="0x0"@},
32934 @{number="51",value="0x0"@},@{number="52",value="0x0"@},
32935 @{number="53",value="0x0"@},@{number="54",value="0x0"@},
32936 @{number="55",value="0x0"@},@{number="56",value="0x0"@},
32937 @{number="57",value="0x0"@},@{number="58",value="0x0"@},
32938 @{number="59",value="0x0"@},@{number="60",value="0x0"@},
32939 @{number="61",value="0x0"@},@{number="62",value="0x0"@},
32940 @{number="63",value="0x0"@},@{number="64",value="0xfe00a300"@},
32941 @{number="65",value="0x29002"@},@{number="66",value="0x202f04b5"@},
32942 @{number="67",value="0xfe0043b0"@},@{number="68",value="0xfe00b3e4"@},
32943 @{number="69",value="0x20002b03"@}]
32948 @subheading The @code{-data-read-memory} Command
32949 @findex -data-read-memory
32951 This command is deprecated, use @code{-data-read-memory-bytes} instead.
32953 @subsubheading Synopsis
32956 -data-read-memory [ -o @var{byte-offset} ]
32957 @var{address} @var{word-format} @var{word-size}
32958 @var{nr-rows} @var{nr-cols} [ @var{aschar} ]
32965 @item @var{address}
32966 An expression specifying the address of the first memory word to be
32967 read. Complex expressions containing embedded white space should be
32968 quoted using the C convention.
32970 @item @var{word-format}
32971 The format to be used to print the memory words. The notation is the
32972 same as for @value{GDBN}'s @code{print} command (@pxref{Output Formats,
32975 @item @var{word-size}
32976 The size of each memory word in bytes.
32978 @item @var{nr-rows}
32979 The number of rows in the output table.
32981 @item @var{nr-cols}
32982 The number of columns in the output table.
32985 If present, indicates that each row should include an @sc{ascii} dump. The
32986 value of @var{aschar} is used as a padding character when a byte is not a
32987 member of the printable @sc{ascii} character set (printable @sc{ascii}
32988 characters are those whose code is between 32 and 126, inclusively).
32990 @item @var{byte-offset}
32991 An offset to add to the @var{address} before fetching memory.
32994 This command displays memory contents as a table of @var{nr-rows} by
32995 @var{nr-cols} words, each word being @var{word-size} bytes. In total,
32996 @code{@var{nr-rows} * @var{nr-cols} * @var{word-size}} bytes are read
32997 (returned as @samp{total-bytes}). Should less than the requested number
32998 of bytes be returned by the target, the missing words are identified
32999 using @samp{N/A}. The number of bytes read from the target is returned
33000 in @samp{nr-bytes} and the starting address used to read memory in
33003 The address of the next/previous row or page is available in
33004 @samp{next-row} and @samp{prev-row}, @samp{next-page} and
33007 @subsubheading @value{GDBN} Command
33009 The corresponding @value{GDBN} command is @samp{x}. @code{gdbtk} has
33010 @samp{gdb_get_mem} memory read command.
33012 @subsubheading Example
33014 Read six bytes of memory starting at @code{bytes+6} but then offset by
33015 @code{-6} bytes. Format as three rows of two columns. One byte per
33016 word. Display each word in hex.
33020 9-data-read-memory -o -6 -- bytes+6 x 1 3 2
33021 9^done,addr="0x00001390",nr-bytes="6",total-bytes="6",
33022 next-row="0x00001396",prev-row="0x0000138e",next-page="0x00001396",
33023 prev-page="0x0000138a",memory=[
33024 @{addr="0x00001390",data=["0x00","0x01"]@},
33025 @{addr="0x00001392",data=["0x02","0x03"]@},
33026 @{addr="0x00001394",data=["0x04","0x05"]@}]
33030 Read two bytes of memory starting at address @code{shorts + 64} and
33031 display as a single word formatted in decimal.
33035 5-data-read-memory shorts+64 d 2 1 1
33036 5^done,addr="0x00001510",nr-bytes="2",total-bytes="2",
33037 next-row="0x00001512",prev-row="0x0000150e",
33038 next-page="0x00001512",prev-page="0x0000150e",memory=[
33039 @{addr="0x00001510",data=["128"]@}]
33043 Read thirty two bytes of memory starting at @code{bytes+16} and format
33044 as eight rows of four columns. Include a string encoding with @samp{x}
33045 used as the non-printable character.
33049 4-data-read-memory bytes+16 x 1 8 4 x
33050 4^done,addr="0x000013a0",nr-bytes="32",total-bytes="32",
33051 next-row="0x000013c0",prev-row="0x0000139c",
33052 next-page="0x000013c0",prev-page="0x00001380",memory=[
33053 @{addr="0x000013a0",data=["0x10","0x11","0x12","0x13"],ascii="xxxx"@},
33054 @{addr="0x000013a4",data=["0x14","0x15","0x16","0x17"],ascii="xxxx"@},
33055 @{addr="0x000013a8",data=["0x18","0x19","0x1a","0x1b"],ascii="xxxx"@},
33056 @{addr="0x000013ac",data=["0x1c","0x1d","0x1e","0x1f"],ascii="xxxx"@},
33057 @{addr="0x000013b0",data=["0x20","0x21","0x22","0x23"],ascii=" !\"#"@},
33058 @{addr="0x000013b4",data=["0x24","0x25","0x26","0x27"],ascii="$%&'"@},
33059 @{addr="0x000013b8",data=["0x28","0x29","0x2a","0x2b"],ascii="()*+"@},
33060 @{addr="0x000013bc",data=["0x2c","0x2d","0x2e","0x2f"],ascii=",-./"@}]
33064 @subheading The @code{-data-read-memory-bytes} Command
33065 @findex -data-read-memory-bytes
33067 @subsubheading Synopsis
33070 -data-read-memory-bytes [ -o @var{offset} ]
33071 @var{address} @var{count}
33078 @item @var{address}
33079 An expression specifying the address of the first addressable memory unit
33080 to be read. Complex expressions containing embedded white space should be
33081 quoted using the C convention.
33084 The number of addressable memory units to read. This should be an integer
33088 The offset relative to @var{address} at which to start reading. This
33089 should be an integer literal. This option is provided so that a frontend
33090 is not required to first evaluate address and then perform address
33091 arithmetics itself.
33095 This command attempts to read all accessible memory regions in the
33096 specified range. First, all regions marked as unreadable in the memory
33097 map (if one is defined) will be skipped. @xref{Memory Region
33098 Attributes}. Second, @value{GDBN} will attempt to read the remaining
33099 regions. For each one, if reading full region results in an errors,
33100 @value{GDBN} will try to read a subset of the region.
33102 In general, every single memory unit in the region may be readable or not,
33103 and the only way to read every readable unit is to try a read at
33104 every address, which is not practical. Therefore, @value{GDBN} will
33105 attempt to read all accessible memory units at either beginning or the end
33106 of the region, using a binary division scheme. This heuristic works
33107 well for reading accross a memory map boundary. Note that if a region
33108 has a readable range that is neither at the beginning or the end,
33109 @value{GDBN} will not read it.
33111 The result record (@pxref{GDB/MI Result Records}) that is output of
33112 the command includes a field named @samp{memory} whose content is a
33113 list of tuples. Each tuple represent a successfully read memory block
33114 and has the following fields:
33118 The start address of the memory block, as hexadecimal literal.
33121 The end address of the memory block, as hexadecimal literal.
33124 The offset of the memory block, as hexadecimal literal, relative to
33125 the start address passed to @code{-data-read-memory-bytes}.
33128 The contents of the memory block, in hex.
33134 @subsubheading @value{GDBN} Command
33136 The corresponding @value{GDBN} command is @samp{x}.
33138 @subsubheading Example
33142 -data-read-memory-bytes &a 10
33143 ^done,memory=[@{begin="0xbffff154",offset="0x00000000",
33145 contents="01000000020000000300"@}]
33150 @subheading The @code{-data-write-memory-bytes} Command
33151 @findex -data-write-memory-bytes
33153 @subsubheading Synopsis
33156 -data-write-memory-bytes @var{address} @var{contents}
33157 -data-write-memory-bytes @var{address} @var{contents} @r{[}@var{count}@r{]}
33164 @item @var{address}
33165 An expression specifying the address of the first addressable memory unit
33166 to be written. Complex expressions containing embedded white space should
33167 be quoted using the C convention.
33169 @item @var{contents}
33170 The hex-encoded data to write. It is an error if @var{contents} does
33171 not represent an integral number of addressable memory units.
33174 Optional argument indicating the number of addressable memory units to be
33175 written. If @var{count} is greater than @var{contents}' length,
33176 @value{GDBN} will repeatedly write @var{contents} until it fills
33177 @var{count} memory units.
33181 @subsubheading @value{GDBN} Command
33183 There's no corresponding @value{GDBN} command.
33185 @subsubheading Example
33189 -data-write-memory-bytes &a "aabbccdd"
33196 -data-write-memory-bytes &a "aabbccdd" 16e
33201 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
33202 @node GDB/MI Tracepoint Commands
33203 @section @sc{gdb/mi} Tracepoint Commands
33205 The commands defined in this section implement MI support for
33206 tracepoints. For detailed introduction, see @ref{Tracepoints}.
33208 @subheading The @code{-trace-find} Command
33209 @findex -trace-find
33211 @subsubheading Synopsis
33214 -trace-find @var{mode} [@var{parameters}@dots{}]
33217 Find a trace frame using criteria defined by @var{mode} and
33218 @var{parameters}. The following table lists permissible
33219 modes and their parameters. For details of operation, see @ref{tfind}.
33224 No parameters are required. Stops examining trace frames.
33227 An integer is required as parameter. Selects tracepoint frame with
33230 @item tracepoint-number
33231 An integer is required as parameter. Finds next
33232 trace frame that corresponds to tracepoint with the specified number.
33235 An address is required as parameter. Finds
33236 next trace frame that corresponds to any tracepoint at the specified
33239 @item pc-inside-range
33240 Two addresses are required as parameters. Finds next trace
33241 frame that corresponds to a tracepoint at an address inside the
33242 specified range. Both bounds are considered to be inside the range.
33244 @item pc-outside-range
33245 Two addresses are required as parameters. Finds
33246 next trace frame that corresponds to a tracepoint at an address outside
33247 the specified range. Both bounds are considered to be inside the range.
33250 Line specification is required as parameter. @xref{Specify Location}.
33251 Finds next trace frame that corresponds to a tracepoint at
33252 the specified location.
33256 If @samp{none} was passed as @var{mode}, the response does not
33257 have fields. Otherwise, the response may have the following fields:
33261 This field has either @samp{0} or @samp{1} as the value, depending
33262 on whether a matching tracepoint was found.
33265 The index of the found traceframe. This field is present iff
33266 the @samp{found} field has value of @samp{1}.
33269 The index of the found tracepoint. This field is present iff
33270 the @samp{found} field has value of @samp{1}.
33273 The information about the frame corresponding to the found trace
33274 frame. This field is present only if a trace frame was found.
33275 @xref{GDB/MI Frame Information}, for description of this field.
33279 @subsubheading @value{GDBN} Command
33281 The corresponding @value{GDBN} command is @samp{tfind}.
33283 @subheading -trace-define-variable
33284 @findex -trace-define-variable
33286 @subsubheading Synopsis
33289 -trace-define-variable @var{name} [ @var{value} ]
33292 Create trace variable @var{name} if it does not exist. If
33293 @var{value} is specified, sets the initial value of the specified
33294 trace variable to that value. Note that the @var{name} should start
33295 with the @samp{$} character.
33297 @subsubheading @value{GDBN} Command
33299 The corresponding @value{GDBN} command is @samp{tvariable}.
33301 @subheading The @code{-trace-frame-collected} Command
33302 @findex -trace-frame-collected
33304 @subsubheading Synopsis
33307 -trace-frame-collected
33308 [--var-print-values @var{var_pval}]
33309 [--comp-print-values @var{comp_pval}]
33310 [--registers-format @var{regformat}]
33311 [--memory-contents]
33314 This command returns the set of collected objects, register names,
33315 trace state variable names, memory ranges and computed expressions
33316 that have been collected at a particular trace frame. The optional
33317 parameters to the command affect the output format in different ways.
33318 See the output description table below for more details.
33320 The reported names can be used in the normal manner to create
33321 varobjs and inspect the objects themselves. The items returned by
33322 this command are categorized so that it is clear which is a variable,
33323 which is a register, which is a trace state variable, which is a
33324 memory range and which is a computed expression.
33326 For instance, if the actions were
33328 collect myVar, myArray[myIndex], myObj.field, myPtr->field, myCount + 2
33329 collect *(int*)0xaf02bef0@@40
33333 the object collected in its entirety would be @code{myVar}. The
33334 object @code{myArray} would be partially collected, because only the
33335 element at index @code{myIndex} would be collected. The remaining
33336 objects would be computed expressions.
33338 An example output would be:
33342 -trace-frame-collected
33344 explicit-variables=[@{name="myVar",value="1"@}],
33345 computed-expressions=[@{name="myArray[myIndex]",value="0"@},
33346 @{name="myObj.field",value="0"@},
33347 @{name="myPtr->field",value="1"@},
33348 @{name="myCount + 2",value="3"@},
33349 @{name="$tvar1 + 1",value="43970027"@}],
33350 registers=[@{number="0",value="0x7fe2c6e79ec8"@},
33351 @{number="1",value="0x0"@},
33352 @{number="2",value="0x4"@},
33354 @{number="125",value="0x0"@}],
33355 tvars=[@{name="$tvar1",current="43970026"@}],
33356 memory=[@{address="0x0000000000602264",length="4"@},
33357 @{address="0x0000000000615bc0",length="4"@}]
33364 @item explicit-variables
33365 The set of objects that have been collected in their entirety (as
33366 opposed to collecting just a few elements of an array or a few struct
33367 members). For each object, its name and value are printed.
33368 The @code{--var-print-values} option affects how or whether the value
33369 field is output. If @var{var_pval} is 0, then print only the names;
33370 if it is 1, print also their values; and if it is 2, print the name,
33371 type and value for simple data types, and the name and type for
33372 arrays, structures and unions.
33374 @item computed-expressions
33375 The set of computed expressions that have been collected at the
33376 current trace frame. The @code{--comp-print-values} option affects
33377 this set like the @code{--var-print-values} option affects the
33378 @code{explicit-variables} set. See above.
33381 The registers that have been collected at the current trace frame.
33382 For each register collected, the name and current value are returned.
33383 The value is formatted according to the @code{--registers-format}
33384 option. See the @command{-data-list-register-values} command for a
33385 list of the allowed formats. The default is @samp{x}.
33388 The trace state variables that have been collected at the current
33389 trace frame. For each trace state variable collected, the name and
33390 current value are returned.
33393 The set of memory ranges that have been collected at the current trace
33394 frame. Its content is a list of tuples. Each tuple represents a
33395 collected memory range and has the following fields:
33399 The start address of the memory range, as hexadecimal literal.
33402 The length of the memory range, as decimal literal.
33405 The contents of the memory block, in hex. This field is only present
33406 if the @code{--memory-contents} option is specified.
33412 @subsubheading @value{GDBN} Command
33414 There is no corresponding @value{GDBN} command.
33416 @subsubheading Example
33418 @subheading -trace-list-variables
33419 @findex -trace-list-variables
33421 @subsubheading Synopsis
33424 -trace-list-variables
33427 Return a table of all defined trace variables. Each element of the
33428 table has the following fields:
33432 The name of the trace variable. This field is always present.
33435 The initial value. This is a 64-bit signed integer. This
33436 field is always present.
33439 The value the trace variable has at the moment. This is a 64-bit
33440 signed integer. This field is absent iff current value is
33441 not defined, for example if the trace was never run, or is
33446 @subsubheading @value{GDBN} Command
33448 The corresponding @value{GDBN} command is @samp{tvariables}.
33450 @subsubheading Example
33454 -trace-list-variables
33455 ^done,trace-variables=@{nr_rows="1",nr_cols="3",
33456 hdr=[@{width="15",alignment="-1",col_name="name",colhdr="Name"@},
33457 @{width="11",alignment="-1",col_name="initial",colhdr="Initial"@},
33458 @{width="11",alignment="-1",col_name="current",colhdr="Current"@}],
33459 body=[variable=@{name="$trace_timestamp",initial="0"@}
33460 variable=@{name="$foo",initial="10",current="15"@}]@}
33464 @subheading -trace-save
33465 @findex -trace-save
33467 @subsubheading Synopsis
33470 -trace-save [ -r ] [ -ctf ] @var{filename}
33473 Saves the collected trace data to @var{filename}. Without the
33474 @samp{-r} option, the data is downloaded from the target and saved
33475 in a local file. With the @samp{-r} option the target is asked
33476 to perform the save.
33478 By default, this command will save the trace in the tfile format. You can
33479 supply the optional @samp{-ctf} argument to save it the CTF format. See
33480 @ref{Trace Files} for more information about CTF.
33482 @subsubheading @value{GDBN} Command
33484 The corresponding @value{GDBN} command is @samp{tsave}.
33487 @subheading -trace-start
33488 @findex -trace-start
33490 @subsubheading Synopsis
33496 Starts a tracing experiment. The result of this command does not
33499 @subsubheading @value{GDBN} Command
33501 The corresponding @value{GDBN} command is @samp{tstart}.
33503 @subheading -trace-status
33504 @findex -trace-status
33506 @subsubheading Synopsis
33512 Obtains the status of a tracing experiment. The result may include
33513 the following fields:
33518 May have a value of either @samp{0}, when no tracing operations are
33519 supported, @samp{1}, when all tracing operations are supported, or
33520 @samp{file} when examining trace file. In the latter case, examining
33521 of trace frame is possible but new tracing experiement cannot be
33522 started. This field is always present.
33525 May have a value of either @samp{0} or @samp{1} depending on whether
33526 tracing experiement is in progress on target. This field is present
33527 if @samp{supported} field is not @samp{0}.
33530 Report the reason why the tracing was stopped last time. This field
33531 may be absent iff tracing was never stopped on target yet. The
33532 value of @samp{request} means the tracing was stopped as result of
33533 the @code{-trace-stop} command. The value of @samp{overflow} means
33534 the tracing buffer is full. The value of @samp{disconnection} means
33535 tracing was automatically stopped when @value{GDBN} has disconnected.
33536 The value of @samp{passcount} means tracing was stopped when a
33537 tracepoint was passed a maximal number of times for that tracepoint.
33538 This field is present if @samp{supported} field is not @samp{0}.
33540 @item stopping-tracepoint
33541 The number of tracepoint whose passcount as exceeded. This field is
33542 present iff the @samp{stop-reason} field has the value of
33546 @itemx frames-created
33547 The @samp{frames} field is a count of the total number of trace frames
33548 in the trace buffer, while @samp{frames-created} is the total created
33549 during the run, including ones that were discarded, such as when a
33550 circular trace buffer filled up. Both fields are optional.
33554 These fields tell the current size of the tracing buffer and the
33555 remaining space. These fields are optional.
33558 The value of the circular trace buffer flag. @code{1} means that the
33559 trace buffer is circular and old trace frames will be discarded if
33560 necessary to make room, @code{0} means that the trace buffer is linear
33564 The value of the disconnected tracing flag. @code{1} means that
33565 tracing will continue after @value{GDBN} disconnects, @code{0} means
33566 that the trace run will stop.
33569 The filename of the trace file being examined. This field is
33570 optional, and only present when examining a trace file.
33574 @subsubheading @value{GDBN} Command
33576 The corresponding @value{GDBN} command is @samp{tstatus}.
33578 @subheading -trace-stop
33579 @findex -trace-stop
33581 @subsubheading Synopsis
33587 Stops a tracing experiment. The result of this command has the same
33588 fields as @code{-trace-status}, except that the @samp{supported} and
33589 @samp{running} fields are not output.
33591 @subsubheading @value{GDBN} Command
33593 The corresponding @value{GDBN} command is @samp{tstop}.
33596 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
33597 @node GDB/MI Symbol Query
33598 @section @sc{gdb/mi} Symbol Query Commands
33602 @subheading The @code{-symbol-info-address} Command
33603 @findex -symbol-info-address
33605 @subsubheading Synopsis
33608 -symbol-info-address @var{symbol}
33611 Describe where @var{symbol} is stored.
33613 @subsubheading @value{GDBN} Command
33615 The corresponding @value{GDBN} command is @samp{info address}.
33617 @subsubheading Example
33621 @subheading The @code{-symbol-info-file} Command
33622 @findex -symbol-info-file
33624 @subsubheading Synopsis
33630 Show the file for the symbol.
33632 @subsubheading @value{GDBN} Command
33634 There's no equivalent @value{GDBN} command. @code{gdbtk} has
33635 @samp{gdb_find_file}.
33637 @subsubheading Example
33641 @subheading The @code{-symbol-info-function} Command
33642 @findex -symbol-info-function
33644 @subsubheading Synopsis
33647 -symbol-info-function
33650 Show which function the symbol lives in.
33652 @subsubheading @value{GDBN} Command
33654 @samp{gdb_get_function} in @code{gdbtk}.
33656 @subsubheading Example
33660 @subheading The @code{-symbol-info-line} Command
33661 @findex -symbol-info-line
33663 @subsubheading Synopsis
33669 Show the core addresses of the code for a source line.
33671 @subsubheading @value{GDBN} Command
33673 The corresponding @value{GDBN} command is @samp{info line}.
33674 @code{gdbtk} has the @samp{gdb_get_line} and @samp{gdb_get_file} commands.
33676 @subsubheading Example
33680 @subheading The @code{-symbol-info-symbol} Command
33681 @findex -symbol-info-symbol
33683 @subsubheading Synopsis
33686 -symbol-info-symbol @var{addr}
33689 Describe what symbol is at location @var{addr}.
33691 @subsubheading @value{GDBN} Command
33693 The corresponding @value{GDBN} command is @samp{info symbol}.
33695 @subsubheading Example
33699 @subheading The @code{-symbol-list-functions} Command
33700 @findex -symbol-list-functions
33702 @subsubheading Synopsis
33705 -symbol-list-functions
33708 List the functions in the executable.
33710 @subsubheading @value{GDBN} Command
33712 @samp{info functions} in @value{GDBN}, @samp{gdb_listfunc} and
33713 @samp{gdb_search} in @code{gdbtk}.
33715 @subsubheading Example
33720 @subheading The @code{-symbol-list-lines} Command
33721 @findex -symbol-list-lines
33723 @subsubheading Synopsis
33726 -symbol-list-lines @var{filename}
33729 Print the list of lines that contain code and their associated program
33730 addresses for the given source filename. The entries are sorted in
33731 ascending PC order.
33733 @subsubheading @value{GDBN} Command
33735 There is no corresponding @value{GDBN} command.
33737 @subsubheading Example
33740 -symbol-list-lines basics.c
33741 ^done,lines=[@{pc="0x08048554",line="7"@},@{pc="0x0804855a",line="8"@}]
33747 @subheading The @code{-symbol-list-types} Command
33748 @findex -symbol-list-types
33750 @subsubheading Synopsis
33756 List all the type names.
33758 @subsubheading @value{GDBN} Command
33760 The corresponding commands are @samp{info types} in @value{GDBN},
33761 @samp{gdb_search} in @code{gdbtk}.
33763 @subsubheading Example
33767 @subheading The @code{-symbol-list-variables} Command
33768 @findex -symbol-list-variables
33770 @subsubheading Synopsis
33773 -symbol-list-variables
33776 List all the global and static variable names.
33778 @subsubheading @value{GDBN} Command
33780 @samp{info variables} in @value{GDBN}, @samp{gdb_search} in @code{gdbtk}.
33782 @subsubheading Example
33786 @subheading The @code{-symbol-locate} Command
33787 @findex -symbol-locate
33789 @subsubheading Synopsis
33795 @subsubheading @value{GDBN} Command
33797 @samp{gdb_loc} in @code{gdbtk}.
33799 @subsubheading Example
33803 @subheading The @code{-symbol-type} Command
33804 @findex -symbol-type
33806 @subsubheading Synopsis
33809 -symbol-type @var{variable}
33812 Show type of @var{variable}.
33814 @subsubheading @value{GDBN} Command
33816 The corresponding @value{GDBN} command is @samp{ptype}, @code{gdbtk} has
33817 @samp{gdb_obj_variable}.
33819 @subsubheading Example
33824 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
33825 @node GDB/MI File Commands
33826 @section @sc{gdb/mi} File Commands
33828 This section describes the GDB/MI commands to specify executable file names
33829 and to read in and obtain symbol table information.
33831 @subheading The @code{-file-exec-and-symbols} Command
33832 @findex -file-exec-and-symbols
33834 @subsubheading Synopsis
33837 -file-exec-and-symbols @var{file}
33840 Specify the executable file to be debugged. This file is the one from
33841 which the symbol table is also read. If no file is specified, the
33842 command clears the executable and symbol information. If breakpoints
33843 are set when using this command with no arguments, @value{GDBN} will produce
33844 error messages. Otherwise, no output is produced, except a completion
33847 @subsubheading @value{GDBN} Command
33849 The corresponding @value{GDBN} command is @samp{file}.
33851 @subsubheading Example
33855 -file-exec-and-symbols /kwikemart/marge/ezannoni/TRUNK/mbx/hello.mbx
33861 @subheading The @code{-file-exec-file} Command
33862 @findex -file-exec-file
33864 @subsubheading Synopsis
33867 -file-exec-file @var{file}
33870 Specify the executable file to be debugged. Unlike
33871 @samp{-file-exec-and-symbols}, the symbol table is @emph{not} read
33872 from this file. If used without argument, @value{GDBN} clears the information
33873 about the executable file. No output is produced, except a completion
33876 @subsubheading @value{GDBN} Command
33878 The corresponding @value{GDBN} command is @samp{exec-file}.
33880 @subsubheading Example
33884 -file-exec-file /kwikemart/marge/ezannoni/TRUNK/mbx/hello.mbx
33891 @subheading The @code{-file-list-exec-sections} Command
33892 @findex -file-list-exec-sections
33894 @subsubheading Synopsis
33897 -file-list-exec-sections
33900 List the sections of the current executable file.
33902 @subsubheading @value{GDBN} Command
33904 The @value{GDBN} command @samp{info file} shows, among the rest, the same
33905 information as this command. @code{gdbtk} has a corresponding command
33906 @samp{gdb_load_info}.
33908 @subsubheading Example
33913 @subheading The @code{-file-list-exec-source-file} Command
33914 @findex -file-list-exec-source-file
33916 @subsubheading Synopsis
33919 -file-list-exec-source-file
33922 List the line number, the current source file, and the absolute path
33923 to the current source file for the current executable. The macro
33924 information field has a value of @samp{1} or @samp{0} depending on
33925 whether or not the file includes preprocessor macro information.
33927 @subsubheading @value{GDBN} Command
33929 The @value{GDBN} equivalent is @samp{info source}
33931 @subsubheading Example
33935 123-file-list-exec-source-file
33936 123^done,line="1",file="foo.c",fullname="/home/bar/foo.c,macro-info="1"
33941 @subheading The @code{-file-list-exec-source-files} Command
33942 @findex -file-list-exec-source-files
33944 @subsubheading Synopsis
33947 -file-list-exec-source-files
33950 List the source files for the current executable.
33952 It will always output both the filename and fullname (absolute file
33953 name) of a source file.
33955 @subsubheading @value{GDBN} Command
33957 The @value{GDBN} equivalent is @samp{info sources}.
33958 @code{gdbtk} has an analogous command @samp{gdb_listfiles}.
33960 @subsubheading Example
33963 -file-list-exec-source-files
33965 @{file=foo.c,fullname=/home/foo.c@},
33966 @{file=/home/bar.c,fullname=/home/bar.c@},
33967 @{file=gdb_could_not_find_fullpath.c@}]
33971 @subheading The @code{-file-list-shared-libraries} Command
33972 @findex -file-list-shared-libraries
33974 @subsubheading Synopsis
33977 -file-list-shared-libraries [ @var{regexp} ]
33980 List the shared libraries in the program.
33981 With a regular expression @var{regexp}, only those libraries whose
33982 names match @var{regexp} are listed.
33984 @subsubheading @value{GDBN} Command
33986 The corresponding @value{GDBN} command is @samp{info shared}. The fields
33987 have a similar meaning to the @code{=library-loaded} notification.
33988 The @code{ranges} field specifies the multiple segments belonging to this
33989 library. Each range has the following fields:
33993 The address defining the inclusive lower bound of the segment.
33995 The address defining the exclusive upper bound of the segment.
33998 @subsubheading Example
34001 -file-list-exec-source-files
34002 ^done,shared-libraries=[
34003 @{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"@}]@},
34004 @{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"@}]@}]
34010 @subheading The @code{-file-list-symbol-files} Command
34011 @findex -file-list-symbol-files
34013 @subsubheading Synopsis
34016 -file-list-symbol-files
34021 @subsubheading @value{GDBN} Command
34023 The corresponding @value{GDBN} command is @samp{info file} (part of it).
34025 @subsubheading Example
34030 @subheading The @code{-file-symbol-file} Command
34031 @findex -file-symbol-file
34033 @subsubheading Synopsis
34036 -file-symbol-file @var{file}
34039 Read symbol table info from the specified @var{file} argument. When
34040 used without arguments, clears @value{GDBN}'s symbol table info. No output is
34041 produced, except for a completion notification.
34043 @subsubheading @value{GDBN} Command
34045 The corresponding @value{GDBN} command is @samp{symbol-file}.
34047 @subsubheading Example
34051 -file-symbol-file /kwikemart/marge/ezannoni/TRUNK/mbx/hello.mbx
34057 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
34058 @node GDB/MI Memory Overlay Commands
34059 @section @sc{gdb/mi} Memory Overlay Commands
34061 The memory overlay commands are not implemented.
34063 @c @subheading -overlay-auto
34065 @c @subheading -overlay-list-mapping-state
34067 @c @subheading -overlay-list-overlays
34069 @c @subheading -overlay-map
34071 @c @subheading -overlay-off
34073 @c @subheading -overlay-on
34075 @c @subheading -overlay-unmap
34077 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
34078 @node GDB/MI Signal Handling Commands
34079 @section @sc{gdb/mi} Signal Handling Commands
34081 Signal handling commands are not implemented.
34083 @c @subheading -signal-handle
34085 @c @subheading -signal-list-handle-actions
34087 @c @subheading -signal-list-signal-types
34091 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
34092 @node GDB/MI Target Manipulation
34093 @section @sc{gdb/mi} Target Manipulation Commands
34096 @subheading The @code{-target-attach} Command
34097 @findex -target-attach
34099 @subsubheading Synopsis
34102 -target-attach @var{pid} | @var{gid} | @var{file}
34105 Attach to a process @var{pid} or a file @var{file} outside of
34106 @value{GDBN}, or a thread group @var{gid}. If attaching to a thread
34107 group, the id previously returned by
34108 @samp{-list-thread-groups --available} must be used.
34110 @subsubheading @value{GDBN} Command
34112 The corresponding @value{GDBN} command is @samp{attach}.
34114 @subsubheading Example
34118 =thread-created,id="1"
34119 *stopped,thread-id="1",frame=@{addr="0xb7f7e410",func="bar",args=[]@}
34125 @subheading The @code{-target-compare-sections} Command
34126 @findex -target-compare-sections
34128 @subsubheading Synopsis
34131 -target-compare-sections [ @var{section} ]
34134 Compare data of section @var{section} on target to the exec file.
34135 Without the argument, all sections are compared.
34137 @subsubheading @value{GDBN} Command
34139 The @value{GDBN} equivalent is @samp{compare-sections}.
34141 @subsubheading Example
34146 @subheading The @code{-target-detach} Command
34147 @findex -target-detach
34149 @subsubheading Synopsis
34152 -target-detach [ @var{pid} | @var{gid} ]
34155 Detach from the remote target which normally resumes its execution.
34156 If either @var{pid} or @var{gid} is specified, detaches from either
34157 the specified process, or specified thread group. There's no output.
34159 @subsubheading @value{GDBN} Command
34161 The corresponding @value{GDBN} command is @samp{detach}.
34163 @subsubheading Example
34173 @subheading The @code{-target-disconnect} Command
34174 @findex -target-disconnect
34176 @subsubheading Synopsis
34182 Disconnect from the remote target. There's no output and the target is
34183 generally not resumed.
34185 @subsubheading @value{GDBN} Command
34187 The corresponding @value{GDBN} command is @samp{disconnect}.
34189 @subsubheading Example
34199 @subheading The @code{-target-download} Command
34200 @findex -target-download
34202 @subsubheading Synopsis
34208 Loads the executable onto the remote target.
34209 It prints out an update message every half second, which includes the fields:
34213 The name of the section.
34215 The size of what has been sent so far for that section.
34217 The size of the section.
34219 The total size of what was sent so far (the current and the previous sections).
34221 The size of the overall executable to download.
34225 Each message is sent as status record (@pxref{GDB/MI Output Syntax, ,
34226 @sc{gdb/mi} Output Syntax}).
34228 In addition, it prints the name and size of the sections, as they are
34229 downloaded. These messages include the following fields:
34233 The name of the section.
34235 The size of the section.
34237 The size of the overall executable to download.
34241 At the end, a summary is printed.
34243 @subsubheading @value{GDBN} Command
34245 The corresponding @value{GDBN} command is @samp{load}.
34247 @subsubheading Example
34249 Note: each status message appears on a single line. Here the messages
34250 have been broken down so that they can fit onto a page.
34255 +download,@{section=".text",section-size="6668",total-size="9880"@}
34256 +download,@{section=".text",section-sent="512",section-size="6668",
34257 total-sent="512",total-size="9880"@}
34258 +download,@{section=".text",section-sent="1024",section-size="6668",
34259 total-sent="1024",total-size="9880"@}
34260 +download,@{section=".text",section-sent="1536",section-size="6668",
34261 total-sent="1536",total-size="9880"@}
34262 +download,@{section=".text",section-sent="2048",section-size="6668",
34263 total-sent="2048",total-size="9880"@}
34264 +download,@{section=".text",section-sent="2560",section-size="6668",
34265 total-sent="2560",total-size="9880"@}
34266 +download,@{section=".text",section-sent="3072",section-size="6668",
34267 total-sent="3072",total-size="9880"@}
34268 +download,@{section=".text",section-sent="3584",section-size="6668",
34269 total-sent="3584",total-size="9880"@}
34270 +download,@{section=".text",section-sent="4096",section-size="6668",
34271 total-sent="4096",total-size="9880"@}
34272 +download,@{section=".text",section-sent="4608",section-size="6668",
34273 total-sent="4608",total-size="9880"@}
34274 +download,@{section=".text",section-sent="5120",section-size="6668",
34275 total-sent="5120",total-size="9880"@}
34276 +download,@{section=".text",section-sent="5632",section-size="6668",
34277 total-sent="5632",total-size="9880"@}
34278 +download,@{section=".text",section-sent="6144",section-size="6668",
34279 total-sent="6144",total-size="9880"@}
34280 +download,@{section=".text",section-sent="6656",section-size="6668",
34281 total-sent="6656",total-size="9880"@}
34282 +download,@{section=".init",section-size="28",total-size="9880"@}
34283 +download,@{section=".fini",section-size="28",total-size="9880"@}
34284 +download,@{section=".data",section-size="3156",total-size="9880"@}
34285 +download,@{section=".data",section-sent="512",section-size="3156",
34286 total-sent="7236",total-size="9880"@}
34287 +download,@{section=".data",section-sent="1024",section-size="3156",
34288 total-sent="7748",total-size="9880"@}
34289 +download,@{section=".data",section-sent="1536",section-size="3156",
34290 total-sent="8260",total-size="9880"@}
34291 +download,@{section=".data",section-sent="2048",section-size="3156",
34292 total-sent="8772",total-size="9880"@}
34293 +download,@{section=".data",section-sent="2560",section-size="3156",
34294 total-sent="9284",total-size="9880"@}
34295 +download,@{section=".data",section-sent="3072",section-size="3156",
34296 total-sent="9796",total-size="9880"@}
34297 ^done,address="0x10004",load-size="9880",transfer-rate="6586",
34304 @subheading The @code{-target-exec-status} Command
34305 @findex -target-exec-status
34307 @subsubheading Synopsis
34310 -target-exec-status
34313 Provide information on the state of the target (whether it is running or
34314 not, for instance).
34316 @subsubheading @value{GDBN} Command
34318 There's no equivalent @value{GDBN} command.
34320 @subsubheading Example
34324 @subheading The @code{-target-list-available-targets} Command
34325 @findex -target-list-available-targets
34327 @subsubheading Synopsis
34330 -target-list-available-targets
34333 List the possible targets to connect to.
34335 @subsubheading @value{GDBN} Command
34337 The corresponding @value{GDBN} command is @samp{help target}.
34339 @subsubheading Example
34343 @subheading The @code{-target-list-current-targets} Command
34344 @findex -target-list-current-targets
34346 @subsubheading Synopsis
34349 -target-list-current-targets
34352 Describe the current target.
34354 @subsubheading @value{GDBN} Command
34356 The corresponding information is printed by @samp{info file} (among
34359 @subsubheading Example
34363 @subheading The @code{-target-list-parameters} Command
34364 @findex -target-list-parameters
34366 @subsubheading Synopsis
34369 -target-list-parameters
34375 @subsubheading @value{GDBN} Command
34379 @subsubheading Example
34382 @subheading The @code{-target-flash-erase} Command
34383 @findex -target-flash-erase
34385 @subsubheading Synopsis
34388 -target-flash-erase
34391 Erases all known flash memory regions on the target.
34393 The corresponding @value{GDBN} command is @samp{flash-erase}.
34395 The output is a list of flash regions that have been erased, with starting
34396 addresses and memory region sizes.
34400 -target-flash-erase
34401 ^done,erased-regions=@{address="0x0",size="0x40000"@}
34405 @subheading The @code{-target-select} Command
34406 @findex -target-select
34408 @subsubheading Synopsis
34411 -target-select @var{type} @var{parameters @dots{}}
34414 Connect @value{GDBN} to the remote target. This command takes two args:
34418 The type of target, for instance @samp{remote}, etc.
34419 @item @var{parameters}
34420 Device names, host names and the like. @xref{Target Commands, ,
34421 Commands for Managing Targets}, for more details.
34424 The output is a connection notification, followed by the address at
34425 which the target program is, in the following form:
34428 ^connected,addr="@var{address}",func="@var{function name}",
34429 args=[@var{arg list}]
34432 @subsubheading @value{GDBN} Command
34434 The corresponding @value{GDBN} command is @samp{target}.
34436 @subsubheading Example
34440 -target-select remote /dev/ttya
34441 ^connected,addr="0xfe00a300",func="??",args=[]
34445 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
34446 @node GDB/MI File Transfer Commands
34447 @section @sc{gdb/mi} File Transfer Commands
34450 @subheading The @code{-target-file-put} Command
34451 @findex -target-file-put
34453 @subsubheading Synopsis
34456 -target-file-put @var{hostfile} @var{targetfile}
34459 Copy file @var{hostfile} from the host system (the machine running
34460 @value{GDBN}) to @var{targetfile} on the target system.
34462 @subsubheading @value{GDBN} Command
34464 The corresponding @value{GDBN} command is @samp{remote put}.
34466 @subsubheading Example
34470 -target-file-put localfile remotefile
34476 @subheading The @code{-target-file-get} Command
34477 @findex -target-file-get
34479 @subsubheading Synopsis
34482 -target-file-get @var{targetfile} @var{hostfile}
34485 Copy file @var{targetfile} from the target system to @var{hostfile}
34486 on the host system.
34488 @subsubheading @value{GDBN} Command
34490 The corresponding @value{GDBN} command is @samp{remote get}.
34492 @subsubheading Example
34496 -target-file-get remotefile localfile
34502 @subheading The @code{-target-file-delete} Command
34503 @findex -target-file-delete
34505 @subsubheading Synopsis
34508 -target-file-delete @var{targetfile}
34511 Delete @var{targetfile} from the target system.
34513 @subsubheading @value{GDBN} Command
34515 The corresponding @value{GDBN} command is @samp{remote delete}.
34517 @subsubheading Example
34521 -target-file-delete remotefile
34527 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
34528 @node GDB/MI Ada Exceptions Commands
34529 @section Ada Exceptions @sc{gdb/mi} Commands
34531 @subheading The @code{-info-ada-exceptions} Command
34532 @findex -info-ada-exceptions
34534 @subsubheading Synopsis
34537 -info-ada-exceptions [ @var{regexp}]
34540 List all Ada exceptions defined within the program being debugged.
34541 With a regular expression @var{regexp}, only those exceptions whose
34542 names match @var{regexp} are listed.
34544 @subsubheading @value{GDBN} Command
34546 The corresponding @value{GDBN} command is @samp{info exceptions}.
34548 @subsubheading Result
34550 The result is a table of Ada exceptions. The following columns are
34551 defined for each exception:
34555 The name of the exception.
34558 The address of the exception.
34562 @subsubheading Example
34565 -info-ada-exceptions aint
34566 ^done,ada-exceptions=@{nr_rows="2",nr_cols="2",
34567 hdr=[@{width="1",alignment="-1",col_name="name",colhdr="Name"@},
34568 @{width="1",alignment="-1",col_name="address",colhdr="Address"@}],
34569 body=[@{name="constraint_error",address="0x0000000000613da0"@},
34570 @{name="const.aint_global_e",address="0x0000000000613b00"@}]@}
34573 @subheading Catching Ada Exceptions
34575 The commands describing how to ask @value{GDBN} to stop when a program
34576 raises an exception are described at @ref{Ada Exception GDB/MI
34577 Catchpoint Commands}.
34580 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
34581 @node GDB/MI Support Commands
34582 @section @sc{gdb/mi} Support Commands
34584 Since new commands and features get regularly added to @sc{gdb/mi},
34585 some commands are available to help front-ends query the debugger
34586 about support for these capabilities. Similarly, it is also possible
34587 to query @value{GDBN} about target support of certain features.
34589 @subheading The @code{-info-gdb-mi-command} Command
34590 @cindex @code{-info-gdb-mi-command}
34591 @findex -info-gdb-mi-command
34593 @subsubheading Synopsis
34596 -info-gdb-mi-command @var{cmd_name}
34599 Query support for the @sc{gdb/mi} command named @var{cmd_name}.
34601 Note that the dash (@code{-}) starting all @sc{gdb/mi} commands
34602 is technically not part of the command name (@pxref{GDB/MI Input
34603 Syntax}), and thus should be omitted in @var{cmd_name}. However,
34604 for ease of use, this command also accepts the form with the leading
34607 @subsubheading @value{GDBN} Command
34609 There is no corresponding @value{GDBN} command.
34611 @subsubheading Result
34613 The result is a tuple. There is currently only one field:
34617 This field is equal to @code{"true"} if the @sc{gdb/mi} command exists,
34618 @code{"false"} otherwise.
34622 @subsubheading Example
34624 Here is an example where the @sc{gdb/mi} command does not exist:
34627 -info-gdb-mi-command unsupported-command
34628 ^done,command=@{exists="false"@}
34632 And here is an example where the @sc{gdb/mi} command is known
34636 -info-gdb-mi-command symbol-list-lines
34637 ^done,command=@{exists="true"@}
34640 @subheading The @code{-list-features} Command
34641 @findex -list-features
34642 @cindex supported @sc{gdb/mi} features, list
34644 Returns a list of particular features of the MI protocol that
34645 this version of gdb implements. A feature can be a command,
34646 or a new field in an output of some command, or even an
34647 important bugfix. While a frontend can sometimes detect presence
34648 of a feature at runtime, it is easier to perform detection at debugger
34651 The command returns a list of strings, with each string naming an
34652 available feature. Each returned string is just a name, it does not
34653 have any internal structure. The list of possible feature names
34659 (gdb) -list-features
34660 ^done,result=["feature1","feature2"]
34663 The current list of features is:
34666 @item frozen-varobjs
34667 Indicates support for the @code{-var-set-frozen} command, as well
34668 as possible presense of the @code{frozen} field in the output
34669 of @code{-varobj-create}.
34670 @item pending-breakpoints
34671 Indicates support for the @option{-f} option to the @code{-break-insert}
34674 Indicates Python scripting support, Python-based
34675 pretty-printing commands, and possible presence of the
34676 @samp{display_hint} field in the output of @code{-var-list-children}
34678 Indicates support for the @code{-thread-info} command.
34679 @item data-read-memory-bytes
34680 Indicates support for the @code{-data-read-memory-bytes} and the
34681 @code{-data-write-memory-bytes} commands.
34682 @item breakpoint-notifications
34683 Indicates that changes to breakpoints and breakpoints created via the
34684 CLI will be announced via async records.
34685 @item ada-task-info
34686 Indicates support for the @code{-ada-task-info} command.
34687 @item language-option
34688 Indicates that all @sc{gdb/mi} commands accept the @option{--language}
34689 option (@pxref{Context management}).
34690 @item info-gdb-mi-command
34691 Indicates support for the @code{-info-gdb-mi-command} command.
34692 @item undefined-command-error-code
34693 Indicates support for the "undefined-command" error code in error result
34694 records, produced when trying to execute an undefined @sc{gdb/mi} command
34695 (@pxref{GDB/MI Result Records}).
34696 @item exec-run-start-option
34697 Indicates that the @code{-exec-run} command supports the @option{--start}
34698 option (@pxref{GDB/MI Program Execution}).
34699 @item data-disassemble-a-option
34700 Indicates that the @code{-data-disassemble} command supports the @option{-a}
34701 option (@pxref{GDB/MI Data Manipulation}).
34704 @subheading The @code{-list-target-features} Command
34705 @findex -list-target-features
34707 Returns a list of particular features that are supported by the
34708 target. Those features affect the permitted MI commands, but
34709 unlike the features reported by the @code{-list-features} command, the
34710 features depend on which target GDB is using at the moment. Whenever
34711 a target can change, due to commands such as @code{-target-select},
34712 @code{-target-attach} or @code{-exec-run}, the list of target features
34713 may change, and the frontend should obtain it again.
34717 (gdb) -list-target-features
34718 ^done,result=["async"]
34721 The current list of features is:
34725 Indicates that the target is capable of asynchronous command
34726 execution, which means that @value{GDBN} will accept further commands
34727 while the target is running.
34730 Indicates that the target is capable of reverse execution.
34731 @xref{Reverse Execution}, for more information.
34735 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
34736 @node GDB/MI Miscellaneous Commands
34737 @section Miscellaneous @sc{gdb/mi} Commands
34739 @c @subheading -gdb-complete
34741 @subheading The @code{-gdb-exit} Command
34744 @subsubheading Synopsis
34750 Exit @value{GDBN} immediately.
34752 @subsubheading @value{GDBN} Command
34754 Approximately corresponds to @samp{quit}.
34756 @subsubheading Example
34766 @subheading The @code{-exec-abort} Command
34767 @findex -exec-abort
34769 @subsubheading Synopsis
34775 Kill the inferior running program.
34777 @subsubheading @value{GDBN} Command
34779 The corresponding @value{GDBN} command is @samp{kill}.
34781 @subsubheading Example
34786 @subheading The @code{-gdb-set} Command
34789 @subsubheading Synopsis
34795 Set an internal @value{GDBN} variable.
34796 @c IS THIS A DOLLAR VARIABLE? OR SOMETHING LIKE ANNOTATE ?????
34798 @subsubheading @value{GDBN} Command
34800 The corresponding @value{GDBN} command is @samp{set}.
34802 @subsubheading Example
34812 @subheading The @code{-gdb-show} Command
34815 @subsubheading Synopsis
34821 Show the current value of a @value{GDBN} variable.
34823 @subsubheading @value{GDBN} Command
34825 The corresponding @value{GDBN} command is @samp{show}.
34827 @subsubheading Example
34836 @c @subheading -gdb-source
34839 @subheading The @code{-gdb-version} Command
34840 @findex -gdb-version
34842 @subsubheading Synopsis
34848 Show version information for @value{GDBN}. Used mostly in testing.
34850 @subsubheading @value{GDBN} Command
34852 The @value{GDBN} equivalent is @samp{show version}. @value{GDBN} by
34853 default shows this information when you start an interactive session.
34855 @subsubheading Example
34857 @c This example modifies the actual output from GDB to avoid overfull
34863 ~Copyright 2000 Free Software Foundation, Inc.
34864 ~GDB is free software, covered by the GNU General Public License, and
34865 ~you are welcome to change it and/or distribute copies of it under
34866 ~ certain conditions.
34867 ~Type "show copying" to see the conditions.
34868 ~There is absolutely no warranty for GDB. Type "show warranty" for
34870 ~This GDB was configured as
34871 "--host=sparc-sun-solaris2.5.1 --target=ppc-eabi".
34876 @subheading The @code{-list-thread-groups} Command
34877 @findex -list-thread-groups
34879 @subheading Synopsis
34882 -list-thread-groups [ --available ] [ --recurse 1 ] [ @var{group} ... ]
34885 Lists thread groups (@pxref{Thread groups}). When a single thread
34886 group is passed as the argument, lists the children of that group.
34887 When several thread group are passed, lists information about those
34888 thread groups. Without any parameters, lists information about all
34889 top-level thread groups.
34891 Normally, thread groups that are being debugged are reported.
34892 With the @samp{--available} option, @value{GDBN} reports thread groups
34893 available on the target.
34895 The output of this command may have either a @samp{threads} result or
34896 a @samp{groups} result. The @samp{thread} result has a list of tuples
34897 as value, with each tuple describing a thread (@pxref{GDB/MI Thread
34898 Information}). The @samp{groups} result has a list of tuples as value,
34899 each tuple describing a thread group. If top-level groups are
34900 requested (that is, no parameter is passed), or when several groups
34901 are passed, the output always has a @samp{groups} result. The format
34902 of the @samp{group} result is described below.
34904 To reduce the number of roundtrips it's possible to list thread groups
34905 together with their children, by passing the @samp{--recurse} option
34906 and the recursion depth. Presently, only recursion depth of 1 is
34907 permitted. If this option is present, then every reported thread group
34908 will also include its children, either as @samp{group} or
34909 @samp{threads} field.
34911 In general, any combination of option and parameters is permitted, with
34912 the following caveats:
34916 When a single thread group is passed, the output will typically
34917 be the @samp{threads} result. Because threads may not contain
34918 anything, the @samp{recurse} option will be ignored.
34921 When the @samp{--available} option is passed, limited information may
34922 be available. In particular, the list of threads of a process might
34923 be inaccessible. Further, specifying specific thread groups might
34924 not give any performance advantage over listing all thread groups.
34925 The frontend should assume that @samp{-list-thread-groups --available}
34926 is always an expensive operation and cache the results.
34930 The @samp{groups} result is a list of tuples, where each tuple may
34931 have the following fields:
34935 Identifier of the thread group. This field is always present.
34936 The identifier is an opaque string; frontends should not try to
34937 convert it to an integer, even though it might look like one.
34940 The type of the thread group. At present, only @samp{process} is a
34944 The target-specific process identifier. This field is only present
34945 for thread groups of type @samp{process} and only if the process exists.
34948 The exit code of this group's last exited thread, formatted in octal.
34949 This field is only present for thread groups of type @samp{process} and
34950 only if the process is not running.
34953 The number of children this thread group has. This field may be
34954 absent for an available thread group.
34957 This field has a list of tuples as value, each tuple describing a
34958 thread. It may be present if the @samp{--recurse} option is
34959 specified, and it's actually possible to obtain the threads.
34962 This field is a list of integers, each identifying a core that one
34963 thread of the group is running on. This field may be absent if
34964 such information is not available.
34967 The name of the executable file that corresponds to this thread group.
34968 The field is only present for thread groups of type @samp{process},
34969 and only if there is a corresponding executable file.
34973 @subheading Example
34977 -list-thread-groups
34978 ^done,groups=[@{id="17",type="process",pid="yyy",num_children="2"@}]
34979 -list-thread-groups 17
34980 ^done,threads=[@{id="2",target-id="Thread 0xb7e14b90 (LWP 21257)",
34981 frame=@{level="0",addr="0xffffe410",func="__kernel_vsyscall",args=[]@},state="running"@},
34982 @{id="1",target-id="Thread 0xb7e156b0 (LWP 21254)",
34983 frame=@{level="0",addr="0x0804891f",func="foo",args=[@{name="i",value="10"@}],
34984 file="/tmp/a.c",fullname="/tmp/a.c",line="158",arch="i386:x86_64"@},state="running"@}]]
34985 -list-thread-groups --available
34986 ^done,groups=[@{id="17",type="process",pid="yyy",num_children="2",cores=[1,2]@}]
34987 -list-thread-groups --available --recurse 1
34988 ^done,groups=[@{id="17", types="process",pid="yyy",num_children="2",cores=[1,2],
34989 threads=[@{id="1",target-id="Thread 0xb7e14b90",cores=[1]@},
34990 @{id="2",target-id="Thread 0xb7e14b90",cores=[2]@}]@},..]
34991 -list-thread-groups --available --recurse 1 17 18
34992 ^done,groups=[@{id="17", types="process",pid="yyy",num_children="2",cores=[1,2],
34993 threads=[@{id="1",target-id="Thread 0xb7e14b90",cores=[1]@},
34994 @{id="2",target-id="Thread 0xb7e14b90",cores=[2]@}]@},...]
34997 @subheading The @code{-info-os} Command
35000 @subsubheading Synopsis
35003 -info-os [ @var{type} ]
35006 If no argument is supplied, the command returns a table of available
35007 operating-system-specific information types. If one of these types is
35008 supplied as an argument @var{type}, then the command returns a table
35009 of data of that type.
35011 The types of information available depend on the target operating
35014 @subsubheading @value{GDBN} Command
35016 The corresponding @value{GDBN} command is @samp{info os}.
35018 @subsubheading Example
35020 When run on a @sc{gnu}/Linux system, the output will look something
35026 ^done,OSDataTable=@{nr_rows="10",nr_cols="3",
35027 hdr=[@{width="10",alignment="-1",col_name="col0",colhdr="Type"@},
35028 @{width="10",alignment="-1",col_name="col1",colhdr="Description"@},
35029 @{width="10",alignment="-1",col_name="col2",colhdr="Title"@}],
35030 body=[item=@{col0="cpus",col1="Listing of all cpus/cores on the system",
35032 item=@{col0="files",col1="Listing of all file descriptors",
35033 col2="File descriptors"@},
35034 item=@{col0="modules",col1="Listing of all loaded kernel modules",
35035 col2="Kernel modules"@},
35036 item=@{col0="msg",col1="Listing of all message queues",
35037 col2="Message queues"@},
35038 item=@{col0="processes",col1="Listing of all processes",
35039 col2="Processes"@},
35040 item=@{col0="procgroups",col1="Listing of all process groups",
35041 col2="Process groups"@},
35042 item=@{col0="semaphores",col1="Listing of all semaphores",
35043 col2="Semaphores"@},
35044 item=@{col0="shm",col1="Listing of all shared-memory regions",
35045 col2="Shared-memory regions"@},
35046 item=@{col0="sockets",col1="Listing of all internet-domain sockets",
35048 item=@{col0="threads",col1="Listing of all threads",
35052 ^done,OSDataTable=@{nr_rows="190",nr_cols="4",
35053 hdr=[@{width="10",alignment="-1",col_name="col0",colhdr="pid"@},
35054 @{width="10",alignment="-1",col_name="col1",colhdr="user"@},
35055 @{width="10",alignment="-1",col_name="col2",colhdr="command"@},
35056 @{width="10",alignment="-1",col_name="col3",colhdr="cores"@}],
35057 body=[item=@{col0="1",col1="root",col2="/sbin/init",col3="0"@},
35058 item=@{col0="2",col1="root",col2="[kthreadd]",col3="1"@},
35059 item=@{col0="3",col1="root",col2="[ksoftirqd/0]",col3="0"@},
35061 item=@{col0="26446",col1="stan",col2="bash",col3="0"@},
35062 item=@{col0="28152",col1="stan",col2="bash",col3="1"@}]@}
35066 (Note that the MI output here includes a @code{"Title"} column that
35067 does not appear in command-line @code{info os}; this column is useful
35068 for MI clients that want to enumerate the types of data, such as in a
35069 popup menu, but is needless clutter on the command line, and
35070 @code{info os} omits it.)
35072 @subheading The @code{-add-inferior} Command
35073 @findex -add-inferior
35075 @subheading Synopsis
35081 Creates a new inferior (@pxref{Inferiors and Programs}). The created
35082 inferior is not associated with any executable. Such association may
35083 be established with the @samp{-file-exec-and-symbols} command
35084 (@pxref{GDB/MI File Commands}). The command response has a single
35085 field, @samp{inferior}, whose value is the identifier of the
35086 thread group corresponding to the new inferior.
35088 @subheading Example
35093 ^done,inferior="i3"
35096 @subheading The @code{-interpreter-exec} Command
35097 @findex -interpreter-exec
35099 @subheading Synopsis
35102 -interpreter-exec @var{interpreter} @var{command}
35104 @anchor{-interpreter-exec}
35106 Execute the specified @var{command} in the given @var{interpreter}.
35108 @subheading @value{GDBN} Command
35110 The corresponding @value{GDBN} command is @samp{interpreter-exec}.
35112 @subheading Example
35116 -interpreter-exec console "break main"
35117 &"During symbol reading, couldn't parse type; debugger out of date?.\n"
35118 &"During symbol reading, bad structure-type format.\n"
35119 ~"Breakpoint 1 at 0x8074fc6: file ../../src/gdb/main.c, line 743.\n"
35124 @subheading The @code{-inferior-tty-set} Command
35125 @findex -inferior-tty-set
35127 @subheading Synopsis
35130 -inferior-tty-set /dev/pts/1
35133 Set terminal for future runs of the program being debugged.
35135 @subheading @value{GDBN} Command
35137 The corresponding @value{GDBN} command is @samp{set inferior-tty} /dev/pts/1.
35139 @subheading Example
35143 -inferior-tty-set /dev/pts/1
35148 @subheading The @code{-inferior-tty-show} Command
35149 @findex -inferior-tty-show
35151 @subheading Synopsis
35157 Show terminal for future runs of program being debugged.
35159 @subheading @value{GDBN} Command
35161 The corresponding @value{GDBN} command is @samp{show inferior-tty}.
35163 @subheading Example
35167 -inferior-tty-set /dev/pts/1
35171 ^done,inferior_tty_terminal="/dev/pts/1"
35175 @subheading The @code{-enable-timings} Command
35176 @findex -enable-timings
35178 @subheading Synopsis
35181 -enable-timings [yes | no]
35184 Toggle the printing of the wallclock, user and system times for an MI
35185 command as a field in its output. This command is to help frontend
35186 developers optimize the performance of their code. No argument is
35187 equivalent to @samp{yes}.
35189 @subheading @value{GDBN} Command
35193 @subheading Example
35201 ^done,bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
35202 addr="0x080484ed",func="main",file="myprog.c",
35203 fullname="/home/nickrob/myprog.c",line="73",thread-groups=["i1"],
35205 time=@{wallclock="0.05185",user="0.00800",system="0.00000"@}
35213 *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",thread-id="0",
35214 frame=@{addr="0x080484ed",func="main",args=[@{name="argc",value="1"@},
35215 @{name="argv",value="0xbfb60364"@}],file="myprog.c",
35216 fullname="/home/nickrob/myprog.c",line="73",arch="i386:x86_64"@}
35220 @subheading The @code{-complete} Command
35223 @subheading Synopsis
35226 -complete @var{command}
35229 Show a list of completions for partially typed CLI @var{command}.
35231 This command is intended for @sc{gdb/mi} frontends that cannot use two separate
35232 CLI and MI channels --- for example: because of lack of PTYs like on Windows or
35233 because @value{GDBN} is used remotely via a SSH connection.
35237 The result consists of two or three fields:
35241 This field contains the completed @var{command}. If @var{command}
35242 has no known completions, this field is omitted.
35245 This field contains a (possibly empty) array of matches. It is always present.
35247 @item max_completions_reached
35248 This field contains @code{1} if number of known completions is above
35249 @code{max-completions} limit (@pxref{Completion}), otherwise it contains
35250 @code{0}. It is always present.
35254 @subheading @value{GDBN} Command
35256 The corresponding @value{GDBN} command is @samp{complete}.
35258 @subheading Example
35263 ^done,completion="break",
35264 matches=["break","break-range"],
35265 max_completions_reached="0"
35268 ^done,completion="b ma",
35269 matches=["b madvise","b main"],max_completions_reached="0"
35271 -complete "b push_b"
35272 ^done,completion="b push_back(",
35274 "b A::push_back(void*)",
35275 "b std::string::push_back(char)",
35276 "b std::vector<int, std::allocator<int> >::push_back(int&&)"],
35277 max_completions_reached="0"
35279 -complete "nonexist"
35280 ^done,matches=[],max_completions_reached="0"
35286 @chapter @value{GDBN} Annotations
35288 This chapter describes annotations in @value{GDBN}. Annotations were
35289 designed to interface @value{GDBN} to graphical user interfaces or other
35290 similar programs which want to interact with @value{GDBN} at a
35291 relatively high level.
35293 The annotation mechanism has largely been superseded by @sc{gdb/mi}
35297 This is Edition @value{EDITION}, @value{DATE}.
35301 * Annotations Overview:: What annotations are; the general syntax.
35302 * Server Prefix:: Issuing a command without affecting user state.
35303 * Prompting:: Annotations marking @value{GDBN}'s need for input.
35304 * Errors:: Annotations for error messages.
35305 * Invalidation:: Some annotations describe things now invalid.
35306 * Annotations for Running::
35307 Whether the program is running, how it stopped, etc.
35308 * Source Annotations:: Annotations describing source code.
35311 @node Annotations Overview
35312 @section What is an Annotation?
35313 @cindex annotations
35315 Annotations start with a newline character, two @samp{control-z}
35316 characters, and the name of the annotation. If there is no additional
35317 information associated with this annotation, the name of the annotation
35318 is followed immediately by a newline. If there is additional
35319 information, the name of the annotation is followed by a space, the
35320 additional information, and a newline. The additional information
35321 cannot contain newline characters.
35323 Any output not beginning with a newline and two @samp{control-z}
35324 characters denotes literal output from @value{GDBN}. Currently there is
35325 no need for @value{GDBN} to output a newline followed by two
35326 @samp{control-z} characters, but if there was such a need, the
35327 annotations could be extended with an @samp{escape} annotation which
35328 means those three characters as output.
35330 The annotation @var{level}, which is specified using the
35331 @option{--annotate} command line option (@pxref{Mode Options}), controls
35332 how much information @value{GDBN} prints together with its prompt,
35333 values of expressions, source lines, and other types of output. Level 0
35334 is for no annotations, level 1 is for use when @value{GDBN} is run as a
35335 subprocess of @sc{gnu} Emacs, level 3 is the maximum annotation suitable
35336 for programs that control @value{GDBN}, and level 2 annotations have
35337 been made obsolete (@pxref{Limitations, , Limitations of the Annotation
35338 Interface, annotate, GDB's Obsolete Annotations}).
35341 @kindex set annotate
35342 @item set annotate @var{level}
35343 The @value{GDBN} command @code{set annotate} sets the level of
35344 annotations to the specified @var{level}.
35346 @item show annotate
35347 @kindex show annotate
35348 Show the current annotation level.
35351 This chapter describes level 3 annotations.
35353 A simple example of starting up @value{GDBN} with annotations is:
35356 $ @kbd{gdb --annotate=3}
35358 Copyright 2003 Free Software Foundation, Inc.
35359 GDB is free software, covered by the GNU General Public License,
35360 and you are welcome to change it and/or distribute copies of it
35361 under certain conditions.
35362 Type "show copying" to see the conditions.
35363 There is absolutely no warranty for GDB. Type "show warranty"
35365 This GDB was configured as "i386-pc-linux-gnu"
35376 Here @samp{quit} is input to @value{GDBN}; the rest is output from
35377 @value{GDBN}. The three lines beginning @samp{^Z^Z} (where @samp{^Z}
35378 denotes a @samp{control-z} character) are annotations; the rest is
35379 output from @value{GDBN}.
35381 @node Server Prefix
35382 @section The Server Prefix
35383 @cindex server prefix
35385 If you prefix a command with @samp{server } then it will not affect
35386 the command history, nor will it affect @value{GDBN}'s notion of which
35387 command to repeat if @key{RET} is pressed on a line by itself. This
35388 means that commands can be run behind a user's back by a front-end in
35389 a transparent manner.
35391 The @code{server } prefix does not affect the recording of values into
35392 the value history; to print a value without recording it into the
35393 value history, use the @code{output} command instead of the
35394 @code{print} command.
35396 Using this prefix also disables confirmation requests
35397 (@pxref{confirmation requests}).
35400 @section Annotation for @value{GDBN} Input
35402 @cindex annotations for prompts
35403 When @value{GDBN} prompts for input, it annotates this fact so it is possible
35404 to know when to send output, when the output from a given command is
35407 Different kinds of input each have a different @dfn{input type}. Each
35408 input type has three annotations: a @code{pre-} annotation, which
35409 denotes the beginning of any prompt which is being output, a plain
35410 annotation, which denotes the end of the prompt, and then a @code{post-}
35411 annotation which denotes the end of any echo which may (or may not) be
35412 associated with the input. For example, the @code{prompt} input type
35413 features the following annotations:
35421 The input types are
35424 @findex pre-prompt annotation
35425 @findex prompt annotation
35426 @findex post-prompt annotation
35428 When @value{GDBN} is prompting for a command (the main @value{GDBN} prompt).
35430 @findex pre-commands annotation
35431 @findex commands annotation
35432 @findex post-commands annotation
35434 When @value{GDBN} prompts for a set of commands, like in the @code{commands}
35435 command. The annotations are repeated for each command which is input.
35437 @findex pre-overload-choice annotation
35438 @findex overload-choice annotation
35439 @findex post-overload-choice annotation
35440 @item overload-choice
35441 When @value{GDBN} wants the user to select between various overloaded functions.
35443 @findex pre-query annotation
35444 @findex query annotation
35445 @findex post-query annotation
35447 When @value{GDBN} wants the user to confirm a potentially dangerous operation.
35449 @findex pre-prompt-for-continue annotation
35450 @findex prompt-for-continue annotation
35451 @findex post-prompt-for-continue annotation
35452 @item prompt-for-continue
35453 When @value{GDBN} is asking the user to press return to continue. Note: Don't
35454 expect this to work well; instead use @code{set height 0} to disable
35455 prompting. This is because the counting of lines is buggy in the
35456 presence of annotations.
35461 @cindex annotations for errors, warnings and interrupts
35463 @findex quit annotation
35468 This annotation occurs right before @value{GDBN} responds to an interrupt.
35470 @findex error annotation
35475 This annotation occurs right before @value{GDBN} responds to an error.
35477 Quit and error annotations indicate that any annotations which @value{GDBN} was
35478 in the middle of may end abruptly. For example, if a
35479 @code{value-history-begin} annotation is followed by a @code{error}, one
35480 cannot expect to receive the matching @code{value-history-end}. One
35481 cannot expect not to receive it either, however; an error annotation
35482 does not necessarily mean that @value{GDBN} is immediately returning all the way
35485 @findex error-begin annotation
35486 A quit or error annotation may be preceded by
35492 Any output between that and the quit or error annotation is the error
35495 Warning messages are not yet annotated.
35496 @c If we want to change that, need to fix warning(), type_error(),
35497 @c range_error(), and possibly other places.
35500 @section Invalidation Notices
35502 @cindex annotations for invalidation messages
35503 The following annotations say that certain pieces of state may have
35507 @findex frames-invalid annotation
35508 @item ^Z^Zframes-invalid
35510 The frames (for example, output from the @code{backtrace} command) may
35513 @findex breakpoints-invalid annotation
35514 @item ^Z^Zbreakpoints-invalid
35516 The breakpoints may have changed. For example, the user just added or
35517 deleted a breakpoint.
35520 @node Annotations for Running
35521 @section Running the Program
35522 @cindex annotations for running programs
35524 @findex starting annotation
35525 @findex stopping annotation
35526 When the program starts executing due to a @value{GDBN} command such as
35527 @code{step} or @code{continue},
35533 is output. When the program stops,
35539 is output. Before the @code{stopped} annotation, a variety of
35540 annotations describe how the program stopped.
35543 @findex exited annotation
35544 @item ^Z^Zexited @var{exit-status}
35545 The program exited, and @var{exit-status} is the exit status (zero for
35546 successful exit, otherwise nonzero).
35548 @findex signalled annotation
35549 @findex signal-name annotation
35550 @findex signal-name-end annotation
35551 @findex signal-string annotation
35552 @findex signal-string-end annotation
35553 @item ^Z^Zsignalled
35554 The program exited with a signal. After the @code{^Z^Zsignalled}, the
35555 annotation continues:
35561 ^Z^Zsignal-name-end
35565 ^Z^Zsignal-string-end
35570 where @var{name} is the name of the signal, such as @code{SIGILL} or
35571 @code{SIGSEGV}, and @var{string} is the explanation of the signal, such
35572 as @code{Illegal Instruction} or @code{Segmentation fault}. The arguments
35573 @var{intro-text}, @var{middle-text}, and @var{end-text} are for the
35574 user's benefit and have no particular format.
35576 @findex signal annotation
35578 The syntax of this annotation is just like @code{signalled}, but @value{GDBN} is
35579 just saying that the program received the signal, not that it was
35580 terminated with it.
35582 @findex breakpoint annotation
35583 @item ^Z^Zbreakpoint @var{number}
35584 The program hit breakpoint number @var{number}.
35586 @findex watchpoint annotation
35587 @item ^Z^Zwatchpoint @var{number}
35588 The program hit watchpoint number @var{number}.
35591 @node Source Annotations
35592 @section Displaying Source
35593 @cindex annotations for source display
35595 @findex source annotation
35596 The following annotation is used instead of displaying source code:
35599 ^Z^Zsource @var{filename}:@var{line}:@var{character}:@var{middle}:@var{addr}
35602 where @var{filename} is an absolute file name indicating which source
35603 file, @var{line} is the line number within that file (where 1 is the
35604 first line in the file), @var{character} is the character position
35605 within the file (where 0 is the first character in the file) (for most
35606 debug formats this will necessarily point to the beginning of a line),
35607 @var{middle} is @samp{middle} if @var{addr} is in the middle of the
35608 line, or @samp{beg} if @var{addr} is at the beginning of the line, and
35609 @var{addr} is the address in the target program associated with the
35610 source which is being displayed. The @var{addr} is in the form @samp{0x}
35611 followed by one or more lowercase hex digits (note that this does not
35612 depend on the language).
35614 @node JIT Interface
35615 @chapter JIT Compilation Interface
35616 @cindex just-in-time compilation
35617 @cindex JIT compilation interface
35619 This chapter documents @value{GDBN}'s @dfn{just-in-time} (JIT) compilation
35620 interface. A JIT compiler is a program or library that generates native
35621 executable code at runtime and executes it, usually in order to achieve good
35622 performance while maintaining platform independence.
35624 Programs that use JIT compilation are normally difficult to debug because
35625 portions of their code are generated at runtime, instead of being loaded from
35626 object files, which is where @value{GDBN} normally finds the program's symbols
35627 and debug information. In order to debug programs that use JIT compilation,
35628 @value{GDBN} has an interface that allows the program to register in-memory
35629 symbol files with @value{GDBN} at runtime.
35631 If you are using @value{GDBN} to debug a program that uses this interface, then
35632 it should work transparently so long as you have not stripped the binary. If
35633 you are developing a JIT compiler, then the interface is documented in the rest
35634 of this chapter. At this time, the only known client of this interface is the
35637 Broadly speaking, the JIT interface mirrors the dynamic loader interface. The
35638 JIT compiler communicates with @value{GDBN} by writing data into a global
35639 variable and calling a fuction at a well-known symbol. When @value{GDBN}
35640 attaches, it reads a linked list of symbol files from the global variable to
35641 find existing code, and puts a breakpoint in the function so that it can find
35642 out about additional code.
35645 * Declarations:: Relevant C struct declarations
35646 * Registering Code:: Steps to register code
35647 * Unregistering Code:: Steps to unregister code
35648 * Custom Debug Info:: Emit debug information in a custom format
35652 @section JIT Declarations
35654 These are the relevant struct declarations that a C program should include to
35655 implement the interface:
35665 struct jit_code_entry
35667 struct jit_code_entry *next_entry;
35668 struct jit_code_entry *prev_entry;
35669 const char *symfile_addr;
35670 uint64_t symfile_size;
35673 struct jit_descriptor
35676 /* This type should be jit_actions_t, but we use uint32_t
35677 to be explicit about the bitwidth. */
35678 uint32_t action_flag;
35679 struct jit_code_entry *relevant_entry;
35680 struct jit_code_entry *first_entry;
35683 /* GDB puts a breakpoint in this function. */
35684 void __attribute__((noinline)) __jit_debug_register_code() @{ @};
35686 /* Make sure to specify the version statically, because the
35687 debugger may check the version before we can set it. */
35688 struct jit_descriptor __jit_debug_descriptor = @{ 1, 0, 0, 0 @};
35691 If the JIT is multi-threaded, then it is important that the JIT synchronize any
35692 modifications to this global data properly, which can easily be done by putting
35693 a global mutex around modifications to these structures.
35695 @node Registering Code
35696 @section Registering Code
35698 To register code with @value{GDBN}, the JIT should follow this protocol:
35702 Generate an object file in memory with symbols and other desired debug
35703 information. The file must include the virtual addresses of the sections.
35706 Create a code entry for the file, which gives the start and size of the symbol
35710 Add it to the linked list in the JIT descriptor.
35713 Point the relevant_entry field of the descriptor at the entry.
35716 Set @code{action_flag} to @code{JIT_REGISTER} and call
35717 @code{__jit_debug_register_code}.
35720 When @value{GDBN} is attached and the breakpoint fires, @value{GDBN} uses the
35721 @code{relevant_entry} pointer so it doesn't have to walk the list looking for
35722 new code. However, the linked list must still be maintained in order to allow
35723 @value{GDBN} to attach to a running process and still find the symbol files.
35725 @node Unregistering Code
35726 @section Unregistering Code
35728 If code is freed, then the JIT should use the following protocol:
35732 Remove the code entry corresponding to the code from the linked list.
35735 Point the @code{relevant_entry} field of the descriptor at the code entry.
35738 Set @code{action_flag} to @code{JIT_UNREGISTER} and call
35739 @code{__jit_debug_register_code}.
35742 If the JIT frees or recompiles code without unregistering it, then @value{GDBN}
35743 and the JIT will leak the memory used for the associated symbol files.
35745 @node Custom Debug Info
35746 @section Custom Debug Info
35747 @cindex custom JIT debug info
35748 @cindex JIT debug info reader
35750 Generating debug information in platform-native file formats (like ELF
35751 or COFF) may be an overkill for JIT compilers; especially if all the
35752 debug info is used for is displaying a meaningful backtrace. The
35753 issue can be resolved by having the JIT writers decide on a debug info
35754 format and also provide a reader that parses the debug info generated
35755 by the JIT compiler. This section gives a brief overview on writing
35756 such a parser. More specific details can be found in the source file
35757 @file{gdb/jit-reader.in}, which is also installed as a header at
35758 @file{@var{includedir}/gdb/jit-reader.h} for easy inclusion.
35760 The reader is implemented as a shared object (so this functionality is
35761 not available on platforms which don't allow loading shared objects at
35762 runtime). Two @value{GDBN} commands, @code{jit-reader-load} and
35763 @code{jit-reader-unload} are provided, to be used to load and unload
35764 the readers from a preconfigured directory. Once loaded, the shared
35765 object is used the parse the debug information emitted by the JIT
35769 * Using JIT Debug Info Readers:: How to use supplied readers correctly
35770 * Writing JIT Debug Info Readers:: Creating a debug-info reader
35773 @node Using JIT Debug Info Readers
35774 @subsection Using JIT Debug Info Readers
35775 @kindex jit-reader-load
35776 @kindex jit-reader-unload
35778 Readers can be loaded and unloaded using the @code{jit-reader-load}
35779 and @code{jit-reader-unload} commands.
35782 @item jit-reader-load @var{reader}
35783 Load the JIT reader named @var{reader}, which is a shared
35784 object specified as either an absolute or a relative file name. In
35785 the latter case, @value{GDBN} will try to load the reader from a
35786 pre-configured directory, usually @file{@var{libdir}/gdb/} on a UNIX
35787 system (here @var{libdir} is the system library directory, often
35788 @file{/usr/local/lib}).
35790 Only one reader can be active at a time; trying to load a second
35791 reader when one is already loaded will result in @value{GDBN}
35792 reporting an error. A new JIT reader can be loaded by first unloading
35793 the current one using @code{jit-reader-unload} and then invoking
35794 @code{jit-reader-load}.
35796 @item jit-reader-unload
35797 Unload the currently loaded JIT reader.
35801 @node Writing JIT Debug Info Readers
35802 @subsection Writing JIT Debug Info Readers
35803 @cindex writing JIT debug info readers
35805 As mentioned, a reader is essentially a shared object conforming to a
35806 certain ABI. This ABI is described in @file{jit-reader.h}.
35808 @file{jit-reader.h} defines the structures, macros and functions
35809 required to write a reader. It is installed (along with
35810 @value{GDBN}), in @file{@var{includedir}/gdb} where @var{includedir} is
35811 the system include directory.
35813 Readers need to be released under a GPL compatible license. A reader
35814 can be declared as released under such a license by placing the macro
35815 @code{GDB_DECLARE_GPL_COMPATIBLE_READER} in a source file.
35817 The entry point for readers is the symbol @code{gdb_init_reader},
35818 which is expected to be a function with the prototype
35820 @findex gdb_init_reader
35822 extern struct gdb_reader_funcs *gdb_init_reader (void);
35825 @cindex @code{struct gdb_reader_funcs}
35827 @code{struct gdb_reader_funcs} contains a set of pointers to callback
35828 functions. These functions are executed to read the debug info
35829 generated by the JIT compiler (@code{read}), to unwind stack frames
35830 (@code{unwind}) and to create canonical frame IDs
35831 (@code{get_Frame_id}). It also has a callback that is called when the
35832 reader is being unloaded (@code{destroy}). The struct looks like this
35835 struct gdb_reader_funcs
35837 /* Must be set to GDB_READER_INTERFACE_VERSION. */
35838 int reader_version;
35840 /* For use by the reader. */
35843 gdb_read_debug_info *read;
35844 gdb_unwind_frame *unwind;
35845 gdb_get_frame_id *get_frame_id;
35846 gdb_destroy_reader *destroy;
35850 @cindex @code{struct gdb_symbol_callbacks}
35851 @cindex @code{struct gdb_unwind_callbacks}
35853 The callbacks are provided with another set of callbacks by
35854 @value{GDBN} to do their job. For @code{read}, these callbacks are
35855 passed in a @code{struct gdb_symbol_callbacks} and for @code{unwind}
35856 and @code{get_frame_id}, in a @code{struct gdb_unwind_callbacks}.
35857 @code{struct gdb_symbol_callbacks} has callbacks to create new object
35858 files and new symbol tables inside those object files. @code{struct
35859 gdb_unwind_callbacks} has callbacks to read registers off the current
35860 frame and to write out the values of the registers in the previous
35861 frame. Both have a callback (@code{target_read}) to read bytes off the
35862 target's address space.
35864 @node In-Process Agent
35865 @chapter In-Process Agent
35866 @cindex debugging agent
35867 The traditional debugging model is conceptually low-speed, but works fine,
35868 because most bugs can be reproduced in debugging-mode execution. However,
35869 as multi-core or many-core processors are becoming mainstream, and
35870 multi-threaded programs become more and more popular, there should be more
35871 and more bugs that only manifest themselves at normal-mode execution, for
35872 example, thread races, because debugger's interference with the program's
35873 timing may conceal the bugs. On the other hand, in some applications,
35874 it is not feasible for the debugger to interrupt the program's execution
35875 long enough for the developer to learn anything helpful about its behavior.
35876 If the program's correctness depends on its real-time behavior, delays
35877 introduced by a debugger might cause the program to fail, even when the
35878 code itself is correct. It is useful to be able to observe the program's
35879 behavior without interrupting it.
35881 Therefore, traditional debugging model is too intrusive to reproduce
35882 some bugs. In order to reduce the interference with the program, we can
35883 reduce the number of operations performed by debugger. The
35884 @dfn{In-Process Agent}, a shared library, is running within the same
35885 process with inferior, and is able to perform some debugging operations
35886 itself. As a result, debugger is only involved when necessary, and
35887 performance of debugging can be improved accordingly. Note that
35888 interference with program can be reduced but can't be removed completely,
35889 because the in-process agent will still stop or slow down the program.
35891 The in-process agent can interpret and execute Agent Expressions
35892 (@pxref{Agent Expressions}) during performing debugging operations. The
35893 agent expressions can be used for different purposes, such as collecting
35894 data in tracepoints, and condition evaluation in breakpoints.
35896 @anchor{Control Agent}
35897 You can control whether the in-process agent is used as an aid for
35898 debugging with the following commands:
35901 @kindex set agent on
35903 Causes the in-process agent to perform some operations on behalf of the
35904 debugger. Just which operations requested by the user will be done
35905 by the in-process agent depends on the its capabilities. For example,
35906 if you request to evaluate breakpoint conditions in the in-process agent,
35907 and the in-process agent has such capability as well, then breakpoint
35908 conditions will be evaluated in the in-process agent.
35910 @kindex set agent off
35911 @item set agent off
35912 Disables execution of debugging operations by the in-process agent. All
35913 of the operations will be performed by @value{GDBN}.
35917 Display the current setting of execution of debugging operations by
35918 the in-process agent.
35922 * In-Process Agent Protocol::
35925 @node In-Process Agent Protocol
35926 @section In-Process Agent Protocol
35927 @cindex in-process agent protocol
35929 The in-process agent is able to communicate with both @value{GDBN} and
35930 GDBserver (@pxref{In-Process Agent}). This section documents the protocol
35931 used for communications between @value{GDBN} or GDBserver and the IPA.
35932 In general, @value{GDBN} or GDBserver sends commands
35933 (@pxref{IPA Protocol Commands}) and data to in-process agent, and then
35934 in-process agent replies back with the return result of the command, or
35935 some other information. The data sent to in-process agent is composed
35936 of primitive data types, such as 4-byte or 8-byte type, and composite
35937 types, which are called objects (@pxref{IPA Protocol Objects}).
35940 * IPA Protocol Objects::
35941 * IPA Protocol Commands::
35944 @node IPA Protocol Objects
35945 @subsection IPA Protocol Objects
35946 @cindex ipa protocol objects
35948 The commands sent to and results received from agent may contain some
35949 complex data types called @dfn{objects}.
35951 The in-process agent is running on the same machine with @value{GDBN}
35952 or GDBserver, so it doesn't have to handle as much differences between
35953 two ends as remote protocol (@pxref{Remote Protocol}) tries to handle.
35954 However, there are still some differences of two ends in two processes:
35958 word size. On some 64-bit machines, @value{GDBN} or GDBserver can be
35959 compiled as a 64-bit executable, while in-process agent is a 32-bit one.
35961 ABI. Some machines may have multiple types of ABI, @value{GDBN} or
35962 GDBserver is compiled with one, and in-process agent is compiled with
35966 Here are the IPA Protocol Objects:
35970 agent expression object. It represents an agent expression
35971 (@pxref{Agent Expressions}).
35972 @anchor{agent expression object}
35974 tracepoint action object. It represents a tracepoint action
35975 (@pxref{Tracepoint Actions,,Tracepoint Action Lists}) to collect registers,
35976 memory, static trace data and to evaluate expression.
35977 @anchor{tracepoint action object}
35979 tracepoint object. It represents a tracepoint (@pxref{Tracepoints}).
35980 @anchor{tracepoint object}
35984 The following table describes important attributes of each IPA protocol
35987 @multitable @columnfractions .30 .20 .50
35988 @headitem Name @tab Size @tab Description
35989 @item @emph{agent expression object} @tab @tab
35990 @item length @tab 4 @tab length of bytes code
35991 @item byte code @tab @var{length} @tab contents of byte code
35992 @item @emph{tracepoint action for collecting memory} @tab @tab
35993 @item 'M' @tab 1 @tab type of tracepoint action
35994 @item addr @tab 8 @tab if @var{basereg} is @samp{-1}, @var{addr} is the
35995 address of the lowest byte to collect, otherwise @var{addr} is the offset
35996 of @var{basereg} for memory collecting.
35997 @item len @tab 8 @tab length of memory for collecting
35998 @item basereg @tab 4 @tab the register number containing the starting
35999 memory address for collecting.
36000 @item @emph{tracepoint action for collecting registers} @tab @tab
36001 @item 'R' @tab 1 @tab type of tracepoint action
36002 @item @emph{tracepoint action for collecting static trace data} @tab @tab
36003 @item 'L' @tab 1 @tab type of tracepoint action
36004 @item @emph{tracepoint action for expression evaluation} @tab @tab
36005 @item 'X' @tab 1 @tab type of tracepoint action
36006 @item agent expression @tab length of @tab @ref{agent expression object}
36007 @item @emph{tracepoint object} @tab @tab
36008 @item number @tab 4 @tab number of tracepoint
36009 @item address @tab 8 @tab address of tracepoint inserted on
36010 @item type @tab 4 @tab type of tracepoint
36011 @item enabled @tab 1 @tab enable or disable of tracepoint
36012 @item step_count @tab 8 @tab step
36013 @item pass_count @tab 8 @tab pass
36014 @item numactions @tab 4 @tab number of tracepoint actions
36015 @item hit count @tab 8 @tab hit count
36016 @item trace frame usage @tab 8 @tab trace frame usage
36017 @item compiled_cond @tab 8 @tab compiled condition
36018 @item orig_size @tab 8 @tab orig size
36019 @item condition @tab 4 if condition is NULL otherwise length of
36020 @ref{agent expression object}
36021 @tab zero if condition is NULL, otherwise is
36022 @ref{agent expression object}
36023 @item actions @tab variable
36024 @tab numactions number of @ref{tracepoint action object}
36027 @node IPA Protocol Commands
36028 @subsection IPA Protocol Commands
36029 @cindex ipa protocol commands
36031 The spaces in each command are delimiters to ease reading this commands
36032 specification. They don't exist in real commands.
36036 @item FastTrace:@var{tracepoint_object} @var{gdb_jump_pad_head}
36037 Installs a new fast tracepoint described by @var{tracepoint_object}
36038 (@pxref{tracepoint object}). The @var{gdb_jump_pad_head}, 8-byte long, is the
36039 head of @dfn{jumppad}, which is used to jump to data collection routine
36044 @item OK @var{target_address} @var{gdb_jump_pad_head} @var{fjump_size} @var{fjump}
36045 @var{target_address} is address of tracepoint in the inferior.
36046 The @var{gdb_jump_pad_head} is updated head of jumppad. Both of
36047 @var{target_address} and @var{gdb_jump_pad_head} are 8-byte long.
36048 The @var{fjump} contains a sequence of instructions jump to jumppad entry.
36049 The @var{fjump_size}, 4-byte long, is the size of @var{fjump}.
36056 Closes the in-process agent. This command is sent when @value{GDBN} or GDBserver
36057 is about to kill inferiors.
36065 @item probe_marker_at:@var{address}
36066 Asks in-process agent to probe the marker at @var{address}.
36073 @item unprobe_marker_at:@var{address}
36074 Asks in-process agent to unprobe the marker at @var{address}.
36078 @chapter Reporting Bugs in @value{GDBN}
36079 @cindex bugs in @value{GDBN}
36080 @cindex reporting bugs in @value{GDBN}
36082 Your bug reports play an essential role in making @value{GDBN} reliable.
36084 Reporting a bug may help you by bringing a solution to your problem, or it
36085 may not. But in any case the principal function of a bug report is to help
36086 the entire community by making the next version of @value{GDBN} work better. Bug
36087 reports are your contribution to the maintenance of @value{GDBN}.
36089 In order for a bug report to serve its purpose, you must include the
36090 information that enables us to fix the bug.
36093 * Bug Criteria:: Have you found a bug?
36094 * Bug Reporting:: How to report bugs
36098 @section Have You Found a Bug?
36099 @cindex bug criteria
36101 If you are not sure whether you have found a bug, here are some guidelines:
36104 @cindex fatal signal
36105 @cindex debugger crash
36106 @cindex crash of debugger
36108 If the debugger gets a fatal signal, for any input whatever, that is a
36109 @value{GDBN} bug. Reliable debuggers never crash.
36111 @cindex error on valid input
36113 If @value{GDBN} produces an error message for valid input, that is a
36114 bug. (Note that if you're cross debugging, the problem may also be
36115 somewhere in the connection to the target.)
36117 @cindex invalid input
36119 If @value{GDBN} does not produce an error message for invalid input,
36120 that is a bug. However, you should note that your idea of
36121 ``invalid input'' might be our idea of ``an extension'' or ``support
36122 for traditional practice''.
36125 If you are an experienced user of debugging tools, your suggestions
36126 for improvement of @value{GDBN} are welcome in any case.
36129 @node Bug Reporting
36130 @section How to Report Bugs
36131 @cindex bug reports
36132 @cindex @value{GDBN} bugs, reporting
36134 A number of companies and individuals offer support for @sc{gnu} products.
36135 If you obtained @value{GDBN} from a support organization, we recommend you
36136 contact that organization first.
36138 You can find contact information for many support companies and
36139 individuals in the file @file{etc/SERVICE} in the @sc{gnu} Emacs
36141 @c should add a web page ref...
36144 @ifset BUGURL_DEFAULT
36145 In any event, we also recommend that you submit bug reports for
36146 @value{GDBN}. The preferred method is to submit them directly using
36147 @uref{http://www.gnu.org/software/gdb/bugs/, @value{GDBN}'s Bugs web
36148 page}. Alternatively, the @email{bug-gdb@@gnu.org, e-mail gateway} can
36151 @strong{Do not send bug reports to @samp{info-gdb}, or to
36152 @samp{help-gdb}, or to any newsgroups.} Most users of @value{GDBN} do
36153 not want to receive bug reports. Those that do have arranged to receive
36156 The mailing list @samp{bug-gdb} has a newsgroup @samp{gnu.gdb.bug} which
36157 serves as a repeater. The mailing list and the newsgroup carry exactly
36158 the same messages. Often people think of posting bug reports to the
36159 newsgroup instead of mailing them. This appears to work, but it has one
36160 problem which can be crucial: a newsgroup posting often lacks a mail
36161 path back to the sender. Thus, if we need to ask for more information,
36162 we may be unable to reach you. For this reason, it is better to send
36163 bug reports to the mailing list.
36165 @ifclear BUGURL_DEFAULT
36166 In any event, we also recommend that you submit bug reports for
36167 @value{GDBN} to @value{BUGURL}.
36171 The fundamental principle of reporting bugs usefully is this:
36172 @strong{report all the facts}. If you are not sure whether to state a
36173 fact or leave it out, state it!
36175 Often people omit facts because they think they know what causes the
36176 problem and assume that some details do not matter. Thus, you might
36177 assume that the name of the variable you use in an example does not matter.
36178 Well, probably it does not, but one cannot be sure. Perhaps the bug is a
36179 stray memory reference which happens to fetch from the location where that
36180 name is stored in memory; perhaps, if the name were different, the contents
36181 of that location would fool the debugger into doing the right thing despite
36182 the bug. Play it safe and give a specific, complete example. That is the
36183 easiest thing for you to do, and the most helpful.
36185 Keep in mind that the purpose of a bug report is to enable us to fix the
36186 bug. It may be that the bug has been reported previously, but neither
36187 you nor we can know that unless your bug report is complete and
36190 Sometimes people give a few sketchy facts and ask, ``Does this ring a
36191 bell?'' Those bug reports are useless, and we urge everyone to
36192 @emph{refuse to respond to them} except to chide the sender to report
36195 To enable us to fix the bug, you should include all these things:
36199 The version of @value{GDBN}. @value{GDBN} announces it if you start
36200 with no arguments; you can also print it at any time using @code{show
36203 Without this, we will not know whether there is any point in looking for
36204 the bug in the current version of @value{GDBN}.
36207 The type of machine you are using, and the operating system name and
36211 The details of the @value{GDBN} build-time configuration.
36212 @value{GDBN} shows these details if you invoke it with the
36213 @option{--configuration} command-line option, or if you type
36214 @code{show configuration} at @value{GDBN}'s prompt.
36217 What compiler (and its version) was used to compile @value{GDBN}---e.g.@:
36218 ``@value{GCC}--2.8.1''.
36221 What compiler (and its version) was used to compile the program you are
36222 debugging---e.g.@: ``@value{GCC}--2.8.1'', or ``HP92453-01 A.10.32.03 HP
36223 C Compiler''. For @value{NGCC}, you can say @kbd{@value{GCC} --version}
36224 to get this information; for other compilers, see the documentation for
36228 The command arguments you gave the compiler to compile your example and
36229 observe the bug. For example, did you use @samp{-O}? To guarantee
36230 you will not omit something important, list them all. A copy of the
36231 Makefile (or the output from make) is sufficient.
36233 If we were to try to guess the arguments, we would probably guess wrong
36234 and then we might not encounter the bug.
36237 A complete input script, and all necessary source files, that will
36241 A description of what behavior you observe that you believe is
36242 incorrect. For example, ``It gets a fatal signal.''
36244 Of course, if the bug is that @value{GDBN} gets a fatal signal, then we
36245 will certainly notice it. But if the bug is incorrect output, we might
36246 not notice unless it is glaringly wrong. You might as well not give us
36247 a chance to make a mistake.
36249 Even if the problem you experience is a fatal signal, you should still
36250 say so explicitly. Suppose something strange is going on, such as, your
36251 copy of @value{GDBN} is out of synch, or you have encountered a bug in
36252 the C library on your system. (This has happened!) Your copy might
36253 crash and ours would not. If you told us to expect a crash, then when
36254 ours fails to crash, we would know that the bug was not happening for
36255 us. If you had not told us to expect a crash, then we would not be able
36256 to draw any conclusion from our observations.
36259 @cindex recording a session script
36260 To collect all this information, you can use a session recording program
36261 such as @command{script}, which is available on many Unix systems.
36262 Just run your @value{GDBN} session inside @command{script} and then
36263 include the @file{typescript} file with your bug report.
36265 Another way to record a @value{GDBN} session is to run @value{GDBN}
36266 inside Emacs and then save the entire buffer to a file.
36269 If you wish to suggest changes to the @value{GDBN} source, send us context
36270 diffs. If you even discuss something in the @value{GDBN} source, refer to
36271 it by context, not by line number.
36273 The line numbers in our development sources will not match those in your
36274 sources. Your line numbers would convey no useful information to us.
36278 Here are some things that are not necessary:
36282 A description of the envelope of the bug.
36284 Often people who encounter a bug spend a lot of time investigating
36285 which changes to the input file will make the bug go away and which
36286 changes will not affect it.
36288 This is often time consuming and not very useful, because the way we
36289 will find the bug is by running a single example under the debugger
36290 with breakpoints, not by pure deduction from a series of examples.
36291 We recommend that you save your time for something else.
36293 Of course, if you can find a simpler example to report @emph{instead}
36294 of the original one, that is a convenience for us. Errors in the
36295 output will be easier to spot, running under the debugger will take
36296 less time, and so on.
36298 However, simplification is not vital; if you do not want to do this,
36299 report the bug anyway and send us the entire test case you used.
36302 A patch for the bug.
36304 A patch for the bug does help us if it is a good one. But do not omit
36305 the necessary information, such as the test case, on the assumption that
36306 a patch is all we need. We might see problems with your patch and decide
36307 to fix the problem another way, or we might not understand it at all.
36309 Sometimes with a program as complicated as @value{GDBN} it is very hard to
36310 construct an example that will make the program follow a certain path
36311 through the code. If you do not send us the example, we will not be able
36312 to construct one, so we will not be able to verify that the bug is fixed.
36314 And if we cannot understand what bug you are trying to fix, or why your
36315 patch should be an improvement, we will not install it. A test case will
36316 help us to understand.
36319 A guess about what the bug is or what it depends on.
36321 Such guesses are usually wrong. Even we cannot guess right about such
36322 things without first using the debugger to find the facts.
36325 @c The readline documentation is distributed with the readline code
36326 @c and consists of the two following files:
36329 @c Use -I with makeinfo to point to the appropriate directory,
36330 @c environment var TEXINPUTS with TeX.
36331 @ifclear SYSTEM_READLINE
36332 @include rluser.texi
36333 @include hsuser.texi
36337 @appendix In Memoriam
36339 The @value{GDBN} project mourns the loss of the following long-time
36344 Fred was a long-standing contributor to @value{GDBN} (1991-2006), and
36345 to Free Software in general. Outside of @value{GDBN}, he was known in
36346 the Amiga world for his series of Fish Disks, and the GeekGadget project.
36348 @item Michael Snyder
36349 Michael was one of the Global Maintainers of the @value{GDBN} project,
36350 with contributions recorded as early as 1996, until 2011. In addition
36351 to his day to day participation, he was a large driving force behind
36352 adding Reverse Debugging to @value{GDBN}.
36355 Beyond their technical contributions to the project, they were also
36356 enjoyable members of the Free Software Community. We will miss them.
36358 @node Formatting Documentation
36359 @appendix Formatting Documentation
36361 @cindex @value{GDBN} reference card
36362 @cindex reference card
36363 The @value{GDBN} 4 release includes an already-formatted reference card, ready
36364 for printing with PostScript or Ghostscript, in the @file{gdb}
36365 subdirectory of the main source directory@footnote{In
36366 @file{gdb-@value{GDBVN}/gdb/refcard.ps} of the version @value{GDBVN}
36367 release.}. If you can use PostScript or Ghostscript with your printer,
36368 you can print the reference card immediately with @file{refcard.ps}.
36370 The release also includes the source for the reference card. You
36371 can format it, using @TeX{}, by typing:
36377 The @value{GDBN} reference card is designed to print in @dfn{landscape}
36378 mode on US ``letter'' size paper;
36379 that is, on a sheet 11 inches wide by 8.5 inches
36380 high. You will need to specify this form of printing as an option to
36381 your @sc{dvi} output program.
36383 @cindex documentation
36385 All the documentation for @value{GDBN} comes as part of the machine-readable
36386 distribution. The documentation is written in Texinfo format, which is
36387 a documentation system that uses a single source file to produce both
36388 on-line information and a printed manual. You can use one of the Info
36389 formatting commands to create the on-line version of the documentation
36390 and @TeX{} (or @code{texi2roff}) to typeset the printed version.
36392 @value{GDBN} includes an already formatted copy of the on-line Info
36393 version of this manual in the @file{gdb} subdirectory. The main Info
36394 file is @file{gdb-@value{GDBVN}/gdb/gdb.info}, and it refers to
36395 subordinate files matching @samp{gdb.info*} in the same directory. If
36396 necessary, you can print out these files, or read them with any editor;
36397 but they are easier to read using the @code{info} subsystem in @sc{gnu}
36398 Emacs or the standalone @code{info} program, available as part of the
36399 @sc{gnu} Texinfo distribution.
36401 If you want to format these Info files yourself, you need one of the
36402 Info formatting programs, such as @code{texinfo-format-buffer} or
36405 If you have @code{makeinfo} installed, and are in the top level
36406 @value{GDBN} source directory (@file{gdb-@value{GDBVN}}, in the case of
36407 version @value{GDBVN}), you can make the Info file by typing:
36414 If you want to typeset and print copies of this manual, you need @TeX{},
36415 a program to print its @sc{dvi} output files, and @file{texinfo.tex}, the
36416 Texinfo definitions file.
36418 @TeX{} is a typesetting program; it does not print files directly, but
36419 produces output files called @sc{dvi} files. To print a typeset
36420 document, you need a program to print @sc{dvi} files. If your system
36421 has @TeX{} installed, chances are it has such a program. The precise
36422 command to use depends on your system; @kbd{lpr -d} is common; another
36423 (for PostScript devices) is @kbd{dvips}. The @sc{dvi} print command may
36424 require a file name without any extension or a @samp{.dvi} extension.
36426 @TeX{} also requires a macro definitions file called
36427 @file{texinfo.tex}. This file tells @TeX{} how to typeset a document
36428 written in Texinfo format. On its own, @TeX{} cannot either read or
36429 typeset a Texinfo file. @file{texinfo.tex} is distributed with GDB
36430 and is located in the @file{gdb-@var{version-number}/texinfo}
36433 If you have @TeX{} and a @sc{dvi} printer program installed, you can
36434 typeset and print this manual. First switch to the @file{gdb}
36435 subdirectory of the main source directory (for example, to
36436 @file{gdb-@value{GDBVN}/gdb}) and type:
36442 Then give @file{gdb.dvi} to your @sc{dvi} printing program.
36444 @node Installing GDB
36445 @appendix Installing @value{GDBN}
36446 @cindex installation
36449 * Requirements:: Requirements for building @value{GDBN}
36450 * Running Configure:: Invoking the @value{GDBN} @file{configure} script
36451 * Separate Objdir:: Compiling @value{GDBN} in another directory
36452 * Config Names:: Specifying names for hosts and targets
36453 * Configure Options:: Summary of options for configure
36454 * System-wide configuration:: Having a system-wide init file
36458 @section Requirements for Building @value{GDBN}
36459 @cindex building @value{GDBN}, requirements for
36461 Building @value{GDBN} requires various tools and packages to be available.
36462 Other packages will be used only if they are found.
36464 @heading Tools/Packages Necessary for Building @value{GDBN}
36466 @item C@t{++}11 compiler
36467 @value{GDBN} is written in C@t{++}11. It should be buildable with any
36468 recent C@t{++}11 compiler, e.g.@: GCC.
36471 @value{GDBN}'s build system relies on features only found in the GNU
36472 make program. Other variants of @code{make} will not work.
36475 @heading Tools/Packages Optional for Building @value{GDBN}
36479 @value{GDBN} can use the Expat XML parsing library. This library may be
36480 included with your operating system distribution; if it is not, you
36481 can get the latest version from @url{http://expat.sourceforge.net}.
36482 The @file{configure} script will search for this library in several
36483 standard locations; if it is installed in an unusual path, you can
36484 use the @option{--with-libexpat-prefix} option to specify its location.
36490 Remote protocol memory maps (@pxref{Memory Map Format})
36492 Target descriptions (@pxref{Target Descriptions})
36494 Remote shared library lists (@xref{Library List Format},
36495 or alternatively @pxref{Library List Format for SVR4 Targets})
36497 MS-Windows shared libraries (@pxref{Shared Libraries})
36499 Traceframe info (@pxref{Traceframe Info Format})
36501 Branch trace (@pxref{Branch Trace Format},
36502 @pxref{Branch Trace Configuration Format})
36506 @value{GDBN} can be scripted using GNU Guile. @xref{Guile}. By
36507 default, @value{GDBN} will be compiled if the Guile libraries are
36508 installed and are found by @file{configure}. You can use the
36509 @code{--with-guile} option to request Guile, and pass either the Guile
36510 version number or the file name of the relevant @code{pkg-config}
36511 program to choose a particular version of Guile.
36514 @value{GDBN}'s features related to character sets (@pxref{Character
36515 Sets}) require a functioning @code{iconv} implementation. If you are
36516 on a GNU system, then this is provided by the GNU C Library. Some
36517 other systems also provide a working @code{iconv}.
36519 If @value{GDBN} is using the @code{iconv} program which is installed
36520 in a non-standard place, you will need to tell @value{GDBN} where to
36521 find it. This is done with @option{--with-iconv-bin} which specifies
36522 the directory that contains the @code{iconv} program. This program is
36523 run in order to make a list of the available character sets.
36525 On systems without @code{iconv}, you can install GNU Libiconv. If
36526 Libiconv is installed in a standard place, @value{GDBN} will
36527 automatically use it if it is needed. If you have previously
36528 installed Libiconv in a non-standard place, you can use the
36529 @option{--with-libiconv-prefix} option to @file{configure}.
36531 @value{GDBN}'s top-level @file{configure} and @file{Makefile} will
36532 arrange to build Libiconv if a directory named @file{libiconv} appears
36533 in the top-most source directory. If Libiconv is built this way, and
36534 if the operating system does not provide a suitable @code{iconv}
36535 implementation, then the just-built library will automatically be used
36536 by @value{GDBN}. One easy way to set this up is to download GNU
36537 Libiconv, unpack it inside the top-level directory of the @value{GDBN}
36538 source tree, and then rename the directory holding the Libiconv source
36539 code to @samp{libiconv}.
36542 @value{GDBN} can support debugging sections that are compressed with
36543 the LZMA library. @xref{MiniDebugInfo}. If this library is not
36544 included with your operating system, you can find it in the xz package
36545 at @url{http://tukaani.org/xz/}. If the LZMA library is available in
36546 the usual place, then the @file{configure} script will use it
36547 automatically. If it is installed in an unusual path, you can use the
36548 @option{--with-lzma-prefix} option to specify its location.
36552 @value{GDBN} can use the GNU MPFR multiple-precision floating-point
36553 library. This library may be included with your operating system
36554 distribution; if it is not, you can get the latest version from
36555 @url{http://www.mpfr.org}. The @file{configure} script will search
36556 for this library in several standard locations; if it is installed
36557 in an unusual path, you can use the @option{--with-libmpfr-prefix}
36558 option to specify its location.
36560 GNU MPFR is used to emulate target floating-point arithmetic during
36561 expression evaluation when the target uses different floating-point
36562 formats than the host. If GNU MPFR it is not available, @value{GDBN}
36563 will fall back to using host floating-point arithmetic.
36566 @value{GDBN} can be scripted using Python language. @xref{Python}.
36567 By default, @value{GDBN} will be compiled if the Python libraries are
36568 installed and are found by @file{configure}. You can use the
36569 @code{--with-python} option to request Python, and pass either the
36570 file name of the relevant @code{python} executable, or the name of the
36571 directory in which Python is installed, to choose a particular
36572 installation of Python.
36575 @cindex compressed debug sections
36576 @value{GDBN} will use the @samp{zlib} library, if available, to read
36577 compressed debug sections. Some linkers, such as GNU gold, are capable
36578 of producing binaries with compressed debug sections. If @value{GDBN}
36579 is compiled with @samp{zlib}, it will be able to read the debug
36580 information in such binaries.
36582 The @samp{zlib} library is likely included with your operating system
36583 distribution; if it is not, you can get the latest version from
36584 @url{http://zlib.net}.
36587 @node Running Configure
36588 @section Invoking the @value{GDBN} @file{configure} Script
36589 @cindex configuring @value{GDBN}
36590 @value{GDBN} comes with a @file{configure} script that automates the process
36591 of preparing @value{GDBN} for installation; you can then use @code{make} to
36592 build the @code{gdb} program.
36594 @c irrelevant in info file; it's as current as the code it lives with.
36595 @footnote{If you have a more recent version of @value{GDBN} than @value{GDBVN},
36596 look at the @file{README} file in the sources; we may have improved the
36597 installation procedures since publishing this manual.}
36600 The @value{GDBN} distribution includes all the source code you need for
36601 @value{GDBN} in a single directory, whose name is usually composed by
36602 appending the version number to @samp{gdb}.
36604 For example, the @value{GDBN} version @value{GDBVN} distribution is in the
36605 @file{gdb-@value{GDBVN}} directory. That directory contains:
36608 @item gdb-@value{GDBVN}/configure @r{(and supporting files)}
36609 script for configuring @value{GDBN} and all its supporting libraries
36611 @item gdb-@value{GDBVN}/gdb
36612 the source specific to @value{GDBN} itself
36614 @item gdb-@value{GDBVN}/bfd
36615 source for the Binary File Descriptor library
36617 @item gdb-@value{GDBVN}/include
36618 @sc{gnu} include files
36620 @item gdb-@value{GDBVN}/libiberty
36621 source for the @samp{-liberty} free software library
36623 @item gdb-@value{GDBVN}/opcodes
36624 source for the library of opcode tables and disassemblers
36626 @item gdb-@value{GDBVN}/readline
36627 source for the @sc{gnu} command-line interface
36630 There may be other subdirectories as well.
36632 The simplest way to configure and build @value{GDBN} is to run @file{configure}
36633 from the @file{gdb-@var{version-number}} source directory, which in
36634 this example is the @file{gdb-@value{GDBVN}} directory.
36636 First switch to the @file{gdb-@var{version-number}} source directory
36637 if you are not already in it; then run @file{configure}. Pass the
36638 identifier for the platform on which @value{GDBN} will run as an
36644 cd gdb-@value{GDBVN}
36649 Running @samp{configure} and then running @code{make} builds the
36650 included supporting libraries, then @code{gdb} itself. The configured
36651 source files, and the binaries, are left in the corresponding source
36655 @file{configure} is a Bourne-shell (@code{/bin/sh}) script; if your
36656 system does not recognize this automatically when you run a different
36657 shell, you may need to run @code{sh} on it explicitly:
36663 You should run the @file{configure} script from the top directory in the
36664 source tree, the @file{gdb-@var{version-number}} directory. If you run
36665 @file{configure} from one of the subdirectories, you will configure only
36666 that subdirectory. That is usually not what you want. In particular,
36667 if you run the first @file{configure} from the @file{gdb} subdirectory
36668 of the @file{gdb-@var{version-number}} directory, you will omit the
36669 configuration of @file{bfd}, @file{readline}, and other sibling
36670 directories of the @file{gdb} subdirectory. This leads to build errors
36671 about missing include files such as @file{bfd/bfd.h}.
36673 You can install @code{@value{GDBN}} anywhere. The best way to do this
36674 is to pass the @code{--prefix} option to @code{configure}, and then
36675 install it with @code{make install}.
36677 @node Separate Objdir
36678 @section Compiling @value{GDBN} in Another Directory
36680 If you want to run @value{GDBN} versions for several host or target machines,
36681 you need a different @code{gdb} compiled for each combination of
36682 host and target. @file{configure} is designed to make this easy by
36683 allowing you to generate each configuration in a separate subdirectory,
36684 rather than in the source directory. If your @code{make} program
36685 handles the @samp{VPATH} feature (@sc{gnu} @code{make} does), running
36686 @code{make} in each of these directories builds the @code{gdb}
36687 program specified there.
36689 To build @code{gdb} in a separate directory, run @file{configure}
36690 with the @samp{--srcdir} option to specify where to find the source.
36691 (You also need to specify a path to find @file{configure}
36692 itself from your working directory. If the path to @file{configure}
36693 would be the same as the argument to @samp{--srcdir}, you can leave out
36694 the @samp{--srcdir} option; it is assumed.)
36696 For example, with version @value{GDBVN}, you can build @value{GDBN} in a
36697 separate directory for a Sun 4 like this:
36701 cd gdb-@value{GDBVN}
36704 ../gdb-@value{GDBVN}/configure
36709 When @file{configure} builds a configuration using a remote source
36710 directory, it creates a tree for the binaries with the same structure
36711 (and using the same names) as the tree under the source directory. In
36712 the example, you'd find the Sun 4 library @file{libiberty.a} in the
36713 directory @file{gdb-sun4/libiberty}, and @value{GDBN} itself in
36714 @file{gdb-sun4/gdb}.
36716 Make sure that your path to the @file{configure} script has just one
36717 instance of @file{gdb} in it. If your path to @file{configure} looks
36718 like @file{../gdb-@value{GDBVN}/gdb/configure}, you are configuring only
36719 one subdirectory of @value{GDBN}, not the whole package. This leads to
36720 build errors about missing include files such as @file{bfd/bfd.h}.
36722 One popular reason to build several @value{GDBN} configurations in separate
36723 directories is to configure @value{GDBN} for cross-compiling (where
36724 @value{GDBN} runs on one machine---the @dfn{host}---while debugging
36725 programs that run on another machine---the @dfn{target}).
36726 You specify a cross-debugging target by
36727 giving the @samp{--target=@var{target}} option to @file{configure}.
36729 When you run @code{make} to build a program or library, you must run
36730 it in a configured directory---whatever directory you were in when you
36731 called @file{configure} (or one of its subdirectories).
36733 The @code{Makefile} that @file{configure} generates in each source
36734 directory also runs recursively. If you type @code{make} in a source
36735 directory such as @file{gdb-@value{GDBVN}} (or in a separate configured
36736 directory configured with @samp{--srcdir=@var{dirname}/gdb-@value{GDBVN}}), you
36737 will build all the required libraries, and then build GDB.
36739 When you have multiple hosts or targets configured in separate
36740 directories, you can run @code{make} on them in parallel (for example,
36741 if they are NFS-mounted on each of the hosts); they will not interfere
36745 @section Specifying Names for Hosts and Targets
36747 The specifications used for hosts and targets in the @file{configure}
36748 script are based on a three-part naming scheme, but some short predefined
36749 aliases are also supported. The full naming scheme encodes three pieces
36750 of information in the following pattern:
36753 @var{architecture}-@var{vendor}-@var{os}
36756 For example, you can use the alias @code{sun4} as a @var{host} argument,
36757 or as the value for @var{target} in a @code{--target=@var{target}}
36758 option. The equivalent full name is @samp{sparc-sun-sunos4}.
36760 The @file{configure} script accompanying @value{GDBN} does not provide
36761 any query facility to list all supported host and target names or
36762 aliases. @file{configure} calls the Bourne shell script
36763 @code{config.sub} to map abbreviations to full names; you can read the
36764 script, if you wish, or you can use it to test your guesses on
36765 abbreviations---for example:
36768 % sh config.sub i386-linux
36770 % sh config.sub alpha-linux
36771 alpha-unknown-linux-gnu
36772 % sh config.sub hp9k700
36774 % sh config.sub sun4
36775 sparc-sun-sunos4.1.1
36776 % sh config.sub sun3
36777 m68k-sun-sunos4.1.1
36778 % sh config.sub i986v
36779 Invalid configuration `i986v': machine `i986v' not recognized
36783 @code{config.sub} is also distributed in the @value{GDBN} source
36784 directory (@file{gdb-@value{GDBVN}}, for version @value{GDBVN}).
36786 @node Configure Options
36787 @section @file{configure} Options
36789 Here is a summary of the @file{configure} options and arguments that
36790 are most often useful for building @value{GDBN}. @file{configure}
36791 also has several other options not listed here. @inforef{Running
36792 configure scripts,,autoconf.info}, for a full
36793 explanation of @file{configure}.
36796 configure @r{[}--help@r{]}
36797 @r{[}--prefix=@var{dir}@r{]}
36798 @r{[}--exec-prefix=@var{dir}@r{]}
36799 @r{[}--srcdir=@var{dirname}@r{]}
36800 @r{[}--target=@var{target}@r{]}
36804 You may introduce options with a single @samp{-} rather than
36805 @samp{--} if you prefer; but you may abbreviate option names if you use
36810 Display a quick summary of how to invoke @file{configure}.
36812 @item --prefix=@var{dir}
36813 Configure the source to install programs and files under directory
36816 @item --exec-prefix=@var{dir}
36817 Configure the source to install programs under directory
36820 @c avoid splitting the warning from the explanation:
36822 @item --srcdir=@var{dirname}
36823 Use this option to make configurations in directories separate from the
36824 @value{GDBN} source directories. Among other things, you can use this to
36825 build (or maintain) several configurations simultaneously, in separate
36826 directories. @file{configure} writes configuration-specific files in
36827 the current directory, but arranges for them to use the source in the
36828 directory @var{dirname}. @file{configure} creates directories under
36829 the working directory in parallel to the source directories below
36832 @item --target=@var{target}
36833 Configure @value{GDBN} for cross-debugging programs running on the specified
36834 @var{target}. Without this option, @value{GDBN} is configured to debug
36835 programs that run on the same machine (@var{host}) as @value{GDBN} itself.
36837 There is no convenient way to generate a list of all available
36838 targets. Also see the @code{--enable-targets} option, below.
36841 There are many other options that are specific to @value{GDBN}. This
36842 lists just the most common ones; there are some very specialized
36843 options not described here.
36846 @item --enable-targets=@r{[}@var{target}@r{]}@dots{}
36847 @itemx --enable-targets=all
36848 Configure @value{GDBN} for cross-debugging programs running on the
36849 specified list of targets. The special value @samp{all} configures
36850 @value{GDBN} for debugging programs running on any target it supports.
36852 @item --with-gdb-datadir=@var{path}
36853 Set the @value{GDBN}-specific data directory. @value{GDBN} will look
36854 here for certain supporting files or scripts. This defaults to the
36855 @file{gdb} subdirectory of @samp{datadi} (which can be set using
36858 @item --with-relocated-sources=@var{dir}
36859 Sets up the default source path substitution rule so that directory
36860 names recorded in debug information will be automatically adjusted for
36861 any directory under @var{dir}. @var{dir} should be a subdirectory of
36862 @value{GDBN}'s configured prefix, the one mentioned in the
36863 @code{--prefix} or @code{--exec-prefix} options to configure. This
36864 option is useful if GDB is supposed to be moved to a different place
36867 @item --enable-64-bit-bfd
36868 Enable 64-bit support in BFD on 32-bit hosts.
36870 @item --disable-gdbmi
36871 Build @value{GDBN} without the GDB/MI machine interface
36875 Build @value{GDBN} with the text-mode full-screen user interface
36876 (TUI). Requires a curses library (ncurses and cursesX are also
36879 @item --with-curses
36880 Use the curses library instead of the termcap library, for text-mode
36881 terminal operations.
36883 @item --with-libunwind-ia64
36884 Use the libunwind library for unwinding function call stack on ia64
36885 target platforms. See http://www.nongnu.org/libunwind/index.html for
36888 @item --with-system-readline
36889 Use the readline library installed on the host, rather than the
36890 library supplied as part of @value{GDBN}.
36892 @item --with-system-zlib
36893 Use the zlib library installed on the host, rather than the library
36894 supplied as part of @value{GDBN}.
36897 Build @value{GDBN} with Expat, a library for XML parsing. (Done by
36898 default if libexpat is installed and found at configure time.) This
36899 library is used to read XML files supplied with @value{GDBN}. If it
36900 is unavailable, some features, such as remote protocol memory maps,
36901 target descriptions, and shared library lists, that are based on XML
36902 files, will not be available in @value{GDBN}. If your host does not
36903 have libexpat installed, you can get the latest version from
36904 `http://expat.sourceforge.net'.
36906 @item --with-libiconv-prefix@r{[}=@var{dir}@r{]}
36908 Build @value{GDBN} with GNU libiconv, a character set encoding
36909 conversion library. This is not done by default, as on GNU systems
36910 the @code{iconv} that is built in to the C library is sufficient. If
36911 your host does not have a working @code{iconv}, you can get the latest
36912 version of GNU iconv from `https://www.gnu.org/software/libiconv/'.
36914 @value{GDBN}'s build system also supports building GNU libiconv as
36915 part of the overall build. @xref{Requirements}.
36918 Build @value{GDBN} with LZMA, a compression library. (Done by default
36919 if liblzma is installed and found at configure time.) LZMA is used by
36920 @value{GDBN}'s "mini debuginfo" feature, which is only useful on
36921 platforms using the ELF object file format. If your host does not
36922 have liblzma installed, you can get the latest version from
36923 `https://tukaani.org/xz/'.
36926 Build @value{GDBN} with GNU MPFR, a library for multiple-precision
36927 floating-point computation with correct rounding. (Done by default if
36928 GNU MPFR is installed and found at configure time.) This library is
36929 used to emulate target floating-point arithmetic during expression
36930 evaluation when the target uses different floating-point formats than
36931 the host. If GNU MPFR is not available, @value{GDBN} will fall back
36932 to using host floating-point arithmetic. If your host does not have
36933 GNU MPFR installed, you can get the latest version from
36934 `http://www.mpfr.org'.
36936 @item --with-python@r{[}=@var{python}@r{]}
36937 Build @value{GDBN} with Python scripting support. (Done by default if
36938 libpython is present and found at configure time.) Python makes
36939 @value{GDBN} scripting much more powerful than the restricted CLI
36940 scripting language. If your host does not have Python installed, you
36941 can find it on `http://www.python.org/download/'. The oldest version
36942 of Python supported by GDB is 2.6. The optional argument @var{python}
36943 is used to find the Python headers and libraries. It can be either
36944 the name of a Python executable, or the name of the directory in which
36945 Python is installed.
36947 @item --with-guile[=GUILE]'
36948 Build @value{GDBN} with GNU Guile scripting support. (Done by default
36949 if libguile is present and found at configure time.) If your host
36950 does not have Guile installed, you can find it at
36951 `https://www.gnu.org/software/guile/'. The optional argument GUILE
36952 can be a version number, which will cause @code{configure} to try to
36953 use that version of Guile; or the file name of a @code{pkg-config}
36954 executable, which will be queried to find the information needed to
36955 compile and link against Guile.
36957 @item --without-included-regex
36958 Don't use the regex library included with @value{GDBN} (as part of the
36959 libiberty library). This is the default on hosts with version 2 of
36962 @item --with-sysroot=@var{dir}
36963 Use @var{dir} as the default system root directory for libraries whose
36964 file names begin with @file{/lib}' or @file{/usr/lib'}. (The value of
36965 @var{dir} can be modified at run time by using the @command{set
36966 sysroot} command.) If @var{dir} is under the @value{GDBN} configured
36967 prefix (set with @code{--prefix} or @code{--exec-prefix options}, the
36968 default system root will be automatically adjusted if and when
36969 @value{GDBN} is moved to a different location.
36971 @item --with-system-gdbinit=@var{file}
36972 Configure @value{GDBN} to automatically load a system-wide init file.
36973 @var{file} should be an absolute file name. If @var{file} is in a
36974 directory under the configured prefix, and @value{GDBN} is moved to
36975 another location after being built, the location of the system-wide
36976 init file will be adjusted accordingly.
36978 @item --enable-build-warnings
36979 When building the @value{GDBN} sources, ask the compiler to warn about
36980 any code which looks even vaguely suspicious. It passes many
36981 different warning flags, depending on the exact version of the
36982 compiler you are using.
36984 @item --enable-werror
36985 Treat compiler warnings as werrors. It adds the @code{-Werror} flag
36986 to the compiler, which will fail the compilation if the compiler
36987 outputs any warning messages.
36989 @item --enable-ubsan
36990 Enable the GCC undefined behavior sanitizer. This is disabled by
36991 default, but passing @code{--enable-ubsan=yes} or
36992 @code{--enable-ubsan=auto} to @code{configure} will enable it. The
36993 undefined behavior sanitizer checks for C@t{++} undefined behavior.
36994 It has a performance cost, so if you are looking at @value{GDBN}'s
36995 performance, you should disable it. The undefined behavior sanitizer
36996 was first introduced in GCC 4.9.
36999 @node System-wide configuration
37000 @section System-wide configuration and settings
37001 @cindex system-wide init file
37003 @value{GDBN} can be configured to have a system-wide init file;
37004 this file will be read and executed at startup (@pxref{Startup, , What
37005 @value{GDBN} does during startup}).
37007 Here is the corresponding configure option:
37010 @item --with-system-gdbinit=@var{file}
37011 Specify that the default location of the system-wide init file is
37015 If @value{GDBN} has been configured with the option @option{--prefix=$prefix},
37016 it may be subject to relocation. Two possible cases:
37020 If the default location of this init file contains @file{$prefix},
37021 it will be subject to relocation. Suppose that the configure options
37022 are @option{--prefix=$prefix --with-system-gdbinit=$prefix/etc/gdbinit};
37023 if @value{GDBN} is moved from @file{$prefix} to @file{$install}, the system
37024 init file is looked for as @file{$install/etc/gdbinit} instead of
37025 @file{$prefix/etc/gdbinit}.
37028 By contrast, if the default location does not contain the prefix,
37029 it will not be relocated. E.g.@: if @value{GDBN} has been configured with
37030 @option{--prefix=/usr/local --with-system-gdbinit=/usr/share/gdb/gdbinit},
37031 then @value{GDBN} will always look for @file{/usr/share/gdb/gdbinit},
37032 wherever @value{GDBN} is installed.
37035 If the configured location of the system-wide init file (as given by the
37036 @option{--with-system-gdbinit} option at configure time) is in the
37037 data-directory (as specified by @option{--with-gdb-datadir} at configure
37038 time) or in one of its subdirectories, then @value{GDBN} will look for the
37039 system-wide init file in the directory specified by the
37040 @option{--data-directory} command-line option.
37041 Note that the system-wide init file is only read once, during @value{GDBN}
37042 initialization. If the data-directory is changed after @value{GDBN} has
37043 started with the @code{set data-directory} command, the file will not be
37047 * System-wide Configuration Scripts:: Installed System-wide Configuration Scripts
37050 @node System-wide Configuration Scripts
37051 @subsection Installed System-wide Configuration Scripts
37052 @cindex system-wide configuration scripts
37054 The @file{system-gdbinit} directory, located inside the data-directory
37055 (as specified by @option{--with-gdb-datadir} at configure time) contains
37056 a number of scripts which can be used as system-wide init files. To
37057 automatically source those scripts at startup, @value{GDBN} should be
37058 configured with @option{--with-system-gdbinit}. Otherwise, any user
37059 should be able to source them by hand as needed.
37061 The following scripts are currently available:
37064 @item @file{elinos.py}
37066 @cindex ELinOS system-wide configuration script
37067 This script is useful when debugging a program on an ELinOS target.
37068 It takes advantage of the environment variables defined in a standard
37069 ELinOS environment in order to determine the location of the system
37070 shared libraries, and then sets the @samp{solib-absolute-prefix}
37071 and @samp{solib-search-path} variables appropriately.
37073 @item @file{wrs-linux.py}
37074 @pindex wrs-linux.py
37075 @cindex Wind River Linux system-wide configuration script
37076 This script is useful when debugging a program on a target running
37077 Wind River Linux. It expects the @env{ENV_PREFIX} to be set to
37078 the host-side sysroot used by the target system.
37082 @node Maintenance Commands
37083 @appendix Maintenance Commands
37084 @cindex maintenance commands
37085 @cindex internal commands
37087 In addition to commands intended for @value{GDBN} users, @value{GDBN}
37088 includes a number of commands intended for @value{GDBN} developers,
37089 that are not documented elsewhere in this manual. These commands are
37090 provided here for reference. (For commands that turn on debugging
37091 messages, see @ref{Debugging Output}.)
37094 @kindex maint agent
37095 @kindex maint agent-eval
37096 @item maint agent @r{[}-at @var{location}@r{,}@r{]} @var{expression}
37097 @itemx maint agent-eval @r{[}-at @var{location}@r{,}@r{]} @var{expression}
37098 Translate the given @var{expression} into remote agent bytecodes.
37099 This command is useful for debugging the Agent Expression mechanism
37100 (@pxref{Agent Expressions}). The @samp{agent} version produces an
37101 expression useful for data collection, such as by tracepoints, while
37102 @samp{maint agent-eval} produces an expression that evaluates directly
37103 to a result. For instance, a collection expression for @code{globa +
37104 globb} will include bytecodes to record four bytes of memory at each
37105 of the addresses of @code{globa} and @code{globb}, while discarding
37106 the result of the addition, while an evaluation expression will do the
37107 addition and return the sum.
37108 If @code{-at} is given, generate remote agent bytecode for @var{location}.
37109 If not, generate remote agent bytecode for current frame PC address.
37111 @kindex maint agent-printf
37112 @item maint agent-printf @var{format},@var{expr},...
37113 Translate the given format string and list of argument expressions
37114 into remote agent bytecodes and display them as a disassembled list.
37115 This command is useful for debugging the agent version of dynamic
37116 printf (@pxref{Dynamic Printf}).
37118 @kindex maint info breakpoints
37119 @item @anchor{maint info breakpoints}maint info breakpoints
37120 Using the same format as @samp{info breakpoints}, display both the
37121 breakpoints you've set explicitly, and those @value{GDBN} is using for
37122 internal purposes. Internal breakpoints are shown with negative
37123 breakpoint numbers. The type column identifies what kind of breakpoint
37128 Normal, explicitly set breakpoint.
37131 Normal, explicitly set watchpoint.
37134 Internal breakpoint, used to handle correctly stepping through
37135 @code{longjmp} calls.
37137 @item longjmp resume
37138 Internal breakpoint at the target of a @code{longjmp}.
37141 Temporary internal breakpoint used by the @value{GDBN} @code{until} command.
37144 Temporary internal breakpoint used by the @value{GDBN} @code{finish} command.
37147 Shared library events.
37151 @kindex maint info btrace
37152 @item maint info btrace
37153 Pint information about raw branch tracing data.
37155 @kindex maint btrace packet-history
37156 @item maint btrace packet-history
37157 Print the raw branch trace packets that are used to compute the
37158 execution history for the @samp{record btrace} command. Both the
37159 information and the format in which it is printed depend on the btrace
37164 For the BTS recording format, print a list of blocks of sequential
37165 code. For each block, the following information is printed:
37169 Newer blocks have higher numbers. The oldest block has number zero.
37170 @item Lowest @samp{PC}
37171 @item Highest @samp{PC}
37175 For the Intel Processor Trace recording format, print a list of
37176 Intel Processor Trace packets. For each packet, the following
37177 information is printed:
37180 @item Packet number
37181 Newer packets have higher numbers. The oldest packet has number zero.
37183 The packet's offset in the trace stream.
37184 @item Packet opcode and payload
37188 @kindex maint btrace clear-packet-history
37189 @item maint btrace clear-packet-history
37190 Discards the cached packet history printed by the @samp{maint btrace
37191 packet-history} command. The history will be computed again when
37194 @kindex maint btrace clear
37195 @item maint btrace clear
37196 Discard the branch trace data. The data will be fetched anew and the
37197 branch trace will be recomputed when needed.
37199 This implicitly truncates the branch trace to a single branch trace
37200 buffer. When updating branch trace incrementally, the branch trace
37201 available to @value{GDBN} may be bigger than a single branch trace
37204 @kindex maint set btrace pt skip-pad
37205 @item maint set btrace pt skip-pad
37206 @kindex maint show btrace pt skip-pad
37207 @item maint show btrace pt skip-pad
37208 Control whether @value{GDBN} will skip PAD packets when computing the
37211 @kindex set displaced-stepping
37212 @kindex show displaced-stepping
37213 @cindex displaced stepping support
37214 @cindex out-of-line single-stepping
37215 @item set displaced-stepping
37216 @itemx show displaced-stepping
37217 Control whether or not @value{GDBN} will do @dfn{displaced stepping}
37218 if the target supports it. Displaced stepping is a way to single-step
37219 over breakpoints without removing them from the inferior, by executing
37220 an out-of-line copy of the instruction that was originally at the
37221 breakpoint location. It is also known as out-of-line single-stepping.
37224 @item set displaced-stepping on
37225 If the target architecture supports it, @value{GDBN} will use
37226 displaced stepping to step over breakpoints.
37228 @item set displaced-stepping off
37229 @value{GDBN} will not use displaced stepping to step over breakpoints,
37230 even if such is supported by the target architecture.
37232 @cindex non-stop mode, and @samp{set displaced-stepping}
37233 @item set displaced-stepping auto
37234 This is the default mode. @value{GDBN} will use displaced stepping
37235 only if non-stop mode is active (@pxref{Non-Stop Mode}) and the target
37236 architecture supports displaced stepping.
37239 @kindex maint check-psymtabs
37240 @item maint check-psymtabs
37241 Check the consistency of currently expanded psymtabs versus symtabs.
37242 Use this to check, for example, whether a symbol is in one but not the other.
37244 @kindex maint check-symtabs
37245 @item maint check-symtabs
37246 Check the consistency of currently expanded symtabs.
37248 @kindex maint expand-symtabs
37249 @item maint expand-symtabs [@var{regexp}]
37250 Expand symbol tables.
37251 If @var{regexp} is specified, only expand symbol tables for file
37252 names matching @var{regexp}.
37254 @kindex maint set catch-demangler-crashes
37255 @kindex maint show catch-demangler-crashes
37256 @cindex demangler crashes
37257 @item maint set catch-demangler-crashes [on|off]
37258 @itemx maint show catch-demangler-crashes
37259 Control whether @value{GDBN} should attempt to catch crashes in the
37260 symbol name demangler. The default is to attempt to catch crashes.
37261 If enabled, the first time a crash is caught, a core file is created,
37262 the offending symbol is displayed and the user is presented with the
37263 option to terminate the current session.
37265 @kindex maint cplus first_component
37266 @item maint cplus first_component @var{name}
37267 Print the first C@t{++} class/namespace component of @var{name}.
37269 @kindex maint cplus namespace
37270 @item maint cplus namespace
37271 Print the list of possible C@t{++} namespaces.
37273 @kindex maint deprecate
37274 @kindex maint undeprecate
37275 @cindex deprecated commands
37276 @item maint deprecate @var{command} @r{[}@var{replacement}@r{]}
37277 @itemx maint undeprecate @var{command}
37278 Deprecate or undeprecate the named @var{command}. Deprecated commands
37279 cause @value{GDBN} to issue a warning when you use them. The optional
37280 argument @var{replacement} says which newer command should be used in
37281 favor of the deprecated one; if it is given, @value{GDBN} will mention
37282 the replacement as part of the warning.
37284 @kindex maint dump-me
37285 @item maint dump-me
37286 @cindex @code{SIGQUIT} signal, dump core of @value{GDBN}
37287 Cause a fatal signal in the debugger and force it to dump its core.
37288 This is supported only on systems which support aborting a program
37289 with the @code{SIGQUIT} signal.
37291 @kindex maint internal-error
37292 @kindex maint internal-warning
37293 @kindex maint demangler-warning
37294 @cindex demangler crashes
37295 @item maint internal-error @r{[}@var{message-text}@r{]}
37296 @itemx maint internal-warning @r{[}@var{message-text}@r{]}
37297 @itemx maint demangler-warning @r{[}@var{message-text}@r{]}
37299 Cause @value{GDBN} to call the internal function @code{internal_error},
37300 @code{internal_warning} or @code{demangler_warning} and hence behave
37301 as though an internal problem has been detected. In addition to
37302 reporting the internal problem, these functions give the user the
37303 opportunity to either quit @value{GDBN} or (for @code{internal_error}
37304 and @code{internal_warning}) create a core file of the current
37305 @value{GDBN} session.
37307 These commands take an optional parameter @var{message-text} that is
37308 used as the text of the error or warning message.
37310 Here's an example of using @code{internal-error}:
37313 (@value{GDBP}) @kbd{maint internal-error testing, 1, 2}
37314 @dots{}/maint.c:121: internal-error: testing, 1, 2
37315 A problem internal to GDB has been detected. Further
37316 debugging may prove unreliable.
37317 Quit this debugging session? (y or n) @kbd{n}
37318 Create a core file? (y or n) @kbd{n}
37322 @cindex @value{GDBN} internal error
37323 @cindex internal errors, control of @value{GDBN} behavior
37324 @cindex demangler crashes
37326 @kindex maint set internal-error
37327 @kindex maint show internal-error
37328 @kindex maint set internal-warning
37329 @kindex maint show internal-warning
37330 @kindex maint set demangler-warning
37331 @kindex maint show demangler-warning
37332 @item maint set internal-error @var{action} [ask|yes|no]
37333 @itemx maint show internal-error @var{action}
37334 @itemx maint set internal-warning @var{action} [ask|yes|no]
37335 @itemx maint show internal-warning @var{action}
37336 @itemx maint set demangler-warning @var{action} [ask|yes|no]
37337 @itemx maint show demangler-warning @var{action}
37338 When @value{GDBN} reports an internal problem (error or warning) it
37339 gives the user the opportunity to both quit @value{GDBN} and create a
37340 core file of the current @value{GDBN} session. These commands let you
37341 override the default behaviour for each particular @var{action},
37342 described in the table below.
37346 You can specify that @value{GDBN} should always (yes) or never (no)
37347 quit. The default is to ask the user what to do.
37350 You can specify that @value{GDBN} should always (yes) or never (no)
37351 create a core file. The default is to ask the user what to do. Note
37352 that there is no @code{corefile} option for @code{demangler-warning}:
37353 demangler warnings always create a core file and this cannot be
37357 @kindex maint packet
37358 @item maint packet @var{text}
37359 If @value{GDBN} is talking to an inferior via the serial protocol,
37360 then this command sends the string @var{text} to the inferior, and
37361 displays the response packet. @value{GDBN} supplies the initial
37362 @samp{$} character, the terminating @samp{#} character, and the
37365 @kindex maint print architecture
37366 @item maint print architecture @r{[}@var{file}@r{]}
37367 Print the entire architecture configuration. The optional argument
37368 @var{file} names the file where the output goes.
37370 @kindex maint print c-tdesc @r{[}@var{file}@r{]}
37371 @item maint print c-tdesc
37372 Print the target description (@pxref{Target Descriptions}) as
37373 a C source file. By default, the target description is for the current
37374 target, but if the optional argument @var{file} is provided, that file
37375 is used to produce the description. The @var{file} should be an XML
37376 document, of the form described in @ref{Target Description Format}.
37377 The created source file is built into @value{GDBN} when @value{GDBN} is
37378 built again. This command is used by developers after they add or
37379 modify XML target descriptions.
37381 @kindex maint check xml-descriptions
37382 @item maint check xml-descriptions @var{dir}
37383 Check that the target descriptions dynamically created by @value{GDBN}
37384 equal the descriptions created from XML files found in @var{dir}.
37386 @anchor{maint check libthread-db}
37387 @kindex maint check libthread-db
37388 @item maint check libthread-db
37389 Run integrity checks on the current inferior's thread debugging
37390 library. This exercises all @code{libthread_db} functionality used by
37391 @value{GDBN} on GNU/Linux systems, and by extension also exercises the
37392 @code{proc_service} functions provided by @value{GDBN} that
37393 @code{libthread_db} uses. Note that parts of the test may be skipped
37394 on some platforms when debugging core files.
37396 @kindex maint print dummy-frames
37397 @item maint print dummy-frames
37398 Prints the contents of @value{GDBN}'s internal dummy-frame stack.
37401 (@value{GDBP}) @kbd{b add}
37403 (@value{GDBP}) @kbd{print add(2,3)}
37404 Breakpoint 2, add (a=2, b=3) at @dots{}
37406 The program being debugged stopped while in a function called from GDB.
37408 (@value{GDBP}) @kbd{maint print dummy-frames}
37409 0xa8206d8: id=@{stack=0xbfffe734,code=0xbfffe73f,!special@}, ptid=process 9353
37413 Takes an optional file parameter.
37415 @kindex maint print registers
37416 @kindex maint print raw-registers
37417 @kindex maint print cooked-registers
37418 @kindex maint print register-groups
37419 @kindex maint print remote-registers
37420 @item maint print registers @r{[}@var{file}@r{]}
37421 @itemx maint print raw-registers @r{[}@var{file}@r{]}
37422 @itemx maint print cooked-registers @r{[}@var{file}@r{]}
37423 @itemx maint print register-groups @r{[}@var{file}@r{]}
37424 @itemx maint print remote-registers @r{[}@var{file}@r{]}
37425 Print @value{GDBN}'s internal register data structures.
37427 The command @code{maint print raw-registers} includes the contents of
37428 the raw register cache; the command @code{maint print
37429 cooked-registers} includes the (cooked) value of all registers,
37430 including registers which aren't available on the target nor visible
37431 to user; the command @code{maint print register-groups} includes the
37432 groups that each register is a member of; and the command @code{maint
37433 print remote-registers} includes the remote target's register numbers
37434 and offsets in the `G' packets.
37436 These commands take an optional parameter, a file name to which to
37437 write the information.
37439 @kindex maint print reggroups
37440 @item maint print reggroups @r{[}@var{file}@r{]}
37441 Print @value{GDBN}'s internal register group data structures. The
37442 optional argument @var{file} tells to what file to write the
37445 The register groups info looks like this:
37448 (@value{GDBP}) @kbd{maint print reggroups}
37461 This command forces @value{GDBN} to flush its internal register cache.
37463 @kindex maint print objfiles
37464 @cindex info for known object files
37465 @item maint print objfiles @r{[}@var{regexp}@r{]}
37466 Print a dump of all known object files.
37467 If @var{regexp} is specified, only print object files whose names
37468 match @var{regexp}. For each object file, this command prints its name,
37469 address in memory, and all of its psymtabs and symtabs.
37471 @kindex maint print user-registers
37472 @cindex user registers
37473 @item maint print user-registers
37474 List all currently available @dfn{user registers}. User registers
37475 typically provide alternate names for actual hardware registers. They
37476 include the four ``standard'' registers @code{$fp}, @code{$pc},
37477 @code{$sp}, and @code{$ps}. @xref{standard registers}. User
37478 registers can be used in expressions in the same way as the canonical
37479 register names, but only the latter are listed by the @code{info
37480 registers} and @code{maint print registers} commands.
37482 @kindex maint print section-scripts
37483 @cindex info for known .debug_gdb_scripts-loaded scripts
37484 @item maint print section-scripts [@var{regexp}]
37485 Print a dump of scripts specified in the @code{.debug_gdb_section} section.
37486 If @var{regexp} is specified, only print scripts loaded by object files
37487 matching @var{regexp}.
37488 For each script, this command prints its name as specified in the objfile,
37489 and the full path if known.
37490 @xref{dotdebug_gdb_scripts section}.
37492 @kindex maint print statistics
37493 @cindex bcache statistics
37494 @item maint print statistics
37495 This command prints, for each object file in the program, various data
37496 about that object file followed by the byte cache (@dfn{bcache})
37497 statistics for the object file. The objfile data includes the number
37498 of minimal, partial, full, and stabs symbols, the number of types
37499 defined by the objfile, the number of as yet unexpanded psym tables,
37500 the number of line tables and string tables, and the amount of memory
37501 used by the various tables. The bcache statistics include the counts,
37502 sizes, and counts of duplicates of all and unique objects, max,
37503 average, and median entry size, total memory used and its overhead and
37504 savings, and various measures of the hash table size and chain
37507 @kindex maint print target-stack
37508 @cindex target stack description
37509 @item maint print target-stack
37510 A @dfn{target} is an interface between the debugger and a particular
37511 kind of file or process. Targets can be stacked in @dfn{strata},
37512 so that more than one target can potentially respond to a request.
37513 In particular, memory accesses will walk down the stack of targets
37514 until they find a target that is interested in handling that particular
37517 This command prints a short description of each layer that was pushed on
37518 the @dfn{target stack}, starting from the top layer down to the bottom one.
37520 @kindex maint print type
37521 @cindex type chain of a data type
37522 @item maint print type @var{expr}
37523 Print the type chain for a type specified by @var{expr}. The argument
37524 can be either a type name or a symbol. If it is a symbol, the type of
37525 that symbol is described. The type chain produced by this command is
37526 a recursive definition of the data type as stored in @value{GDBN}'s
37527 data structures, including its flags and contained types.
37529 @kindex maint selftest
37531 @item maint selftest @r{[}@var{filter}@r{]}
37532 Run any self tests that were compiled in to @value{GDBN}. This will
37533 print a message showing how many tests were run, and how many failed.
37534 If a @var{filter} is passed, only the tests with @var{filter} in their
37537 @kindex maint info selftests
37539 @item maint info selftests
37540 List the selftests compiled in to @value{GDBN}.
37542 @kindex maint set dwarf always-disassemble
37543 @kindex maint show dwarf always-disassemble
37544 @item maint set dwarf always-disassemble
37545 @item maint show dwarf always-disassemble
37546 Control the behavior of @code{info address} when using DWARF debugging
37549 The default is @code{off}, which means that @value{GDBN} should try to
37550 describe a variable's location in an easily readable format. When
37551 @code{on}, @value{GDBN} will instead display the DWARF location
37552 expression in an assembly-like format. Note that some locations are
37553 too complex for @value{GDBN} to describe simply; in this case you will
37554 always see the disassembly form.
37556 Here is an example of the resulting disassembly:
37559 (gdb) info addr argc
37560 Symbol "argc" is a complex DWARF expression:
37564 For more information on these expressions, see
37565 @uref{http://www.dwarfstd.org/, the DWARF standard}.
37567 @kindex maint set dwarf max-cache-age
37568 @kindex maint show dwarf max-cache-age
37569 @item maint set dwarf max-cache-age
37570 @itemx maint show dwarf max-cache-age
37571 Control the DWARF compilation unit cache.
37573 @cindex DWARF compilation units cache
37574 In object files with inter-compilation-unit references, such as those
37575 produced by the GCC option @samp{-feliminate-dwarf2-dups}, the DWARF
37576 reader needs to frequently refer to previously read compilation units.
37577 This setting controls how long a compilation unit will remain in the
37578 cache if it is not referenced. A higher limit means that cached
37579 compilation units will be stored in memory longer, and more total
37580 memory will be used. Setting it to zero disables caching, which will
37581 slow down @value{GDBN} startup, but reduce memory consumption.
37583 @kindex maint set dwarf unwinders
37584 @kindex maint show dwarf unwinders
37585 @item maint set dwarf unwinders
37586 @itemx maint show dwarf unwinders
37587 Control use of the DWARF frame unwinders.
37589 @cindex DWARF frame unwinders
37590 Many targets that support DWARF debugging use @value{GDBN}'s DWARF
37591 frame unwinders to build the backtrace. Many of these targets will
37592 also have a second mechanism for building the backtrace for use in
37593 cases where DWARF information is not available, this second mechanism
37594 is often an analysis of a function's prologue.
37596 In order to extend testing coverage of the second level stack
37597 unwinding mechanisms it is helpful to be able to disable the DWARF
37598 stack unwinders, this can be done with this switch.
37600 In normal use of @value{GDBN} disabling the DWARF unwinders is not
37601 advisable, there are cases that are better handled through DWARF than
37602 prologue analysis, and the debug experience is likely to be better
37603 with the DWARF frame unwinders enabled.
37605 If DWARF frame unwinders are not supported for a particular target
37606 architecture, then enabling this flag does not cause them to be used.
37607 @kindex maint set profile
37608 @kindex maint show profile
37609 @cindex profiling GDB
37610 @item maint set profile
37611 @itemx maint show profile
37612 Control profiling of @value{GDBN}.
37614 Profiling will be disabled until you use the @samp{maint set profile}
37615 command to enable it. When you enable profiling, the system will begin
37616 collecting timing and execution count data; when you disable profiling or
37617 exit @value{GDBN}, the results will be written to a log file. Remember that
37618 if you use profiling, @value{GDBN} will overwrite the profiling log file
37619 (often called @file{gmon.out}). If you have a record of important profiling
37620 data in a @file{gmon.out} file, be sure to move it to a safe location.
37622 Configuring with @samp{--enable-profiling} arranges for @value{GDBN} to be
37623 compiled with the @samp{-pg} compiler option.
37625 @kindex maint set show-debug-regs
37626 @kindex maint show show-debug-regs
37627 @cindex hardware debug registers
37628 @item maint set show-debug-regs
37629 @itemx maint show show-debug-regs
37630 Control whether to show variables that mirror the hardware debug
37631 registers. Use @code{on} to enable, @code{off} to disable. If
37632 enabled, the debug registers values are shown when @value{GDBN} inserts or
37633 removes a hardware breakpoint or watchpoint, and when the inferior
37634 triggers a hardware-assisted breakpoint or watchpoint.
37636 @kindex maint set show-all-tib
37637 @kindex maint show show-all-tib
37638 @item maint set show-all-tib
37639 @itemx maint show show-all-tib
37640 Control whether to show all non zero areas within a 1k block starting
37641 at thread local base, when using the @samp{info w32 thread-information-block}
37644 @kindex maint set target-async
37645 @kindex maint show target-async
37646 @item maint set target-async
37647 @itemx maint show target-async
37648 This controls whether @value{GDBN} targets operate in synchronous or
37649 asynchronous mode (@pxref{Background Execution}). Normally the
37650 default is asynchronous, if it is available; but this can be changed
37651 to more easily debug problems occurring only in synchronous mode.
37653 @kindex maint set target-non-stop @var{mode} [on|off|auto]
37654 @kindex maint show target-non-stop
37655 @item maint set target-non-stop
37656 @itemx maint show target-non-stop
37658 This controls whether @value{GDBN} targets always operate in non-stop
37659 mode even if @code{set non-stop} is @code{off} (@pxref{Non-Stop
37660 Mode}). The default is @code{auto}, meaning non-stop mode is enabled
37661 if supported by the target.
37664 @item maint set target-non-stop auto
37665 This is the default mode. @value{GDBN} controls the target in
37666 non-stop mode if the target supports it.
37668 @item maint set target-non-stop on
37669 @value{GDBN} controls the target in non-stop mode even if the target
37670 does not indicate support.
37672 @item maint set target-non-stop off
37673 @value{GDBN} does not control the target in non-stop mode even if the
37674 target supports it.
37677 @kindex maint set per-command
37678 @kindex maint show per-command
37679 @item maint set per-command
37680 @itemx maint show per-command
37681 @cindex resources used by commands
37683 @value{GDBN} can display the resources used by each command.
37684 This is useful in debugging performance problems.
37687 @item maint set per-command space [on|off]
37688 @itemx maint show per-command space
37689 Enable or disable the printing of the memory used by GDB for each command.
37690 If enabled, @value{GDBN} will display how much memory each command
37691 took, following the command's own output.
37692 This can also be requested by invoking @value{GDBN} with the
37693 @option{--statistics} command-line switch (@pxref{Mode Options}).
37695 @item maint set per-command time [on|off]
37696 @itemx maint show per-command time
37697 Enable or disable the printing of the execution time of @value{GDBN}
37699 If enabled, @value{GDBN} will display how much time it
37700 took to execute each command, following the command's own output.
37701 Both CPU time and wallclock time are printed.
37702 Printing both is useful when trying to determine whether the cost is
37703 CPU or, e.g., disk/network latency.
37704 Note that the CPU time printed is for @value{GDBN} only, it does not include
37705 the execution time of the inferior because there's no mechanism currently
37706 to compute how much time was spent by @value{GDBN} and how much time was
37707 spent by the program been debugged.
37708 This can also be requested by invoking @value{GDBN} with the
37709 @option{--statistics} command-line switch (@pxref{Mode Options}).
37711 @item maint set per-command symtab [on|off]
37712 @itemx maint show per-command symtab
37713 Enable or disable the printing of basic symbol table statistics
37715 If enabled, @value{GDBN} will display the following information:
37719 number of symbol tables
37721 number of primary symbol tables
37723 number of blocks in the blockvector
37727 @kindex maint set check-libthread-db
37728 @kindex maint show check-libthread-db
37729 @item maint set check-libthread-db [on|off]
37730 @itemx maint show check-libthread-db
37731 Control whether @value{GDBN} should run integrity checks on inferior
37732 specific thread debugging libraries as they are loaded. The default
37733 is not to perform such checks. If any check fails @value{GDBN} will
37734 unload the library and continue searching for a suitable candidate as
37735 described in @ref{set libthread-db-search-path}. For more information
37736 about the tests, see @ref{maint check libthread-db}.
37738 @kindex maint space
37739 @cindex memory used by commands
37740 @item maint space @var{value}
37741 An alias for @code{maint set per-command space}.
37742 A non-zero value enables it, zero disables it.
37745 @cindex time of command execution
37746 @item maint time @var{value}
37747 An alias for @code{maint set per-command time}.
37748 A non-zero value enables it, zero disables it.
37750 @kindex maint translate-address
37751 @item maint translate-address @r{[}@var{section}@r{]} @var{addr}
37752 Find the symbol stored at the location specified by the address
37753 @var{addr} and an optional section name @var{section}. If found,
37754 @value{GDBN} prints the name of the closest symbol and an offset from
37755 the symbol's location to the specified address. This is similar to
37756 the @code{info address} command (@pxref{Symbols}), except that this
37757 command also allows to find symbols in other sections.
37759 If section was not specified, the section in which the symbol was found
37760 is also printed. For dynamically linked executables, the name of
37761 executable or shared library containing the symbol is printed as well.
37763 @kindex maint test-options
37764 @item maint test-options require-delimiter
37765 @itemx maint test-options unknown-is-error
37766 @itemx maint test-options unknown-is-operand
37767 These commands are used by the testsuite to validate the command
37768 options framework. The @code{require-delimiter} variant requires a
37769 double-dash delimiter to indicate end of options. The
37770 @code{unknown-is-error} and @code{unknown-is-operand} do not. The
37771 @code{unknown-is-error} variant throws an error on unknown option,
37772 while @code{unknown-is-operand} treats unknown options as the start of
37773 the command's operands. When run, the commands output the result of
37774 the processed options. When completed, the commands store the
37775 internal result of completion in a variable exposed by the @code{maint
37776 show test-options-completion-result} command.
37778 @kindex maint show test-options-completion-result
37779 @item maint show test-options-completion-result
37780 Shows the result of completing the @code{maint test-options}
37781 subcommands. This is used by the testsuite to validate completion
37782 support in the command options framework.
37784 @kindex maint set test-settings
37785 @kindex maint show test-settings
37786 @item maint set test-settings @var{kind}
37787 @itemx maint show test-settings @var{kind}
37788 These are representative commands for each @var{kind} of setting type
37789 @value{GDBN} supports. They are used by the testsuite for exercising
37790 the settings infrastructure.
37793 @item maint with @var{setting} [@var{value}] [-- @var{command}]
37794 Like the @code{with} command, but works with @code{maintenance set}
37795 variables. This is used by the testsuite to exercise the @code{with}
37796 command's infrastructure.
37800 The following command is useful for non-interactive invocations of
37801 @value{GDBN}, such as in the test suite.
37804 @item set watchdog @var{nsec}
37805 @kindex set watchdog
37806 @cindex watchdog timer
37807 @cindex timeout for commands
37808 Set the maximum number of seconds @value{GDBN} will wait for the
37809 target operation to finish. If this time expires, @value{GDBN}
37810 reports and error and the command is aborted.
37812 @item show watchdog
37813 Show the current setting of the target wait timeout.
37816 @node Remote Protocol
37817 @appendix @value{GDBN} Remote Serial Protocol
37822 * Stop Reply Packets::
37823 * General Query Packets::
37824 * Architecture-Specific Protocol Details::
37825 * Tracepoint Packets::
37826 * Host I/O Packets::
37828 * Notification Packets::
37829 * Remote Non-Stop::
37830 * Packet Acknowledgment::
37832 * File-I/O Remote Protocol Extension::
37833 * Library List Format::
37834 * Library List Format for SVR4 Targets::
37835 * Memory Map Format::
37836 * Thread List Format::
37837 * Traceframe Info Format::
37838 * Branch Trace Format::
37839 * Branch Trace Configuration Format::
37845 There may be occasions when you need to know something about the
37846 protocol---for example, if there is only one serial port to your target
37847 machine, you might want your program to do something special if it
37848 recognizes a packet meant for @value{GDBN}.
37850 In the examples below, @samp{->} and @samp{<-} are used to indicate
37851 transmitted and received data, respectively.
37853 @cindex protocol, @value{GDBN} remote serial
37854 @cindex serial protocol, @value{GDBN} remote
37855 @cindex remote serial protocol
37856 All @value{GDBN} commands and responses (other than acknowledgments
37857 and notifications, see @ref{Notification Packets}) are sent as a
37858 @var{packet}. A @var{packet} is introduced with the character
37859 @samp{$}, the actual @var{packet-data}, and the terminating character
37860 @samp{#} followed by a two-digit @var{checksum}:
37863 @code{$}@var{packet-data}@code{#}@var{checksum}
37867 @cindex checksum, for @value{GDBN} remote
37869 The two-digit @var{checksum} is computed as the modulo 256 sum of all
37870 characters between the leading @samp{$} and the trailing @samp{#} (an
37871 eight bit unsigned checksum).
37873 Implementors should note that prior to @value{GDBN} 5.0 the protocol
37874 specification also included an optional two-digit @var{sequence-id}:
37877 @code{$}@var{sequence-id}@code{:}@var{packet-data}@code{#}@var{checksum}
37880 @cindex sequence-id, for @value{GDBN} remote
37882 That @var{sequence-id} was appended to the acknowledgment. @value{GDBN}
37883 has never output @var{sequence-id}s. Stubs that handle packets added
37884 since @value{GDBN} 5.0 must not accept @var{sequence-id}.
37886 When either the host or the target machine receives a packet, the first
37887 response expected is an acknowledgment: either @samp{+} (to indicate
37888 the package was received correctly) or @samp{-} (to request
37892 -> @code{$}@var{packet-data}@code{#}@var{checksum}
37897 The @samp{+}/@samp{-} acknowledgments can be disabled
37898 once a connection is established.
37899 @xref{Packet Acknowledgment}, for details.
37901 The host (@value{GDBN}) sends @var{command}s, and the target (the
37902 debugging stub incorporated in your program) sends a @var{response}. In
37903 the case of step and continue @var{command}s, the response is only sent
37904 when the operation has completed, and the target has again stopped all
37905 threads in all attached processes. This is the default all-stop mode
37906 behavior, but the remote protocol also supports @value{GDBN}'s non-stop
37907 execution mode; see @ref{Remote Non-Stop}, for details.
37909 @var{packet-data} consists of a sequence of characters with the
37910 exception of @samp{#} and @samp{$} (see @samp{X} packet for additional
37913 @cindex remote protocol, field separator
37914 Fields within the packet should be separated using @samp{,} @samp{;} or
37915 @samp{:}. Except where otherwise noted all numbers are represented in
37916 @sc{hex} with leading zeros suppressed.
37918 Implementors should note that prior to @value{GDBN} 5.0, the character
37919 @samp{:} could not appear as the third character in a packet (as it
37920 would potentially conflict with the @var{sequence-id}).
37922 @cindex remote protocol, binary data
37923 @anchor{Binary Data}
37924 Binary data in most packets is encoded either as two hexadecimal
37925 digits per byte of binary data. This allowed the traditional remote
37926 protocol to work over connections which were only seven-bit clean.
37927 Some packets designed more recently assume an eight-bit clean
37928 connection, and use a more efficient encoding to send and receive
37931 The binary data representation uses @code{7d} (@sc{ascii} @samp{@}})
37932 as an escape character. Any escaped byte is transmitted as the escape
37933 character followed by the original character XORed with @code{0x20}.
37934 For example, the byte @code{0x7d} would be transmitted as the two
37935 bytes @code{0x7d 0x5d}. The bytes @code{0x23} (@sc{ascii} @samp{#}),
37936 @code{0x24} (@sc{ascii} @samp{$}), and @code{0x7d} (@sc{ascii}
37937 @samp{@}}) must always be escaped. Responses sent by the stub
37938 must also escape @code{0x2a} (@sc{ascii} @samp{*}), so that it
37939 is not interpreted as the start of a run-length encoded sequence
37942 Response @var{data} can be run-length encoded to save space.
37943 Run-length encoding replaces runs of identical characters with one
37944 instance of the repeated character, followed by a @samp{*} and a
37945 repeat count. The repeat count is itself sent encoded, to avoid
37946 binary characters in @var{data}: a value of @var{n} is sent as
37947 @code{@var{n}+29}. For a repeat count greater or equal to 3, this
37948 produces a printable @sc{ascii} character, e.g.@: a space (@sc{ascii}
37949 code 32) for a repeat count of 3. (This is because run-length
37950 encoding starts to win for counts 3 or more.) Thus, for example,
37951 @samp{0* } is a run-length encoding of ``0000'': the space character
37952 after @samp{*} means repeat the leading @code{0} @w{@code{32 - 29 =
37955 The printable characters @samp{#} and @samp{$} or with a numeric value
37956 greater than 126 must not be used. Runs of six repeats (@samp{#}) or
37957 seven repeats (@samp{$}) can be expanded using a repeat count of only
37958 five (@samp{"}). For example, @samp{00000000} can be encoded as
37961 The error response returned for some packets includes a two character
37962 error number. That number is not well defined.
37964 @cindex empty response, for unsupported packets
37965 For any @var{command} not supported by the stub, an empty response
37966 (@samp{$#00}) should be returned. That way it is possible to extend the
37967 protocol. A newer @value{GDBN} can tell if a packet is supported based
37970 At a minimum, a stub is required to support the @samp{g} and @samp{G}
37971 commands for register access, and the @samp{m} and @samp{M} commands
37972 for memory access. Stubs that only control single-threaded targets
37973 can implement run control with the @samp{c} (continue), and @samp{s}
37974 (step) commands. Stubs that support multi-threading targets should
37975 support the @samp{vCont} command. All other commands are optional.
37980 The following table provides a complete list of all currently defined
37981 @var{command}s and their corresponding response @var{data}.
37982 @xref{File-I/O Remote Protocol Extension}, for details about the File
37983 I/O extension of the remote protocol.
37985 Each packet's description has a template showing the packet's overall
37986 syntax, followed by an explanation of the packet's meaning. We
37987 include spaces in some of the templates for clarity; these are not
37988 part of the packet's syntax. No @value{GDBN} packet uses spaces to
37989 separate its components. For example, a template like @samp{foo
37990 @var{bar} @var{baz}} describes a packet beginning with the three ASCII
37991 bytes @samp{foo}, followed by a @var{bar}, followed directly by a
37992 @var{baz}. @value{GDBN} does not transmit a space character between the
37993 @samp{foo} and the @var{bar}, or between the @var{bar} and the
37996 @cindex @var{thread-id}, in remote protocol
37997 @anchor{thread-id syntax}
37998 Several packets and replies include a @var{thread-id} field to identify
37999 a thread. Normally these are positive numbers with a target-specific
38000 interpretation, formatted as big-endian hex strings. A @var{thread-id}
38001 can also be a literal @samp{-1} to indicate all threads, or @samp{0} to
38004 In addition, the remote protocol supports a multiprocess feature in
38005 which the @var{thread-id} syntax is extended to optionally include both
38006 process and thread ID fields, as @samp{p@var{pid}.@var{tid}}.
38007 The @var{pid} (process) and @var{tid} (thread) components each have the
38008 format described above: a positive number with target-specific
38009 interpretation formatted as a big-endian hex string, literal @samp{-1}
38010 to indicate all processes or threads (respectively), or @samp{0} to
38011 indicate an arbitrary process or thread. Specifying just a process, as
38012 @samp{p@var{pid}}, is equivalent to @samp{p@var{pid}.-1}. It is an
38013 error to specify all processes but a specific thread, such as
38014 @samp{p-1.@var{tid}}. Note that the @samp{p} prefix is @emph{not} used
38015 for those packets and replies explicitly documented to include a process
38016 ID, rather than a @var{thread-id}.
38018 The multiprocess @var{thread-id} syntax extensions are only used if both
38019 @value{GDBN} and the stub report support for the @samp{multiprocess}
38020 feature using @samp{qSupported}. @xref{multiprocess extensions}, for
38023 Note that all packet forms beginning with an upper- or lower-case
38024 letter, other than those described here, are reserved for future use.
38026 Here are the packet descriptions.
38031 @cindex @samp{!} packet
38032 @anchor{extended mode}
38033 Enable extended mode. In extended mode, the remote server is made
38034 persistent. The @samp{R} packet is used to restart the program being
38040 The remote target both supports and has enabled extended mode.
38044 @cindex @samp{?} packet
38046 Indicate the reason the target halted. The reply is the same as for
38047 step and continue. This packet has a special interpretation when the
38048 target is in non-stop mode; see @ref{Remote Non-Stop}.
38051 @xref{Stop Reply Packets}, for the reply specifications.
38053 @item A @var{arglen},@var{argnum},@var{arg},@dots{}
38054 @cindex @samp{A} packet
38055 Initialized @code{argv[]} array passed into program. @var{arglen}
38056 specifies the number of bytes in the hex encoded byte stream
38057 @var{arg}. See @code{gdbserver} for more details.
38062 The arguments were set.
38068 @cindex @samp{b} packet
38069 (Don't use this packet; its behavior is not well-defined.)
38070 Change the serial line speed to @var{baud}.
38072 JTC: @emph{When does the transport layer state change? When it's
38073 received, or after the ACK is transmitted. In either case, there are
38074 problems if the command or the acknowledgment packet is dropped.}
38076 Stan: @emph{If people really wanted to add something like this, and get
38077 it working for the first time, they ought to modify ser-unix.c to send
38078 some kind of out-of-band message to a specially-setup stub and have the
38079 switch happen "in between" packets, so that from remote protocol's point
38080 of view, nothing actually happened.}
38082 @item B @var{addr},@var{mode}
38083 @cindex @samp{B} packet
38084 Set (@var{mode} is @samp{S}) or clear (@var{mode} is @samp{C}) a
38085 breakpoint at @var{addr}.
38087 Don't use this packet. Use the @samp{Z} and @samp{z} packets instead
38088 (@pxref{insert breakpoint or watchpoint packet}).
38090 @cindex @samp{bc} packet
38093 Backward continue. Execute the target system in reverse. No parameter.
38094 @xref{Reverse Execution}, for more information.
38097 @xref{Stop Reply Packets}, for the reply specifications.
38099 @cindex @samp{bs} packet
38102 Backward single step. Execute one instruction in reverse. No parameter.
38103 @xref{Reverse Execution}, for more information.
38106 @xref{Stop Reply Packets}, for the reply specifications.
38108 @item c @r{[}@var{addr}@r{]}
38109 @cindex @samp{c} packet
38110 Continue at @var{addr}, which is the address to resume. If @var{addr}
38111 is omitted, resume at current address.
38113 This packet is deprecated for multi-threading support. @xref{vCont
38117 @xref{Stop Reply Packets}, for the reply specifications.
38119 @item C @var{sig}@r{[};@var{addr}@r{]}
38120 @cindex @samp{C} packet
38121 Continue with signal @var{sig} (hex signal number). If
38122 @samp{;@var{addr}} is omitted, resume at same address.
38124 This packet is deprecated for multi-threading support. @xref{vCont
38128 @xref{Stop Reply Packets}, for the reply specifications.
38131 @cindex @samp{d} packet
38134 Don't use this packet; instead, define a general set packet
38135 (@pxref{General Query Packets}).
38139 @cindex @samp{D} packet
38140 The first form of the packet is used to detach @value{GDBN} from the
38141 remote system. It is sent to the remote target
38142 before @value{GDBN} disconnects via the @code{detach} command.
38144 The second form, including a process ID, is used when multiprocess
38145 protocol extensions are enabled (@pxref{multiprocess extensions}), to
38146 detach only a specific process. The @var{pid} is specified as a
38147 big-endian hex string.
38157 @item F @var{RC},@var{EE},@var{CF};@var{XX}
38158 @cindex @samp{F} packet
38159 A reply from @value{GDBN} to an @samp{F} packet sent by the target.
38160 This is part of the File-I/O protocol extension. @xref{File-I/O
38161 Remote Protocol Extension}, for the specification.
38164 @anchor{read registers packet}
38165 @cindex @samp{g} packet
38166 Read general registers.
38170 @item @var{XX@dots{}}
38171 Each byte of register data is described by two hex digits. The bytes
38172 with the register are transmitted in target byte order. The size of
38173 each register and their position within the @samp{g} packet are
38174 determined by the @value{GDBN} internal gdbarch functions
38175 @code{DEPRECATED_REGISTER_RAW_SIZE} and @code{gdbarch_register_name}.
38177 When reading registers from a trace frame (@pxref{Analyze Collected
38178 Data,,Using the Collected Data}), the stub may also return a string of
38179 literal @samp{x}'s in place of the register data digits, to indicate
38180 that the corresponding register has not been collected, thus its value
38181 is unavailable. For example, for an architecture with 4 registers of
38182 4 bytes each, the following reply indicates to @value{GDBN} that
38183 registers 0 and 2 have not been collected, while registers 1 and 3
38184 have been collected, and both have zero value:
38188 <- @code{xxxxxxxx00000000xxxxxxxx00000000}
38195 @item G @var{XX@dots{}}
38196 @cindex @samp{G} packet
38197 Write general registers. @xref{read registers packet}, for a
38198 description of the @var{XX@dots{}} data.
38208 @item H @var{op} @var{thread-id}
38209 @cindex @samp{H} packet
38210 Set thread for subsequent operations (@samp{m}, @samp{M}, @samp{g},
38211 @samp{G}, et.al.). Depending on the operation to be performed, @var{op}
38212 should be @samp{c} for step and continue operations (note that this
38213 is deprecated, supporting the @samp{vCont} command is a better
38214 option), and @samp{g} for other operations. The thread designator
38215 @var{thread-id} has the format and interpretation described in
38216 @ref{thread-id syntax}.
38227 @c 'H': How restrictive (or permissive) is the thread model. If a
38228 @c thread is selected and stopped, are other threads allowed
38229 @c to continue to execute? As I mentioned above, I think the
38230 @c semantics of each command when a thread is selected must be
38231 @c described. For example:
38233 @c 'g': If the stub supports threads and a specific thread is
38234 @c selected, returns the register block from that thread;
38235 @c otherwise returns current registers.
38237 @c 'G' If the stub supports threads and a specific thread is
38238 @c selected, sets the registers of the register block of
38239 @c that thread; otherwise sets current registers.
38241 @item i @r{[}@var{addr}@r{[},@var{nnn}@r{]]}
38242 @anchor{cycle step packet}
38243 @cindex @samp{i} packet
38244 Step the remote target by a single clock cycle. If @samp{,@var{nnn}} is
38245 present, cycle step @var{nnn} cycles. If @var{addr} is present, cycle
38246 step starting at that address.
38249 @cindex @samp{I} packet
38250 Signal, then cycle step. @xref{step with signal packet}. @xref{cycle
38254 @cindex @samp{k} packet
38257 The exact effect of this packet is not specified.
38259 For a bare-metal target, it may power cycle or reset the target
38260 system. For that reason, the @samp{k} packet has no reply.
38262 For a single-process target, it may kill that process if possible.
38264 A multiple-process target may choose to kill just one process, or all
38265 that are under @value{GDBN}'s control. For more precise control, use
38266 the vKill packet (@pxref{vKill packet}).
38268 If the target system immediately closes the connection in response to
38269 @samp{k}, @value{GDBN} does not consider the lack of packet
38270 acknowledgment to be an error, and assumes the kill was successful.
38272 If connected using @kbd{target extended-remote}, and the target does
38273 not close the connection in response to a kill request, @value{GDBN}
38274 probes the target state as if a new connection was opened
38275 (@pxref{? packet}).
38277 @item m @var{addr},@var{length}
38278 @cindex @samp{m} packet
38279 Read @var{length} addressable memory units starting at address @var{addr}
38280 (@pxref{addressable memory unit}). Note that @var{addr} may not be aligned to
38281 any particular boundary.
38283 The stub need not use any particular size or alignment when gathering
38284 data from memory for the response; even if @var{addr} is word-aligned
38285 and @var{length} is a multiple of the word size, the stub is free to
38286 use byte accesses, or not. For this reason, this packet may not be
38287 suitable for accessing memory-mapped I/O devices.
38288 @cindex alignment of remote memory accesses
38289 @cindex size of remote memory accesses
38290 @cindex memory, alignment and size of remote accesses
38294 @item @var{XX@dots{}}
38295 Memory contents; each byte is transmitted as a two-digit hexadecimal number.
38296 The reply may contain fewer addressable memory units than requested if the
38297 server was able to read only part of the region of memory.
38302 @item M @var{addr},@var{length}:@var{XX@dots{}}
38303 @cindex @samp{M} packet
38304 Write @var{length} addressable memory units starting at address @var{addr}
38305 (@pxref{addressable memory unit}). The data is given by @var{XX@dots{}}; each
38306 byte is transmitted as a two-digit hexadecimal number.
38313 for an error (this includes the case where only part of the data was
38318 @cindex @samp{p} packet
38319 Read the value of register @var{n}; @var{n} is in hex.
38320 @xref{read registers packet}, for a description of how the returned
38321 register value is encoded.
38325 @item @var{XX@dots{}}
38326 the register's value
38330 Indicating an unrecognized @var{query}.
38333 @item P @var{n@dots{}}=@var{r@dots{}}
38334 @anchor{write register packet}
38335 @cindex @samp{P} packet
38336 Write register @var{n@dots{}} with value @var{r@dots{}}. The register
38337 number @var{n} is in hexadecimal, and @var{r@dots{}} contains two hex
38338 digits for each byte in the register (target byte order).
38348 @item q @var{name} @var{params}@dots{}
38349 @itemx Q @var{name} @var{params}@dots{}
38350 @cindex @samp{q} packet
38351 @cindex @samp{Q} packet
38352 General query (@samp{q}) and set (@samp{Q}). These packets are
38353 described fully in @ref{General Query Packets}.
38356 @cindex @samp{r} packet
38357 Reset the entire system.
38359 Don't use this packet; use the @samp{R} packet instead.
38362 @cindex @samp{R} packet
38363 Restart the program being debugged. The @var{XX}, while needed, is ignored.
38364 This packet is only available in extended mode (@pxref{extended mode}).
38366 The @samp{R} packet has no reply.
38368 @item s @r{[}@var{addr}@r{]}
38369 @cindex @samp{s} packet
38370 Single step, resuming at @var{addr}. If
38371 @var{addr} is omitted, resume at same address.
38373 This packet is deprecated for multi-threading support. @xref{vCont
38377 @xref{Stop Reply Packets}, for the reply specifications.
38379 @item S @var{sig}@r{[};@var{addr}@r{]}
38380 @anchor{step with signal packet}
38381 @cindex @samp{S} packet
38382 Step with signal. This is analogous to the @samp{C} packet, but
38383 requests a single-step, rather than a normal resumption of execution.
38385 This packet is deprecated for multi-threading support. @xref{vCont
38389 @xref{Stop Reply Packets}, for the reply specifications.
38391 @item t @var{addr}:@var{PP},@var{MM}
38392 @cindex @samp{t} packet
38393 Search backwards starting at address @var{addr} for a match with pattern
38394 @var{PP} and mask @var{MM}, both of which are are 4 byte long.
38395 There must be at least 3 digits in @var{addr}.
38397 @item T @var{thread-id}
38398 @cindex @samp{T} packet
38399 Find out if the thread @var{thread-id} is alive. @xref{thread-id syntax}.
38404 thread is still alive
38410 Packets starting with @samp{v} are identified by a multi-letter name,
38411 up to the first @samp{;} or @samp{?} (or the end of the packet).
38413 @item vAttach;@var{pid}
38414 @cindex @samp{vAttach} packet
38415 Attach to a new process with the specified process ID @var{pid}.
38416 The process ID is a
38417 hexadecimal integer identifying the process. In all-stop mode, all
38418 threads in the attached process are stopped; in non-stop mode, it may be
38419 attached without being stopped if that is supported by the target.
38421 @c In non-stop mode, on a successful vAttach, the stub should set the
38422 @c current thread to a thread of the newly-attached process. After
38423 @c attaching, GDB queries for the attached process's thread ID with qC.
38424 @c Also note that, from a user perspective, whether or not the
38425 @c target is stopped on attach in non-stop mode depends on whether you
38426 @c use the foreground or background version of the attach command, not
38427 @c on what vAttach does; GDB does the right thing with respect to either
38428 @c stopping or restarting threads.
38430 This packet is only available in extended mode (@pxref{extended mode}).
38436 @item @r{Any stop packet}
38437 for success in all-stop mode (@pxref{Stop Reply Packets})
38439 for success in non-stop mode (@pxref{Remote Non-Stop})
38442 @item vCont@r{[};@var{action}@r{[}:@var{thread-id}@r{]]}@dots{}
38443 @cindex @samp{vCont} packet
38444 @anchor{vCont packet}
38445 Resume the inferior, specifying different actions for each thread.
38447 For each inferior thread, the leftmost action with a matching
38448 @var{thread-id} is applied. Threads that don't match any action
38449 remain in their current state. Thread IDs are specified using the
38450 syntax described in @ref{thread-id syntax}. If multiprocess
38451 extensions (@pxref{multiprocess extensions}) are supported, actions
38452 can be specified to match all threads in a process by using the
38453 @samp{p@var{pid}.-1} form of the @var{thread-id}. An action with no
38454 @var{thread-id} matches all threads. Specifying no actions is an
38457 Currently supported actions are:
38463 Continue with signal @var{sig}. The signal @var{sig} should be two hex digits.
38467 Step with signal @var{sig}. The signal @var{sig} should be two hex digits.
38470 @item r @var{start},@var{end}
38471 Step once, and then keep stepping as long as the thread stops at
38472 addresses between @var{start} (inclusive) and @var{end} (exclusive).
38473 The remote stub reports a stop reply when either the thread goes out
38474 of the range or is stopped due to an unrelated reason, such as hitting
38475 a breakpoint. @xref{range stepping}.
38477 If the range is empty (@var{start} == @var{end}), then the action
38478 becomes equivalent to the @samp{s} action. In other words,
38479 single-step once, and report the stop (even if the stepped instruction
38480 jumps to @var{start}).
38482 (A stop reply may be sent at any point even if the PC is still within
38483 the stepping range; for example, it is valid to implement this packet
38484 in a degenerate way as a single instruction step operation.)
38488 The optional argument @var{addr} normally associated with the
38489 @samp{c}, @samp{C}, @samp{s}, and @samp{S} packets is
38490 not supported in @samp{vCont}.
38492 The @samp{t} action is only relevant in non-stop mode
38493 (@pxref{Remote Non-Stop}) and may be ignored by the stub otherwise.
38494 A stop reply should be generated for any affected thread not already stopped.
38495 When a thread is stopped by means of a @samp{t} action,
38496 the corresponding stop reply should indicate that the thread has stopped with
38497 signal @samp{0}, regardless of whether the target uses some other signal
38498 as an implementation detail.
38500 The server must ignore @samp{c}, @samp{C}, @samp{s}, @samp{S}, and
38501 @samp{r} actions for threads that are already running. Conversely,
38502 the server must ignore @samp{t} actions for threads that are already
38505 @emph{Note:} In non-stop mode, a thread is considered running until
38506 @value{GDBN} acknowleges an asynchronous stop notification for it with
38507 the @samp{vStopped} packet (@pxref{Remote Non-Stop}).
38509 The stub must support @samp{vCont} if it reports support for
38510 multiprocess extensions (@pxref{multiprocess extensions}).
38513 @xref{Stop Reply Packets}, for the reply specifications.
38516 @cindex @samp{vCont?} packet
38517 Request a list of actions supported by the @samp{vCont} packet.
38521 @item vCont@r{[};@var{action}@dots{}@r{]}
38522 The @samp{vCont} packet is supported. Each @var{action} is a supported
38523 command in the @samp{vCont} packet.
38525 The @samp{vCont} packet is not supported.
38528 @anchor{vCtrlC packet}
38530 @cindex @samp{vCtrlC} packet
38531 Interrupt remote target as if a control-C was pressed on the remote
38532 terminal. This is the equivalent to reacting to the @code{^C}
38533 (@samp{\003}, the control-C character) character in all-stop mode
38534 while the target is running, except this works in non-stop mode.
38535 @xref{interrupting remote targets}, for more info on the all-stop
38546 @item vFile:@var{operation}:@var{parameter}@dots{}
38547 @cindex @samp{vFile} packet
38548 Perform a file operation on the target system. For details,
38549 see @ref{Host I/O Packets}.
38551 @item vFlashErase:@var{addr},@var{length}
38552 @cindex @samp{vFlashErase} packet
38553 Direct the stub to erase @var{length} bytes of flash starting at
38554 @var{addr}. The region may enclose any number of flash blocks, but
38555 its start and end must fall on block boundaries, as indicated by the
38556 flash block size appearing in the memory map (@pxref{Memory Map
38557 Format}). @value{GDBN} groups flash memory programming operations
38558 together, and sends a @samp{vFlashDone} request after each group; the
38559 stub is allowed to delay erase operation until the @samp{vFlashDone}
38560 packet is received.
38570 @item vFlashWrite:@var{addr}:@var{XX@dots{}}
38571 @cindex @samp{vFlashWrite} packet
38572 Direct the stub to write data to flash address @var{addr}. The data
38573 is passed in binary form using the same encoding as for the @samp{X}
38574 packet (@pxref{Binary Data}). The memory ranges specified by
38575 @samp{vFlashWrite} packets preceding a @samp{vFlashDone} packet must
38576 not overlap, and must appear in order of increasing addresses
38577 (although @samp{vFlashErase} packets for higher addresses may already
38578 have been received; the ordering is guaranteed only between
38579 @samp{vFlashWrite} packets). If a packet writes to an address that was
38580 neither erased by a preceding @samp{vFlashErase} packet nor by some other
38581 target-specific method, the results are unpredictable.
38589 for vFlashWrite addressing non-flash memory
38595 @cindex @samp{vFlashDone} packet
38596 Indicate to the stub that flash programming operation is finished.
38597 The stub is permitted to delay or batch the effects of a group of
38598 @samp{vFlashErase} and @samp{vFlashWrite} packets until a
38599 @samp{vFlashDone} packet is received. The contents of the affected
38600 regions of flash memory are unpredictable until the @samp{vFlashDone}
38601 request is completed.
38603 @item vKill;@var{pid}
38604 @cindex @samp{vKill} packet
38605 @anchor{vKill packet}
38606 Kill the process with the specified process ID @var{pid}, which is a
38607 hexadecimal integer identifying the process. This packet is used in
38608 preference to @samp{k} when multiprocess protocol extensions are
38609 supported; see @ref{multiprocess extensions}.
38619 @item vMustReplyEmpty
38620 @cindex @samp{vMustReplyEmpty} packet
38621 The correct reply to an unknown @samp{v} packet is to return the empty
38622 string, however, some older versions of @command{gdbserver} would
38623 incorrectly return @samp{OK} for unknown @samp{v} packets.
38625 The @samp{vMustReplyEmpty} is used as a feature test to check how
38626 @command{gdbserver} handles unknown packets, it is important that this
38627 packet be handled in the same way as other unknown @samp{v} packets.
38628 If this packet is handled differently to other unknown @samp{v}
38629 packets then it is possile that @value{GDBN} may run into problems in
38630 other areas, specifically around use of @samp{vFile:setfs:}.
38632 @item vRun;@var{filename}@r{[};@var{argument}@r{]}@dots{}
38633 @cindex @samp{vRun} packet
38634 Run the program @var{filename}, passing it each @var{argument} on its
38635 command line. The file and arguments are hex-encoded strings. If
38636 @var{filename} is an empty string, the stub may use a default program
38637 (e.g.@: the last program run). The program is created in the stopped
38640 @c FIXME: What about non-stop mode?
38642 This packet is only available in extended mode (@pxref{extended mode}).
38648 @item @r{Any stop packet}
38649 for success (@pxref{Stop Reply Packets})
38653 @cindex @samp{vStopped} packet
38654 @xref{Notification Packets}.
38656 @item X @var{addr},@var{length}:@var{XX@dots{}}
38658 @cindex @samp{X} packet
38659 Write data to memory, where the data is transmitted in binary.
38660 Memory is specified by its address @var{addr} and number of addressable memory
38661 units @var{length} (@pxref{addressable memory unit});
38662 @samp{@var{XX}@dots{}} is binary data (@pxref{Binary Data}).
38672 @item z @var{type},@var{addr},@var{kind}
38673 @itemx Z @var{type},@var{addr},@var{kind}
38674 @anchor{insert breakpoint or watchpoint packet}
38675 @cindex @samp{z} packet
38676 @cindex @samp{Z} packets
38677 Insert (@samp{Z}) or remove (@samp{z}) a @var{type} breakpoint or
38678 watchpoint starting at address @var{address} of kind @var{kind}.
38680 Each breakpoint and watchpoint packet @var{type} is documented
38683 @emph{Implementation notes: A remote target shall return an empty string
38684 for an unrecognized breakpoint or watchpoint packet @var{type}. A
38685 remote target shall support either both or neither of a given
38686 @samp{Z@var{type}@dots{}} and @samp{z@var{type}@dots{}} packet pair. To
38687 avoid potential problems with duplicate packets, the operations should
38688 be implemented in an idempotent way.}
38690 @item z0,@var{addr},@var{kind}
38691 @itemx Z0,@var{addr},@var{kind}@r{[};@var{cond_list}@dots{}@r{]}@r{[};cmds:@var{persist},@var{cmd_list}@dots{}@r{]}
38692 @cindex @samp{z0} packet
38693 @cindex @samp{Z0} packet
38694 Insert (@samp{Z0}) or remove (@samp{z0}) a software breakpoint at address
38695 @var{addr} of type @var{kind}.
38697 A software breakpoint is implemented by replacing the instruction at
38698 @var{addr} with a software breakpoint or trap instruction. The
38699 @var{kind} is target-specific and typically indicates the size of the
38700 breakpoint in bytes that should be inserted. E.g., the @sc{arm} and
38701 @sc{mips} can insert either a 2 or 4 byte breakpoint. Some
38702 architectures have additional meanings for @var{kind}
38703 (@pxref{Architecture-Specific Protocol Details}); if no
38704 architecture-specific value is being used, it should be @samp{0}.
38705 @var{kind} is hex-encoded. @var{cond_list} is an optional list of
38706 conditional expressions in bytecode form that should be evaluated on
38707 the target's side. These are the conditions that should be taken into
38708 consideration when deciding if the breakpoint trigger should be
38709 reported back to @value{GDBN}.
38711 See also the @samp{swbreak} stop reason (@pxref{swbreak stop reason})
38712 for how to best report a software breakpoint event to @value{GDBN}.
38714 The @var{cond_list} parameter is comprised of a series of expressions,
38715 concatenated without separators. Each expression has the following form:
38719 @item X @var{len},@var{expr}
38720 @var{len} is the length of the bytecode expression and @var{expr} is the
38721 actual conditional expression in bytecode form.
38725 The optional @var{cmd_list} parameter introduces commands that may be
38726 run on the target, rather than being reported back to @value{GDBN}.
38727 The parameter starts with a numeric flag @var{persist}; if the flag is
38728 nonzero, then the breakpoint may remain active and the commands
38729 continue to be run even when @value{GDBN} disconnects from the target.
38730 Following this flag is a series of expressions concatenated with no
38731 separators. Each expression has the following form:
38735 @item X @var{len},@var{expr}
38736 @var{len} is the length of the bytecode expression and @var{expr} is the
38737 actual commands expression in bytecode form.
38741 @emph{Implementation note: It is possible for a target to copy or move
38742 code that contains software breakpoints (e.g., when implementing
38743 overlays). The behavior of this packet, in the presence of such a
38744 target, is not defined.}
38756 @item z1,@var{addr},@var{kind}
38757 @itemx Z1,@var{addr},@var{kind}@r{[};@var{cond_list}@dots{}@r{]}@r{[};cmds:@var{persist},@var{cmd_list}@dots{}@r{]}
38758 @cindex @samp{z1} packet
38759 @cindex @samp{Z1} packet
38760 Insert (@samp{Z1}) or remove (@samp{z1}) a hardware breakpoint at
38761 address @var{addr}.
38763 A hardware breakpoint is implemented using a mechanism that is not
38764 dependent on being able to modify the target's memory. The
38765 @var{kind}, @var{cond_list}, and @var{cmd_list} arguments have the
38766 same meaning as in @samp{Z0} packets.
38768 @emph{Implementation note: A hardware breakpoint is not affected by code
38781 @item z2,@var{addr},@var{kind}
38782 @itemx Z2,@var{addr},@var{kind}
38783 @cindex @samp{z2} packet
38784 @cindex @samp{Z2} packet
38785 Insert (@samp{Z2}) or remove (@samp{z2}) a write watchpoint at @var{addr}.
38786 The number of bytes to watch is specified by @var{kind}.
38798 @item z3,@var{addr},@var{kind}
38799 @itemx Z3,@var{addr},@var{kind}
38800 @cindex @samp{z3} packet
38801 @cindex @samp{Z3} packet
38802 Insert (@samp{Z3}) or remove (@samp{z3}) a read watchpoint at @var{addr}.
38803 The number of bytes to watch is specified by @var{kind}.
38815 @item z4,@var{addr},@var{kind}
38816 @itemx Z4,@var{addr},@var{kind}
38817 @cindex @samp{z4} packet
38818 @cindex @samp{Z4} packet
38819 Insert (@samp{Z4}) or remove (@samp{z4}) an access watchpoint at @var{addr}.
38820 The number of bytes to watch is specified by @var{kind}.
38834 @node Stop Reply Packets
38835 @section Stop Reply Packets
38836 @cindex stop reply packets
38838 The @samp{C}, @samp{c}, @samp{S}, @samp{s}, @samp{vCont},
38839 @samp{vAttach}, @samp{vRun}, @samp{vStopped}, and @samp{?} packets can
38840 receive any of the below as a reply. Except for @samp{?}
38841 and @samp{vStopped}, that reply is only returned
38842 when the target halts. In the below the exact meaning of @dfn{signal
38843 number} is defined by the header @file{include/gdb/signals.h} in the
38844 @value{GDBN} source code.
38846 In non-stop mode, the server will simply reply @samp{OK} to commands
38847 such as @samp{vCont}; any stop will be the subject of a future
38848 notification. @xref{Remote Non-Stop}.
38850 As in the description of request packets, we include spaces in the
38851 reply templates for clarity; these are not part of the reply packet's
38852 syntax. No @value{GDBN} stop reply packet uses spaces to separate its
38858 The program received signal number @var{AA} (a two-digit hexadecimal
38859 number). This is equivalent to a @samp{T} response with no
38860 @var{n}:@var{r} pairs.
38862 @item T @var{AA} @var{n1}:@var{r1};@var{n2}:@var{r2};@dots{}
38863 @cindex @samp{T} packet reply
38864 The program received signal number @var{AA} (a two-digit hexadecimal
38865 number). This is equivalent to an @samp{S} response, except that the
38866 @samp{@var{n}:@var{r}} pairs can carry values of important registers
38867 and other information directly in the stop reply packet, reducing
38868 round-trip latency. Single-step and breakpoint traps are reported
38869 this way. Each @samp{@var{n}:@var{r}} pair is interpreted as follows:
38873 If @var{n} is a hexadecimal number, it is a register number, and the
38874 corresponding @var{r} gives that register's value. The data @var{r} is a
38875 series of bytes in target byte order, with each byte given by a
38876 two-digit hex number.
38879 If @var{n} is @samp{thread}, then @var{r} is the @var{thread-id} of
38880 the stopped thread, as specified in @ref{thread-id syntax}.
38883 If @var{n} is @samp{core}, then @var{r} is the hexadecimal number of
38884 the core on which the stop event was detected.
38887 If @var{n} is a recognized @dfn{stop reason}, it describes a more
38888 specific event that stopped the target. The currently defined stop
38889 reasons are listed below. The @var{aa} should be @samp{05}, the trap
38890 signal. At most one stop reason should be present.
38893 Otherwise, @value{GDBN} should ignore this @samp{@var{n}:@var{r}} pair
38894 and go on to the next; this allows us to extend the protocol in the
38898 The currently defined stop reasons are:
38904 The packet indicates a watchpoint hit, and @var{r} is the data address, in
38907 @item syscall_entry
38908 @itemx syscall_return
38909 The packet indicates a syscall entry or return, and @var{r} is the
38910 syscall number, in hex.
38912 @cindex shared library events, remote reply
38914 The packet indicates that the loaded libraries have changed.
38915 @value{GDBN} should use @samp{qXfer:libraries:read} to fetch a new
38916 list of loaded libraries. The @var{r} part is ignored.
38918 @cindex replay log events, remote reply
38920 The packet indicates that the target cannot continue replaying
38921 logged execution events, because it has reached the end (or the
38922 beginning when executing backward) of the log. The value of @var{r}
38923 will be either @samp{begin} or @samp{end}. @xref{Reverse Execution},
38924 for more information.
38927 @anchor{swbreak stop reason}
38928 The packet indicates a software breakpoint instruction was executed,
38929 irrespective of whether it was @value{GDBN} that planted the
38930 breakpoint or the breakpoint is hardcoded in the program. The @var{r}
38931 part must be left empty.
38933 On some architectures, such as x86, at the architecture level, when a
38934 breakpoint instruction executes the program counter points at the
38935 breakpoint address plus an offset. On such targets, the stub is
38936 responsible for adjusting the PC to point back at the breakpoint
38939 This packet should not be sent by default; older @value{GDBN} versions
38940 did not support it. @value{GDBN} requests it, by supplying an
38941 appropriate @samp{qSupported} feature (@pxref{qSupported}). The
38942 remote stub must also supply the appropriate @samp{qSupported} feature
38943 indicating support.
38945 This packet is required for correct non-stop mode operation.
38948 The packet indicates the target stopped for a hardware breakpoint.
38949 The @var{r} part must be left empty.
38951 The same remarks about @samp{qSupported} and non-stop mode above
38954 @cindex fork events, remote reply
38956 The packet indicates that @code{fork} was called, and @var{r}
38957 is the thread ID of the new child process. Refer to
38958 @ref{thread-id syntax} for the format of the @var{thread-id}
38959 field. This packet is only applicable to targets that support
38962 This packet should not be sent by default; older @value{GDBN} versions
38963 did not support it. @value{GDBN} requests it, by supplying an
38964 appropriate @samp{qSupported} feature (@pxref{qSupported}). The
38965 remote stub must also supply the appropriate @samp{qSupported} feature
38966 indicating support.
38968 @cindex vfork events, remote reply
38970 The packet indicates that @code{vfork} was called, and @var{r}
38971 is the thread ID of the new child process. Refer to
38972 @ref{thread-id syntax} for the format of the @var{thread-id}
38973 field. This packet is only applicable to targets that support
38976 This packet should not be sent by default; older @value{GDBN} versions
38977 did not support it. @value{GDBN} requests it, by supplying an
38978 appropriate @samp{qSupported} feature (@pxref{qSupported}). The
38979 remote stub must also supply the appropriate @samp{qSupported} feature
38980 indicating support.
38982 @cindex vforkdone events, remote reply
38984 The packet indicates that a child process created by a vfork
38985 has either called @code{exec} or terminated, so that the
38986 address spaces of the parent and child process are no longer
38987 shared. The @var{r} part is ignored. This packet is only
38988 applicable to targets that support vforkdone events.
38990 This packet should not be sent by default; older @value{GDBN} versions
38991 did not support it. @value{GDBN} requests it, by supplying an
38992 appropriate @samp{qSupported} feature (@pxref{qSupported}). The
38993 remote stub must also supply the appropriate @samp{qSupported} feature
38994 indicating support.
38996 @cindex exec events, remote reply
38998 The packet indicates that @code{execve} was called, and @var{r}
38999 is the absolute pathname of the file that was executed, in hex.
39000 This packet is only applicable to targets that support exec events.
39002 This packet should not be sent by default; older @value{GDBN} versions
39003 did not support it. @value{GDBN} requests it, by supplying an
39004 appropriate @samp{qSupported} feature (@pxref{qSupported}). The
39005 remote stub must also supply the appropriate @samp{qSupported} feature
39006 indicating support.
39008 @cindex thread create event, remote reply
39009 @anchor{thread create event}
39011 The packet indicates that the thread was just created. The new thread
39012 is stopped until @value{GDBN} sets it running with a resumption packet
39013 (@pxref{vCont packet}). This packet should not be sent by default;
39014 @value{GDBN} requests it with the @ref{QThreadEvents} packet. See
39015 also the @samp{w} (@pxref{thread exit event}) remote reply below. The
39016 @var{r} part is ignored.
39021 @itemx W @var{AA} ; process:@var{pid}
39022 The process exited, and @var{AA} is the exit status. This is only
39023 applicable to certain targets.
39025 The second form of the response, including the process ID of the
39026 exited process, can be used only when @value{GDBN} has reported
39027 support for multiprocess protocol extensions; see @ref{multiprocess
39028 extensions}. Both @var{AA} and @var{pid} are formatted as big-endian
39032 @itemx X @var{AA} ; process:@var{pid}
39033 The process terminated with signal @var{AA}.
39035 The second form of the response, including the process ID of the
39036 terminated process, can be used only when @value{GDBN} has reported
39037 support for multiprocess protocol extensions; see @ref{multiprocess
39038 extensions}. Both @var{AA} and @var{pid} are formatted as big-endian
39041 @anchor{thread exit event}
39042 @cindex thread exit event, remote reply
39043 @item w @var{AA} ; @var{tid}
39045 The thread exited, and @var{AA} is the exit status. This response
39046 should not be sent by default; @value{GDBN} requests it with the
39047 @ref{QThreadEvents} packet. See also @ref{thread create event} above.
39048 @var{AA} is formatted as a big-endian hex string.
39051 There are no resumed threads left in the target. In other words, even
39052 though the process is alive, the last resumed thread has exited. For
39053 example, say the target process has two threads: thread 1 and thread
39054 2. The client leaves thread 1 stopped, and resumes thread 2, which
39055 subsequently exits. At this point, even though the process is still
39056 alive, and thus no @samp{W} stop reply is sent, no thread is actually
39057 executing either. The @samp{N} stop reply thus informs the client
39058 that it can stop waiting for stop replies. This packet should not be
39059 sent by default; older @value{GDBN} versions did not support it.
39060 @value{GDBN} requests it, by supplying an appropriate
39061 @samp{qSupported} feature (@pxref{qSupported}). The remote stub must
39062 also supply the appropriate @samp{qSupported} feature indicating
39065 @item O @var{XX}@dots{}
39066 @samp{@var{XX}@dots{}} is hex encoding of @sc{ascii} data, to be
39067 written as the program's console output. This can happen at any time
39068 while the program is running and the debugger should continue to wait
39069 for @samp{W}, @samp{T}, etc. This reply is not permitted in non-stop mode.
39071 @item F @var{call-id},@var{parameter}@dots{}
39072 @var{call-id} is the identifier which says which host system call should
39073 be called. This is just the name of the function. Translation into the
39074 correct system call is only applicable as it's defined in @value{GDBN}.
39075 @xref{File-I/O Remote Protocol Extension}, for a list of implemented
39078 @samp{@var{parameter}@dots{}} is a list of parameters as defined for
39079 this very system call.
39081 The target replies with this packet when it expects @value{GDBN} to
39082 call a host system call on behalf of the target. @value{GDBN} replies
39083 with an appropriate @samp{F} packet and keeps up waiting for the next
39084 reply packet from the target. The latest @samp{C}, @samp{c}, @samp{S}
39085 or @samp{s} action is expected to be continued. @xref{File-I/O Remote
39086 Protocol Extension}, for more details.
39090 @node General Query Packets
39091 @section General Query Packets
39092 @cindex remote query requests
39094 Packets starting with @samp{q} are @dfn{general query packets};
39095 packets starting with @samp{Q} are @dfn{general set packets}. General
39096 query and set packets are a semi-unified form for retrieving and
39097 sending information to and from the stub.
39099 The initial letter of a query or set packet is followed by a name
39100 indicating what sort of thing the packet applies to. For example,
39101 @value{GDBN} may use a @samp{qSymbol} packet to exchange symbol
39102 definitions with the stub. These packet names follow some
39107 The name must not contain commas, colons or semicolons.
39109 Most @value{GDBN} query and set packets have a leading upper case
39112 The names of custom vendor packets should use a company prefix, in
39113 lower case, followed by a period. For example, packets designed at
39114 the Acme Corporation might begin with @samp{qacme.foo} (for querying
39115 foos) or @samp{Qacme.bar} (for setting bars).
39118 The name of a query or set packet should be separated from any
39119 parameters by a @samp{:}; the parameters themselves should be
39120 separated by @samp{,} or @samp{;}. Stubs must be careful to match the
39121 full packet name, and check for a separator or the end of the packet,
39122 in case two packet names share a common prefix. New packets should not begin
39123 with @samp{qC}, @samp{qP}, or @samp{qL}@footnote{The @samp{qP} and @samp{qL}
39124 packets predate these conventions, and have arguments without any terminator
39125 for the packet name; we suspect they are in widespread use in places that
39126 are difficult to upgrade. The @samp{qC} packet has no arguments, but some
39127 existing stubs (e.g.@: RedBoot) are known to not check for the end of the
39130 Like the descriptions of the other packets, each description here
39131 has a template showing the packet's overall syntax, followed by an
39132 explanation of the packet's meaning. We include spaces in some of the
39133 templates for clarity; these are not part of the packet's syntax. No
39134 @value{GDBN} packet uses spaces to separate its components.
39136 Here are the currently defined query and set packets:
39142 Turn on or off the agent as a helper to perform some debugging operations
39143 delegated from @value{GDBN} (@pxref{Control Agent}).
39145 @item QAllow:@var{op}:@var{val}@dots{}
39146 @cindex @samp{QAllow} packet
39147 Specify which operations @value{GDBN} expects to request of the
39148 target, as a semicolon-separated list of operation name and value
39149 pairs. Possible values for @var{op} include @samp{WriteReg},
39150 @samp{WriteMem}, @samp{InsertBreak}, @samp{InsertTrace},
39151 @samp{InsertFastTrace}, and @samp{Stop}. @var{val} is either 0,
39152 indicating that @value{GDBN} will not request the operation, or 1,
39153 indicating that it may. (The target can then use this to set up its
39154 own internals optimally, for instance if the debugger never expects to
39155 insert breakpoints, it may not need to install its own trap handler.)
39158 @cindex current thread, remote request
39159 @cindex @samp{qC} packet
39160 Return the current thread ID.
39164 @item QC @var{thread-id}
39165 Where @var{thread-id} is a thread ID as documented in
39166 @ref{thread-id syntax}.
39167 @item @r{(anything else)}
39168 Any other reply implies the old thread ID.
39171 @item qCRC:@var{addr},@var{length}
39172 @cindex CRC of memory block, remote request
39173 @cindex @samp{qCRC} packet
39174 @anchor{qCRC packet}
39175 Compute the CRC checksum of a block of memory using CRC-32 defined in
39176 IEEE 802.3. The CRC is computed byte at a time, taking the most
39177 significant bit of each byte first. The initial pattern code
39178 @code{0xffffffff} is used to ensure leading zeros affect the CRC.
39180 @emph{Note:} This is the same CRC used in validating separate debug
39181 files (@pxref{Separate Debug Files, , Debugging Information in Separate
39182 Files}). However the algorithm is slightly different. When validating
39183 separate debug files, the CRC is computed taking the @emph{least}
39184 significant bit of each byte first, and the final result is inverted to
39185 detect trailing zeros.
39190 An error (such as memory fault)
39191 @item C @var{crc32}
39192 The specified memory region's checksum is @var{crc32}.
39195 @item QDisableRandomization:@var{value}
39196 @cindex disable address space randomization, remote request
39197 @cindex @samp{QDisableRandomization} packet
39198 Some target operating systems will randomize the virtual address space
39199 of the inferior process as a security feature, but provide a feature
39200 to disable such randomization, e.g.@: to allow for a more deterministic
39201 debugging experience. On such systems, this packet with a @var{value}
39202 of 1 directs the target to disable address space randomization for
39203 processes subsequently started via @samp{vRun} packets, while a packet
39204 with a @var{value} of 0 tells the target to enable address space
39207 This packet is only available in extended mode (@pxref{extended mode}).
39212 The request succeeded.
39215 An error occurred. The error number @var{nn} is given as hex digits.
39218 An empty reply indicates that @samp{QDisableRandomization} is not supported
39222 This packet is not probed by default; the remote stub must request it,
39223 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
39224 This should only be done on targets that actually support disabling
39225 address space randomization.
39227 @item QStartupWithShell:@var{value}
39228 @cindex startup with shell, remote request
39229 @cindex @samp{QStartupWithShell} packet
39230 On UNIX-like targets, it is possible to start the inferior using a
39231 shell program. This is the default behavior on both @value{GDBN} and
39232 @command{gdbserver} (@pxref{set startup-with-shell}). This packet is
39233 used to inform @command{gdbserver} whether it should start the
39234 inferior using a shell or not.
39236 If @var{value} is @samp{0}, @command{gdbserver} will not use a shell
39237 to start the inferior. If @var{value} is @samp{1},
39238 @command{gdbserver} will use a shell to start the inferior. All other
39239 values are considered an error.
39241 This packet is only available in extended mode (@pxref{extended
39247 The request succeeded.
39250 An error occurred. The error number @var{nn} is given as hex digits.
39253 This packet is not probed by default; the remote stub must request it,
39254 by supplying an appropriate @samp{qSupported} response
39255 (@pxref{qSupported}). This should only be done on targets that
39256 actually support starting the inferior using a shell.
39258 Use of this packet is controlled by the @code{set startup-with-shell}
39259 command; @pxref{set startup-with-shell}.
39261 @item QEnvironmentHexEncoded:@var{hex-value}
39262 @anchor{QEnvironmentHexEncoded}
39263 @cindex set environment variable, remote request
39264 @cindex @samp{QEnvironmentHexEncoded} packet
39265 On UNIX-like targets, it is possible to set environment variables that
39266 will be passed to the inferior during the startup process. This
39267 packet is used to inform @command{gdbserver} of an environment
39268 variable that has been defined by the user on @value{GDBN} (@pxref{set
39271 The packet is composed by @var{hex-value}, an hex encoded
39272 representation of the @var{name=value} format representing an
39273 environment variable. The name of the environment variable is
39274 represented by @var{name}, and the value to be assigned to the
39275 environment variable is represented by @var{value}. If the variable
39276 has no value (i.e., the value is @code{null}), then @var{value} will
39279 This packet is only available in extended mode (@pxref{extended
39285 The request succeeded.
39288 This packet is not probed by default; the remote stub must request it,
39289 by supplying an appropriate @samp{qSupported} response
39290 (@pxref{qSupported}). This should only be done on targets that
39291 actually support passing environment variables to the starting
39294 This packet is related to the @code{set environment} command;
39295 @pxref{set environment}.
39297 @item QEnvironmentUnset:@var{hex-value}
39298 @anchor{QEnvironmentUnset}
39299 @cindex unset environment variable, remote request
39300 @cindex @samp{QEnvironmentUnset} packet
39301 On UNIX-like targets, it is possible to unset environment variables
39302 before starting the inferior in the remote target. This packet is
39303 used to inform @command{gdbserver} of an environment variable that has
39304 been unset by the user on @value{GDBN} (@pxref{unset environment}).
39306 The packet is composed by @var{hex-value}, an hex encoded
39307 representation of the name of the environment variable to be unset.
39309 This packet is only available in extended mode (@pxref{extended
39315 The request succeeded.
39318 This packet is not probed by default; the remote stub must request it,
39319 by supplying an appropriate @samp{qSupported} response
39320 (@pxref{qSupported}). This should only be done on targets that
39321 actually support passing environment variables to the starting
39324 This packet is related to the @code{unset environment} command;
39325 @pxref{unset environment}.
39327 @item QEnvironmentReset
39328 @anchor{QEnvironmentReset}
39329 @cindex reset environment, remote request
39330 @cindex @samp{QEnvironmentReset} packet
39331 On UNIX-like targets, this packet is used to reset the state of
39332 environment variables in the remote target before starting the
39333 inferior. In this context, reset means unsetting all environment
39334 variables that were previously set by the user (i.e., were not
39335 initially present in the environment). It is sent to
39336 @command{gdbserver} before the @samp{QEnvironmentHexEncoded}
39337 (@pxref{QEnvironmentHexEncoded}) and the @samp{QEnvironmentUnset}
39338 (@pxref{QEnvironmentUnset}) packets.
39340 This packet is only available in extended mode (@pxref{extended
39346 The request succeeded.
39349 This packet is not probed by default; the remote stub must request it,
39350 by supplying an appropriate @samp{qSupported} response
39351 (@pxref{qSupported}). This should only be done on targets that
39352 actually support passing environment variables to the starting
39355 @item QSetWorkingDir:@r{[}@var{directory}@r{]}
39356 @anchor{QSetWorkingDir packet}
39357 @cindex set working directory, remote request
39358 @cindex @samp{QSetWorkingDir} packet
39359 This packet is used to inform the remote server of the intended
39360 current working directory for programs that are going to be executed.
39362 The packet is composed by @var{directory}, an hex encoded
39363 representation of the directory that the remote inferior will use as
39364 its current working directory. If @var{directory} is an empty string,
39365 the remote server should reset the inferior's current working
39366 directory to its original, empty value.
39368 This packet is only available in extended mode (@pxref{extended
39374 The request succeeded.
39378 @itemx qsThreadInfo
39379 @cindex list active threads, remote request
39380 @cindex @samp{qfThreadInfo} packet
39381 @cindex @samp{qsThreadInfo} packet
39382 Obtain a list of all active thread IDs from the target (OS). Since there
39383 may be too many active threads to fit into one reply packet, this query
39384 works iteratively: it may require more than one query/reply sequence to
39385 obtain the entire list of threads. The first query of the sequence will
39386 be the @samp{qfThreadInfo} query; subsequent queries in the
39387 sequence will be the @samp{qsThreadInfo} query.
39389 NOTE: This packet replaces the @samp{qL} query (see below).
39393 @item m @var{thread-id}
39395 @item m @var{thread-id},@var{thread-id}@dots{}
39396 a comma-separated list of thread IDs
39398 (lower case letter @samp{L}) denotes end of list.
39401 In response to each query, the target will reply with a list of one or
39402 more thread IDs, separated by commas.
39403 @value{GDBN} will respond to each reply with a request for more thread
39404 ids (using the @samp{qs} form of the query), until the target responds
39405 with @samp{l} (lower-case ell, for @dfn{last}).
39406 Refer to @ref{thread-id syntax}, for the format of the @var{thread-id}
39409 @emph{Note: @value{GDBN} will send the @code{qfThreadInfo} query during the
39410 initial connection with the remote target, and the very first thread ID
39411 mentioned in the reply will be stopped by @value{GDBN} in a subsequent
39412 message. Therefore, the stub should ensure that the first thread ID in
39413 the @code{qfThreadInfo} reply is suitable for being stopped by @value{GDBN}.}
39415 @item qGetTLSAddr:@var{thread-id},@var{offset},@var{lm}
39416 @cindex get thread-local storage address, remote request
39417 @cindex @samp{qGetTLSAddr} packet
39418 Fetch the address associated with thread local storage specified
39419 by @var{thread-id}, @var{offset}, and @var{lm}.
39421 @var{thread-id} is the thread ID associated with the
39422 thread for which to fetch the TLS address. @xref{thread-id syntax}.
39424 @var{offset} is the (big endian, hex encoded) offset associated with the
39425 thread local variable. (This offset is obtained from the debug
39426 information associated with the variable.)
39428 @var{lm} is the (big endian, hex encoded) OS/ABI-specific encoding of the
39429 load module associated with the thread local storage. For example,
39430 a @sc{gnu}/Linux system will pass the link map address of the shared
39431 object associated with the thread local storage under consideration.
39432 Other operating environments may choose to represent the load module
39433 differently, so the precise meaning of this parameter will vary.
39437 @item @var{XX}@dots{}
39438 Hex encoded (big endian) bytes representing the address of the thread
39439 local storage requested.
39442 An error occurred. The error number @var{nn} is given as hex digits.
39445 An empty reply indicates that @samp{qGetTLSAddr} is not supported by the stub.
39448 @item qGetTIBAddr:@var{thread-id}
39449 @cindex get thread information block address
39450 @cindex @samp{qGetTIBAddr} packet
39451 Fetch address of the Windows OS specific Thread Information Block.
39453 @var{thread-id} is the thread ID associated with the thread.
39457 @item @var{XX}@dots{}
39458 Hex encoded (big endian) bytes representing the linear address of the
39459 thread information block.
39462 An error occured. This means that either the thread was not found, or the
39463 address could not be retrieved.
39466 An empty reply indicates that @samp{qGetTIBAddr} is not supported by the stub.
39469 @item qL @var{startflag} @var{threadcount} @var{nextthread}
39470 Obtain thread information from RTOS. Where: @var{startflag} (one hex
39471 digit) is one to indicate the first query and zero to indicate a
39472 subsequent query; @var{threadcount} (two hex digits) is the maximum
39473 number of threads the response packet can contain; and @var{nextthread}
39474 (eight hex digits), for subsequent queries (@var{startflag} is zero), is
39475 returned in the response as @var{argthread}.
39477 Don't use this packet; use the @samp{qfThreadInfo} query instead (see above).
39481 @item qM @var{count} @var{done} @var{argthread} @var{thread}@dots{}
39482 Where: @var{count} (two hex digits) is the number of threads being
39483 returned; @var{done} (one hex digit) is zero to indicate more threads
39484 and one indicates no further threads; @var{argthreadid} (eight hex
39485 digits) is @var{nextthread} from the request packet; @var{thread}@dots{}
39486 is a sequence of thread IDs, @var{threadid} (eight hex
39487 digits), from the target. See @code{remote.c:parse_threadlist_response()}.
39491 @cindex section offsets, remote request
39492 @cindex @samp{qOffsets} packet
39493 Get section offsets that the target used when relocating the downloaded
39498 @item Text=@var{xxx};Data=@var{yyy}@r{[};Bss=@var{zzz}@r{]}
39499 Relocate the @code{Text} section by @var{xxx} from its original address.
39500 Relocate the @code{Data} section by @var{yyy} from its original address.
39501 If the object file format provides segment information (e.g.@: @sc{elf}
39502 @samp{PT_LOAD} program headers), @value{GDBN} will relocate entire
39503 segments by the supplied offsets.
39505 @emph{Note: while a @code{Bss} offset may be included in the response,
39506 @value{GDBN} ignores this and instead applies the @code{Data} offset
39507 to the @code{Bss} section.}
39509 @item TextSeg=@var{xxx}@r{[};DataSeg=@var{yyy}@r{]}
39510 Relocate the first segment of the object file, which conventionally
39511 contains program code, to a starting address of @var{xxx}. If
39512 @samp{DataSeg} is specified, relocate the second segment, which
39513 conventionally contains modifiable data, to a starting address of
39514 @var{yyy}. @value{GDBN} will report an error if the object file
39515 does not contain segment information, or does not contain at least
39516 as many segments as mentioned in the reply. Extra segments are
39517 kept at fixed offsets relative to the last relocated segment.
39520 @item qP @var{mode} @var{thread-id}
39521 @cindex thread information, remote request
39522 @cindex @samp{qP} packet
39523 Returns information on @var{thread-id}. Where: @var{mode} is a hex
39524 encoded 32 bit mode; @var{thread-id} is a thread ID
39525 (@pxref{thread-id syntax}).
39527 Don't use this packet; use the @samp{qThreadExtraInfo} query instead
39530 Reply: see @code{remote.c:remote_unpack_thread_info_response()}.
39534 @cindex non-stop mode, remote request
39535 @cindex @samp{QNonStop} packet
39537 Enter non-stop (@samp{QNonStop:1}) or all-stop (@samp{QNonStop:0}) mode.
39538 @xref{Remote Non-Stop}, for more information.
39543 The request succeeded.
39546 An error occurred. The error number @var{nn} is given as hex digits.
39549 An empty reply indicates that @samp{QNonStop} is not supported by
39553 This packet is not probed by default; the remote stub must request it,
39554 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
39555 Use of this packet is controlled by the @code{set non-stop} command;
39556 @pxref{Non-Stop Mode}.
39558 @item QCatchSyscalls:1 @r{[};@var{sysno}@r{]}@dots{}
39559 @itemx QCatchSyscalls:0
39560 @cindex catch syscalls from inferior, remote request
39561 @cindex @samp{QCatchSyscalls} packet
39562 @anchor{QCatchSyscalls}
39563 Enable (@samp{QCatchSyscalls:1}) or disable (@samp{QCatchSyscalls:0})
39564 catching syscalls from the inferior process.
39566 For @samp{QCatchSyscalls:1}, each listed syscall @var{sysno} (encoded
39567 in hex) should be reported to @value{GDBN}. If no syscall @var{sysno}
39568 is listed, every system call should be reported.
39570 Note that if a syscall not in the list is reported, @value{GDBN} will
39571 still filter the event according to its own list from all corresponding
39572 @code{catch syscall} commands. However, it is more efficient to only
39573 report the requested syscalls.
39575 Multiple @samp{QCatchSyscalls:1} packets do not combine; any earlier
39576 @samp{QCatchSyscalls:1} list is completely replaced by the new list.
39578 If the inferior process execs, the state of @samp{QCatchSyscalls} is
39579 kept for the new process too. On targets where exec may affect syscall
39580 numbers, for example with exec between 32 and 64-bit processes, the
39581 client should send a new packet with the new syscall list.
39586 The request succeeded.
39589 An error occurred. @var{nn} are hex digits.
39592 An empty reply indicates that @samp{QCatchSyscalls} is not supported by
39596 Use of this packet is controlled by the @code{set remote catch-syscalls}
39597 command (@pxref{Remote Configuration, set remote catch-syscalls}).
39598 This packet is not probed by default; the remote stub must request it,
39599 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
39601 @item QPassSignals: @var{signal} @r{[};@var{signal}@r{]}@dots{}
39602 @cindex pass signals to inferior, remote request
39603 @cindex @samp{QPassSignals} packet
39604 @anchor{QPassSignals}
39605 Each listed @var{signal} should be passed directly to the inferior process.
39606 Signals are numbered identically to continue packets and stop replies
39607 (@pxref{Stop Reply Packets}). Each @var{signal} list item should be
39608 strictly greater than the previous item. These signals do not need to stop
39609 the inferior, or be reported to @value{GDBN}. All other signals should be
39610 reported to @value{GDBN}. Multiple @samp{QPassSignals} packets do not
39611 combine; any earlier @samp{QPassSignals} list is completely replaced by the
39612 new list. This packet improves performance when using @samp{handle
39613 @var{signal} nostop noprint pass}.
39618 The request succeeded.
39621 An error occurred. The error number @var{nn} is given as hex digits.
39624 An empty reply indicates that @samp{QPassSignals} is not supported by
39628 Use of this packet is controlled by the @code{set remote pass-signals}
39629 command (@pxref{Remote Configuration, set remote pass-signals}).
39630 This packet is not probed by default; the remote stub must request it,
39631 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
39633 @item QProgramSignals: @var{signal} @r{[};@var{signal}@r{]}@dots{}
39634 @cindex signals the inferior may see, remote request
39635 @cindex @samp{QProgramSignals} packet
39636 @anchor{QProgramSignals}
39637 Each listed @var{signal} may be delivered to the inferior process.
39638 Others should be silently discarded.
39640 In some cases, the remote stub may need to decide whether to deliver a
39641 signal to the program or not without @value{GDBN} involvement. One
39642 example of that is while detaching --- the program's threads may have
39643 stopped for signals that haven't yet had a chance of being reported to
39644 @value{GDBN}, and so the remote stub can use the signal list specified
39645 by this packet to know whether to deliver or ignore those pending
39648 This does not influence whether to deliver a signal as requested by a
39649 resumption packet (@pxref{vCont packet}).
39651 Signals are numbered identically to continue packets and stop replies
39652 (@pxref{Stop Reply Packets}). Each @var{signal} list item should be
39653 strictly greater than the previous item. Multiple
39654 @samp{QProgramSignals} packets do not combine; any earlier
39655 @samp{QProgramSignals} list is completely replaced by the new list.
39660 The request succeeded.
39663 An error occurred. The error number @var{nn} is given as hex digits.
39666 An empty reply indicates that @samp{QProgramSignals} is not supported
39670 Use of this packet is controlled by the @code{set remote program-signals}
39671 command (@pxref{Remote Configuration, set remote program-signals}).
39672 This packet is not probed by default; the remote stub must request it,
39673 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
39675 @anchor{QThreadEvents}
39676 @item QThreadEvents:1
39677 @itemx QThreadEvents:0
39678 @cindex thread create/exit events, remote request
39679 @cindex @samp{QThreadEvents} packet
39681 Enable (@samp{QThreadEvents:1}) or disable (@samp{QThreadEvents:0})
39682 reporting of thread create and exit events. @xref{thread create
39683 event}, for the reply specifications. For example, this is used in
39684 non-stop mode when @value{GDBN} stops a set of threads and
39685 synchronously waits for the their corresponding stop replies. Without
39686 exit events, if one of the threads exits, @value{GDBN} would hang
39687 forever not knowing that it should no longer expect a stop for that
39688 same thread. @value{GDBN} does not enable this feature unless the
39689 stub reports that it supports it by including @samp{QThreadEvents+} in
39690 its @samp{qSupported} reply.
39695 The request succeeded.
39698 An error occurred. The error number @var{nn} is given as hex digits.
39701 An empty reply indicates that @samp{QThreadEvents} is not supported by
39705 Use of this packet is controlled by the @code{set remote thread-events}
39706 command (@pxref{Remote Configuration, set remote thread-events}).
39708 @item qRcmd,@var{command}
39709 @cindex execute remote command, remote request
39710 @cindex @samp{qRcmd} packet
39711 @var{command} (hex encoded) is passed to the local interpreter for
39712 execution. Invalid commands should be reported using the output
39713 string. Before the final result packet, the target may also respond
39714 with a number of intermediate @samp{O@var{output}} console output
39715 packets. @emph{Implementors should note that providing access to a
39716 stubs's interpreter may have security implications}.
39721 A command response with no output.
39723 A command response with the hex encoded output string @var{OUTPUT}.
39725 Indicate a badly formed request.
39727 An empty reply indicates that @samp{qRcmd} is not recognized.
39730 (Note that the @code{qRcmd} packet's name is separated from the
39731 command by a @samp{,}, not a @samp{:}, contrary to the naming
39732 conventions above. Please don't use this packet as a model for new
39735 @item qSearch:memory:@var{address};@var{length};@var{search-pattern}
39736 @cindex searching memory, in remote debugging
39738 @cindex @samp{qSearch:memory} packet
39740 @cindex @samp{qSearch memory} packet
39741 @anchor{qSearch memory}
39742 Search @var{length} bytes at @var{address} for @var{search-pattern}.
39743 Both @var{address} and @var{length} are encoded in hex;
39744 @var{search-pattern} is a sequence of bytes, also hex encoded.
39749 The pattern was not found.
39751 The pattern was found at @var{address}.
39753 A badly formed request or an error was encountered while searching memory.
39755 An empty reply indicates that @samp{qSearch:memory} is not recognized.
39758 @item QStartNoAckMode
39759 @cindex @samp{QStartNoAckMode} packet
39760 @anchor{QStartNoAckMode}
39761 Request that the remote stub disable the normal @samp{+}/@samp{-}
39762 protocol acknowledgments (@pxref{Packet Acknowledgment}).
39767 The stub has switched to no-acknowledgment mode.
39768 @value{GDBN} acknowledges this reponse,
39769 but neither the stub nor @value{GDBN} shall send or expect further
39770 @samp{+}/@samp{-} acknowledgments in the current connection.
39772 An empty reply indicates that the stub does not support no-acknowledgment mode.
39775 @item qSupported @r{[}:@var{gdbfeature} @r{[};@var{gdbfeature}@r{]}@dots{} @r{]}
39776 @cindex supported packets, remote query
39777 @cindex features of the remote protocol
39778 @cindex @samp{qSupported} packet
39779 @anchor{qSupported}
39780 Tell the remote stub about features supported by @value{GDBN}, and
39781 query the stub for features it supports. This packet allows
39782 @value{GDBN} and the remote stub to take advantage of each others'
39783 features. @samp{qSupported} also consolidates multiple feature probes
39784 at startup, to improve @value{GDBN} performance---a single larger
39785 packet performs better than multiple smaller probe packets on
39786 high-latency links. Some features may enable behavior which must not
39787 be on by default, e.g.@: because it would confuse older clients or
39788 stubs. Other features may describe packets which could be
39789 automatically probed for, but are not. These features must be
39790 reported before @value{GDBN} will use them. This ``default
39791 unsupported'' behavior is not appropriate for all packets, but it
39792 helps to keep the initial connection time under control with new
39793 versions of @value{GDBN} which support increasing numbers of packets.
39797 @item @var{stubfeature} @r{[};@var{stubfeature}@r{]}@dots{}
39798 The stub supports or does not support each returned @var{stubfeature},
39799 depending on the form of each @var{stubfeature} (see below for the
39802 An empty reply indicates that @samp{qSupported} is not recognized,
39803 or that no features needed to be reported to @value{GDBN}.
39806 The allowed forms for each feature (either a @var{gdbfeature} in the
39807 @samp{qSupported} packet, or a @var{stubfeature} in the response)
39811 @item @var{name}=@var{value}
39812 The remote protocol feature @var{name} is supported, and associated
39813 with the specified @var{value}. The format of @var{value} depends
39814 on the feature, but it must not include a semicolon.
39816 The remote protocol feature @var{name} is supported, and does not
39817 need an associated value.
39819 The remote protocol feature @var{name} is not supported.
39821 The remote protocol feature @var{name} may be supported, and
39822 @value{GDBN} should auto-detect support in some other way when it is
39823 needed. This form will not be used for @var{gdbfeature} notifications,
39824 but may be used for @var{stubfeature} responses.
39827 Whenever the stub receives a @samp{qSupported} request, the
39828 supplied set of @value{GDBN} features should override any previous
39829 request. This allows @value{GDBN} to put the stub in a known
39830 state, even if the stub had previously been communicating with
39831 a different version of @value{GDBN}.
39833 The following values of @var{gdbfeature} (for the packet sent by @value{GDBN})
39838 This feature indicates whether @value{GDBN} supports multiprocess
39839 extensions to the remote protocol. @value{GDBN} does not use such
39840 extensions unless the stub also reports that it supports them by
39841 including @samp{multiprocess+} in its @samp{qSupported} reply.
39842 @xref{multiprocess extensions}, for details.
39845 This feature indicates that @value{GDBN} supports the XML target
39846 description. If the stub sees @samp{xmlRegisters=} with target
39847 specific strings separated by a comma, it will report register
39851 This feature indicates whether @value{GDBN} supports the
39852 @samp{qRelocInsn} packet (@pxref{Tracepoint Packets,,Relocate
39853 instruction reply packet}).
39856 This feature indicates whether @value{GDBN} supports the swbreak stop
39857 reason in stop replies. @xref{swbreak stop reason}, for details.
39860 This feature indicates whether @value{GDBN} supports the hwbreak stop
39861 reason in stop replies. @xref{swbreak stop reason}, for details.
39864 This feature indicates whether @value{GDBN} supports fork event
39865 extensions to the remote protocol. @value{GDBN} does not use such
39866 extensions unless the stub also reports that it supports them by
39867 including @samp{fork-events+} in its @samp{qSupported} reply.
39870 This feature indicates whether @value{GDBN} supports vfork event
39871 extensions to the remote protocol. @value{GDBN} does not use such
39872 extensions unless the stub also reports that it supports them by
39873 including @samp{vfork-events+} in its @samp{qSupported} reply.
39876 This feature indicates whether @value{GDBN} supports exec event
39877 extensions to the remote protocol. @value{GDBN} does not use such
39878 extensions unless the stub also reports that it supports them by
39879 including @samp{exec-events+} in its @samp{qSupported} reply.
39881 @item vContSupported
39882 This feature indicates whether @value{GDBN} wants to know the
39883 supported actions in the reply to @samp{vCont?} packet.
39886 Stubs should ignore any unknown values for
39887 @var{gdbfeature}. Any @value{GDBN} which sends a @samp{qSupported}
39888 packet supports receiving packets of unlimited length (earlier
39889 versions of @value{GDBN} may reject overly long responses). Additional values
39890 for @var{gdbfeature} may be defined in the future to let the stub take
39891 advantage of new features in @value{GDBN}, e.g.@: incompatible
39892 improvements in the remote protocol---the @samp{multiprocess} feature is
39893 an example of such a feature. The stub's reply should be independent
39894 of the @var{gdbfeature} entries sent by @value{GDBN}; first @value{GDBN}
39895 describes all the features it supports, and then the stub replies with
39896 all the features it supports.
39898 Similarly, @value{GDBN} will silently ignore unrecognized stub feature
39899 responses, as long as each response uses one of the standard forms.
39901 Some features are flags. A stub which supports a flag feature
39902 should respond with a @samp{+} form response. Other features
39903 require values, and the stub should respond with an @samp{=}
39906 Each feature has a default value, which @value{GDBN} will use if
39907 @samp{qSupported} is not available or if the feature is not mentioned
39908 in the @samp{qSupported} response. The default values are fixed; a
39909 stub is free to omit any feature responses that match the defaults.
39911 Not all features can be probed, but for those which can, the probing
39912 mechanism is useful: in some cases, a stub's internal
39913 architecture may not allow the protocol layer to know some information
39914 about the underlying target in advance. This is especially common in
39915 stubs which may be configured for multiple targets.
39917 These are the currently defined stub features and their properties:
39919 @multitable @columnfractions 0.35 0.2 0.12 0.2
39920 @c NOTE: The first row should be @headitem, but we do not yet require
39921 @c a new enough version of Texinfo (4.7) to use @headitem.
39923 @tab Value Required
39927 @item @samp{PacketSize}
39932 @item @samp{qXfer:auxv:read}
39937 @item @samp{qXfer:btrace:read}
39942 @item @samp{qXfer:btrace-conf:read}
39947 @item @samp{qXfer:exec-file:read}
39952 @item @samp{qXfer:features:read}
39957 @item @samp{qXfer:libraries:read}
39962 @item @samp{qXfer:libraries-svr4:read}
39967 @item @samp{augmented-libraries-svr4-read}
39972 @item @samp{qXfer:memory-map:read}
39977 @item @samp{qXfer:sdata:read}
39982 @item @samp{qXfer:spu:read}
39987 @item @samp{qXfer:spu:write}
39992 @item @samp{qXfer:siginfo:read}
39997 @item @samp{qXfer:siginfo:write}
40002 @item @samp{qXfer:threads:read}
40007 @item @samp{qXfer:traceframe-info:read}
40012 @item @samp{qXfer:uib:read}
40017 @item @samp{qXfer:fdpic:read}
40022 @item @samp{Qbtrace:off}
40027 @item @samp{Qbtrace:bts}
40032 @item @samp{Qbtrace:pt}
40037 @item @samp{Qbtrace-conf:bts:size}
40042 @item @samp{Qbtrace-conf:pt:size}
40047 @item @samp{QNonStop}
40052 @item @samp{QCatchSyscalls}
40057 @item @samp{QPassSignals}
40062 @item @samp{QStartNoAckMode}
40067 @item @samp{multiprocess}
40072 @item @samp{ConditionalBreakpoints}
40077 @item @samp{ConditionalTracepoints}
40082 @item @samp{ReverseContinue}
40087 @item @samp{ReverseStep}
40092 @item @samp{TracepointSource}
40097 @item @samp{QAgent}
40102 @item @samp{QAllow}
40107 @item @samp{QDisableRandomization}
40112 @item @samp{EnableDisableTracepoints}
40117 @item @samp{QTBuffer:size}
40122 @item @samp{tracenz}
40127 @item @samp{BreakpointCommands}
40132 @item @samp{swbreak}
40137 @item @samp{hwbreak}
40142 @item @samp{fork-events}
40147 @item @samp{vfork-events}
40152 @item @samp{exec-events}
40157 @item @samp{QThreadEvents}
40162 @item @samp{no-resumed}
40169 These are the currently defined stub features, in more detail:
40172 @cindex packet size, remote protocol
40173 @item PacketSize=@var{bytes}
40174 The remote stub can accept packets up to at least @var{bytes} in
40175 length. @value{GDBN} will send packets up to this size for bulk
40176 transfers, and will never send larger packets. This is a limit on the
40177 data characters in the packet, including the frame and checksum.
40178 There is no trailing NUL byte in a remote protocol packet; if the stub
40179 stores packets in a NUL-terminated format, it should allow an extra
40180 byte in its buffer for the NUL. If this stub feature is not supported,
40181 @value{GDBN} guesses based on the size of the @samp{g} packet response.
40183 @item qXfer:auxv:read
40184 The remote stub understands the @samp{qXfer:auxv:read} packet
40185 (@pxref{qXfer auxiliary vector read}).
40187 @item qXfer:btrace:read
40188 The remote stub understands the @samp{qXfer:btrace:read}
40189 packet (@pxref{qXfer btrace read}).
40191 @item qXfer:btrace-conf:read
40192 The remote stub understands the @samp{qXfer:btrace-conf:read}
40193 packet (@pxref{qXfer btrace-conf read}).
40195 @item qXfer:exec-file:read
40196 The remote stub understands the @samp{qXfer:exec-file:read} packet
40197 (@pxref{qXfer executable filename read}).
40199 @item qXfer:features:read
40200 The remote stub understands the @samp{qXfer:features:read} packet
40201 (@pxref{qXfer target description read}).
40203 @item qXfer:libraries:read
40204 The remote stub understands the @samp{qXfer:libraries:read} packet
40205 (@pxref{qXfer library list read}).
40207 @item qXfer:libraries-svr4:read
40208 The remote stub understands the @samp{qXfer:libraries-svr4:read} packet
40209 (@pxref{qXfer svr4 library list read}).
40211 @item augmented-libraries-svr4-read
40212 The remote stub understands the augmented form of the
40213 @samp{qXfer:libraries-svr4:read} packet
40214 (@pxref{qXfer svr4 library list read}).
40216 @item qXfer:memory-map:read
40217 The remote stub understands the @samp{qXfer:memory-map:read} packet
40218 (@pxref{qXfer memory map read}).
40220 @item qXfer:sdata:read
40221 The remote stub understands the @samp{qXfer:sdata:read} packet
40222 (@pxref{qXfer sdata read}).
40224 @item qXfer:spu:read
40225 The remote stub understands the @samp{qXfer:spu:read} packet
40226 (@pxref{qXfer spu read}).
40228 @item qXfer:spu:write
40229 The remote stub understands the @samp{qXfer:spu:write} packet
40230 (@pxref{qXfer spu write}).
40232 @item qXfer:siginfo:read
40233 The remote stub understands the @samp{qXfer:siginfo:read} packet
40234 (@pxref{qXfer siginfo read}).
40236 @item qXfer:siginfo:write
40237 The remote stub understands the @samp{qXfer:siginfo:write} packet
40238 (@pxref{qXfer siginfo write}).
40240 @item qXfer:threads:read
40241 The remote stub understands the @samp{qXfer:threads:read} packet
40242 (@pxref{qXfer threads read}).
40244 @item qXfer:traceframe-info:read
40245 The remote stub understands the @samp{qXfer:traceframe-info:read}
40246 packet (@pxref{qXfer traceframe info read}).
40248 @item qXfer:uib:read
40249 The remote stub understands the @samp{qXfer:uib:read}
40250 packet (@pxref{qXfer unwind info block}).
40252 @item qXfer:fdpic:read
40253 The remote stub understands the @samp{qXfer:fdpic:read}
40254 packet (@pxref{qXfer fdpic loadmap read}).
40257 The remote stub understands the @samp{QNonStop} packet
40258 (@pxref{QNonStop}).
40260 @item QCatchSyscalls
40261 The remote stub understands the @samp{QCatchSyscalls} packet
40262 (@pxref{QCatchSyscalls}).
40265 The remote stub understands the @samp{QPassSignals} packet
40266 (@pxref{QPassSignals}).
40268 @item QStartNoAckMode
40269 The remote stub understands the @samp{QStartNoAckMode} packet and
40270 prefers to operate in no-acknowledgment mode. @xref{Packet Acknowledgment}.
40273 @anchor{multiprocess extensions}
40274 @cindex multiprocess extensions, in remote protocol
40275 The remote stub understands the multiprocess extensions to the remote
40276 protocol syntax. The multiprocess extensions affect the syntax of
40277 thread IDs in both packets and replies (@pxref{thread-id syntax}), and
40278 add process IDs to the @samp{D} packet and @samp{W} and @samp{X}
40279 replies. Note that reporting this feature indicates support for the
40280 syntactic extensions only, not that the stub necessarily supports
40281 debugging of more than one process at a time. The stub must not use
40282 multiprocess extensions in packet replies unless @value{GDBN} has also
40283 indicated it supports them in its @samp{qSupported} request.
40285 @item qXfer:osdata:read
40286 The remote stub understands the @samp{qXfer:osdata:read} packet
40287 ((@pxref{qXfer osdata read}).
40289 @item ConditionalBreakpoints
40290 The target accepts and implements evaluation of conditional expressions
40291 defined for breakpoints. The target will only report breakpoint triggers
40292 when such conditions are true (@pxref{Conditions, ,Break Conditions}).
40294 @item ConditionalTracepoints
40295 The remote stub accepts and implements conditional expressions defined
40296 for tracepoints (@pxref{Tracepoint Conditions}).
40298 @item ReverseContinue
40299 The remote stub accepts and implements the reverse continue packet
40303 The remote stub accepts and implements the reverse step packet
40306 @item TracepointSource
40307 The remote stub understands the @samp{QTDPsrc} packet that supplies
40308 the source form of tracepoint definitions.
40311 The remote stub understands the @samp{QAgent} packet.
40314 The remote stub understands the @samp{QAllow} packet.
40316 @item QDisableRandomization
40317 The remote stub understands the @samp{QDisableRandomization} packet.
40319 @item StaticTracepoint
40320 @cindex static tracepoints, in remote protocol
40321 The remote stub supports static tracepoints.
40323 @item InstallInTrace
40324 @anchor{install tracepoint in tracing}
40325 The remote stub supports installing tracepoint in tracing.
40327 @item EnableDisableTracepoints
40328 The remote stub supports the @samp{QTEnable} (@pxref{QTEnable}) and
40329 @samp{QTDisable} (@pxref{QTDisable}) packets that allow tracepoints
40330 to be enabled and disabled while a trace experiment is running.
40332 @item QTBuffer:size
40333 The remote stub supports the @samp{QTBuffer:size} (@pxref{QTBuffer-size})
40334 packet that allows to change the size of the trace buffer.
40337 @cindex string tracing, in remote protocol
40338 The remote stub supports the @samp{tracenz} bytecode for collecting strings.
40339 See @ref{Bytecode Descriptions} for details about the bytecode.
40341 @item BreakpointCommands
40342 @cindex breakpoint commands, in remote protocol
40343 The remote stub supports running a breakpoint's command list itself,
40344 rather than reporting the hit to @value{GDBN}.
40347 The remote stub understands the @samp{Qbtrace:off} packet.
40350 The remote stub understands the @samp{Qbtrace:bts} packet.
40353 The remote stub understands the @samp{Qbtrace:pt} packet.
40355 @item Qbtrace-conf:bts:size
40356 The remote stub understands the @samp{Qbtrace-conf:bts:size} packet.
40358 @item Qbtrace-conf:pt:size
40359 The remote stub understands the @samp{Qbtrace-conf:pt:size} packet.
40362 The remote stub reports the @samp{swbreak} stop reason for memory
40366 The remote stub reports the @samp{hwbreak} stop reason for hardware
40370 The remote stub reports the @samp{fork} stop reason for fork events.
40373 The remote stub reports the @samp{vfork} stop reason for vfork events
40374 and vforkdone events.
40377 The remote stub reports the @samp{exec} stop reason for exec events.
40379 @item vContSupported
40380 The remote stub reports the supported actions in the reply to
40381 @samp{vCont?} packet.
40383 @item QThreadEvents
40384 The remote stub understands the @samp{QThreadEvents} packet.
40387 The remote stub reports the @samp{N} stop reply.
40392 @cindex symbol lookup, remote request
40393 @cindex @samp{qSymbol} packet
40394 Notify the target that @value{GDBN} is prepared to serve symbol lookup
40395 requests. Accept requests from the target for the values of symbols.
40400 The target does not need to look up any (more) symbols.
40401 @item qSymbol:@var{sym_name}
40402 The target requests the value of symbol @var{sym_name} (hex encoded).
40403 @value{GDBN} may provide the value by using the
40404 @samp{qSymbol:@var{sym_value}:@var{sym_name}} message, described
40408 @item qSymbol:@var{sym_value}:@var{sym_name}
40409 Set the value of @var{sym_name} to @var{sym_value}.
40411 @var{sym_name} (hex encoded) is the name of a symbol whose value the
40412 target has previously requested.
40414 @var{sym_value} (hex) is the value for symbol @var{sym_name}. If
40415 @value{GDBN} cannot supply a value for @var{sym_name}, then this field
40421 The target does not need to look up any (more) symbols.
40422 @item qSymbol:@var{sym_name}
40423 The target requests the value of a new symbol @var{sym_name} (hex
40424 encoded). @value{GDBN} will continue to supply the values of symbols
40425 (if available), until the target ceases to request them.
40430 @itemx QTDisconnected
40437 @itemx qTMinFTPILen
40439 @xref{Tracepoint Packets}.
40441 @item qThreadExtraInfo,@var{thread-id}
40442 @cindex thread attributes info, remote request
40443 @cindex @samp{qThreadExtraInfo} packet
40444 Obtain from the target OS a printable string description of thread
40445 attributes for the thread @var{thread-id}; see @ref{thread-id syntax},
40446 for the forms of @var{thread-id}. This
40447 string may contain anything that the target OS thinks is interesting
40448 for @value{GDBN} to tell the user about the thread. The string is
40449 displayed in @value{GDBN}'s @code{info threads} display. Some
40450 examples of possible thread extra info strings are @samp{Runnable}, or
40451 @samp{Blocked on Mutex}.
40455 @item @var{XX}@dots{}
40456 Where @samp{@var{XX}@dots{}} is a hex encoding of @sc{ascii} data,
40457 comprising the printable string containing the extra information about
40458 the thread's attributes.
40461 (Note that the @code{qThreadExtraInfo} packet's name is separated from
40462 the command by a @samp{,}, not a @samp{:}, contrary to the naming
40463 conventions above. Please don't use this packet as a model for new
40482 @xref{Tracepoint Packets}.
40484 @item qXfer:@var{object}:read:@var{annex}:@var{offset},@var{length}
40485 @cindex read special object, remote request
40486 @cindex @samp{qXfer} packet
40487 @anchor{qXfer read}
40488 Read uninterpreted bytes from the target's special data area
40489 identified by the keyword @var{object}. Request @var{length} bytes
40490 starting at @var{offset} bytes into the data. The content and
40491 encoding of @var{annex} is specific to @var{object}; it can supply
40492 additional details about what data to access.
40497 Data @var{data} (@pxref{Binary Data}) has been read from the
40498 target. There may be more data at a higher address (although
40499 it is permitted to return @samp{m} even for the last valid
40500 block of data, as long as at least one byte of data was read).
40501 It is possible for @var{data} to have fewer bytes than the @var{length} in the
40505 Data @var{data} (@pxref{Binary Data}) has been read from the target.
40506 There is no more data to be read. It is possible for @var{data} to
40507 have fewer bytes than the @var{length} in the request.
40510 The @var{offset} in the request is at the end of the data.
40511 There is no more data to be read.
40514 The request was malformed, or @var{annex} was invalid.
40517 The offset was invalid, or there was an error encountered reading the data.
40518 The @var{nn} part is a hex-encoded @code{errno} value.
40521 An empty reply indicates the @var{object} string was not recognized by
40522 the stub, or that the object does not support reading.
40525 Here are the specific requests of this form defined so far. All the
40526 @samp{qXfer:@var{object}:read:@dots{}} requests use the same reply
40527 formats, listed above.
40530 @item qXfer:auxv:read::@var{offset},@var{length}
40531 @anchor{qXfer auxiliary vector read}
40532 Access the target's @dfn{auxiliary vector}. @xref{OS Information,
40533 auxiliary vector}. Note @var{annex} must be empty.
40535 This packet is not probed by default; the remote stub must request it,
40536 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
40538 @item qXfer:btrace:read:@var{annex}:@var{offset},@var{length}
40539 @anchor{qXfer btrace read}
40541 Return a description of the current branch trace.
40542 @xref{Branch Trace Format}. The annex part of the generic @samp{qXfer}
40543 packet may have one of the following values:
40547 Returns all available branch trace.
40550 Returns all available branch trace if the branch trace changed since
40551 the last read request.
40554 Returns the new branch trace since the last read request. Adds a new
40555 block to the end of the trace that begins at zero and ends at the source
40556 location of the first branch in the trace buffer. This extra block is
40557 used to stitch traces together.
40559 If the trace buffer overflowed, returns an error indicating the overflow.
40562 This packet is not probed by default; the remote stub must request it
40563 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
40565 @item qXfer:btrace-conf:read::@var{offset},@var{length}
40566 @anchor{qXfer btrace-conf read}
40568 Return a description of the current branch trace configuration.
40569 @xref{Branch Trace Configuration Format}.
40571 This packet is not probed by default; the remote stub must request it
40572 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
40574 @item qXfer:exec-file:read:@var{annex}:@var{offset},@var{length}
40575 @anchor{qXfer executable filename read}
40576 Return the full absolute name of the file that was executed to create
40577 a process running on the remote system. The annex specifies the
40578 numeric process ID of the process to query, encoded as a hexadecimal
40579 number. If the annex part is empty the remote stub should return the
40580 filename corresponding to the currently executing process.
40582 This packet is not probed by default; the remote stub must request it,
40583 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
40585 @item qXfer:features:read:@var{annex}:@var{offset},@var{length}
40586 @anchor{qXfer target description read}
40587 Access the @dfn{target description}. @xref{Target Descriptions}. The
40588 annex specifies which XML document to access. The main description is
40589 always loaded from the @samp{target.xml} annex.
40591 This packet is not probed by default; the remote stub must request it,
40592 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
40594 @item qXfer:libraries:read:@var{annex}:@var{offset},@var{length}
40595 @anchor{qXfer library list read}
40596 Access the target's list of loaded libraries. @xref{Library List Format}.
40597 The annex part of the generic @samp{qXfer} packet must be empty
40598 (@pxref{qXfer read}).
40600 Targets which maintain a list of libraries in the program's memory do
40601 not need to implement this packet; it is designed for platforms where
40602 the operating system manages the list of loaded libraries.
40604 This packet is not probed by default; the remote stub must request it,
40605 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
40607 @item qXfer:libraries-svr4:read:@var{annex}:@var{offset},@var{length}
40608 @anchor{qXfer svr4 library list read}
40609 Access the target's list of loaded libraries when the target is an SVR4
40610 platform. @xref{Library List Format for SVR4 Targets}. The annex part
40611 of the generic @samp{qXfer} packet must be empty unless the remote
40612 stub indicated it supports the augmented form of this packet
40613 by supplying an appropriate @samp{qSupported} response
40614 (@pxref{qXfer read}, @ref{qSupported}).
40616 This packet is optional for better performance on SVR4 targets.
40617 @value{GDBN} uses memory read packets to read the SVR4 library list otherwise.
40619 This packet is not probed by default; the remote stub must request it,
40620 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
40622 If the remote stub indicates it supports the augmented form of this
40623 packet then the annex part of the generic @samp{qXfer} packet may
40624 contain a semicolon-separated list of @samp{@var{name}=@var{value}}
40625 arguments. The currently supported arguments are:
40628 @item start=@var{address}
40629 A hexadecimal number specifying the address of the @samp{struct
40630 link_map} to start reading the library list from. If unset or zero
40631 then the first @samp{struct link_map} in the library list will be
40632 chosen as the starting point.
40634 @item prev=@var{address}
40635 A hexadecimal number specifying the address of the @samp{struct
40636 link_map} immediately preceding the @samp{struct link_map}
40637 specified by the @samp{start} argument. If unset or zero then
40638 the remote stub will expect that no @samp{struct link_map}
40639 exists prior to the starting point.
40643 Arguments that are not understood by the remote stub will be silently
40646 @item qXfer:memory-map:read::@var{offset},@var{length}
40647 @anchor{qXfer memory map read}
40648 Access the target's @dfn{memory-map}. @xref{Memory Map Format}. The
40649 annex part of the generic @samp{qXfer} packet must be empty
40650 (@pxref{qXfer read}).
40652 This packet is not probed by default; the remote stub must request it,
40653 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
40655 @item qXfer:sdata:read::@var{offset},@var{length}
40656 @anchor{qXfer sdata read}
40658 Read contents of the extra collected static tracepoint marker
40659 information. The annex part of the generic @samp{qXfer} packet must
40660 be empty (@pxref{qXfer read}). @xref{Tracepoint Actions,,Tracepoint
40663 This packet is not probed by default; the remote stub must request it,
40664 by supplying an appropriate @samp{qSupported} response
40665 (@pxref{qSupported}).
40667 @item qXfer:siginfo:read::@var{offset},@var{length}
40668 @anchor{qXfer siginfo read}
40669 Read contents of the extra signal information on the target
40670 system. The annex part of the generic @samp{qXfer} packet must be
40671 empty (@pxref{qXfer read}).
40673 This packet is not probed by default; the remote stub must request it,
40674 by supplying an appropriate @samp{qSupported} response
40675 (@pxref{qSupported}).
40677 @item qXfer:spu:read:@var{annex}:@var{offset},@var{length}
40678 @anchor{qXfer spu read}
40679 Read contents of an @code{spufs} file on the target system. The
40680 annex specifies which file to read; it must be of the form
40681 @file{@var{id}/@var{name}}, where @var{id} specifies an SPU context ID
40682 in the target process, and @var{name} identifes the @code{spufs} file
40683 in that context to be accessed.
40685 This packet is not probed by default; the remote stub must request it,
40686 by supplying an appropriate @samp{qSupported} response
40687 (@pxref{qSupported}).
40689 @item qXfer:threads:read::@var{offset},@var{length}
40690 @anchor{qXfer threads read}
40691 Access the list of threads on target. @xref{Thread List Format}. The
40692 annex part of the generic @samp{qXfer} packet must be empty
40693 (@pxref{qXfer read}).
40695 This packet is not probed by default; the remote stub must request it,
40696 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
40698 @item qXfer:traceframe-info:read::@var{offset},@var{length}
40699 @anchor{qXfer traceframe info read}
40701 Return a description of the current traceframe's contents.
40702 @xref{Traceframe Info Format}. The annex part of the generic
40703 @samp{qXfer} packet must be empty (@pxref{qXfer read}).
40705 This packet is not probed by default; the remote stub must request it,
40706 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
40708 @item qXfer:uib:read:@var{pc}:@var{offset},@var{length}
40709 @anchor{qXfer unwind info block}
40711 Return the unwind information block for @var{pc}. This packet is used
40712 on OpenVMS/ia64 to ask the kernel unwind information.
40714 This packet is not probed by default.
40716 @item qXfer:fdpic:read:@var{annex}:@var{offset},@var{length}
40717 @anchor{qXfer fdpic loadmap read}
40718 Read contents of @code{loadmap}s on the target system. The
40719 annex, either @samp{exec} or @samp{interp}, specifies which @code{loadmap},
40720 executable @code{loadmap} or interpreter @code{loadmap} to read.
40722 This packet is not probed by default; the remote stub must request it,
40723 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
40725 @item qXfer:osdata:read::@var{offset},@var{length}
40726 @anchor{qXfer osdata read}
40727 Access the target's @dfn{operating system information}.
40728 @xref{Operating System Information}.
40732 @item qXfer:@var{object}:write:@var{annex}:@var{offset}:@var{data}@dots{}
40733 @cindex write data into object, remote request
40734 @anchor{qXfer write}
40735 Write uninterpreted bytes into the target's special data area
40736 identified by the keyword @var{object}, starting at @var{offset} bytes
40737 into the data. The binary-encoded data (@pxref{Binary Data}) to be
40738 written is given by @var{data}@dots{}. The content and encoding of @var{annex}
40739 is specific to @var{object}; it can supply additional details about what data
40745 @var{nn} (hex encoded) is the number of bytes written.
40746 This may be fewer bytes than supplied in the request.
40749 The request was malformed, or @var{annex} was invalid.
40752 The offset was invalid, or there was an error encountered writing the data.
40753 The @var{nn} part is a hex-encoded @code{errno} value.
40756 An empty reply indicates the @var{object} string was not
40757 recognized by the stub, or that the object does not support writing.
40760 Here are the specific requests of this form defined so far. All the
40761 @samp{qXfer:@var{object}:write:@dots{}} requests use the same reply
40762 formats, listed above.
40765 @item qXfer:siginfo:write::@var{offset}:@var{data}@dots{}
40766 @anchor{qXfer siginfo write}
40767 Write @var{data} to the extra signal information on the target system.
40768 The annex part of the generic @samp{qXfer} packet must be
40769 empty (@pxref{qXfer write}).
40771 This packet is not probed by default; the remote stub must request it,
40772 by supplying an appropriate @samp{qSupported} response
40773 (@pxref{qSupported}).
40775 @item qXfer:spu:write:@var{annex}:@var{offset}:@var{data}@dots{}
40776 @anchor{qXfer spu write}
40777 Write @var{data} to an @code{spufs} file on the target system. The
40778 annex specifies which file to write; it must be of the form
40779 @file{@var{id}/@var{name}}, where @var{id} specifies an SPU context ID
40780 in the target process, and @var{name} identifes the @code{spufs} file
40781 in that context to be accessed.
40783 This packet is not probed by default; the remote stub must request it,
40784 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
40787 @item qXfer:@var{object}:@var{operation}:@dots{}
40788 Requests of this form may be added in the future. When a stub does
40789 not recognize the @var{object} keyword, or its support for
40790 @var{object} does not recognize the @var{operation} keyword, the stub
40791 must respond with an empty packet.
40793 @item qAttached:@var{pid}
40794 @cindex query attached, remote request
40795 @cindex @samp{qAttached} packet
40796 Return an indication of whether the remote server attached to an
40797 existing process or created a new process. When the multiprocess
40798 protocol extensions are supported (@pxref{multiprocess extensions}),
40799 @var{pid} is an integer in hexadecimal format identifying the target
40800 process. Otherwise, @value{GDBN} will omit the @var{pid} field and
40801 the query packet will be simplified as @samp{qAttached}.
40803 This query is used, for example, to know whether the remote process
40804 should be detached or killed when a @value{GDBN} session is ended with
40805 the @code{quit} command.
40810 The remote server attached to an existing process.
40812 The remote server created a new process.
40814 A badly formed request or an error was encountered.
40818 Enable branch tracing for the current thread using Branch Trace Store.
40823 Branch tracing has been enabled.
40825 A badly formed request or an error was encountered.
40829 Enable branch tracing for the current thread using Intel Processor Trace.
40834 Branch tracing has been enabled.
40836 A badly formed request or an error was encountered.
40840 Disable branch tracing for the current thread.
40845 Branch tracing has been disabled.
40847 A badly formed request or an error was encountered.
40850 @item Qbtrace-conf:bts:size=@var{value}
40851 Set the requested ring buffer size for new threads that use the
40852 btrace recording method in bts format.
40857 The ring buffer size has been set.
40859 A badly formed request or an error was encountered.
40862 @item Qbtrace-conf:pt:size=@var{value}
40863 Set the requested ring buffer size for new threads that use the
40864 btrace recording method in pt format.
40869 The ring buffer size has been set.
40871 A badly formed request or an error was encountered.
40876 @node Architecture-Specific Protocol Details
40877 @section Architecture-Specific Protocol Details
40879 This section describes how the remote protocol is applied to specific
40880 target architectures. Also see @ref{Standard Target Features}, for
40881 details of XML target descriptions for each architecture.
40884 * ARM-Specific Protocol Details::
40885 * MIPS-Specific Protocol Details::
40888 @node ARM-Specific Protocol Details
40889 @subsection @acronym{ARM}-specific Protocol Details
40892 * ARM Breakpoint Kinds::
40895 @node ARM Breakpoint Kinds
40896 @subsubsection @acronym{ARM} Breakpoint Kinds
40897 @cindex breakpoint kinds, @acronym{ARM}
40899 These breakpoint kinds are defined for the @samp{Z0} and @samp{Z1} packets.
40904 16-bit Thumb mode breakpoint.
40907 32-bit Thumb mode (Thumb-2) breakpoint.
40910 32-bit @acronym{ARM} mode breakpoint.
40914 @node MIPS-Specific Protocol Details
40915 @subsection @acronym{MIPS}-specific Protocol Details
40918 * MIPS Register packet Format::
40919 * MIPS Breakpoint Kinds::
40922 @node MIPS Register packet Format
40923 @subsubsection @acronym{MIPS} Register Packet Format
40924 @cindex register packet format, @acronym{MIPS}
40926 The following @code{g}/@code{G} packets have previously been defined.
40927 In the below, some thirty-two bit registers are transferred as
40928 sixty-four bits. Those registers should be zero/sign extended (which?)
40929 to fill the space allocated. Register bytes are transferred in target
40930 byte order. The two nibbles within a register byte are transferred
40931 most-significant -- least-significant.
40936 All registers are transferred as thirty-two bit quantities in the order:
40937 32 general-purpose; sr; lo; hi; bad; cause; pc; 32 floating-point
40938 registers; fsr; fir; fp.
40941 All registers are transferred as sixty-four bit quantities (including
40942 thirty-two bit registers such as @code{sr}). The ordering is the same
40947 @node MIPS Breakpoint Kinds
40948 @subsubsection @acronym{MIPS} Breakpoint Kinds
40949 @cindex breakpoint kinds, @acronym{MIPS}
40951 These breakpoint kinds are defined for the @samp{Z0} and @samp{Z1} packets.
40956 16-bit @acronym{MIPS16} mode breakpoint.
40959 16-bit @acronym{microMIPS} mode breakpoint.
40962 32-bit standard @acronym{MIPS} mode breakpoint.
40965 32-bit @acronym{microMIPS} mode breakpoint.
40969 @node Tracepoint Packets
40970 @section Tracepoint Packets
40971 @cindex tracepoint packets
40972 @cindex packets, tracepoint
40974 Here we describe the packets @value{GDBN} uses to implement
40975 tracepoints (@pxref{Tracepoints}).
40979 @item QTDP:@var{n}:@var{addr}:@var{ena}:@var{step}:@var{pass}[:F@var{flen}][:X@var{len},@var{bytes}]@r{[}-@r{]}
40980 @cindex @samp{QTDP} packet
40981 Create a new tracepoint, number @var{n}, at @var{addr}. If @var{ena}
40982 is @samp{E}, then the tracepoint is enabled; if it is @samp{D}, then
40983 the tracepoint is disabled. The @var{step} gives the tracepoint's step
40984 count, and @var{pass} gives its pass count. If an @samp{F} is present,
40985 then the tracepoint is to be a fast tracepoint, and the @var{flen} is
40986 the number of bytes that the target should copy elsewhere to make room
40987 for the tracepoint. If an @samp{X} is present, it introduces a
40988 tracepoint condition, which consists of a hexadecimal length, followed
40989 by a comma and hex-encoded bytes, in a manner similar to action
40990 encodings as described below. If the trailing @samp{-} is present,
40991 further @samp{QTDP} packets will follow to specify this tracepoint's
40997 The packet was understood and carried out.
40999 @xref{Tracepoint Packets,,Relocate instruction reply packet}.
41001 The packet was not recognized.
41004 @item QTDP:-@var{n}:@var{addr}:@r{[}S@r{]}@var{action}@dots{}@r{[}-@r{]}
41005 Define actions to be taken when a tracepoint is hit. The @var{n} and
41006 @var{addr} must be the same as in the initial @samp{QTDP} packet for
41007 this tracepoint. This packet may only be sent immediately after
41008 another @samp{QTDP} packet that ended with a @samp{-}. If the
41009 trailing @samp{-} is present, further @samp{QTDP} packets will follow,
41010 specifying more actions for this tracepoint.
41012 In the series of action packets for a given tracepoint, at most one
41013 can have an @samp{S} before its first @var{action}. If such a packet
41014 is sent, it and the following packets define ``while-stepping''
41015 actions. Any prior packets define ordinary actions --- that is, those
41016 taken when the tracepoint is first hit. If no action packet has an
41017 @samp{S}, then all the packets in the series specify ordinary
41018 tracepoint actions.
41020 The @samp{@var{action}@dots{}} portion of the packet is a series of
41021 actions, concatenated without separators. Each action has one of the
41027 Collect the registers whose bits are set in @var{mask},
41028 a hexadecimal number whose @var{i}'th bit is set if register number
41029 @var{i} should be collected. (The least significant bit is numbered
41030 zero.) Note that @var{mask} may be any number of digits long; it may
41031 not fit in a 32-bit word.
41033 @item M @var{basereg},@var{offset},@var{len}
41034 Collect @var{len} bytes of memory starting at the address in register
41035 number @var{basereg}, plus @var{offset}. If @var{basereg} is
41036 @samp{-1}, then the range has a fixed address: @var{offset} is the
41037 address of the lowest byte to collect. The @var{basereg},
41038 @var{offset}, and @var{len} parameters are all unsigned hexadecimal
41039 values (the @samp{-1} value for @var{basereg} is a special case).
41041 @item X @var{len},@var{expr}
41042 Evaluate @var{expr}, whose length is @var{len}, and collect memory as
41043 it directs. The agent expression @var{expr} is as described in
41044 @ref{Agent Expressions}. Each byte of the expression is encoded as a
41045 two-digit hex number in the packet; @var{len} is the number of bytes
41046 in the expression (and thus one-half the number of hex digits in the
41051 Any number of actions may be packed together in a single @samp{QTDP}
41052 packet, as long as the packet does not exceed the maximum packet
41053 length (400 bytes, for many stubs). There may be only one @samp{R}
41054 action per tracepoint, and it must precede any @samp{M} or @samp{X}
41055 actions. Any registers referred to by @samp{M} and @samp{X} actions
41056 must be collected by a preceding @samp{R} action. (The
41057 ``while-stepping'' actions are treated as if they were attached to a
41058 separate tracepoint, as far as these restrictions are concerned.)
41063 The packet was understood and carried out.
41065 @xref{Tracepoint Packets,,Relocate instruction reply packet}.
41067 The packet was not recognized.
41070 @item QTDPsrc:@var{n}:@var{addr}:@var{type}:@var{start}:@var{slen}:@var{bytes}
41071 @cindex @samp{QTDPsrc} packet
41072 Specify a source string of tracepoint @var{n} at address @var{addr}.
41073 This is useful to get accurate reproduction of the tracepoints
41074 originally downloaded at the beginning of the trace run. The @var{type}
41075 is the name of the tracepoint part, such as @samp{cond} for the
41076 tracepoint's conditional expression (see below for a list of types), while
41077 @var{bytes} is the string, encoded in hexadecimal.
41079 @var{start} is the offset of the @var{bytes} within the overall source
41080 string, while @var{slen} is the total length of the source string.
41081 This is intended for handling source strings that are longer than will
41082 fit in a single packet.
41083 @c Add detailed example when this info is moved into a dedicated
41084 @c tracepoint descriptions section.
41086 The available string types are @samp{at} for the location,
41087 @samp{cond} for the conditional, and @samp{cmd} for an action command.
41088 @value{GDBN} sends a separate packet for each command in the action
41089 list, in the same order in which the commands are stored in the list.
41091 The target does not need to do anything with source strings except
41092 report them back as part of the replies to the @samp{qTfP}/@samp{qTsP}
41095 Although this packet is optional, and @value{GDBN} will only send it
41096 if the target replies with @samp{TracepointSource} @xref{General
41097 Query Packets}, it makes both disconnected tracing and trace files
41098 much easier to use. Otherwise the user must be careful that the
41099 tracepoints in effect while looking at trace frames are identical to
41100 the ones in effect during the trace run; even a small discrepancy
41101 could cause @samp{tdump} not to work, or a particular trace frame not
41104 @item QTDV:@var{n}:@var{value}:@var{builtin}:@var{name}
41105 @cindex define trace state variable, remote request
41106 @cindex @samp{QTDV} packet
41107 Create a new trace state variable, number @var{n}, with an initial
41108 value of @var{value}, which is a 64-bit signed integer. Both @var{n}
41109 and @var{value} are encoded as hexadecimal values. @value{GDBN} has
41110 the option of not using this packet for initial values of zero; the
41111 target should simply create the trace state variables as they are
41112 mentioned in expressions. The value @var{builtin} should be 1 (one)
41113 if the trace state variable is builtin and 0 (zero) if it is not builtin.
41114 @value{GDBN} only sets @var{builtin} to 1 if a previous @samp{qTfV} or
41115 @samp{qTsV} packet had it set. The contents of @var{name} is the
41116 hex-encoded name (without the leading @samp{$}) of the trace state
41119 @item QTFrame:@var{n}
41120 @cindex @samp{QTFrame} packet
41121 Select the @var{n}'th tracepoint frame from the buffer, and use the
41122 register and memory contents recorded there to answer subsequent
41123 request packets from @value{GDBN}.
41125 A successful reply from the stub indicates that the stub has found the
41126 requested frame. The response is a series of parts, concatenated
41127 without separators, describing the frame we selected. Each part has
41128 one of the following forms:
41132 The selected frame is number @var{n} in the trace frame buffer;
41133 @var{f} is a hexadecimal number. If @var{f} is @samp{-1}, then there
41134 was no frame matching the criteria in the request packet.
41137 The selected trace frame records a hit of tracepoint number @var{t};
41138 @var{t} is a hexadecimal number.
41142 @item QTFrame:pc:@var{addr}
41143 Like @samp{QTFrame:@var{n}}, but select the first tracepoint frame after the
41144 currently selected frame whose PC is @var{addr};
41145 @var{addr} is a hexadecimal number.
41147 @item QTFrame:tdp:@var{t}
41148 Like @samp{QTFrame:@var{n}}, but select the first tracepoint frame after the
41149 currently selected frame that is a hit of tracepoint @var{t}; @var{t}
41150 is a hexadecimal number.
41152 @item QTFrame:range:@var{start}:@var{end}
41153 Like @samp{QTFrame:@var{n}}, but select the first tracepoint frame after the
41154 currently selected frame whose PC is between @var{start} (inclusive)
41155 and @var{end} (inclusive); @var{start} and @var{end} are hexadecimal
41158 @item QTFrame:outside:@var{start}:@var{end}
41159 Like @samp{QTFrame:range:@var{start}:@var{end}}, but select the first
41160 frame @emph{outside} the given range of addresses (exclusive).
41163 @cindex @samp{qTMinFTPILen} packet
41164 This packet requests the minimum length of instruction at which a fast
41165 tracepoint (@pxref{Set Tracepoints}) may be placed. For instance, on
41166 the 32-bit x86 architecture, it is possible to use a 4-byte jump, but
41167 it depends on the target system being able to create trampolines in
41168 the first 64K of memory, which might or might not be possible for that
41169 system. So the reply to this packet will be 4 if it is able to
41176 The minimum instruction length is currently unknown.
41178 The minimum instruction length is @var{length}, where @var{length}
41179 is a hexadecimal number greater or equal to 1. A reply
41180 of 1 means that a fast tracepoint may be placed on any instruction
41181 regardless of size.
41183 An error has occurred.
41185 An empty reply indicates that the request is not supported by the stub.
41189 @cindex @samp{QTStart} packet
41190 Begin the tracepoint experiment. Begin collecting data from
41191 tracepoint hits in the trace frame buffer. This packet supports the
41192 @samp{qRelocInsn} reply (@pxref{Tracepoint Packets,,Relocate
41193 instruction reply packet}).
41196 @cindex @samp{QTStop} packet
41197 End the tracepoint experiment. Stop collecting trace frames.
41199 @item QTEnable:@var{n}:@var{addr}
41201 @cindex @samp{QTEnable} packet
41202 Enable tracepoint @var{n} at address @var{addr} in a started tracepoint
41203 experiment. If the tracepoint was previously disabled, then collection
41204 of data from it will resume.
41206 @item QTDisable:@var{n}:@var{addr}
41208 @cindex @samp{QTDisable} packet
41209 Disable tracepoint @var{n} at address @var{addr} in a started tracepoint
41210 experiment. No more data will be collected from the tracepoint unless
41211 @samp{QTEnable:@var{n}:@var{addr}} is subsequently issued.
41214 @cindex @samp{QTinit} packet
41215 Clear the table of tracepoints, and empty the trace frame buffer.
41217 @item QTro:@var{start1},@var{end1}:@var{start2},@var{end2}:@dots{}
41218 @cindex @samp{QTro} packet
41219 Establish the given ranges of memory as ``transparent''. The stub
41220 will answer requests for these ranges from memory's current contents,
41221 if they were not collected as part of the tracepoint hit.
41223 @value{GDBN} uses this to mark read-only regions of memory, like those
41224 containing program code. Since these areas never change, they should
41225 still have the same contents they did when the tracepoint was hit, so
41226 there's no reason for the stub to refuse to provide their contents.
41228 @item QTDisconnected:@var{value}
41229 @cindex @samp{QTDisconnected} packet
41230 Set the choice to what to do with the tracing run when @value{GDBN}
41231 disconnects from the target. A @var{value} of 1 directs the target to
41232 continue the tracing run, while 0 tells the target to stop tracing if
41233 @value{GDBN} is no longer in the picture.
41236 @cindex @samp{qTStatus} packet
41237 Ask the stub if there is a trace experiment running right now.
41239 The reply has the form:
41243 @item T@var{running}@r{[};@var{field}@r{]}@dots{}
41244 @var{running} is a single digit @code{1} if the trace is presently
41245 running, or @code{0} if not. It is followed by semicolon-separated
41246 optional fields that an agent may use to report additional status.
41250 If the trace is not running, the agent may report any of several
41251 explanations as one of the optional fields:
41256 No trace has been run yet.
41258 @item tstop[:@var{text}]:0
41259 The trace was stopped by a user-originated stop command. The optional
41260 @var{text} field is a user-supplied string supplied as part of the
41261 stop command (for instance, an explanation of why the trace was
41262 stopped manually). It is hex-encoded.
41265 The trace stopped because the trace buffer filled up.
41267 @item tdisconnected:0
41268 The trace stopped because @value{GDBN} disconnected from the target.
41270 @item tpasscount:@var{tpnum}
41271 The trace stopped because tracepoint @var{tpnum} exceeded its pass count.
41273 @item terror:@var{text}:@var{tpnum}
41274 The trace stopped because tracepoint @var{tpnum} had an error. The
41275 string @var{text} is available to describe the nature of the error
41276 (for instance, a divide by zero in the condition expression); it
41280 The trace stopped for some other reason.
41284 Additional optional fields supply statistical and other information.
41285 Although not required, they are extremely useful for users monitoring
41286 the progress of a trace run. If a trace has stopped, and these
41287 numbers are reported, they must reflect the state of the just-stopped
41292 @item tframes:@var{n}
41293 The number of trace frames in the buffer.
41295 @item tcreated:@var{n}
41296 The total number of trace frames created during the run. This may
41297 be larger than the trace frame count, if the buffer is circular.
41299 @item tsize:@var{n}
41300 The total size of the trace buffer, in bytes.
41302 @item tfree:@var{n}
41303 The number of bytes still unused in the buffer.
41305 @item circular:@var{n}
41306 The value of the circular trace buffer flag. @code{1} means that the
41307 trace buffer is circular and old trace frames will be discarded if
41308 necessary to make room, @code{0} means that the trace buffer is linear
41311 @item disconn:@var{n}
41312 The value of the disconnected tracing flag. @code{1} means that
41313 tracing will continue after @value{GDBN} disconnects, @code{0} means
41314 that the trace run will stop.
41318 @item qTP:@var{tp}:@var{addr}
41319 @cindex tracepoint status, remote request
41320 @cindex @samp{qTP} packet
41321 Ask the stub for the current state of tracepoint number @var{tp} at
41322 address @var{addr}.
41326 @item V@var{hits}:@var{usage}
41327 The tracepoint has been hit @var{hits} times so far during the trace
41328 run, and accounts for @var{usage} in the trace buffer. Note that
41329 @code{while-stepping} steps are not counted as separate hits, but the
41330 steps' space consumption is added into the usage number.
41334 @item qTV:@var{var}
41335 @cindex trace state variable value, remote request
41336 @cindex @samp{qTV} packet
41337 Ask the stub for the value of the trace state variable number @var{var}.
41342 The value of the variable is @var{value}. This will be the current
41343 value of the variable if the user is examining a running target, or a
41344 saved value if the variable was collected in the trace frame that the
41345 user is looking at. Note that multiple requests may result in
41346 different reply values, such as when requesting values while the
41347 program is running.
41350 The value of the variable is unknown. This would occur, for example,
41351 if the user is examining a trace frame in which the requested variable
41356 @cindex @samp{qTfP} packet
41358 @cindex @samp{qTsP} packet
41359 These packets request data about tracepoints that are being used by
41360 the target. @value{GDBN} sends @code{qTfP} to get the first piece
41361 of data, and multiple @code{qTsP} to get additional pieces. Replies
41362 to these packets generally take the form of the @code{QTDP} packets
41363 that define tracepoints. (FIXME add detailed syntax)
41366 @cindex @samp{qTfV} packet
41368 @cindex @samp{qTsV} packet
41369 These packets request data about trace state variables that are on the
41370 target. @value{GDBN} sends @code{qTfV} to get the first vari of data,
41371 and multiple @code{qTsV} to get additional variables. Replies to
41372 these packets follow the syntax of the @code{QTDV} packets that define
41373 trace state variables.
41379 @cindex @samp{qTfSTM} packet
41380 @cindex @samp{qTsSTM} packet
41381 These packets request data about static tracepoint markers that exist
41382 in the target program. @value{GDBN} sends @code{qTfSTM} to get the
41383 first piece of data, and multiple @code{qTsSTM} to get additional
41384 pieces. Replies to these packets take the following form:
41388 @item m @var{address}:@var{id}:@var{extra}
41390 @item m @var{address}:@var{id}:@var{extra},@var{address}:@var{id}:@var{extra}@dots{}
41391 a comma-separated list of markers
41393 (lower case letter @samp{L}) denotes end of list.
41395 An error occurred. The error number @var{nn} is given as hex digits.
41397 An empty reply indicates that the request is not supported by the
41401 The @var{address} is encoded in hex;
41402 @var{id} and @var{extra} are strings encoded in hex.
41404 In response to each query, the target will reply with a list of one or
41405 more markers, separated by commas. @value{GDBN} will respond to each
41406 reply with a request for more markers (using the @samp{qs} form of the
41407 query), until the target responds with @samp{l} (lower-case ell, for
41410 @item qTSTMat:@var{address}
41412 @cindex @samp{qTSTMat} packet
41413 This packets requests data about static tracepoint markers in the
41414 target program at @var{address}. Replies to this packet follow the
41415 syntax of the @samp{qTfSTM} and @code{qTsSTM} packets that list static
41416 tracepoint markers.
41418 @item QTSave:@var{filename}
41419 @cindex @samp{QTSave} packet
41420 This packet directs the target to save trace data to the file name
41421 @var{filename} in the target's filesystem. The @var{filename} is encoded
41422 as a hex string; the interpretation of the file name (relative vs
41423 absolute, wild cards, etc) is up to the target.
41425 @item qTBuffer:@var{offset},@var{len}
41426 @cindex @samp{qTBuffer} packet
41427 Return up to @var{len} bytes of the current contents of trace buffer,
41428 starting at @var{offset}. The trace buffer is treated as if it were
41429 a contiguous collection of traceframes, as per the trace file format.
41430 The reply consists as many hex-encoded bytes as the target can deliver
41431 in a packet; it is not an error to return fewer than were asked for.
41432 A reply consisting of just @code{l} indicates that no bytes are
41435 @item QTBuffer:circular:@var{value}
41436 This packet directs the target to use a circular trace buffer if
41437 @var{value} is 1, or a linear buffer if the value is 0.
41439 @item QTBuffer:size:@var{size}
41440 @anchor{QTBuffer-size}
41441 @cindex @samp{QTBuffer size} packet
41442 This packet directs the target to make the trace buffer be of size
41443 @var{size} if possible. A value of @code{-1} tells the target to
41444 use whatever size it prefers.
41446 @item QTNotes:@r{[}@var{type}:@var{text}@r{]}@r{[};@var{type}:@var{text}@r{]}@dots{}
41447 @cindex @samp{QTNotes} packet
41448 This packet adds optional textual notes to the trace run. Allowable
41449 types include @code{user}, @code{notes}, and @code{tstop}, the
41450 @var{text} fields are arbitrary strings, hex-encoded.
41454 @subsection Relocate instruction reply packet
41455 When installing fast tracepoints in memory, the target may need to
41456 relocate the instruction currently at the tracepoint address to a
41457 different address in memory. For most instructions, a simple copy is
41458 enough, but, for example, call instructions that implicitly push the
41459 return address on the stack, and relative branches or other
41460 PC-relative instructions require offset adjustment, so that the effect
41461 of executing the instruction at a different address is the same as if
41462 it had executed in the original location.
41464 In response to several of the tracepoint packets, the target may also
41465 respond with a number of intermediate @samp{qRelocInsn} request
41466 packets before the final result packet, to have @value{GDBN} handle
41467 this relocation operation. If a packet supports this mechanism, its
41468 documentation will explicitly say so. See for example the above
41469 descriptions for the @samp{QTStart} and @samp{QTDP} packets. The
41470 format of the request is:
41473 @item qRelocInsn:@var{from};@var{to}
41475 This requests @value{GDBN} to copy instruction at address @var{from}
41476 to address @var{to}, possibly adjusted so that executing the
41477 instruction at @var{to} has the same effect as executing it at
41478 @var{from}. @value{GDBN} writes the adjusted instruction to target
41479 memory starting at @var{to}.
41484 @item qRelocInsn:@var{adjusted_size}
41485 Informs the stub the relocation is complete. The @var{adjusted_size} is
41486 the length in bytes of resulting relocated instruction sequence.
41488 A badly formed request was detected, or an error was encountered while
41489 relocating the instruction.
41492 @node Host I/O Packets
41493 @section Host I/O Packets
41494 @cindex Host I/O, remote protocol
41495 @cindex file transfer, remote protocol
41497 The @dfn{Host I/O} packets allow @value{GDBN} to perform I/O
41498 operations on the far side of a remote link. For example, Host I/O is
41499 used to upload and download files to a remote target with its own
41500 filesystem. Host I/O uses the same constant values and data structure
41501 layout as the target-initiated File-I/O protocol. However, the
41502 Host I/O packets are structured differently. The target-initiated
41503 protocol relies on target memory to store parameters and buffers.
41504 Host I/O requests are initiated by @value{GDBN}, and the
41505 target's memory is not involved. @xref{File-I/O Remote Protocol
41506 Extension}, for more details on the target-initiated protocol.
41508 The Host I/O request packets all encode a single operation along with
41509 its arguments. They have this format:
41513 @item vFile:@var{operation}: @var{parameter}@dots{}
41514 @var{operation} is the name of the particular request; the target
41515 should compare the entire packet name up to the second colon when checking
41516 for a supported operation. The format of @var{parameter} depends on
41517 the operation. Numbers are always passed in hexadecimal. Negative
41518 numbers have an explicit minus sign (i.e.@: two's complement is not
41519 used). Strings (e.g.@: filenames) are encoded as a series of
41520 hexadecimal bytes. The last argument to a system call may be a
41521 buffer of escaped binary data (@pxref{Binary Data}).
41525 The valid responses to Host I/O packets are:
41529 @item F @var{result} [, @var{errno}] [; @var{attachment}]
41530 @var{result} is the integer value returned by this operation, usually
41531 non-negative for success and -1 for errors. If an error has occured,
41532 @var{errno} will be included in the result specifying a
41533 value defined by the File-I/O protocol (@pxref{Errno Values}). For
41534 operations which return data, @var{attachment} supplies the data as a
41535 binary buffer. Binary buffers in response packets are escaped in the
41536 normal way (@pxref{Binary Data}). See the individual packet
41537 documentation for the interpretation of @var{result} and
41541 An empty response indicates that this operation is not recognized.
41545 These are the supported Host I/O operations:
41548 @item vFile:open: @var{filename}, @var{flags}, @var{mode}
41549 Open a file at @var{filename} and return a file descriptor for it, or
41550 return -1 if an error occurs. The @var{filename} is a string,
41551 @var{flags} is an integer indicating a mask of open flags
41552 (@pxref{Open Flags}), and @var{mode} is an integer indicating a mask
41553 of mode bits to use if the file is created (@pxref{mode_t Values}).
41554 @xref{open}, for details of the open flags and mode values.
41556 @item vFile:close: @var{fd}
41557 Close the open file corresponding to @var{fd} and return 0, or
41558 -1 if an error occurs.
41560 @item vFile:pread: @var{fd}, @var{count}, @var{offset}
41561 Read data from the open file corresponding to @var{fd}. Up to
41562 @var{count} bytes will be read from the file, starting at @var{offset}
41563 relative to the start of the file. The target may read fewer bytes;
41564 common reasons include packet size limits and an end-of-file
41565 condition. The number of bytes read is returned. Zero should only be
41566 returned for a successful read at the end of the file, or if
41567 @var{count} was zero.
41569 The data read should be returned as a binary attachment on success.
41570 If zero bytes were read, the response should include an empty binary
41571 attachment (i.e.@: a trailing semicolon). The return value is the
41572 number of target bytes read; the binary attachment may be longer if
41573 some characters were escaped.
41575 @item vFile:pwrite: @var{fd}, @var{offset}, @var{data}
41576 Write @var{data} (a binary buffer) to the open file corresponding
41577 to @var{fd}. Start the write at @var{offset} from the start of the
41578 file. Unlike many @code{write} system calls, there is no
41579 separate @var{count} argument; the length of @var{data} in the
41580 packet is used. @samp{vFile:write} returns the number of bytes written,
41581 which may be shorter than the length of @var{data}, or -1 if an
41584 @item vFile:fstat: @var{fd}
41585 Get information about the open file corresponding to @var{fd}.
41586 On success the information is returned as a binary attachment
41587 and the return value is the size of this attachment in bytes.
41588 If an error occurs the return value is -1. The format of the
41589 returned binary attachment is as described in @ref{struct stat}.
41591 @item vFile:unlink: @var{filename}
41592 Delete the file at @var{filename} on the target. Return 0,
41593 or -1 if an error occurs. The @var{filename} is a string.
41595 @item vFile:readlink: @var{filename}
41596 Read value of symbolic link @var{filename} on the target. Return
41597 the number of bytes read, or -1 if an error occurs.
41599 The data read should be returned as a binary attachment on success.
41600 If zero bytes were read, the response should include an empty binary
41601 attachment (i.e.@: a trailing semicolon). The return value is the
41602 number of target bytes read; the binary attachment may be longer if
41603 some characters were escaped.
41605 @item vFile:setfs: @var{pid}
41606 Select the filesystem on which @code{vFile} operations with
41607 @var{filename} arguments will operate. This is required for
41608 @value{GDBN} to be able to access files on remote targets where
41609 the remote stub does not share a common filesystem with the
41612 If @var{pid} is nonzero, select the filesystem as seen by process
41613 @var{pid}. If @var{pid} is zero, select the filesystem as seen by
41614 the remote stub. Return 0 on success, or -1 if an error occurs.
41615 If @code{vFile:setfs:} indicates success, the selected filesystem
41616 remains selected until the next successful @code{vFile:setfs:}
41622 @section Interrupts
41623 @cindex interrupts (remote protocol)
41624 @anchor{interrupting remote targets}
41626 In all-stop mode, when a program on the remote target is running,
41627 @value{GDBN} may attempt to interrupt it by sending a @samp{Ctrl-C},
41628 @code{BREAK} or a @code{BREAK} followed by @code{g}, control of which
41629 is specified via @value{GDBN}'s @samp{interrupt-sequence}.
41631 The precise meaning of @code{BREAK} is defined by the transport
41632 mechanism and may, in fact, be undefined. @value{GDBN} does not
41633 currently define a @code{BREAK} mechanism for any of the network
41634 interfaces except for TCP, in which case @value{GDBN} sends the
41635 @code{telnet} BREAK sequence.
41637 @samp{Ctrl-C}, on the other hand, is defined and implemented for all
41638 transport mechanisms. It is represented by sending the single byte
41639 @code{0x03} without any of the usual packet overhead described in
41640 the Overview section (@pxref{Overview}). When a @code{0x03} byte is
41641 transmitted as part of a packet, it is considered to be packet data
41642 and does @emph{not} represent an interrupt. E.g., an @samp{X} packet
41643 (@pxref{X packet}), used for binary downloads, may include an unescaped
41644 @code{0x03} as part of its packet.
41646 @code{BREAK} followed by @code{g} is also known as Magic SysRq g.
41647 When Linux kernel receives this sequence from serial port,
41648 it stops execution and connects to gdb.
41650 In non-stop mode, because packet resumptions are asynchronous
41651 (@pxref{vCont packet}), @value{GDBN} is always free to send a remote
41652 command to the remote stub, even when the target is running. For that
41653 reason, @value{GDBN} instead sends a regular packet (@pxref{vCtrlC
41654 packet}) with the usual packet framing instead of the single byte
41657 Stubs are not required to recognize these interrupt mechanisms and the
41658 precise meaning associated with receipt of the interrupt is
41659 implementation defined. If the target supports debugging of multiple
41660 threads and/or processes, it should attempt to interrupt all
41661 currently-executing threads and processes.
41662 If the stub is successful at interrupting the
41663 running program, it should send one of the stop
41664 reply packets (@pxref{Stop Reply Packets}) to @value{GDBN} as a result
41665 of successfully stopping the program in all-stop mode, and a stop reply
41666 for each stopped thread in non-stop mode.
41667 Interrupts received while the
41668 program is stopped are queued and the program will be interrupted when
41669 it is resumed next time.
41671 @node Notification Packets
41672 @section Notification Packets
41673 @cindex notification packets
41674 @cindex packets, notification
41676 The @value{GDBN} remote serial protocol includes @dfn{notifications},
41677 packets that require no acknowledgment. Both the GDB and the stub
41678 may send notifications (although the only notifications defined at
41679 present are sent by the stub). Notifications carry information
41680 without incurring the round-trip latency of an acknowledgment, and so
41681 are useful for low-impact communications where occasional packet loss
41684 A notification packet has the form @samp{% @var{data} #
41685 @var{checksum}}, where @var{data} is the content of the notification,
41686 and @var{checksum} is a checksum of @var{data}, computed and formatted
41687 as for ordinary @value{GDBN} packets. A notification's @var{data}
41688 never contains @samp{$}, @samp{%} or @samp{#} characters. Upon
41689 receiving a notification, the recipient sends no @samp{+} or @samp{-}
41690 to acknowledge the notification's receipt or to report its corruption.
41692 Every notification's @var{data} begins with a name, which contains no
41693 colon characters, followed by a colon character.
41695 Recipients should silently ignore corrupted notifications and
41696 notifications they do not understand. Recipients should restart
41697 timeout periods on receipt of a well-formed notification, whether or
41698 not they understand it.
41700 Senders should only send the notifications described here when this
41701 protocol description specifies that they are permitted. In the
41702 future, we may extend the protocol to permit existing notifications in
41703 new contexts; this rule helps older senders avoid confusing newer
41706 (Older versions of @value{GDBN} ignore bytes received until they see
41707 the @samp{$} byte that begins an ordinary packet, so new stubs may
41708 transmit notifications without fear of confusing older clients. There
41709 are no notifications defined for @value{GDBN} to send at the moment, but we
41710 assume that most older stubs would ignore them, as well.)
41712 Each notification is comprised of three parts:
41714 @item @var{name}:@var{event}
41715 The notification packet is sent by the side that initiates the
41716 exchange (currently, only the stub does that), with @var{event}
41717 carrying the specific information about the notification, and
41718 @var{name} specifying the name of the notification.
41720 The acknowledge sent by the other side, usually @value{GDBN}, to
41721 acknowledge the exchange and request the event.
41724 The purpose of an asynchronous notification mechanism is to report to
41725 @value{GDBN} that something interesting happened in the remote stub.
41727 The remote stub may send notification @var{name}:@var{event}
41728 at any time, but @value{GDBN} acknowledges the notification when
41729 appropriate. The notification event is pending before @value{GDBN}
41730 acknowledges. Only one notification at a time may be pending; if
41731 additional events occur before @value{GDBN} has acknowledged the
41732 previous notification, they must be queued by the stub for later
41733 synchronous transmission in response to @var{ack} packets from
41734 @value{GDBN}. Because the notification mechanism is unreliable,
41735 the stub is permitted to resend a notification if it believes
41736 @value{GDBN} may not have received it.
41738 Specifically, notifications may appear when @value{GDBN} is not
41739 otherwise reading input from the stub, or when @value{GDBN} is
41740 expecting to read a normal synchronous response or a
41741 @samp{+}/@samp{-} acknowledgment to a packet it has sent.
41742 Notification packets are distinct from any other communication from
41743 the stub so there is no ambiguity.
41745 After receiving a notification, @value{GDBN} shall acknowledge it by
41746 sending a @var{ack} packet as a regular, synchronous request to the
41747 stub. Such acknowledgment is not required to happen immediately, as
41748 @value{GDBN} is permitted to send other, unrelated packets to the
41749 stub first, which the stub should process normally.
41751 Upon receiving a @var{ack} packet, if the stub has other queued
41752 events to report to @value{GDBN}, it shall respond by sending a
41753 normal @var{event}. @value{GDBN} shall then send another @var{ack}
41754 packet to solicit further responses; again, it is permitted to send
41755 other, unrelated packets as well which the stub should process
41758 If the stub receives a @var{ack} packet and there are no additional
41759 @var{event} to report, the stub shall return an @samp{OK} response.
41760 At this point, @value{GDBN} has finished processing a notification
41761 and the stub has completed sending any queued events. @value{GDBN}
41762 won't accept any new notifications until the final @samp{OK} is
41763 received . If further notification events occur, the stub shall send
41764 a new notification, @value{GDBN} shall accept the notification, and
41765 the process shall be repeated.
41767 The process of asynchronous notification can be illustrated by the
41770 <- @code{%Stop:T0505:98e7ffbf;04:4ce6ffbf;08:b1b6e54c;thread:p7526.7526;core:0;}
41773 <- @code{T0505:68f37db7;04:40f37db7;08:63850408;thread:p7526.7528;core:0;}
41775 <- @code{T0505:68e3fdb6;04:40e3fdb6;08:63850408;thread:p7526.7529;core:0;}
41780 The following notifications are defined:
41781 @multitable @columnfractions 0.12 0.12 0.38 0.38
41790 @tab @var{reply}. The @var{reply} has the form of a stop reply, as
41791 described in @ref{Stop Reply Packets}. Refer to @ref{Remote Non-Stop},
41792 for information on how these notifications are acknowledged by
41794 @tab Report an asynchronous stop event in non-stop mode.
41798 @node Remote Non-Stop
41799 @section Remote Protocol Support for Non-Stop Mode
41801 @value{GDBN}'s remote protocol supports non-stop debugging of
41802 multi-threaded programs, as described in @ref{Non-Stop Mode}. If the stub
41803 supports non-stop mode, it should report that to @value{GDBN} by including
41804 @samp{QNonStop+} in its @samp{qSupported} response (@pxref{qSupported}).
41806 @value{GDBN} typically sends a @samp{QNonStop} packet only when
41807 establishing a new connection with the stub. Entering non-stop mode
41808 does not alter the state of any currently-running threads, but targets
41809 must stop all threads in any already-attached processes when entering
41810 all-stop mode. @value{GDBN} uses the @samp{?} packet as necessary to
41811 probe the target state after a mode change.
41813 In non-stop mode, when an attached process encounters an event that
41814 would otherwise be reported with a stop reply, it uses the
41815 asynchronous notification mechanism (@pxref{Notification Packets}) to
41816 inform @value{GDBN}. In contrast to all-stop mode, where all threads
41817 in all processes are stopped when a stop reply is sent, in non-stop
41818 mode only the thread reporting the stop event is stopped. That is,
41819 when reporting a @samp{S} or @samp{T} response to indicate completion
41820 of a step operation, hitting a breakpoint, or a fault, only the
41821 affected thread is stopped; any other still-running threads continue
41822 to run. When reporting a @samp{W} or @samp{X} response, all running
41823 threads belonging to other attached processes continue to run.
41825 In non-stop mode, the target shall respond to the @samp{?} packet as
41826 follows. First, any incomplete stop reply notification/@samp{vStopped}
41827 sequence in progress is abandoned. The target must begin a new
41828 sequence reporting stop events for all stopped threads, whether or not
41829 it has previously reported those events to @value{GDBN}. The first
41830 stop reply is sent as a synchronous reply to the @samp{?} packet, and
41831 subsequent stop replies are sent as responses to @samp{vStopped} packets
41832 using the mechanism described above. The target must not send
41833 asynchronous stop reply notifications until the sequence is complete.
41834 If all threads are running when the target receives the @samp{?} packet,
41835 or if the target is not attached to any process, it shall respond
41838 If the stub supports non-stop mode, it should also support the
41839 @samp{swbreak} stop reason if software breakpoints are supported, and
41840 the @samp{hwbreak} stop reason if hardware breakpoints are supported
41841 (@pxref{swbreak stop reason}). This is because given the asynchronous
41842 nature of non-stop mode, between the time a thread hits a breakpoint
41843 and the time the event is finally processed by @value{GDBN}, the
41844 breakpoint may have already been removed from the target. Due to
41845 this, @value{GDBN} needs to be able to tell whether a trap stop was
41846 caused by a delayed breakpoint event, which should be ignored, as
41847 opposed to a random trap signal, which should be reported to the user.
41848 Note the @samp{swbreak} feature implies that the target is responsible
41849 for adjusting the PC when a software breakpoint triggers, if
41850 necessary, such as on the x86 architecture.
41852 @node Packet Acknowledgment
41853 @section Packet Acknowledgment
41855 @cindex acknowledgment, for @value{GDBN} remote
41856 @cindex packet acknowledgment, for @value{GDBN} remote
41857 By default, when either the host or the target machine receives a packet,
41858 the first response expected is an acknowledgment: either @samp{+} (to indicate
41859 the package was received correctly) or @samp{-} (to request retransmission).
41860 This mechanism allows the @value{GDBN} remote protocol to operate over
41861 unreliable transport mechanisms, such as a serial line.
41863 In cases where the transport mechanism is itself reliable (such as a pipe or
41864 TCP connection), the @samp{+}/@samp{-} acknowledgments are redundant.
41865 It may be desirable to disable them in that case to reduce communication
41866 overhead, or for other reasons. This can be accomplished by means of the
41867 @samp{QStartNoAckMode} packet; @pxref{QStartNoAckMode}.
41869 When in no-acknowledgment mode, neither the stub nor @value{GDBN} shall send or
41870 expect @samp{+}/@samp{-} protocol acknowledgments. The packet
41871 and response format still includes the normal checksum, as described in
41872 @ref{Overview}, but the checksum may be ignored by the receiver.
41874 If the stub supports @samp{QStartNoAckMode} and prefers to operate in
41875 no-acknowledgment mode, it should report that to @value{GDBN}
41876 by including @samp{QStartNoAckMode+} in its response to @samp{qSupported};
41877 @pxref{qSupported}.
41878 If @value{GDBN} also supports @samp{QStartNoAckMode} and it has not been
41879 disabled via the @code{set remote noack-packet off} command
41880 (@pxref{Remote Configuration}),
41881 @value{GDBN} may then send a @samp{QStartNoAckMode} packet to the stub.
41882 Only then may the stub actually turn off packet acknowledgments.
41883 @value{GDBN} sends a final @samp{+} acknowledgment of the stub's @samp{OK}
41884 response, which can be safely ignored by the stub.
41886 Note that @code{set remote noack-packet} command only affects negotiation
41887 between @value{GDBN} and the stub when subsequent connections are made;
41888 it does not affect the protocol acknowledgment state for any current
41890 Since @samp{+}/@samp{-} acknowledgments are enabled by default when a
41891 new connection is established,
41892 there is also no protocol request to re-enable the acknowledgments
41893 for the current connection, once disabled.
41898 Example sequence of a target being re-started. Notice how the restart
41899 does not get any direct output:
41904 @emph{target restarts}
41907 <- @code{T001:1234123412341234}
41911 Example sequence of a target being stepped by a single instruction:
41914 -> @code{G1445@dots{}}
41919 <- @code{T001:1234123412341234}
41923 <- @code{1455@dots{}}
41927 @node File-I/O Remote Protocol Extension
41928 @section File-I/O Remote Protocol Extension
41929 @cindex File-I/O remote protocol extension
41932 * File-I/O Overview::
41933 * Protocol Basics::
41934 * The F Request Packet::
41935 * The F Reply Packet::
41936 * The Ctrl-C Message::
41938 * List of Supported Calls::
41939 * Protocol-specific Representation of Datatypes::
41941 * File-I/O Examples::
41944 @node File-I/O Overview
41945 @subsection File-I/O Overview
41946 @cindex file-i/o overview
41948 The @dfn{File I/O remote protocol extension} (short: File-I/O) allows the
41949 target to use the host's file system and console I/O to perform various
41950 system calls. System calls on the target system are translated into a
41951 remote protocol packet to the host system, which then performs the needed
41952 actions and returns a response packet to the target system.
41953 This simulates file system operations even on targets that lack file systems.
41955 The protocol is defined to be independent of both the host and target systems.
41956 It uses its own internal representation of datatypes and values. Both
41957 @value{GDBN} and the target's @value{GDBN} stub are responsible for
41958 translating the system-dependent value representations into the internal
41959 protocol representations when data is transmitted.
41961 The communication is synchronous. A system call is possible only when
41962 @value{GDBN} is waiting for a response from the @samp{C}, @samp{c}, @samp{S}
41963 or @samp{s} packets. While @value{GDBN} handles the request for a system call,
41964 the target is stopped to allow deterministic access to the target's
41965 memory. Therefore File-I/O is not interruptible by target signals. On
41966 the other hand, it is possible to interrupt File-I/O by a user interrupt
41967 (@samp{Ctrl-C}) within @value{GDBN}.
41969 The target's request to perform a host system call does not finish
41970 the latest @samp{C}, @samp{c}, @samp{S} or @samp{s} action. That means,
41971 after finishing the system call, the target returns to continuing the
41972 previous activity (continue, step). No additional continue or step
41973 request from @value{GDBN} is required.
41976 (@value{GDBP}) continue
41977 <- target requests 'system call X'
41978 target is stopped, @value{GDBN} executes system call
41979 -> @value{GDBN} returns result
41980 ... target continues, @value{GDBN} returns to wait for the target
41981 <- target hits breakpoint and sends a Txx packet
41984 The protocol only supports I/O on the console and to regular files on
41985 the host file system. Character or block special devices, pipes,
41986 named pipes, sockets or any other communication method on the host
41987 system are not supported by this protocol.
41989 File I/O is not supported in non-stop mode.
41991 @node Protocol Basics
41992 @subsection Protocol Basics
41993 @cindex protocol basics, file-i/o
41995 The File-I/O protocol uses the @code{F} packet as the request as well
41996 as reply packet. Since a File-I/O system call can only occur when
41997 @value{GDBN} is waiting for a response from the continuing or stepping target,
41998 the File-I/O request is a reply that @value{GDBN} has to expect as a result
41999 of a previous @samp{C}, @samp{c}, @samp{S} or @samp{s} packet.
42000 This @code{F} packet contains all information needed to allow @value{GDBN}
42001 to call the appropriate host system call:
42005 A unique identifier for the requested system call.
42008 All parameters to the system call. Pointers are given as addresses
42009 in the target memory address space. Pointers to strings are given as
42010 pointer/length pair. Numerical values are given as they are.
42011 Numerical control flags are given in a protocol-specific representation.
42015 At this point, @value{GDBN} has to perform the following actions.
42019 If the parameters include pointer values to data needed as input to a
42020 system call, @value{GDBN} requests this data from the target with a
42021 standard @code{m} packet request. This additional communication has to be
42022 expected by the target implementation and is handled as any other @code{m}
42026 @value{GDBN} translates all value from protocol representation to host
42027 representation as needed. Datatypes are coerced into the host types.
42030 @value{GDBN} calls the system call.
42033 It then coerces datatypes back to protocol representation.
42036 If the system call is expected to return data in buffer space specified
42037 by pointer parameters to the call, the data is transmitted to the
42038 target using a @code{M} or @code{X} packet. This packet has to be expected
42039 by the target implementation and is handled as any other @code{M} or @code{X}
42044 Eventually @value{GDBN} replies with another @code{F} packet which contains all
42045 necessary information for the target to continue. This at least contains
42052 @code{errno}, if has been changed by the system call.
42059 After having done the needed type and value coercion, the target continues
42060 the latest continue or step action.
42062 @node The F Request Packet
42063 @subsection The @code{F} Request Packet
42064 @cindex file-i/o request packet
42065 @cindex @code{F} request packet
42067 The @code{F} request packet has the following format:
42070 @item F@var{call-id},@var{parameter@dots{}}
42072 @var{call-id} is the identifier to indicate the host system call to be called.
42073 This is just the name of the function.
42075 @var{parameter@dots{}} are the parameters to the system call.
42076 Parameters are hexadecimal integer values, either the actual values in case
42077 of scalar datatypes, pointers to target buffer space in case of compound
42078 datatypes and unspecified memory areas, or pointer/length pairs in case
42079 of string parameters. These are appended to the @var{call-id} as a
42080 comma-delimited list. All values are transmitted in ASCII
42081 string representation, pointer/length pairs separated by a slash.
42087 @node The F Reply Packet
42088 @subsection The @code{F} Reply Packet
42089 @cindex file-i/o reply packet
42090 @cindex @code{F} reply packet
42092 The @code{F} reply packet has the following format:
42096 @item F@var{retcode},@var{errno},@var{Ctrl-C flag};@var{call-specific attachment}
42098 @var{retcode} is the return code of the system call as hexadecimal value.
42100 @var{errno} is the @code{errno} set by the call, in protocol-specific
42102 This parameter can be omitted if the call was successful.
42104 @var{Ctrl-C flag} is only sent if the user requested a break. In this
42105 case, @var{errno} must be sent as well, even if the call was successful.
42106 The @var{Ctrl-C flag} itself consists of the character @samp{C}:
42113 or, if the call was interrupted before the host call has been performed:
42120 assuming 4 is the protocol-specific representation of @code{EINTR}.
42125 @node The Ctrl-C Message
42126 @subsection The @samp{Ctrl-C} Message
42127 @cindex ctrl-c message, in file-i/o protocol
42129 If the @samp{Ctrl-C} flag is set in the @value{GDBN}
42130 reply packet (@pxref{The F Reply Packet}),
42131 the target should behave as if it had
42132 gotten a break message. The meaning for the target is ``system call
42133 interrupted by @code{SIGINT}''. Consequentially, the target should actually stop
42134 (as with a break message) and return to @value{GDBN} with a @code{T02}
42137 It's important for the target to know in which
42138 state the system call was interrupted. There are two possible cases:
42142 The system call hasn't been performed on the host yet.
42145 The system call on the host has been finished.
42149 These two states can be distinguished by the target by the value of the
42150 returned @code{errno}. If it's the protocol representation of @code{EINTR}, the system
42151 call hasn't been performed. This is equivalent to the @code{EINTR} handling
42152 on POSIX systems. In any other case, the target may presume that the
42153 system call has been finished --- successfully or not --- and should behave
42154 as if the break message arrived right after the system call.
42156 @value{GDBN} must behave reliably. If the system call has not been called
42157 yet, @value{GDBN} may send the @code{F} reply immediately, setting @code{EINTR} as
42158 @code{errno} in the packet. If the system call on the host has been finished
42159 before the user requests a break, the full action must be finished by
42160 @value{GDBN}. This requires sending @code{M} or @code{X} packets as necessary.
42161 The @code{F} packet may only be sent when either nothing has happened
42162 or the full action has been completed.
42165 @subsection Console I/O
42166 @cindex console i/o as part of file-i/o
42168 By default and if not explicitly closed by the target system, the file
42169 descriptors 0, 1 and 2 are connected to the @value{GDBN} console. Output
42170 on the @value{GDBN} console is handled as any other file output operation
42171 (@code{write(1, @dots{})} or @code{write(2, @dots{})}). Console input is handled
42172 by @value{GDBN} so that after the target read request from file descriptor
42173 0 all following typing is buffered until either one of the following
42178 The user types @kbd{Ctrl-c}. The behaviour is as explained above, and the
42180 system call is treated as finished.
42183 The user presses @key{RET}. This is treated as end of input with a trailing
42187 The user types @kbd{Ctrl-d}. This is treated as end of input. No trailing
42188 character (neither newline nor @samp{Ctrl-D}) is appended to the input.
42192 If the user has typed more characters than fit in the buffer given to
42193 the @code{read} call, the trailing characters are buffered in @value{GDBN} until
42194 either another @code{read(0, @dots{})} is requested by the target, or debugging
42195 is stopped at the user's request.
42198 @node List of Supported Calls
42199 @subsection List of Supported Calls
42200 @cindex list of supported file-i/o calls
42217 @unnumberedsubsubsec open
42218 @cindex open, file-i/o system call
42223 int open(const char *pathname, int flags);
42224 int open(const char *pathname, int flags, mode_t mode);
42228 @samp{Fopen,@var{pathptr}/@var{len},@var{flags},@var{mode}}
42231 @var{flags} is the bitwise @code{OR} of the following values:
42235 If the file does not exist it will be created. The host
42236 rules apply as far as file ownership and time stamps
42240 When used with @code{O_CREAT}, if the file already exists it is
42241 an error and open() fails.
42244 If the file already exists and the open mode allows
42245 writing (@code{O_RDWR} or @code{O_WRONLY} is given) it will be
42246 truncated to zero length.
42249 The file is opened in append mode.
42252 The file is opened for reading only.
42255 The file is opened for writing only.
42258 The file is opened for reading and writing.
42262 Other bits are silently ignored.
42266 @var{mode} is the bitwise @code{OR} of the following values:
42270 User has read permission.
42273 User has write permission.
42276 Group has read permission.
42279 Group has write permission.
42282 Others have read permission.
42285 Others have write permission.
42289 Other bits are silently ignored.
42292 @item Return value:
42293 @code{open} returns the new file descriptor or -1 if an error
42300 @var{pathname} already exists and @code{O_CREAT} and @code{O_EXCL} were used.
42303 @var{pathname} refers to a directory.
42306 The requested access is not allowed.
42309 @var{pathname} was too long.
42312 A directory component in @var{pathname} does not exist.
42315 @var{pathname} refers to a device, pipe, named pipe or socket.
42318 @var{pathname} refers to a file on a read-only filesystem and
42319 write access was requested.
42322 @var{pathname} is an invalid pointer value.
42325 No space on device to create the file.
42328 The process already has the maximum number of files open.
42331 The limit on the total number of files open on the system
42335 The call was interrupted by the user.
42341 @unnumberedsubsubsec close
42342 @cindex close, file-i/o system call
42351 @samp{Fclose,@var{fd}}
42353 @item Return value:
42354 @code{close} returns zero on success, or -1 if an error occurred.
42360 @var{fd} isn't a valid open file descriptor.
42363 The call was interrupted by the user.
42369 @unnumberedsubsubsec read
42370 @cindex read, file-i/o system call
42375 int read(int fd, void *buf, unsigned int count);
42379 @samp{Fread,@var{fd},@var{bufptr},@var{count}}
42381 @item Return value:
42382 On success, the number of bytes read is returned.
42383 Zero indicates end of file. If count is zero, read
42384 returns zero as well. On error, -1 is returned.
42390 @var{fd} is not a valid file descriptor or is not open for
42394 @var{bufptr} is an invalid pointer value.
42397 The call was interrupted by the user.
42403 @unnumberedsubsubsec write
42404 @cindex write, file-i/o system call
42409 int write(int fd, const void *buf, unsigned int count);
42413 @samp{Fwrite,@var{fd},@var{bufptr},@var{count}}
42415 @item Return value:
42416 On success, the number of bytes written are returned.
42417 Zero indicates nothing was written. On error, -1
42424 @var{fd} is not a valid file descriptor or is not open for
42428 @var{bufptr} is an invalid pointer value.
42431 An attempt was made to write a file that exceeds the
42432 host-specific maximum file size allowed.
42435 No space on device to write the data.
42438 The call was interrupted by the user.
42444 @unnumberedsubsubsec lseek
42445 @cindex lseek, file-i/o system call
42450 long lseek (int fd, long offset, int flag);
42454 @samp{Flseek,@var{fd},@var{offset},@var{flag}}
42456 @var{flag} is one of:
42460 The offset is set to @var{offset} bytes.
42463 The offset is set to its current location plus @var{offset}
42467 The offset is set to the size of the file plus @var{offset}
42471 @item Return value:
42472 On success, the resulting unsigned offset in bytes from
42473 the beginning of the file is returned. Otherwise, a
42474 value of -1 is returned.
42480 @var{fd} is not a valid open file descriptor.
42483 @var{fd} is associated with the @value{GDBN} console.
42486 @var{flag} is not a proper value.
42489 The call was interrupted by the user.
42495 @unnumberedsubsubsec rename
42496 @cindex rename, file-i/o system call
42501 int rename(const char *oldpath, const char *newpath);
42505 @samp{Frename,@var{oldpathptr}/@var{len},@var{newpathptr}/@var{len}}
42507 @item Return value:
42508 On success, zero is returned. On error, -1 is returned.
42514 @var{newpath} is an existing directory, but @var{oldpath} is not a
42518 @var{newpath} is a non-empty directory.
42521 @var{oldpath} or @var{newpath} is a directory that is in use by some
42525 An attempt was made to make a directory a subdirectory
42529 A component used as a directory in @var{oldpath} or new
42530 path is not a directory. Or @var{oldpath} is a directory
42531 and @var{newpath} exists but is not a directory.
42534 @var{oldpathptr} or @var{newpathptr} are invalid pointer values.
42537 No access to the file or the path of the file.
42541 @var{oldpath} or @var{newpath} was too long.
42544 A directory component in @var{oldpath} or @var{newpath} does not exist.
42547 The file is on a read-only filesystem.
42550 The device containing the file has no room for the new
42554 The call was interrupted by the user.
42560 @unnumberedsubsubsec unlink
42561 @cindex unlink, file-i/o system call
42566 int unlink(const char *pathname);
42570 @samp{Funlink,@var{pathnameptr}/@var{len}}
42572 @item Return value:
42573 On success, zero is returned. On error, -1 is returned.
42579 No access to the file or the path of the file.
42582 The system does not allow unlinking of directories.
42585 The file @var{pathname} cannot be unlinked because it's
42586 being used by another process.
42589 @var{pathnameptr} is an invalid pointer value.
42592 @var{pathname} was too long.
42595 A directory component in @var{pathname} does not exist.
42598 A component of the path is not a directory.
42601 The file is on a read-only filesystem.
42604 The call was interrupted by the user.
42610 @unnumberedsubsubsec stat/fstat
42611 @cindex fstat, file-i/o system call
42612 @cindex stat, file-i/o system call
42617 int stat(const char *pathname, struct stat *buf);
42618 int fstat(int fd, struct stat *buf);
42622 @samp{Fstat,@var{pathnameptr}/@var{len},@var{bufptr}}@*
42623 @samp{Ffstat,@var{fd},@var{bufptr}}
42625 @item Return value:
42626 On success, zero is returned. On error, -1 is returned.
42632 @var{fd} is not a valid open file.
42635 A directory component in @var{pathname} does not exist or the
42636 path is an empty string.
42639 A component of the path is not a directory.
42642 @var{pathnameptr} is an invalid pointer value.
42645 No access to the file or the path of the file.
42648 @var{pathname} was too long.
42651 The call was interrupted by the user.
42657 @unnumberedsubsubsec gettimeofday
42658 @cindex gettimeofday, file-i/o system call
42663 int gettimeofday(struct timeval *tv, void *tz);
42667 @samp{Fgettimeofday,@var{tvptr},@var{tzptr}}
42669 @item Return value:
42670 On success, 0 is returned, -1 otherwise.
42676 @var{tz} is a non-NULL pointer.
42679 @var{tvptr} and/or @var{tzptr} is an invalid pointer value.
42685 @unnumberedsubsubsec isatty
42686 @cindex isatty, file-i/o system call
42691 int isatty(int fd);
42695 @samp{Fisatty,@var{fd}}
42697 @item Return value:
42698 Returns 1 if @var{fd} refers to the @value{GDBN} console, 0 otherwise.
42704 The call was interrupted by the user.
42709 Note that the @code{isatty} call is treated as a special case: it returns
42710 1 to the target if the file descriptor is attached
42711 to the @value{GDBN} console, 0 otherwise. Implementing through system calls
42712 would require implementing @code{ioctl} and would be more complex than
42717 @unnumberedsubsubsec system
42718 @cindex system, file-i/o system call
42723 int system(const char *command);
42727 @samp{Fsystem,@var{commandptr}/@var{len}}
42729 @item Return value:
42730 If @var{len} is zero, the return value indicates whether a shell is
42731 available. A zero return value indicates a shell is not available.
42732 For non-zero @var{len}, the value returned is -1 on error and the
42733 return status of the command otherwise. Only the exit status of the
42734 command is returned, which is extracted from the host's @code{system}
42735 return value by calling @code{WEXITSTATUS(retval)}. In case
42736 @file{/bin/sh} could not be executed, 127 is returned.
42742 The call was interrupted by the user.
42747 @value{GDBN} takes over the full task of calling the necessary host calls
42748 to perform the @code{system} call. The return value of @code{system} on
42749 the host is simplified before it's returned
42750 to the target. Any termination signal information from the child process
42751 is discarded, and the return value consists
42752 entirely of the exit status of the called command.
42754 Due to security concerns, the @code{system} call is by default refused
42755 by @value{GDBN}. The user has to allow this call explicitly with the
42756 @code{set remote system-call-allowed 1} command.
42759 @item set remote system-call-allowed
42760 @kindex set remote system-call-allowed
42761 Control whether to allow the @code{system} calls in the File I/O
42762 protocol for the remote target. The default is zero (disabled).
42764 @item show remote system-call-allowed
42765 @kindex show remote system-call-allowed
42766 Show whether the @code{system} calls are allowed in the File I/O
42770 @node Protocol-specific Representation of Datatypes
42771 @subsection Protocol-specific Representation of Datatypes
42772 @cindex protocol-specific representation of datatypes, in file-i/o protocol
42775 * Integral Datatypes::
42777 * Memory Transfer::
42782 @node Integral Datatypes
42783 @unnumberedsubsubsec Integral Datatypes
42784 @cindex integral datatypes, in file-i/o protocol
42786 The integral datatypes used in the system calls are @code{int},
42787 @code{unsigned int}, @code{long}, @code{unsigned long},
42788 @code{mode_t}, and @code{time_t}.
42790 @code{int}, @code{unsigned int}, @code{mode_t} and @code{time_t} are
42791 implemented as 32 bit values in this protocol.
42793 @code{long} and @code{unsigned long} are implemented as 64 bit types.
42795 @xref{Limits}, for corresponding MIN and MAX values (similar to those
42796 in @file{limits.h}) to allow range checking on host and target.
42798 @code{time_t} datatypes are defined as seconds since the Epoch.
42800 All integral datatypes transferred as part of a memory read or write of a
42801 structured datatype e.g.@: a @code{struct stat} have to be given in big endian
42804 @node Pointer Values
42805 @unnumberedsubsubsec Pointer Values
42806 @cindex pointer values, in file-i/o protocol
42808 Pointers to target data are transmitted as they are. An exception
42809 is made for pointers to buffers for which the length isn't
42810 transmitted as part of the function call, namely strings. Strings
42811 are transmitted as a pointer/length pair, both as hex values, e.g.@:
42818 which is a pointer to data of length 18 bytes at position 0x1aaf.
42819 The length is defined as the full string length in bytes, including
42820 the trailing null byte. For example, the string @code{"hello world"}
42821 at address 0x123456 is transmitted as
42827 @node Memory Transfer
42828 @unnumberedsubsubsec Memory Transfer
42829 @cindex memory transfer, in file-i/o protocol
42831 Structured data which is transferred using a memory read or write (for
42832 example, a @code{struct stat}) is expected to be in a protocol-specific format
42833 with all scalar multibyte datatypes being big endian. Translation to
42834 this representation needs to be done both by the target before the @code{F}
42835 packet is sent, and by @value{GDBN} before
42836 it transfers memory to the target. Transferred pointers to structured
42837 data should point to the already-coerced data at any time.
42841 @unnumberedsubsubsec struct stat
42842 @cindex struct stat, in file-i/o protocol
42844 The buffer of type @code{struct stat} used by the target and @value{GDBN}
42845 is defined as follows:
42849 unsigned int st_dev; /* device */
42850 unsigned int st_ino; /* inode */
42851 mode_t st_mode; /* protection */
42852 unsigned int st_nlink; /* number of hard links */
42853 unsigned int st_uid; /* user ID of owner */
42854 unsigned int st_gid; /* group ID of owner */
42855 unsigned int st_rdev; /* device type (if inode device) */
42856 unsigned long st_size; /* total size, in bytes */
42857 unsigned long st_blksize; /* blocksize for filesystem I/O */
42858 unsigned long st_blocks; /* number of blocks allocated */
42859 time_t st_atime; /* time of last access */
42860 time_t st_mtime; /* time of last modification */
42861 time_t st_ctime; /* time of last change */
42865 The integral datatypes conform to the definitions given in the
42866 appropriate section (see @ref{Integral Datatypes}, for details) so this
42867 structure is of size 64 bytes.
42869 The values of several fields have a restricted meaning and/or
42875 A value of 0 represents a file, 1 the console.
42878 No valid meaning for the target. Transmitted unchanged.
42881 Valid mode bits are described in @ref{Constants}. Any other
42882 bits have currently no meaning for the target.
42887 No valid meaning for the target. Transmitted unchanged.
42892 These values have a host and file system dependent
42893 accuracy. Especially on Windows hosts, the file system may not
42894 support exact timing values.
42897 The target gets a @code{struct stat} of the above representation and is
42898 responsible for coercing it to the target representation before
42901 Note that due to size differences between the host, target, and protocol
42902 representations of @code{struct stat} members, these members could eventually
42903 get truncated on the target.
42905 @node struct timeval
42906 @unnumberedsubsubsec struct timeval
42907 @cindex struct timeval, in file-i/o protocol
42909 The buffer of type @code{struct timeval} used by the File-I/O protocol
42910 is defined as follows:
42914 time_t tv_sec; /* second */
42915 long tv_usec; /* microsecond */
42919 The integral datatypes conform to the definitions given in the
42920 appropriate section (see @ref{Integral Datatypes}, for details) so this
42921 structure is of size 8 bytes.
42924 @subsection Constants
42925 @cindex constants, in file-i/o protocol
42927 The following values are used for the constants inside of the
42928 protocol. @value{GDBN} and target are responsible for translating these
42929 values before and after the call as needed.
42940 @unnumberedsubsubsec Open Flags
42941 @cindex open flags, in file-i/o protocol
42943 All values are given in hexadecimal representation.
42955 @node mode_t Values
42956 @unnumberedsubsubsec mode_t Values
42957 @cindex mode_t values, in file-i/o protocol
42959 All values are given in octal representation.
42976 @unnumberedsubsubsec Errno Values
42977 @cindex errno values, in file-i/o protocol
42979 All values are given in decimal representation.
43004 @code{EUNKNOWN} is used as a fallback error value if a host system returns
43005 any error value not in the list of supported error numbers.
43008 @unnumberedsubsubsec Lseek Flags
43009 @cindex lseek flags, in file-i/o protocol
43018 @unnumberedsubsubsec Limits
43019 @cindex limits, in file-i/o protocol
43021 All values are given in decimal representation.
43024 INT_MIN -2147483648
43026 UINT_MAX 4294967295
43027 LONG_MIN -9223372036854775808
43028 LONG_MAX 9223372036854775807
43029 ULONG_MAX 18446744073709551615
43032 @node File-I/O Examples
43033 @subsection File-I/O Examples
43034 @cindex file-i/o examples
43036 Example sequence of a write call, file descriptor 3, buffer is at target
43037 address 0x1234, 6 bytes should be written:
43040 <- @code{Fwrite,3,1234,6}
43041 @emph{request memory read from target}
43044 @emph{return "6 bytes written"}
43048 Example sequence of a read call, file descriptor 3, buffer is at target
43049 address 0x1234, 6 bytes should be read:
43052 <- @code{Fread,3,1234,6}
43053 @emph{request memory write to target}
43054 -> @code{X1234,6:XXXXXX}
43055 @emph{return "6 bytes read"}
43059 Example sequence of a read call, call fails on the host due to invalid
43060 file descriptor (@code{EBADF}):
43063 <- @code{Fread,3,1234,6}
43067 Example sequence of a read call, user presses @kbd{Ctrl-c} before syscall on
43071 <- @code{Fread,3,1234,6}
43076 Example sequence of a read call, user presses @kbd{Ctrl-c} after syscall on
43080 <- @code{Fread,3,1234,6}
43081 -> @code{X1234,6:XXXXXX}
43085 @node Library List Format
43086 @section Library List Format
43087 @cindex library list format, remote protocol
43089 On some platforms, a dynamic loader (e.g.@: @file{ld.so}) runs in the
43090 same process as your application to manage libraries. In this case,
43091 @value{GDBN} can use the loader's symbol table and normal memory
43092 operations to maintain a list of shared libraries. On other
43093 platforms, the operating system manages loaded libraries.
43094 @value{GDBN} can not retrieve the list of currently loaded libraries
43095 through memory operations, so it uses the @samp{qXfer:libraries:read}
43096 packet (@pxref{qXfer library list read}) instead. The remote stub
43097 queries the target's operating system and reports which libraries
43100 The @samp{qXfer:libraries:read} packet returns an XML document which
43101 lists loaded libraries and their offsets. Each library has an
43102 associated name and one or more segment or section base addresses,
43103 which report where the library was loaded in memory.
43105 For the common case of libraries that are fully linked binaries, the
43106 library should have a list of segments. If the target supports
43107 dynamic linking of a relocatable object file, its library XML element
43108 should instead include a list of allocated sections. The segment or
43109 section bases are start addresses, not relocation offsets; they do not
43110 depend on the library's link-time base addresses.
43112 @value{GDBN} must be linked with the Expat library to support XML
43113 library lists. @xref{Expat}.
43115 A simple memory map, with one loaded library relocated by a single
43116 offset, looks like this:
43120 <library name="/lib/libc.so.6">
43121 <segment address="0x10000000"/>
43126 Another simple memory map, with one loaded library with three
43127 allocated sections (.text, .data, .bss), looks like this:
43131 <library name="sharedlib.o">
43132 <section address="0x10000000"/>
43133 <section address="0x20000000"/>
43134 <section address="0x30000000"/>
43139 The format of a library list is described by this DTD:
43142 <!-- library-list: Root element with versioning -->
43143 <!ELEMENT library-list (library)*>
43144 <!ATTLIST library-list version CDATA #FIXED "1.0">
43145 <!ELEMENT library (segment*, section*)>
43146 <!ATTLIST library name CDATA #REQUIRED>
43147 <!ELEMENT segment EMPTY>
43148 <!ATTLIST segment address CDATA #REQUIRED>
43149 <!ELEMENT section EMPTY>
43150 <!ATTLIST section address CDATA #REQUIRED>
43153 In addition, segments and section descriptors cannot be mixed within a
43154 single library element, and you must supply at least one segment or
43155 section for each library.
43157 @node Library List Format for SVR4 Targets
43158 @section Library List Format for SVR4 Targets
43159 @cindex library list format, remote protocol
43161 On SVR4 platforms @value{GDBN} can use the symbol table of a dynamic loader
43162 (e.g.@: @file{ld.so}) and normal memory operations to maintain a list of
43163 shared libraries. Still a special library list provided by this packet is
43164 more efficient for the @value{GDBN} remote protocol.
43166 The @samp{qXfer:libraries-svr4:read} packet returns an XML document which lists
43167 loaded libraries and their SVR4 linker parameters. For each library on SVR4
43168 target, the following parameters are reported:
43172 @code{name}, the absolute file name from the @code{l_name} field of
43173 @code{struct link_map}.
43175 @code{lm} with address of @code{struct link_map} used for TLS
43176 (Thread Local Storage) access.
43178 @code{l_addr}, the displacement as read from the field @code{l_addr} of
43179 @code{struct link_map}. For prelinked libraries this is not an absolute
43180 memory address. It is a displacement of absolute memory address against
43181 address the file was prelinked to during the library load.
43183 @code{l_ld}, which is memory address of the @code{PT_DYNAMIC} segment
43186 Additionally the single @code{main-lm} attribute specifies address of
43187 @code{struct link_map} used for the main executable. This parameter is used
43188 for TLS access and its presence is optional.
43190 @value{GDBN} must be linked with the Expat library to support XML
43191 SVR4 library lists. @xref{Expat}.
43193 A simple memory map, with two loaded libraries (which do not use prelink),
43197 <library-list-svr4 version="1.0" main-lm="0xe4f8f8">
43198 <library name="/lib/ld-linux.so.2" lm="0xe4f51c" l_addr="0xe2d000"
43200 <library name="/lib/libc.so.6" lm="0xe4fbe8" l_addr="0x154000"
43202 </library-list-svr>
43205 The format of an SVR4 library list is described by this DTD:
43208 <!-- library-list-svr4: Root element with versioning -->
43209 <!ELEMENT library-list-svr4 (library)*>
43210 <!ATTLIST library-list-svr4 version CDATA #FIXED "1.0">
43211 <!ATTLIST library-list-svr4 main-lm CDATA #IMPLIED>
43212 <!ELEMENT library EMPTY>
43213 <!ATTLIST library name CDATA #REQUIRED>
43214 <!ATTLIST library lm CDATA #REQUIRED>
43215 <!ATTLIST library l_addr CDATA #REQUIRED>
43216 <!ATTLIST library l_ld CDATA #REQUIRED>
43219 @node Memory Map Format
43220 @section Memory Map Format
43221 @cindex memory map format
43223 To be able to write into flash memory, @value{GDBN} needs to obtain a
43224 memory map from the target. This section describes the format of the
43227 The memory map is obtained using the @samp{qXfer:memory-map:read}
43228 (@pxref{qXfer memory map read}) packet and is an XML document that
43229 lists memory regions.
43231 @value{GDBN} must be linked with the Expat library to support XML
43232 memory maps. @xref{Expat}.
43234 The top-level structure of the document is shown below:
43237 <?xml version="1.0"?>
43238 <!DOCTYPE memory-map
43239 PUBLIC "+//IDN gnu.org//DTD GDB Memory Map V1.0//EN"
43240 "http://sourceware.org/gdb/gdb-memory-map.dtd">
43246 Each region can be either:
43251 A region of RAM starting at @var{addr} and extending for @var{length}
43255 <memory type="ram" start="@var{addr}" length="@var{length}"/>
43260 A region of read-only memory:
43263 <memory type="rom" start="@var{addr}" length="@var{length}"/>
43268 A region of flash memory, with erasure blocks @var{blocksize}
43272 <memory type="flash" start="@var{addr}" length="@var{length}">
43273 <property name="blocksize">@var{blocksize}</property>
43279 Regions must not overlap. @value{GDBN} assumes that areas of memory not covered
43280 by the memory map are RAM, and uses the ordinary @samp{M} and @samp{X}
43281 packets to write to addresses in such ranges.
43283 The formal DTD for memory map format is given below:
43286 <!-- ................................................... -->
43287 <!-- Memory Map XML DTD ................................ -->
43288 <!-- File: memory-map.dtd .............................. -->
43289 <!-- .................................... .............. -->
43290 <!-- memory-map.dtd -->
43291 <!-- memory-map: Root element with versioning -->
43292 <!ELEMENT memory-map (memory)*>
43293 <!ATTLIST memory-map version CDATA #FIXED "1.0.0">
43294 <!ELEMENT memory (property)*>
43295 <!-- memory: Specifies a memory region,
43296 and its type, or device. -->
43297 <!ATTLIST memory type (ram|rom|flash) #REQUIRED
43298 start CDATA #REQUIRED
43299 length CDATA #REQUIRED>
43300 <!-- property: Generic attribute tag -->
43301 <!ELEMENT property (#PCDATA | property)*>
43302 <!ATTLIST property name (blocksize) #REQUIRED>
43305 @node Thread List Format
43306 @section Thread List Format
43307 @cindex thread list format
43309 To efficiently update the list of threads and their attributes,
43310 @value{GDBN} issues the @samp{qXfer:threads:read} packet
43311 (@pxref{qXfer threads read}) and obtains the XML document with
43312 the following structure:
43315 <?xml version="1.0"?>
43317 <thread id="id" core="0" name="name">
43318 ... description ...
43323 Each @samp{thread} element must have the @samp{id} attribute that
43324 identifies the thread (@pxref{thread-id syntax}). The
43325 @samp{core} attribute, if present, specifies which processor core
43326 the thread was last executing on. The @samp{name} attribute, if
43327 present, specifies the human-readable name of the thread. The content
43328 of the of @samp{thread} element is interpreted as human-readable
43329 auxiliary information. The @samp{handle} attribute, if present,
43330 is a hex encoded representation of the thread handle.
43333 @node Traceframe Info Format
43334 @section Traceframe Info Format
43335 @cindex traceframe info format
43337 To be able to know which objects in the inferior can be examined when
43338 inspecting a tracepoint hit, @value{GDBN} needs to obtain the list of
43339 memory ranges, registers and trace state variables that have been
43340 collected in a traceframe.
43342 This list is obtained using the @samp{qXfer:traceframe-info:read}
43343 (@pxref{qXfer traceframe info read}) packet and is an XML document.
43345 @value{GDBN} must be linked with the Expat library to support XML
43346 traceframe info discovery. @xref{Expat}.
43348 The top-level structure of the document is shown below:
43351 <?xml version="1.0"?>
43352 <!DOCTYPE traceframe-info
43353 PUBLIC "+//IDN gnu.org//DTD GDB Memory Map V1.0//EN"
43354 "http://sourceware.org/gdb/gdb-traceframe-info.dtd">
43360 Each traceframe block can be either:
43365 A region of collected memory starting at @var{addr} and extending for
43366 @var{length} bytes from there:
43369 <memory start="@var{addr}" length="@var{length}"/>
43373 A block indicating trace state variable numbered @var{number} has been
43377 <tvar id="@var{number}"/>
43382 The formal DTD for the traceframe info format is given below:
43385 <!ELEMENT traceframe-info (memory | tvar)* >
43386 <!ATTLIST traceframe-info version CDATA #FIXED "1.0">
43388 <!ELEMENT memory EMPTY>
43389 <!ATTLIST memory start CDATA #REQUIRED
43390 length CDATA #REQUIRED>
43392 <!ATTLIST tvar id CDATA #REQUIRED>
43395 @node Branch Trace Format
43396 @section Branch Trace Format
43397 @cindex branch trace format
43399 In order to display the branch trace of an inferior thread,
43400 @value{GDBN} needs to obtain the list of branches. This list is
43401 represented as list of sequential code blocks that are connected via
43402 branches. The code in each block has been executed sequentially.
43404 This list is obtained using the @samp{qXfer:btrace:read}
43405 (@pxref{qXfer btrace read}) packet and is an XML document.
43407 @value{GDBN} must be linked with the Expat library to support XML
43408 traceframe info discovery. @xref{Expat}.
43410 The top-level structure of the document is shown below:
43413 <?xml version="1.0"?>
43415 PUBLIC "+//IDN gnu.org//DTD GDB Branch Trace V1.0//EN"
43416 "http://sourceware.org/gdb/gdb-btrace.dtd">
43425 A block of sequentially executed instructions starting at @var{begin}
43426 and ending at @var{end}:
43429 <block begin="@var{begin}" end="@var{end}"/>
43434 The formal DTD for the branch trace format is given below:
43437 <!ELEMENT btrace (block* | pt) >
43438 <!ATTLIST btrace version CDATA #FIXED "1.0">
43440 <!ELEMENT block EMPTY>
43441 <!ATTLIST block begin CDATA #REQUIRED
43442 end CDATA #REQUIRED>
43444 <!ELEMENT pt (pt-config?, raw?)>
43446 <!ELEMENT pt-config (cpu?)>
43448 <!ELEMENT cpu EMPTY>
43449 <!ATTLIST cpu vendor CDATA #REQUIRED
43450 family CDATA #REQUIRED
43451 model CDATA #REQUIRED
43452 stepping CDATA #REQUIRED>
43454 <!ELEMENT raw (#PCDATA)>
43457 @node Branch Trace Configuration Format
43458 @section Branch Trace Configuration Format
43459 @cindex branch trace configuration format
43461 For each inferior thread, @value{GDBN} can obtain the branch trace
43462 configuration using the @samp{qXfer:btrace-conf:read}
43463 (@pxref{qXfer btrace-conf read}) packet.
43465 The configuration describes the branch trace format and configuration
43466 settings for that format. The following information is described:
43470 This thread uses the @dfn{Branch Trace Store} (@acronym{BTS}) format.
43473 The size of the @acronym{BTS} ring buffer in bytes.
43476 This thread uses the @dfn{Intel Processor Trace} (@acronym{Intel
43480 The size of the @acronym{Intel PT} ring buffer in bytes.
43484 @value{GDBN} must be linked with the Expat library to support XML
43485 branch trace configuration discovery. @xref{Expat}.
43487 The formal DTD for the branch trace configuration format is given below:
43490 <!ELEMENT btrace-conf (bts?, pt?)>
43491 <!ATTLIST btrace-conf version CDATA #FIXED "1.0">
43493 <!ELEMENT bts EMPTY>
43494 <!ATTLIST bts size CDATA #IMPLIED>
43496 <!ELEMENT pt EMPTY>
43497 <!ATTLIST pt size CDATA #IMPLIED>
43500 @include agentexpr.texi
43502 @node Target Descriptions
43503 @appendix Target Descriptions
43504 @cindex target descriptions
43506 One of the challenges of using @value{GDBN} to debug embedded systems
43507 is that there are so many minor variants of each processor
43508 architecture in use. It is common practice for vendors to start with
43509 a standard processor core --- ARM, PowerPC, or @acronym{MIPS}, for example ---
43510 and then make changes to adapt it to a particular market niche. Some
43511 architectures have hundreds of variants, available from dozens of
43512 vendors. This leads to a number of problems:
43516 With so many different customized processors, it is difficult for
43517 the @value{GDBN} maintainers to keep up with the changes.
43519 Since individual variants may have short lifetimes or limited
43520 audiences, it may not be worthwhile to carry information about every
43521 variant in the @value{GDBN} source tree.
43523 When @value{GDBN} does support the architecture of the embedded system
43524 at hand, the task of finding the correct architecture name to give the
43525 @command{set architecture} command can be error-prone.
43528 To address these problems, the @value{GDBN} remote protocol allows a
43529 target system to not only identify itself to @value{GDBN}, but to
43530 actually describe its own features. This lets @value{GDBN} support
43531 processor variants it has never seen before --- to the extent that the
43532 descriptions are accurate, and that @value{GDBN} understands them.
43534 @value{GDBN} must be linked with the Expat library to support XML
43535 target descriptions. @xref{Expat}.
43538 * Retrieving Descriptions:: How descriptions are fetched from a target.
43539 * Target Description Format:: The contents of a target description.
43540 * Predefined Target Types:: Standard types available for target
43542 * Enum Target Types:: How to define enum target types.
43543 * Standard Target Features:: Features @value{GDBN} knows about.
43546 @node Retrieving Descriptions
43547 @section Retrieving Descriptions
43549 Target descriptions can be read from the target automatically, or
43550 specified by the user manually. The default behavior is to read the
43551 description from the target. @value{GDBN} retrieves it via the remote
43552 protocol using @samp{qXfer} requests (@pxref{General Query Packets,
43553 qXfer}). The @var{annex} in the @samp{qXfer} packet will be
43554 @samp{target.xml}. The contents of the @samp{target.xml} annex are an
43555 XML document, of the form described in @ref{Target Description
43558 Alternatively, you can specify a file to read for the target description.
43559 If a file is set, the target will not be queried. The commands to
43560 specify a file are:
43563 @cindex set tdesc filename
43564 @item set tdesc filename @var{path}
43565 Read the target description from @var{path}.
43567 @cindex unset tdesc filename
43568 @item unset tdesc filename
43569 Do not read the XML target description from a file. @value{GDBN}
43570 will use the description supplied by the current target.
43572 @cindex show tdesc filename
43573 @item show tdesc filename
43574 Show the filename to read for a target description, if any.
43578 @node Target Description Format
43579 @section Target Description Format
43580 @cindex target descriptions, XML format
43582 A target description annex is an @uref{http://www.w3.org/XML/, XML}
43583 document which complies with the Document Type Definition provided in
43584 the @value{GDBN} sources in @file{gdb/features/gdb-target.dtd}. This
43585 means you can use generally available tools like @command{xmllint} to
43586 check that your feature descriptions are well-formed and valid.
43587 However, to help people unfamiliar with XML write descriptions for
43588 their targets, we also describe the grammar here.
43590 Target descriptions can identify the architecture of the remote target
43591 and (for some architectures) provide information about custom register
43592 sets. They can also identify the OS ABI of the remote target.
43593 @value{GDBN} can use this information to autoconfigure for your
43594 target, or to warn you if you connect to an unsupported target.
43596 Here is a simple target description:
43599 <target version="1.0">
43600 <architecture>i386:x86-64</architecture>
43605 This minimal description only says that the target uses
43606 the x86-64 architecture.
43608 A target description has the following overall form, with [ ] marking
43609 optional elements and @dots{} marking repeatable elements. The elements
43610 are explained further below.
43613 <?xml version="1.0"?>
43614 <!DOCTYPE target SYSTEM "gdb-target.dtd">
43615 <target version="1.0">
43616 @r{[}@var{architecture}@r{]}
43617 @r{[}@var{osabi}@r{]}
43618 @r{[}@var{compatible}@r{]}
43619 @r{[}@var{feature}@dots{}@r{]}
43624 The description is generally insensitive to whitespace and line
43625 breaks, under the usual common-sense rules. The XML version
43626 declaration and document type declaration can generally be omitted
43627 (@value{GDBN} does not require them), but specifying them may be
43628 useful for XML validation tools. The @samp{version} attribute for
43629 @samp{<target>} may also be omitted, but we recommend
43630 including it; if future versions of @value{GDBN} use an incompatible
43631 revision of @file{gdb-target.dtd}, they will detect and report
43632 the version mismatch.
43634 @subsection Inclusion
43635 @cindex target descriptions, inclusion
43638 @cindex <xi:include>
43641 It can sometimes be valuable to split a target description up into
43642 several different annexes, either for organizational purposes, or to
43643 share files between different possible target descriptions. You can
43644 divide a description into multiple files by replacing any element of
43645 the target description with an inclusion directive of the form:
43648 <xi:include href="@var{document}"/>
43652 When @value{GDBN} encounters an element of this form, it will retrieve
43653 the named XML @var{document}, and replace the inclusion directive with
43654 the contents of that document. If the current description was read
43655 using @samp{qXfer}, then so will be the included document;
43656 @var{document} will be interpreted as the name of an annex. If the
43657 current description was read from a file, @value{GDBN} will look for
43658 @var{document} as a file in the same directory where it found the
43659 original description.
43661 @subsection Architecture
43662 @cindex <architecture>
43664 An @samp{<architecture>} element has this form:
43667 <architecture>@var{arch}</architecture>
43670 @var{arch} is one of the architectures from the set accepted by
43671 @code{set architecture} (@pxref{Targets, ,Specifying a Debugging Target}).
43674 @cindex @code{<osabi>}
43676 This optional field was introduced in @value{GDBN} version 7.0.
43677 Previous versions of @value{GDBN} ignore it.
43679 An @samp{<osabi>} element has this form:
43682 <osabi>@var{abi-name}</osabi>
43685 @var{abi-name} is an OS ABI name from the same selection accepted by
43686 @w{@code{set osabi}} (@pxref{ABI, ,Configuring the Current ABI}).
43688 @subsection Compatible Architecture
43689 @cindex @code{<compatible>}
43691 This optional field was introduced in @value{GDBN} version 7.0.
43692 Previous versions of @value{GDBN} ignore it.
43694 A @samp{<compatible>} element has this form:
43697 <compatible>@var{arch}</compatible>
43700 @var{arch} is one of the architectures from the set accepted by
43701 @code{set architecture} (@pxref{Targets, ,Specifying a Debugging Target}).
43703 A @samp{<compatible>} element is used to specify that the target
43704 is able to run binaries in some other than the main target architecture
43705 given by the @samp{<architecture>} element. For example, on the
43706 Cell Broadband Engine, the main architecture is @code{powerpc:common}
43707 or @code{powerpc:common64}, but the system is able to run binaries
43708 in the @code{spu} architecture as well. The way to describe this
43709 capability with @samp{<compatible>} is as follows:
43712 <architecture>powerpc:common</architecture>
43713 <compatible>spu</compatible>
43716 @subsection Features
43719 Each @samp{<feature>} describes some logical portion of the target
43720 system. Features are currently used to describe available CPU
43721 registers and the types of their contents. A @samp{<feature>} element
43725 <feature name="@var{name}">
43726 @r{[}@var{type}@dots{}@r{]}
43732 Each feature's name should be unique within the description. The name
43733 of a feature does not matter unless @value{GDBN} has some special
43734 knowledge of the contents of that feature; if it does, the feature
43735 should have its standard name. @xref{Standard Target Features}.
43739 Any register's value is a collection of bits which @value{GDBN} must
43740 interpret. The default interpretation is a two's complement integer,
43741 but other types can be requested by name in the register description.
43742 Some predefined types are provided by @value{GDBN} (@pxref{Predefined
43743 Target Types}), and the description can define additional composite
43746 Each type element must have an @samp{id} attribute, which gives
43747 a unique (within the containing @samp{<feature>}) name to the type.
43748 Types must be defined before they are used.
43751 Some targets offer vector registers, which can be treated as arrays
43752 of scalar elements. These types are written as @samp{<vector>} elements,
43753 specifying the array element type, @var{type}, and the number of elements,
43757 <vector id="@var{id}" type="@var{type}" count="@var{count}"/>
43761 If a register's value is usefully viewed in multiple ways, define it
43762 with a union type containing the useful representations. The
43763 @samp{<union>} element contains one or more @samp{<field>} elements,
43764 each of which has a @var{name} and a @var{type}:
43767 <union id="@var{id}">
43768 <field name="@var{name}" type="@var{type}"/>
43775 If a register's value is composed from several separate values, define
43776 it with either a structure type or a flags type.
43777 A flags type may only contain bitfields.
43778 A structure type may either contain only bitfields or contain no bitfields.
43779 If the value contains only bitfields, its total size in bytes must be
43782 Non-bitfield values have a @var{name} and @var{type}.
43785 <struct id="@var{id}">
43786 <field name="@var{name}" type="@var{type}"/>
43791 Both @var{name} and @var{type} values are required.
43792 No implicit padding is added.
43794 Bitfield values have a @var{name}, @var{start}, @var{end} and @var{type}.
43797 <struct id="@var{id}" size="@var{size}">
43798 <field name="@var{name}" start="@var{start}" end="@var{end}" type="@var{type}"/>
43804 <flags id="@var{id}" size="@var{size}">
43805 <field name="@var{name}" start="@var{start}" end="@var{end}" type="@var{type}"/>
43810 The @var{name} value is required.
43811 Bitfield values may be named with the empty string, @samp{""},
43812 in which case the field is ``filler'' and its value is not printed.
43813 Not all bits need to be specified, so ``filler'' fields are optional.
43815 The @var{start} and @var{end} values are required, and @var{type}
43817 The field's @var{start} must be less than or equal to its @var{end},
43818 and zero represents the least significant bit.
43820 The default value of @var{type} is @code{bool} for single bit fields,
43821 and an unsigned integer otherwise.
43823 Which to choose? Structures or flags?
43825 Registers defined with @samp{flags} have these advantages over
43826 defining them with @samp{struct}:
43830 Arithmetic may be performed on them as if they were integers.
43832 They are printed in a more readable fashion.
43835 Registers defined with @samp{struct} have one advantage over
43836 defining them with @samp{flags}:
43840 One can fetch individual fields like in @samp{C}.
43843 (gdb) print $my_struct_reg.field3
43849 @subsection Registers
43852 Each register is represented as an element with this form:
43855 <reg name="@var{name}"
43856 bitsize="@var{size}"
43857 @r{[}regnum="@var{num}"@r{]}
43858 @r{[}save-restore="@var{save-restore}"@r{]}
43859 @r{[}type="@var{type}"@r{]}
43860 @r{[}group="@var{group}"@r{]}/>
43864 The components are as follows:
43869 The register's name; it must be unique within the target description.
43872 The register's size, in bits.
43875 The register's number. If omitted, a register's number is one greater
43876 than that of the previous register (either in the current feature or in
43877 a preceding feature); the first register in the target description
43878 defaults to zero. This register number is used to read or write
43879 the register; e.g.@: it is used in the remote @code{p} and @code{P}
43880 packets, and registers appear in the @code{g} and @code{G} packets
43881 in order of increasing register number.
43884 Whether the register should be preserved across inferior function
43885 calls; this must be either @code{yes} or @code{no}. The default is
43886 @code{yes}, which is appropriate for most registers except for
43887 some system control registers; this is not related to the target's
43891 The type of the register. It may be a predefined type, a type
43892 defined in the current feature, or one of the special types @code{int}
43893 and @code{float}. @code{int} is an integer type of the correct size
43894 for @var{bitsize}, and @code{float} is a floating point type (in the
43895 architecture's normal floating point format) of the correct size for
43896 @var{bitsize}. The default is @code{int}.
43899 The register group to which this register belongs. It can be one of the
43900 standard register groups @code{general}, @code{float}, @code{vector} or an
43901 arbitrary string. Group names should be limited to alphanumeric characters.
43902 If a group name is made up of multiple words the words may be separated by
43903 hyphens; e.g.@: @code{special-group} or @code{ultra-special-group}. If no
43904 @var{group} is specified, @value{GDBN} will not display the register in
43905 @code{info registers}.
43909 @node Predefined Target Types
43910 @section Predefined Target Types
43911 @cindex target descriptions, predefined types
43913 Type definitions in the self-description can build up composite types
43914 from basic building blocks, but can not define fundamental types. Instead,
43915 standard identifiers are provided by @value{GDBN} for the fundamental
43916 types. The currently supported types are:
43921 Boolean type, occupying a single bit.
43929 Signed integer types holding the specified number of bits.
43937 Unsigned integer types holding the specified number of bits.
43941 Pointers to unspecified code and data. The program counter and
43942 any dedicated return address register may be marked as code
43943 pointers; printing a code pointer converts it into a symbolic
43944 address. The stack pointer and any dedicated address registers
43945 may be marked as data pointers.
43948 Single precision IEEE floating point.
43951 Double precision IEEE floating point.
43954 The 12-byte extended precision format used by ARM FPA registers.
43957 The 10-byte extended precision format used by x87 registers.
43960 32bit @sc{eflags} register used by x86.
43963 32bit @sc{mxcsr} register used by x86.
43967 @node Enum Target Types
43968 @section Enum Target Types
43969 @cindex target descriptions, enum types
43971 Enum target types are useful in @samp{struct} and @samp{flags}
43972 register descriptions. @xref{Target Description Format}.
43974 Enum types have a name, size and a list of name/value pairs.
43977 <enum id="@var{id}" size="@var{size}">
43978 <evalue name="@var{name}" value="@var{value}"/>
43983 Enums must be defined before they are used.
43986 <enum id="levels_type" size="4">
43987 <evalue name="low" value="0"/>
43988 <evalue name="high" value="1"/>
43990 <flags id="flags_type" size="4">
43991 <field name="X" start="0"/>
43992 <field name="LEVEL" start="1" end="1" type="levels_type"/>
43994 <reg name="flags" bitsize="32" type="flags_type"/>
43997 Given that description, a value of 3 for the @samp{flags} register
43998 would be printed as:
44001 (gdb) info register flags
44002 flags 0x3 [ X LEVEL=high ]
44005 @node Standard Target Features
44006 @section Standard Target Features
44007 @cindex target descriptions, standard features
44009 A target description must contain either no registers or all the
44010 target's registers. If the description contains no registers, then
44011 @value{GDBN} will assume a default register layout, selected based on
44012 the architecture. If the description contains any registers, the
44013 default layout will not be used; the standard registers must be
44014 described in the target description, in such a way that @value{GDBN}
44015 can recognize them.
44017 This is accomplished by giving specific names to feature elements
44018 which contain standard registers. @value{GDBN} will look for features
44019 with those names and verify that they contain the expected registers;
44020 if any known feature is missing required registers, or if any required
44021 feature is missing, @value{GDBN} will reject the target
44022 description. You can add additional registers to any of the
44023 standard features --- @value{GDBN} will display them just as if
44024 they were added to an unrecognized feature.
44026 This section lists the known features and their expected contents.
44027 Sample XML documents for these features are included in the
44028 @value{GDBN} source tree, in the directory @file{gdb/features}.
44030 Names recognized by @value{GDBN} should include the name of the
44031 company or organization which selected the name, and the overall
44032 architecture to which the feature applies; so e.g.@: the feature
44033 containing ARM core registers is named @samp{org.gnu.gdb.arm.core}.
44035 The names of registers are not case sensitive for the purpose
44036 of recognizing standard features, but @value{GDBN} will only display
44037 registers using the capitalization used in the description.
44040 * AArch64 Features::
44044 * MicroBlaze Features::
44048 * Nios II Features::
44049 * OpenRISC 1000 Features::
44050 * PowerPC Features::
44051 * RISC-V Features::
44052 * S/390 and System z Features::
44058 @node AArch64 Features
44059 @subsection AArch64 Features
44060 @cindex target descriptions, AArch64 features
44062 The @samp{org.gnu.gdb.aarch64.core} feature is required for AArch64
44063 targets. It should contain registers @samp{x0} through @samp{x30},
44064 @samp{sp}, @samp{pc}, and @samp{cpsr}.
44066 The @samp{org.gnu.gdb.aarch64.fpu} feature is optional. If present,
44067 it should contain registers @samp{v0} through @samp{v31}, @samp{fpsr},
44070 The @samp{org.gnu.gdb.aarch64.sve} feature is optional. If present,
44071 it should contain registers @samp{z0} through @samp{z31}, @samp{p0}
44072 through @samp{p15}, @samp{ffr} and @samp{vg}.
44074 The @samp{org.gnu.gdb.aarch64.pauth} feature is optional. If present,
44075 it should contain registers @samp{pauth_dmask} and @samp{pauth_cmask}.
44078 @subsection ARC Features
44079 @cindex target descriptions, ARC Features
44081 ARC processors are highly configurable, so even core registers and their number
44082 are not completely predetermined. In addition flags and PC registers which are
44083 important to @value{GDBN} are not ``core'' registers in ARC. It is required
44084 that one of the core registers features is present.
44085 @samp{org.gnu.gdb.arc.aux-minimal} feature is mandatory.
44087 The @samp{org.gnu.gdb.arc.core.v2} feature is required for ARC EM and ARC HS
44088 targets with a normal register file. It should contain registers @samp{r0}
44089 through @samp{r25}, @samp{gp}, @samp{fp}, @samp{sp}, @samp{r30}, @samp{blink},
44090 @samp{lp_count} and @samp{pcl}. This feature may contain register @samp{ilink}
44091 and any of extension core registers @samp{r32} through @samp{r59/acch}.
44092 @samp{ilink} and extension core registers are not available to read/write, when
44093 debugging GNU/Linux applications, thus @samp{ilink} is made optional.
44095 The @samp{org.gnu.gdb.arc.core-reduced.v2} feature is required for ARC EM and
44096 ARC HS targets with a reduced register file. It should contain registers
44097 @samp{r0} through @samp{r3}, @samp{r10} through @samp{r15}, @samp{gp},
44098 @samp{fp}, @samp{sp}, @samp{r30}, @samp{blink}, @samp{lp_count} and @samp{pcl}.
44099 This feature may contain register @samp{ilink} and any of extension core
44100 registers @samp{r32} through @samp{r59/acch}.
44102 The @samp{org.gnu.gdb.arc.core.arcompact} feature is required for ARCompact
44103 targets with a normal register file. It should contain registers @samp{r0}
44104 through @samp{r25}, @samp{gp}, @samp{fp}, @samp{sp}, @samp{r30}, @samp{blink},
44105 @samp{lp_count} and @samp{pcl}. This feature may contain registers
44106 @samp{ilink1}, @samp{ilink2} and any of extension core registers @samp{r32}
44107 through @samp{r59/acch}. @samp{ilink1} and @samp{ilink2} and extension core
44108 registers are not available when debugging GNU/Linux applications. The only
44109 difference with @samp{org.gnu.gdb.arc.core.v2} feature is in the names of
44110 @samp{ilink1} and @samp{ilink2} registers and that @samp{r30} is mandatory in
44111 ARC v2, but @samp{ilink2} is optional on ARCompact.
44113 The @samp{org.gnu.gdb.arc.aux-minimal} feature is required for all ARC
44114 targets. It should contain registers @samp{pc} and @samp{status32}.
44117 @subsection ARM Features
44118 @cindex target descriptions, ARM features
44120 The @samp{org.gnu.gdb.arm.core} feature is required for non-M-profile
44122 It should contain registers @samp{r0} through @samp{r13}, @samp{sp},
44123 @samp{lr}, @samp{pc}, and @samp{cpsr}.
44125 For M-profile targets (e.g. Cortex-M3), the @samp{org.gnu.gdb.arm.core}
44126 feature is replaced by @samp{org.gnu.gdb.arm.m-profile}. It should contain
44127 registers @samp{r0} through @samp{r13}, @samp{sp}, @samp{lr}, @samp{pc},
44130 The @samp{org.gnu.gdb.arm.fpa} feature is optional. If present, it
44131 should contain registers @samp{f0} through @samp{f7} and @samp{fps}.
44133 The @samp{org.gnu.gdb.xscale.iwmmxt} feature is optional. If present,
44134 it should contain at least registers @samp{wR0} through @samp{wR15} and
44135 @samp{wCGR0} through @samp{wCGR3}. The @samp{wCID}, @samp{wCon},
44136 @samp{wCSSF}, and @samp{wCASF} registers are optional.
44138 The @samp{org.gnu.gdb.arm.vfp} feature is optional. If present, it
44139 should contain at least registers @samp{d0} through @samp{d15}. If
44140 they are present, @samp{d16} through @samp{d31} should also be included.
44141 @value{GDBN} will synthesize the single-precision registers from
44142 halves of the double-precision registers.
44144 The @samp{org.gnu.gdb.arm.neon} feature is optional. It does not
44145 need to contain registers; it instructs @value{GDBN} to display the
44146 VFP double-precision registers as vectors and to synthesize the
44147 quad-precision registers from pairs of double-precision registers.
44148 If this feature is present, @samp{org.gnu.gdb.arm.vfp} must also
44149 be present and include 32 double-precision registers.
44151 @node i386 Features
44152 @subsection i386 Features
44153 @cindex target descriptions, i386 features
44155 The @samp{org.gnu.gdb.i386.core} feature is required for i386/amd64
44156 targets. It should describe the following registers:
44160 @samp{eax} through @samp{edi} plus @samp{eip} for i386
44162 @samp{rax} through @samp{r15} plus @samp{rip} for amd64
44164 @samp{eflags}, @samp{cs}, @samp{ss}, @samp{ds}, @samp{es},
44165 @samp{fs}, @samp{gs}
44167 @samp{st0} through @samp{st7}
44169 @samp{fctrl}, @samp{fstat}, @samp{ftag}, @samp{fiseg}, @samp{fioff},
44170 @samp{foseg}, @samp{fooff} and @samp{fop}
44173 The register sets may be different, depending on the target.
44175 The @samp{org.gnu.gdb.i386.sse} feature is optional. It should
44176 describe registers:
44180 @samp{xmm0} through @samp{xmm7} for i386
44182 @samp{xmm0} through @samp{xmm15} for amd64
44187 The @samp{org.gnu.gdb.i386.avx} feature is optional and requires the
44188 @samp{org.gnu.gdb.i386.sse} feature. It should
44189 describe the upper 128 bits of @sc{ymm} registers:
44193 @samp{ymm0h} through @samp{ymm7h} for i386
44195 @samp{ymm0h} through @samp{ymm15h} for amd64
44198 The @samp{org.gnu.gdb.i386.mpx} is an optional feature representing Intel
44199 Memory Protection Extension (MPX). It should describe the following registers:
44203 @samp{bnd0raw} through @samp{bnd3raw} for i386 and amd64.
44205 @samp{bndcfgu} and @samp{bndstatus} for i386 and amd64.
44208 The @samp{org.gnu.gdb.i386.linux} feature is optional. It should
44209 describe a single register, @samp{orig_eax}.
44211 The @samp{org.gnu.gdb.i386.segments} feature is optional. It should
44212 describe two system registers: @samp{fs_base} and @samp{gs_base}.
44214 The @samp{org.gnu.gdb.i386.avx512} feature is optional and requires the
44215 @samp{org.gnu.gdb.i386.avx} feature. It should
44216 describe additional @sc{xmm} registers:
44220 @samp{xmm16h} through @samp{xmm31h}, only valid for amd64.
44223 It should describe the upper 128 bits of additional @sc{ymm} registers:
44227 @samp{ymm16h} through @samp{ymm31h}, only valid for amd64.
44231 describe the upper 256 bits of @sc{zmm} registers:
44235 @samp{zmm0h} through @samp{zmm7h} for i386.
44237 @samp{zmm0h} through @samp{zmm15h} for amd64.
44241 describe the additional @sc{zmm} registers:
44245 @samp{zmm16h} through @samp{zmm31h}, only valid for amd64.
44248 The @samp{org.gnu.gdb.i386.pkeys} feature is optional. It should
44249 describe a single register, @samp{pkru}. It is a 32-bit register
44250 valid for i386 and amd64.
44252 @node MicroBlaze Features
44253 @subsection MicroBlaze Features
44254 @cindex target descriptions, MicroBlaze features
44256 The @samp{org.gnu.gdb.microblaze.core} feature is required for MicroBlaze
44257 targets. It should contain registers @samp{r0} through @samp{r31},
44258 @samp{rpc}, @samp{rmsr}, @samp{rear}, @samp{resr}, @samp{rfsr}, @samp{rbtr},
44259 @samp{rpvr}, @samp{rpvr1} through @samp{rpvr11}, @samp{redr}, @samp{rpid},
44260 @samp{rzpr}, @samp{rtlbx}, @samp{rtlbsx}, @samp{rtlblo}, and @samp{rtlbhi}.
44262 The @samp{org.gnu.gdb.microblaze.stack-protect} feature is optional.
44263 If present, it should contain registers @samp{rshr} and @samp{rslr}
44265 @node MIPS Features
44266 @subsection @acronym{MIPS} Features
44267 @cindex target descriptions, @acronym{MIPS} features
44269 The @samp{org.gnu.gdb.mips.cpu} feature is required for @acronym{MIPS} targets.
44270 It should contain registers @samp{r0} through @samp{r31}, @samp{lo},
44271 @samp{hi}, and @samp{pc}. They may be 32-bit or 64-bit depending
44274 The @samp{org.gnu.gdb.mips.cp0} feature is also required. It should
44275 contain at least the @samp{status}, @samp{badvaddr}, and @samp{cause}
44276 registers. They may be 32-bit or 64-bit depending on the target.
44278 The @samp{org.gnu.gdb.mips.fpu} feature is currently required, though
44279 it may be optional in a future version of @value{GDBN}. It should
44280 contain registers @samp{f0} through @samp{f31}, @samp{fcsr}, and
44281 @samp{fir}. They may be 32-bit or 64-bit depending on the target.
44283 The @samp{org.gnu.gdb.mips.dsp} feature is optional. It should
44284 contain registers @samp{hi1} through @samp{hi3}, @samp{lo1} through
44285 @samp{lo3}, and @samp{dspctl}. The @samp{dspctl} register should
44286 be 32-bit and the rest may be 32-bit or 64-bit depending on the target.
44288 The @samp{org.gnu.gdb.mips.linux} feature is optional. It should
44289 contain a single register, @samp{restart}, which is used by the
44290 Linux kernel to control restartable syscalls.
44292 @node M68K Features
44293 @subsection M68K Features
44294 @cindex target descriptions, M68K features
44297 @item @samp{org.gnu.gdb.m68k.core}
44298 @itemx @samp{org.gnu.gdb.coldfire.core}
44299 @itemx @samp{org.gnu.gdb.fido.core}
44300 One of those features must be always present.
44301 The feature that is present determines which flavor of m68k is
44302 used. The feature that is present should contain registers
44303 @samp{d0} through @samp{d7}, @samp{a0} through @samp{a5}, @samp{fp},
44304 @samp{sp}, @samp{ps} and @samp{pc}.
44306 @item @samp{org.gnu.gdb.coldfire.fp}
44307 This feature is optional. If present, it should contain registers
44308 @samp{fp0} through @samp{fp7}, @samp{fpcontrol}, @samp{fpstatus} and
44312 @node NDS32 Features
44313 @subsection NDS32 Features
44314 @cindex target descriptions, NDS32 features
44316 The @samp{org.gnu.gdb.nds32.core} feature is required for NDS32
44317 targets. It should contain at least registers @samp{r0} through
44318 @samp{r10}, @samp{r15}, @samp{fp}, @samp{gp}, @samp{lp}, @samp{sp},
44321 The @samp{org.gnu.gdb.nds32.fpu} feature is optional. If present,
44322 it should contain 64-bit double-precision floating-point registers
44323 @samp{fd0} through @emph{fdN}, which should be @samp{fd3}, @samp{fd7},
44324 @samp{fd15}, or @samp{fd31} based on the FPU configuration implemented.
44326 @emph{Note:} The first sixteen 64-bit double-precision floating-point
44327 registers are overlapped with the thirty-two 32-bit single-precision
44328 floating-point registers. The 32-bit single-precision registers, if
44329 not being listed explicitly, will be synthesized from halves of the
44330 overlapping 64-bit double-precision registers. Listing 32-bit
44331 single-precision registers explicitly is deprecated, and the
44332 support to it could be totally removed some day.
44334 @node Nios II Features
44335 @subsection Nios II Features
44336 @cindex target descriptions, Nios II features
44338 The @samp{org.gnu.gdb.nios2.cpu} feature is required for Nios II
44339 targets. It should contain the 32 core registers (@samp{zero},
44340 @samp{at}, @samp{r2} through @samp{r23}, @samp{et} through @samp{ra}),
44341 @samp{pc}, and the 16 control registers (@samp{status} through
44344 @node OpenRISC 1000 Features
44345 @subsection Openrisc 1000 Features
44346 @cindex target descriptions, OpenRISC 1000 features
44348 The @samp{org.gnu.gdb.or1k.group0} feature is required for OpenRISC 1000
44349 targets. It should contain the 32 general purpose registers (@samp{r0}
44350 through @samp{r31}), @samp{ppc}, @samp{npc} and @samp{sr}.
44352 @node PowerPC Features
44353 @subsection PowerPC Features
44354 @cindex target descriptions, PowerPC features
44356 The @samp{org.gnu.gdb.power.core} feature is required for PowerPC
44357 targets. It should contain registers @samp{r0} through @samp{r31},
44358 @samp{pc}, @samp{msr}, @samp{cr}, @samp{lr}, @samp{ctr}, and
44359 @samp{xer}. They may be 32-bit or 64-bit depending on the target.
44361 The @samp{org.gnu.gdb.power.fpu} feature is optional. It should
44362 contain registers @samp{f0} through @samp{f31} and @samp{fpscr}.
44364 The @samp{org.gnu.gdb.power.altivec} feature is optional. It should
44365 contain registers @samp{vr0} through @samp{vr31}, @samp{vscr}, and
44366 @samp{vrsave}. @value{GDBN} will define pseudo-registers @samp{v0}
44367 through @samp{v31} as aliases for the corresponding @samp{vrX}
44370 The @samp{org.gnu.gdb.power.vsx} feature is optional. It should
44371 contain registers @samp{vs0h} through @samp{vs31h}. @value{GDBN} will
44372 combine these registers with the floating point registers (@samp{f0}
44373 through @samp{f31}) and the altivec registers (@samp{vr0} through
44374 @samp{vr31}) to present the 128-bit wide registers @samp{vs0} through
44375 @samp{vs63}, the set of vector-scalar registers for POWER7.
44376 Therefore, this feature requires both @samp{org.gnu.gdb.power.fpu} and
44377 @samp{org.gnu.gdb.power.altivec}.
44379 The @samp{org.gnu.gdb.power.spe} feature is optional. It should
44380 contain registers @samp{ev0h} through @samp{ev31h}, @samp{acc}, and
44381 @samp{spefscr}. SPE targets should provide 32-bit registers in
44382 @samp{org.gnu.gdb.power.core} and provide the upper halves in
44383 @samp{ev0h} through @samp{ev31h}. @value{GDBN} will combine
44384 these to present registers @samp{ev0} through @samp{ev31} to the
44387 The @samp{org.gnu.gdb.power.ppr} feature is optional. It should
44388 contain the 64-bit register @samp{ppr}.
44390 The @samp{org.gnu.gdb.power.dscr} feature is optional. It should
44391 contain the 64-bit register @samp{dscr}.
44393 The @samp{org.gnu.gdb.power.tar} feature is optional. It should
44394 contain the 64-bit register @samp{tar}.
44396 The @samp{org.gnu.gdb.power.ebb} feature is optional. It should
44397 contain registers @samp{bescr}, @samp{ebbhr} and @samp{ebbrr}, all
44400 The @samp{org.gnu.gdb.power.linux.pmu} feature is optional. It should
44401 contain registers @samp{mmcr0}, @samp{mmcr2}, @samp{siar}, @samp{sdar}
44402 and @samp{sier}, all 64-bit wide. This is the subset of the isa 2.07
44403 server PMU registers provided by @sc{gnu}/Linux.
44405 The @samp{org.gnu.gdb.power.htm.spr} feature is optional. It should
44406 contain registers @samp{tfhar}, @samp{texasr} and @samp{tfiar}, all
44409 The @samp{org.gnu.gdb.power.htm.core} feature is optional. It should
44410 contain the checkpointed general-purpose registers @samp{cr0} through
44411 @samp{cr31}, as well as the checkpointed registers @samp{clr} and
44412 @samp{cctr}. These registers may all be either 32-bit or 64-bit
44413 depending on the target. It should also contain the checkpointed
44414 registers @samp{ccr} and @samp{cxer}, which should both be 32-bit
44417 The @samp{org.gnu.gdb.power.htm.fpu} feature is optional. It should
44418 contain the checkpointed 64-bit floating-point registers @samp{cf0}
44419 through @samp{cf31}, as well as the checkpointed 64-bit register
44422 The @samp{org.gnu.gdb.power.htm.altivec} feature is optional. It
44423 should contain the checkpointed altivec registers @samp{cvr0} through
44424 @samp{cvr31}, all 128-bit wide. It should also contain the
44425 checkpointed registers @samp{cvscr} and @samp{cvrsave}, both 32-bit
44428 The @samp{org.gnu.gdb.power.htm.vsx} feature is optional. It should
44429 contain registers @samp{cvs0h} through @samp{cvs31h}. @value{GDBN}
44430 will combine these registers with the checkpointed floating point
44431 registers (@samp{cf0} through @samp{cf31}) and the checkpointed
44432 altivec registers (@samp{cvr0} through @samp{cvr31}) to present the
44433 128-bit wide checkpointed vector-scalar registers @samp{cvs0} through
44434 @samp{cvs63}. Therefore, this feature requires both
44435 @samp{org.gnu.gdb.power.htm.altivec} and
44436 @samp{org.gnu.gdb.power.htm.fpu}.
44438 The @samp{org.gnu.gdb.power.htm.ppr} feature is optional. It should
44439 contain the 64-bit checkpointed register @samp{cppr}.
44441 The @samp{org.gnu.gdb.power.htm.dscr} feature is optional. It should
44442 contain the 64-bit checkpointed register @samp{cdscr}.
44444 The @samp{org.gnu.gdb.power.htm.tar} feature is optional. It should
44445 contain the 64-bit checkpointed register @samp{ctar}.
44448 @node RISC-V Features
44449 @subsection RISC-V Features
44450 @cindex target descriptions, RISC-V Features
44452 The @samp{org.gnu.gdb.riscv.cpu} feature is required for RISC-V
44453 targets. It should contain the registers @samp{x0} through
44454 @samp{x31}, and @samp{pc}. Either the architectural names (@samp{x0},
44455 @samp{x1}, etc) can be used, or the ABI names (@samp{zero}, @samp{ra},
44458 The @samp{org.gnu.gdb.riscv.fpu} feature is optional. If present, it
44459 should contain registers @samp{f0} through @samp{f31}, @samp{fflags},
44460 @samp{frm}, and @samp{fcsr}. As with the cpu feature, either the
44461 architectural register names, or the ABI names can be used.
44463 The @samp{org.gnu.gdb.riscv.virtual} feature is optional. If present,
44464 it should contain registers that are not backed by real registers on
44465 the target, but are instead virtual, where the register value is
44466 derived from other target state. In many ways these are like
44467 @value{GDBN}s pseudo-registers, except implemented by the target.
44468 Currently the only register expected in this set is the one byte
44469 @samp{priv} register that contains the target's privilege level in the
44470 least significant two bits.
44472 The @samp{org.gnu.gdb.riscv.csr} feature is optional. If present, it
44473 should contain all of the target's standard CSRs. Standard CSRs are
44474 those defined in the RISC-V specification documents. There is some
44475 overlap between this feature and the fpu feature; the @samp{fflags},
44476 @samp{frm}, and @samp{fcsr} registers could be in either feature. The
44477 expectation is that these registers will be in the fpu feature if the
44478 target has floating point hardware, but can be moved into the csr
44479 feature if the target has the floating point control registers, but no
44480 other floating point hardware.
44482 @node S/390 and System z Features
44483 @subsection S/390 and System z Features
44484 @cindex target descriptions, S/390 features
44485 @cindex target descriptions, System z features
44487 The @samp{org.gnu.gdb.s390.core} feature is required for S/390 and
44488 System z targets. It should contain the PSW and the 16 general
44489 registers. In particular, System z targets should provide the 64-bit
44490 registers @samp{pswm}, @samp{pswa}, and @samp{r0} through @samp{r15}.
44491 S/390 targets should provide the 32-bit versions of these registers.
44492 A System z target that runs in 31-bit addressing mode should provide
44493 32-bit versions of @samp{pswm} and @samp{pswa}, as well as the general
44494 register's upper halves @samp{r0h} through @samp{r15h}, and their
44495 lower halves @samp{r0l} through @samp{r15l}.
44497 The @samp{org.gnu.gdb.s390.fpr} feature is required. It should
44498 contain the 64-bit registers @samp{f0} through @samp{f15}, and
44501 The @samp{org.gnu.gdb.s390.acr} feature is required. It should
44502 contain the 32-bit registers @samp{acr0} through @samp{acr15}.
44504 The @samp{org.gnu.gdb.s390.linux} feature is optional. It should
44505 contain the register @samp{orig_r2}, which is 64-bit wide on System z
44506 targets and 32-bit otherwise. In addition, the feature may contain
44507 the @samp{last_break} register, whose width depends on the addressing
44508 mode, as well as the @samp{system_call} register, which is always
44511 The @samp{org.gnu.gdb.s390.tdb} feature is optional. It should
44512 contain the 64-bit registers @samp{tdb0}, @samp{tac}, @samp{tct},
44513 @samp{atia}, and @samp{tr0} through @samp{tr15}.
44515 The @samp{org.gnu.gdb.s390.vx} feature is optional. It should contain
44516 64-bit wide registers @samp{v0l} through @samp{v15l}, which will be
44517 combined by @value{GDBN} with the floating point registers @samp{f0}
44518 through @samp{f15} to present the 128-bit wide vector registers
44519 @samp{v0} through @samp{v15}. In addition, this feature should
44520 contain the 128-bit wide vector registers @samp{v16} through
44523 The @samp{org.gnu.gdb.s390.gs} feature is optional. It should contain
44524 the 64-bit wide guarded-storage-control registers @samp{gsd},
44525 @samp{gssm}, and @samp{gsepla}.
44527 The @samp{org.gnu.gdb.s390.gsbc} feature is optional. It should contain
44528 the 64-bit wide guarded-storage broadcast control registers
44529 @samp{bc_gsd}, @samp{bc_gssm}, and @samp{bc_gsepla}.
44531 @node Sparc Features
44532 @subsection Sparc Features
44533 @cindex target descriptions, sparc32 features
44534 @cindex target descriptions, sparc64 features
44535 The @samp{org.gnu.gdb.sparc.cpu} feature is required for sparc32/sparc64
44536 targets. It should describe the following registers:
44540 @samp{g0} through @samp{g7}
44542 @samp{o0} through @samp{o7}
44544 @samp{l0} through @samp{l7}
44546 @samp{i0} through @samp{i7}
44549 They may be 32-bit or 64-bit depending on the target.
44551 Also the @samp{org.gnu.gdb.sparc.fpu} feature is required for sparc32/sparc64
44552 targets. It should describe the following registers:
44556 @samp{f0} through @samp{f31}
44558 @samp{f32} through @samp{f62} for sparc64
44561 The @samp{org.gnu.gdb.sparc.cp0} feature is required for sparc32/sparc64
44562 targets. It should describe the following registers:
44566 @samp{y}, @samp{psr}, @samp{wim}, @samp{tbr}, @samp{pc}, @samp{npc},
44567 @samp{fsr}, and @samp{csr} for sparc32
44569 @samp{pc}, @samp{npc}, @samp{state}, @samp{fsr}, @samp{fprs}, and @samp{y}
44573 @node TIC6x Features
44574 @subsection TMS320C6x Features
44575 @cindex target descriptions, TIC6x features
44576 @cindex target descriptions, TMS320C6x features
44577 The @samp{org.gnu.gdb.tic6x.core} feature is required for TMS320C6x
44578 targets. It should contain registers @samp{A0} through @samp{A15},
44579 registers @samp{B0} through @samp{B15}, @samp{CSR} and @samp{PC}.
44581 The @samp{org.gnu.gdb.tic6x.gp} feature is optional. It should
44582 contain registers @samp{A16} through @samp{A31} and @samp{B16}
44583 through @samp{B31}.
44585 The @samp{org.gnu.gdb.tic6x.c6xp} feature is optional. It should
44586 contain registers @samp{TSR}, @samp{ILC} and @samp{RILC}.
44588 @node Operating System Information
44589 @appendix Operating System Information
44590 @cindex operating system information
44596 Users of @value{GDBN} often wish to obtain information about the state of
44597 the operating system running on the target---for example the list of
44598 processes, or the list of open files. This section describes the
44599 mechanism that makes it possible. This mechanism is similar to the
44600 target features mechanism (@pxref{Target Descriptions}), but focuses
44601 on a different aspect of target.
44603 Operating system information is retrived from the target via the
44604 remote protocol, using @samp{qXfer} requests (@pxref{qXfer osdata
44605 read}). The object name in the request should be @samp{osdata}, and
44606 the @var{annex} identifies the data to be fetched.
44609 @appendixsection Process list
44610 @cindex operating system information, process list
44612 When requesting the process list, the @var{annex} field in the
44613 @samp{qXfer} request should be @samp{processes}. The returned data is
44614 an XML document. The formal syntax of this document is defined in
44615 @file{gdb/features/osdata.dtd}.
44617 An example document is:
44620 <?xml version="1.0"?>
44621 <!DOCTYPE target SYSTEM "osdata.dtd">
44622 <osdata type="processes">
44624 <column name="pid">1</column>
44625 <column name="user">root</column>
44626 <column name="command">/sbin/init</column>
44627 <column name="cores">1,2,3</column>
44632 Each item should include a column whose name is @samp{pid}. The value
44633 of that column should identify the process on the target. The
44634 @samp{user} and @samp{command} columns are optional, and will be
44635 displayed by @value{GDBN}. The @samp{cores} column, if present,
44636 should contain a comma-separated list of cores that this process
44637 is running on. Target may provide additional columns,
44638 which @value{GDBN} currently ignores.
44640 @node Trace File Format
44641 @appendix Trace File Format
44642 @cindex trace file format
44644 The trace file comes in three parts: a header, a textual description
44645 section, and a trace frame section with binary data.
44647 The header has the form @code{\x7fTRACE0\n}. The first byte is
44648 @code{0x7f} so as to indicate that the file contains binary data,
44649 while the @code{0} is a version number that may have different values
44652 The description section consists of multiple lines of @sc{ascii} text
44653 separated by newline characters (@code{0xa}). The lines may include a
44654 variety of optional descriptive or context-setting information, such
44655 as tracepoint definitions or register set size. @value{GDBN} will
44656 ignore any line that it does not recognize. An empty line marks the end
44661 Specifies the size of a register block in bytes. This is equal to the
44662 size of a @code{g} packet payload in the remote protocol. @var{size}
44663 is an ascii decimal number. There should be only one such line in
44664 a single trace file.
44666 @item status @var{status}
44667 Trace status. @var{status} has the same format as a @code{qTStatus}
44668 remote packet reply. There should be only one such line in a single trace
44671 @item tp @var{payload}
44672 Tracepoint definition. The @var{payload} has the same format as
44673 @code{qTfP}/@code{qTsP} remote packet reply payload. A single tracepoint
44674 may take multiple lines of definition, corresponding to the multiple
44677 @item tsv @var{payload}
44678 Trace state variable definition. The @var{payload} has the same format as
44679 @code{qTfV}/@code{qTsV} remote packet reply payload. A single variable
44680 may take multiple lines of definition, corresponding to the multiple
44683 @item tdesc @var{payload}
44684 Target description in XML format. The @var{payload} is a single line of
44685 the XML file. All such lines should be concatenated together to get
44686 the original XML file. This file is in the same format as @code{qXfer}
44687 @code{features} payload, and corresponds to the main @code{target.xml}
44688 file. Includes are not allowed.
44692 The trace frame section consists of a number of consecutive frames.
44693 Each frame begins with a two-byte tracepoint number, followed by a
44694 four-byte size giving the amount of data in the frame. The data in
44695 the frame consists of a number of blocks, each introduced by a
44696 character indicating its type (at least register, memory, and trace
44697 state variable). The data in this section is raw binary, not a
44698 hexadecimal or other encoding; its endianness matches the target's
44701 @c FIXME bi-arch may require endianness/arch info in description section
44704 @item R @var{bytes}
44705 Register block. The number and ordering of bytes matches that of a
44706 @code{g} packet in the remote protocol. Note that these are the
44707 actual bytes, in target order, not a hexadecimal encoding.
44709 @item M @var{address} @var{length} @var{bytes}...
44710 Memory block. This is a contiguous block of memory, at the 8-byte
44711 address @var{address}, with a 2-byte length @var{length}, followed by
44712 @var{length} bytes.
44714 @item V @var{number} @var{value}
44715 Trace state variable block. This records the 8-byte signed value
44716 @var{value} of trace state variable numbered @var{number}.
44720 Future enhancements of the trace file format may include additional types
44723 @node Index Section Format
44724 @appendix @code{.gdb_index} section format
44725 @cindex .gdb_index section format
44726 @cindex index section format
44728 This section documents the index section that is created by @code{save
44729 gdb-index} (@pxref{Index Files}). The index section is
44730 DWARF-specific; some knowledge of DWARF is assumed in this
44733 The mapped index file format is designed to be directly
44734 @code{mmap}able on any architecture. In most cases, a datum is
44735 represented using a little-endian 32-bit integer value, called an
44736 @code{offset_type}. Big endian machines must byte-swap the values
44737 before using them. Exceptions to this rule are noted. The data is
44738 laid out such that alignment is always respected.
44740 A mapped index consists of several areas, laid out in order.
44744 The file header. This is a sequence of values, of @code{offset_type}
44745 unless otherwise noted:
44749 The version number, currently 8. Versions 1, 2 and 3 are obsolete.
44750 Version 4 uses a different hashing function from versions 5 and 6.
44751 Version 6 includes symbols for inlined functions, whereas versions 4
44752 and 5 do not. Version 7 adds attributes to the CU indices in the
44753 symbol table. Version 8 specifies that symbols from DWARF type units
44754 (@samp{DW_TAG_type_unit}) refer to the type unit's symbol table and not the
44755 compilation unit (@samp{DW_TAG_comp_unit}) using the type.
44757 @value{GDBN} will only read version 4, 5, or 6 indices
44758 by specifying @code{set use-deprecated-index-sections on}.
44759 GDB has a workaround for potentially broken version 7 indices so it is
44760 currently not flagged as deprecated.
44763 The offset, from the start of the file, of the CU list.
44766 The offset, from the start of the file, of the types CU list. Note
44767 that this area can be empty, in which case this offset will be equal
44768 to the next offset.
44771 The offset, from the start of the file, of the address area.
44774 The offset, from the start of the file, of the symbol table.
44777 The offset, from the start of the file, of the constant pool.
44781 The CU list. This is a sequence of pairs of 64-bit little-endian
44782 values, sorted by the CU offset. The first element in each pair is
44783 the offset of a CU in the @code{.debug_info} section. The second
44784 element in each pair is the length of that CU. References to a CU
44785 elsewhere in the map are done using a CU index, which is just the
44786 0-based index into this table. Note that if there are type CUs, then
44787 conceptually CUs and type CUs form a single list for the purposes of
44791 The types CU list. This is a sequence of triplets of 64-bit
44792 little-endian values. In a triplet, the first value is the CU offset,
44793 the second value is the type offset in the CU, and the third value is
44794 the type signature. The types CU list is not sorted.
44797 The address area. The address area consists of a sequence of address
44798 entries. Each address entry has three elements:
44802 The low address. This is a 64-bit little-endian value.
44805 The high address. This is a 64-bit little-endian value. Like
44806 @code{DW_AT_high_pc}, the value is one byte beyond the end.
44809 The CU index. This is an @code{offset_type} value.
44813 The symbol table. This is an open-addressed hash table. The size of
44814 the hash table is always a power of 2.
44816 Each slot in the hash table consists of a pair of @code{offset_type}
44817 values. The first value is the offset of the symbol's name in the
44818 constant pool. The second value is the offset of the CU vector in the
44821 If both values are 0, then this slot in the hash table is empty. This
44822 is ok because while 0 is a valid constant pool index, it cannot be a
44823 valid index for both a string and a CU vector.
44825 The hash value for a table entry is computed by applying an
44826 iterative hash function to the symbol's name. Starting with an
44827 initial value of @code{r = 0}, each (unsigned) character @samp{c} in
44828 the string is incorporated into the hash using the formula depending on the
44833 The formula is @code{r = r * 67 + c - 113}.
44835 @item Versions 5 to 7
44836 The formula is @code{r = r * 67 + tolower (c) - 113}.
44839 The terminating @samp{\0} is not incorporated into the hash.
44841 The step size used in the hash table is computed via
44842 @code{((hash * 17) & (size - 1)) | 1}, where @samp{hash} is the hash
44843 value, and @samp{size} is the size of the hash table. The step size
44844 is used to find the next candidate slot when handling a hash
44847 The names of C@t{++} symbols in the hash table are canonicalized. We
44848 don't currently have a simple description of the canonicalization
44849 algorithm; if you intend to create new index sections, you must read
44853 The constant pool. This is simply a bunch of bytes. It is organized
44854 so that alignment is correct: CU vectors are stored first, followed by
44857 A CU vector in the constant pool is a sequence of @code{offset_type}
44858 values. The first value is the number of CU indices in the vector.
44859 Each subsequent value is the index and symbol attributes of a CU in
44860 the CU list. This element in the hash table is used to indicate which
44861 CUs define the symbol and how the symbol is used.
44862 See below for the format of each CU index+attributes entry.
44864 A string in the constant pool is zero-terminated.
44867 Attributes were added to CU index values in @code{.gdb_index} version 7.
44868 If a symbol has multiple uses within a CU then there is one
44869 CU index+attributes value for each use.
44871 The format of each CU index+attributes entry is as follows
44877 This is the index of the CU in the CU list.
44879 These bits are reserved for future purposes and must be zero.
44881 The kind of the symbol in the CU.
44885 This value is reserved and should not be used.
44886 By reserving zero the full @code{offset_type} value is backwards compatible
44887 with previous versions of the index.
44889 The symbol is a type.
44891 The symbol is a variable or an enum value.
44893 The symbol is a function.
44895 Any other kind of symbol.
44897 These values are reserved.
44901 This bit is zero if the value is global and one if it is static.
44903 The determination of whether a symbol is global or static is complicated.
44904 The authorative reference is the file @file{dwarf2read.c} in
44905 @value{GDBN} sources.
44909 This pseudo-code describes the computation of a symbol's kind and
44910 global/static attributes in the index.
44913 is_external = get_attribute (die, DW_AT_external);
44914 language = get_attribute (cu_die, DW_AT_language);
44917 case DW_TAG_typedef:
44918 case DW_TAG_base_type:
44919 case DW_TAG_subrange_type:
44923 case DW_TAG_enumerator:
44925 is_static = language != CPLUS;
44927 case DW_TAG_subprogram:
44929 is_static = ! (is_external || language == ADA);
44931 case DW_TAG_constant:
44933 is_static = ! is_external;
44935 case DW_TAG_variable:
44937 is_static = ! is_external;
44939 case DW_TAG_namespace:
44943 case DW_TAG_class_type:
44944 case DW_TAG_interface_type:
44945 case DW_TAG_structure_type:
44946 case DW_TAG_union_type:
44947 case DW_TAG_enumeration_type:
44949 is_static = language != CPLUS;
44957 @appendix Manual pages
44961 * gdb man:: The GNU Debugger man page
44962 * gdbserver man:: Remote Server for the GNU Debugger man page
44963 * gcore man:: Generate a core file of a running program
44964 * gdbinit man:: gdbinit scripts
44965 * gdb-add-index man:: Add index files to speed up GDB
44971 @c man title gdb The GNU Debugger
44973 @c man begin SYNOPSIS gdb
44974 gdb [@option{-help}] [@option{-nh}] [@option{-nx}] [@option{-q}]
44975 [@option{-batch}] [@option{-cd=}@var{dir}] [@option{-f}]
44976 [@option{-b}@w{ }@var{bps}]
44977 [@option{-tty=}@var{dev}] [@option{-s} @var{symfile}]
44978 [@option{-e}@w{ }@var{prog}] [@option{-se}@w{ }@var{prog}]
44979 [@option{-c}@w{ }@var{core}] [@option{-p}@w{ }@var{procID}]
44980 [@option{-x}@w{ }@var{cmds}] [@option{-d}@w{ }@var{dir}]
44981 [@var{prog}|@var{prog} @var{procID}|@var{prog} @var{core}]
44984 @c man begin DESCRIPTION gdb
44985 The purpose of a debugger such as @value{GDBN} is to allow you to see what is
44986 going on ``inside'' another program while it executes -- or what another
44987 program was doing at the moment it crashed.
44989 @value{GDBN} can do four main kinds of things (plus other things in support of
44990 these) to help you catch bugs in the act:
44994 Start your program, specifying anything that might affect its behavior.
44997 Make your program stop on specified conditions.
45000 Examine what has happened, when your program has stopped.
45003 Change things in your program, so you can experiment with correcting the
45004 effects of one bug and go on to learn about another.
45007 You can use @value{GDBN} to debug programs written in C, C@t{++}, Fortran and
45010 @value{GDBN} is invoked with the shell command @code{gdb}. Once started, it reads
45011 commands from the terminal until you tell it to exit with the @value{GDBN}
45012 command @code{quit}. You can get online help from @value{GDBN} itself
45013 by using the command @code{help}.
45015 You can run @code{gdb} with no arguments or options; but the most
45016 usual way to start @value{GDBN} is with one argument or two, specifying an
45017 executable program as the argument:
45023 You can also start with both an executable program and a core file specified:
45029 You can, instead, specify a process ID as a second argument or use option
45030 @code{-p}, if you want to debug a running process:
45038 would attach @value{GDBN} to process @code{1234}. With option @option{-p} you
45039 can omit the @var{program} filename.
45041 Here are some of the most frequently needed @value{GDBN} commands:
45043 @c pod2man highlights the right hand side of the @item lines.
45045 @item break [@var{file}:]@var{function}
45046 Set a breakpoint at @var{function} (in @var{file}).
45048 @item run [@var{arglist}]
45049 Start your program (with @var{arglist}, if specified).
45052 Backtrace: display the program stack.
45054 @item print @var{expr}
45055 Display the value of an expression.
45058 Continue running your program (after stopping, e.g. at a breakpoint).
45061 Execute next program line (after stopping); step @emph{over} any
45062 function calls in the line.
45064 @item edit [@var{file}:]@var{function}
45065 look at the program line where it is presently stopped.
45067 @item list [@var{file}:]@var{function}
45068 type the text of the program in the vicinity of where it is presently stopped.
45071 Execute next program line (after stopping); step @emph{into} any
45072 function calls in the line.
45074 @item help [@var{name}]
45075 Show information about @value{GDBN} command @var{name}, or general information
45076 about using @value{GDBN}.
45079 Exit from @value{GDBN}.
45083 For full details on @value{GDBN},
45084 see @cite{Using GDB: A Guide to the GNU Source-Level Debugger},
45085 by Richard M. Stallman and Roland H. Pesch. The same text is available online
45086 as the @code{gdb} entry in the @code{info} program.
45090 @c man begin OPTIONS gdb
45091 Any arguments other than options specify an executable
45092 file and core file (or process ID); that is, the first argument
45093 encountered with no
45094 associated option flag is equivalent to a @option{-se} option, and the second,
45095 if any, is equivalent to a @option{-c} option if it's the name of a file.
45097 both long and short forms; both are shown here. The long forms are also
45098 recognized if you truncate them, so long as enough of the option is
45099 present to be unambiguous. (If you prefer, you can flag option
45100 arguments with @option{+} rather than @option{-}, though we illustrate the
45101 more usual convention.)
45103 All the options and command line arguments you give are processed
45104 in sequential order. The order makes a difference when the @option{-x}
45110 List all options, with brief explanations.
45112 @item -symbols=@var{file}
45113 @itemx -s @var{file}
45114 Read symbol table from file @var{file}.
45117 Enable writing into executable and core files.
45119 @item -exec=@var{file}
45120 @itemx -e @var{file}
45121 Use file @var{file} as the executable file to execute when
45122 appropriate, and for examining pure data in conjunction with a core
45125 @item -se=@var{file}
45126 Read symbol table from file @var{file} and use it as the executable
45129 @item -core=@var{file}
45130 @itemx -c @var{file}
45131 Use file @var{file} as a core dump to examine.
45133 @item -command=@var{file}
45134 @itemx -x @var{file}
45135 Execute @value{GDBN} commands from file @var{file}.
45137 @item -ex @var{command}
45138 Execute given @value{GDBN} @var{command}.
45140 @item -directory=@var{directory}
45141 @itemx -d @var{directory}
45142 Add @var{directory} to the path to search for source files.
45145 Do not execute commands from @file{~/.gdbinit}.
45149 Do not execute commands from any @file{.gdbinit} initialization files.
45153 ``Quiet''. Do not print the introductory and copyright messages. These
45154 messages are also suppressed in batch mode.
45157 Run in batch mode. Exit with status @code{0} after processing all the command
45158 files specified with @option{-x} (and @file{.gdbinit}, if not inhibited).
45159 Exit with nonzero status if an error occurs in executing the @value{GDBN}
45160 commands in the command files.
45162 Batch mode may be useful for running @value{GDBN} as a filter, for example to
45163 download and run a program on another computer; in order to make this
45164 more useful, the message
45167 Program exited normally.
45171 (which is ordinarily issued whenever a program running under @value{GDBN} control
45172 terminates) is not issued when running in batch mode.
45174 @item -cd=@var{directory}
45175 Run @value{GDBN} using @var{directory} as its working directory,
45176 instead of the current directory.
45180 Emacs sets this option when it runs @value{GDBN} as a subprocess. It tells
45181 @value{GDBN} to output the full file name and line number in a standard,
45182 recognizable fashion each time a stack frame is displayed (which
45183 includes each time the program stops). This recognizable format looks
45184 like two @samp{\032} characters, followed by the file name, line number
45185 and character position separated by colons, and a newline. The
45186 Emacs-to-@value{GDBN} interface program uses the two @samp{\032}
45187 characters as a signal to display the source code for the frame.
45190 Set the line speed (baud rate or bits per second) of any serial
45191 interface used by @value{GDBN} for remote debugging.
45193 @item -tty=@var{device}
45194 Run using @var{device} for your program's standard input and output.
45198 @c man begin SEEALSO gdb
45200 The full documentation for @value{GDBN} is maintained as a Texinfo manual.
45201 If the @code{info} and @code{gdb} programs and @value{GDBN}'s Texinfo
45202 documentation are properly installed at your site, the command
45209 should give you access to the complete manual.
45211 @cite{Using GDB: A Guide to the GNU Source-Level Debugger},
45212 Richard M. Stallman and Roland H. Pesch, July 1991.
45216 @node gdbserver man
45217 @heading gdbserver man
45219 @c man title gdbserver Remote Server for the GNU Debugger
45221 @c man begin SYNOPSIS gdbserver
45222 gdbserver @var{comm} @var{prog} [@var{args}@dots{}]
45224 gdbserver --attach @var{comm} @var{pid}
45226 gdbserver --multi @var{comm}
45230 @c man begin DESCRIPTION gdbserver
45231 @command{gdbserver} is a program that allows you to run @value{GDBN} on a different machine
45232 than the one which is running the program being debugged.
45235 @subheading Usage (server (target) side)
45238 Usage (server (target) side):
45241 First, you need to have a copy of the program you want to debug put onto
45242 the target system. The program can be stripped to save space if needed, as
45243 @command{gdbserver} doesn't care about symbols. All symbol handling is taken care of by
45244 the @value{GDBN} running on the host system.
45246 To use the server, you log on to the target system, and run the @command{gdbserver}
45247 program. You must tell it (a) how to communicate with @value{GDBN}, (b) the name of
45248 your program, and (c) its arguments. The general syntax is:
45251 target> gdbserver @var{comm} @var{program} [@var{args} ...]
45254 For example, using a serial port, you might say:
45258 @c @file would wrap it as F</dev/com1>.
45259 target> gdbserver /dev/com1 emacs foo.txt
45262 target> gdbserver @file{/dev/com1} emacs foo.txt
45266 This tells @command{gdbserver} to debug emacs with an argument of foo.txt, and
45267 to communicate with @value{GDBN} via @file{/dev/com1}. @command{gdbserver} now
45268 waits patiently for the host @value{GDBN} to communicate with it.
45270 To use a TCP connection, you could say:
45273 target> gdbserver host:2345 emacs foo.txt
45276 This says pretty much the same thing as the last example, except that we are
45277 going to communicate with the @code{host} @value{GDBN} via TCP. The @code{host:2345} argument means
45278 that we are expecting to see a TCP connection from @code{host} to local TCP port
45279 2345. (Currently, the @code{host} part is ignored.) You can choose any number you
45280 want for the port number as long as it does not conflict with any existing TCP
45281 ports on the target system. This same port number must be used in the host
45282 @value{GDBN}s @code{target remote} command, which will be described shortly. Note that if
45283 you chose a port number that conflicts with another service, @command{gdbserver} will
45284 print an error message and exit.
45286 @command{gdbserver} can also attach to running programs.
45287 This is accomplished via the @option{--attach} argument. The syntax is:
45290 target> gdbserver --attach @var{comm} @var{pid}
45293 @var{pid} is the process ID of a currently running process. It isn't
45294 necessary to point @command{gdbserver} at a binary for the running process.
45296 To start @code{gdbserver} without supplying an initial command to run
45297 or process ID to attach, use the @option{--multi} command line option.
45298 In such case you should connect using @kbd{target extended-remote} to start
45299 the program you want to debug.
45302 target> gdbserver --multi @var{comm}
45306 @subheading Usage (host side)
45312 You need an unstripped copy of the target program on your host system, since
45313 @value{GDBN} needs to examine its symbol tables and such. Start up @value{GDBN} as you normally
45314 would, with the target program as the first argument. (You may need to use the
45315 @option{--baud} option if the serial line is running at anything except 9600 baud.)
45316 That is @code{gdb TARGET-PROG}, or @code{gdb --baud BAUD TARGET-PROG}. After that, the only
45317 new command you need to know about is @code{target remote}
45318 (or @code{target extended-remote}). Its argument is either
45319 a device name (usually a serial device, like @file{/dev/ttyb}), or a @code{HOST:PORT}
45320 descriptor. For example:
45324 @c @file would wrap it as F</dev/ttyb>.
45325 (gdb) target remote /dev/ttyb
45328 (gdb) target remote @file{/dev/ttyb}
45333 communicates with the server via serial line @file{/dev/ttyb}, and:
45336 (gdb) target remote the-target:2345
45340 communicates via a TCP connection to port 2345 on host `the-target', where
45341 you previously started up @command{gdbserver} with the same port number. Note that for
45342 TCP connections, you must start up @command{gdbserver} prior to using the `target remote'
45343 command, otherwise you may get an error that looks something like
45344 `Connection refused'.
45346 @command{gdbserver} can also debug multiple inferiors at once,
45349 the @value{GDBN} manual in node @code{Inferiors and Programs}
45350 -- shell command @code{info -f gdb -n 'Inferiors and Programs'}.
45353 @ref{Inferiors and Programs}.
45355 In such case use the @code{extended-remote} @value{GDBN} command variant:
45358 (gdb) target extended-remote the-target:2345
45361 The @command{gdbserver} option @option{--multi} may or may not be used in such
45365 @c man begin OPTIONS gdbserver
45366 There are three different modes for invoking @command{gdbserver}:
45371 Debug a specific program specified by its program name:
45374 gdbserver @var{comm} @var{prog} [@var{args}@dots{}]
45377 The @var{comm} parameter specifies how should the server communicate
45378 with @value{GDBN}; it is either a device name (to use a serial line),
45379 a TCP port number (@code{:1234}), or @code{-} or @code{stdio} to use
45380 stdin/stdout of @code{gdbserver}. Specify the name of the program to
45381 debug in @var{prog}. Any remaining arguments will be passed to the
45382 program verbatim. When the program exits, @value{GDBN} will close the
45383 connection, and @code{gdbserver} will exit.
45386 Debug a specific program by specifying the process ID of a running
45390 gdbserver --attach @var{comm} @var{pid}
45393 The @var{comm} parameter is as described above. Supply the process ID
45394 of a running program in @var{pid}; @value{GDBN} will do everything
45395 else. Like with the previous mode, when the process @var{pid} exits,
45396 @value{GDBN} will close the connection, and @code{gdbserver} will exit.
45399 Multi-process mode -- debug more than one program/process:
45402 gdbserver --multi @var{comm}
45405 In this mode, @value{GDBN} can instruct @command{gdbserver} which
45406 command(s) to run. Unlike the other 2 modes, @value{GDBN} will not
45407 close the connection when a process being debugged exits, so you can
45408 debug several processes in the same session.
45411 In each of the modes you may specify these options:
45416 List all options, with brief explanations.
45419 This option causes @command{gdbserver} to print its version number and exit.
45422 @command{gdbserver} will attach to a running program. The syntax is:
45425 target> gdbserver --attach @var{comm} @var{pid}
45428 @var{pid} is the process ID of a currently running process. It isn't
45429 necessary to point @command{gdbserver} at a binary for the running process.
45432 To start @code{gdbserver} without supplying an initial command to run
45433 or process ID to attach, use this command line option.
45434 Then you can connect using @kbd{target extended-remote} and start
45435 the program you want to debug. The syntax is:
45438 target> gdbserver --multi @var{comm}
45442 Instruct @code{gdbserver} to display extra status information about the debugging
45444 This option is intended for @code{gdbserver} development and for bug reports to
45447 @item --remote-debug
45448 Instruct @code{gdbserver} to display remote protocol debug output.
45449 This option is intended for @code{gdbserver} development and for bug reports to
45452 @item --debug-file=@var{filename}
45453 Instruct @code{gdbserver} to send any debug output to the given @var{filename}.
45454 This option is intended for @code{gdbserver} development and for bug reports to
45457 @item --debug-format=option1@r{[},option2,...@r{]}
45458 Instruct @code{gdbserver} to include extra information in each line
45459 of debugging output.
45460 @xref{Other Command-Line Arguments for gdbserver}.
45463 Specify a wrapper to launch programs
45464 for debugging. The option should be followed by the name of the
45465 wrapper, then any command-line arguments to pass to the wrapper, then
45466 @kbd{--} indicating the end of the wrapper arguments.
45469 By default, @command{gdbserver} keeps the listening TCP port open, so that
45470 additional connections are possible. However, if you start @code{gdbserver}
45471 with the @option{--once} option, it will stop listening for any further
45472 connection attempts after connecting to the first @value{GDBN} session.
45474 @c --disable-packet is not documented for users.
45476 @c --disable-randomization and --no-disable-randomization are superseded by
45477 @c QDisableRandomization.
45482 @c man begin SEEALSO gdbserver
45484 The full documentation for @value{GDBN} is maintained as a Texinfo manual.
45485 If the @code{info} and @code{gdb} programs and @value{GDBN}'s Texinfo
45486 documentation are properly installed at your site, the command
45492 should give you access to the complete manual.
45494 @cite{Using GDB: A Guide to the GNU Source-Level Debugger},
45495 Richard M. Stallman and Roland H. Pesch, July 1991.
45502 @c man title gcore Generate a core file of a running program
45505 @c man begin SYNOPSIS gcore
45506 gcore [-a] [-o @var{prefix}] @var{pid1} [@var{pid2}...@var{pidN}]
45510 @c man begin DESCRIPTION gcore
45511 Generate core dumps of one or more running programs with process IDs
45512 @var{pid1}, @var{pid2}, etc. A core file produced by @command{gcore}
45513 is equivalent to one produced by the kernel when the process crashes
45514 (and when @kbd{ulimit -c} was used to set up an appropriate core dump
45515 limit). However, unlike after a crash, after @command{gcore} finishes
45516 its job the program remains running without any change.
45519 @c man begin OPTIONS gcore
45522 Dump all memory mappings. The actual effect of this option depends on
45523 the Operating System. On @sc{gnu}/Linux, it will disable
45524 @code{use-coredump-filter} (@pxref{set use-coredump-filter}) and
45525 enable @code{dump-excluded-mappings} (@pxref{set
45526 dump-excluded-mappings}).
45528 @item -o @var{prefix}
45529 The optional argument @var{prefix} specifies the prefix to be used
45530 when composing the file names of the core dumps. The file name is
45531 composed as @file{@var{prefix}.@var{pid}}, where @var{pid} is the
45532 process ID of the running program being analyzed by @command{gcore}.
45533 If not specified, @var{prefix} defaults to @var{gcore}.
45537 @c man begin SEEALSO gcore
45539 The full documentation for @value{GDBN} is maintained as a Texinfo manual.
45540 If the @code{info} and @code{gdb} programs and @value{GDBN}'s Texinfo
45541 documentation are properly installed at your site, the command
45548 should give you access to the complete manual.
45550 @cite{Using GDB: A Guide to the GNU Source-Level Debugger},
45551 Richard M. Stallman and Roland H. Pesch, July 1991.
45558 @c man title gdbinit GDB initialization scripts
45561 @c man begin SYNOPSIS gdbinit
45562 @ifset SYSTEM_GDBINIT
45563 @value{SYSTEM_GDBINIT}
45572 @c man begin DESCRIPTION gdbinit
45573 These files contain @value{GDBN} commands to automatically execute during
45574 @value{GDBN} startup. The lines of contents are canned sequences of commands,
45577 the @value{GDBN} manual in node @code{Sequences}
45578 -- shell command @code{info -f gdb -n Sequences}.
45584 Please read more in
45586 the @value{GDBN} manual in node @code{Startup}
45587 -- shell command @code{info -f gdb -n Startup}.
45594 @ifset SYSTEM_GDBINIT
45595 @item @value{SYSTEM_GDBINIT}
45597 @ifclear SYSTEM_GDBINIT
45598 @item (not enabled with @code{--with-system-gdbinit} during compilation)
45600 System-wide initialization file. It is executed unless user specified
45601 @value{GDBN} option @code{-nx} or @code{-n}.
45604 the @value{GDBN} manual in node @code{System-wide configuration}
45605 -- shell command @code{info -f gdb -n 'System-wide configuration'}.
45608 @ref{System-wide configuration}.
45612 User initialization file. It is executed unless user specified
45613 @value{GDBN} options @code{-nx}, @code{-n} or @code{-nh}.
45616 Initialization file for current directory. It may need to be enabled with
45617 @value{GDBN} security command @code{set auto-load local-gdbinit}.
45620 the @value{GDBN} manual in node @code{Init File in the Current Directory}
45621 -- shell command @code{info -f gdb -n 'Init File in the Current Directory'}.
45624 @ref{Init File in the Current Directory}.
45629 @c man begin SEEALSO gdbinit
45631 gdb(1), @code{info -f gdb -n Startup}
45633 The full documentation for @value{GDBN} is maintained as a Texinfo manual.
45634 If the @code{info} and @code{gdb} programs and @value{GDBN}'s Texinfo
45635 documentation are properly installed at your site, the command
45641 should give you access to the complete manual.
45643 @cite{Using GDB: A Guide to the GNU Source-Level Debugger},
45644 Richard M. Stallman and Roland H. Pesch, July 1991.
45648 @node gdb-add-index man
45649 @heading gdb-add-index
45650 @pindex gdb-add-index
45651 @anchor{gdb-add-index}
45653 @c man title gdb-add-index Add index files to speed up GDB
45655 @c man begin SYNOPSIS gdb-add-index
45656 gdb-add-index @var{filename}
45659 @c man begin DESCRIPTION gdb-add-index
45660 When @value{GDBN} finds a symbol file, it scans the symbols in the
45661 file in order to construct an internal symbol table. This lets most
45662 @value{GDBN} operations work quickly--at the cost of a delay early on.
45663 For large programs, this delay can be quite lengthy, so @value{GDBN}
45664 provides a way to build an index, which speeds up startup.
45666 To determine whether a file contains such an index, use the command
45667 @kbd{readelf -S filename}: the index is stored in a section named
45668 @code{.gdb_index}. The index file can only be produced on systems
45669 which use ELF binaries and DWARF debug information (i.e., sections
45670 named @code{.debug_*}).
45672 @command{gdb-add-index} uses @value{GDBN} and @command{objdump} found
45673 in the @env{PATH} environment variable. If you want to use different
45674 versions of these programs, you can specify them through the
45675 @env{GDB} and @env{OBJDUMP} environment variables.
45679 the @value{GDBN} manual in node @code{Index Files}
45680 -- shell command @kbd{info -f gdb -n "Index Files"}.
45687 @c man begin SEEALSO gdb-add-index
45689 The full documentation for @value{GDBN} is maintained as a Texinfo manual.
45690 If the @code{info} and @code{gdb} programs and @value{GDBN}'s Texinfo
45691 documentation are properly installed at your site, the command
45697 should give you access to the complete manual.
45699 @cite{Using GDB: A Guide to the GNU Source-Level Debugger},
45700 Richard M. Stallman and Roland H. Pesch, July 1991.
45706 @node GNU Free Documentation License
45707 @appendix GNU Free Documentation License
45710 @node Concept Index
45711 @unnumbered Concept Index
45715 @node Command and Variable Index
45716 @unnumbered Command, Variable, and Function Index
45721 % I think something like @@colophon should be in texinfo. In the
45723 \long\def\colophon{\hbox to0pt{}\vfill
45724 \centerline{The body of this manual is set in}
45725 \centerline{\fontname\tenrm,}
45726 \centerline{with headings in {\bf\fontname\tenbf}}
45727 \centerline{and examples in {\tt\fontname\tentt}.}
45728 \centerline{{\it\fontname\tenit\/},}
45729 \centerline{{\bf\fontname\tenbf}, and}
45730 \centerline{{\sl\fontname\tensl\/}}
45731 \centerline{are used for emphasis.}\vfill}
45733 % Blame: doc@@cygnus.com, 1991.