1 \input texinfo @c -*-texinfo-*-
2 @c Copyright (C) 1988-2017 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{} 1988-2017 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) 1988-2017 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.
550 @chapter A Sample @value{GDBN} Session
552 You can use this manual at your leisure to read all about @value{GDBN}.
553 However, a handful of commands are enough to get started using the
554 debugger. This chapter illustrates those commands.
557 In this sample session, we emphasize user input like this: @b{input},
558 to make it easier to pick out from the surrounding output.
561 @c FIXME: this example may not be appropriate for some configs, where
562 @c FIXME...primary interest is in remote use.
564 One of the preliminary versions of @sc{gnu} @code{m4} (a generic macro
565 processor) exhibits the following bug: sometimes, when we change its
566 quote strings from the default, the commands used to capture one macro
567 definition within another stop working. In the following short @code{m4}
568 session, we define a macro @code{foo} which expands to @code{0000}; we
569 then use the @code{m4} built-in @code{defn} to define @code{bar} as the
570 same thing. However, when we change the open quote string to
571 @code{<QUOTE>} and the close quote string to @code{<UNQUOTE>}, the same
572 procedure fails to define a new synonym @code{baz}:
581 @b{define(bar,defn(`foo'))}
585 @b{changequote(<QUOTE>,<UNQUOTE>)}
587 @b{define(baz,defn(<QUOTE>foo<UNQUOTE>))}
590 m4: End of input: 0: fatal error: EOF in string
594 Let us use @value{GDBN} to try to see what is going on.
597 $ @b{@value{GDBP} m4}
598 @c FIXME: this falsifies the exact text played out, to permit smallbook
599 @c FIXME... format to come out better.
600 @value{GDBN} is free software and you are welcome to distribute copies
601 of it under certain conditions; type "show copying" to see
603 There is absolutely no warranty for @value{GDBN}; type "show warranty"
606 @value{GDBN} @value{GDBVN}, Copyright 1999 Free Software Foundation, Inc...
611 @value{GDBN} reads only enough symbol data to know where to find the
612 rest when needed; as a result, the first prompt comes up very quickly.
613 We now tell @value{GDBN} to use a narrower display width than usual, so
614 that examples fit in this manual.
617 (@value{GDBP}) @b{set width 70}
621 We need to see how the @code{m4} built-in @code{changequote} works.
622 Having looked at the source, we know the relevant subroutine is
623 @code{m4_changequote}, so we set a breakpoint there with the @value{GDBN}
624 @code{break} command.
627 (@value{GDBP}) @b{break m4_changequote}
628 Breakpoint 1 at 0x62f4: file builtin.c, line 879.
632 Using the @code{run} command, we start @code{m4} running under @value{GDBN}
633 control; as long as control does not reach the @code{m4_changequote}
634 subroutine, the program runs as usual:
637 (@value{GDBP}) @b{run}
638 Starting program: /work/Editorial/gdb/gnu/m4/m4
646 To trigger the breakpoint, we call @code{changequote}. @value{GDBN}
647 suspends execution of @code{m4}, displaying information about the
648 context where it stops.
651 @b{changequote(<QUOTE>,<UNQUOTE>)}
653 Breakpoint 1, m4_changequote (argc=3, argv=0x33c70)
655 879 if (bad_argc(TOKEN_DATA_TEXT(argv[0]),argc,1,3))
659 Now we use the command @code{n} (@code{next}) to advance execution to
660 the next line of the current function.
664 882 set_quotes((argc >= 2) ? TOKEN_DATA_TEXT(argv[1])\
669 @code{set_quotes} looks like a promising subroutine. We can go into it
670 by using the command @code{s} (@code{step}) instead of @code{next}.
671 @code{step} goes to the next line to be executed in @emph{any}
672 subroutine, so it steps into @code{set_quotes}.
676 set_quotes (lq=0x34c78 "<QUOTE>", rq=0x34c88 "<UNQUOTE>")
678 530 if (lquote != def_lquote)
682 The display that shows the subroutine where @code{m4} is now
683 suspended (and its arguments) is called a stack frame display. It
684 shows a summary of the stack. We can use the @code{backtrace}
685 command (which can also be spelled @code{bt}), to see where we are
686 in the stack as a whole: the @code{backtrace} command displays a
687 stack frame for each active subroutine.
690 (@value{GDBP}) @b{bt}
691 #0 set_quotes (lq=0x34c78 "<QUOTE>", rq=0x34c88 "<UNQUOTE>")
693 #1 0x6344 in m4_changequote (argc=3, argv=0x33c70)
695 #2 0x8174 in expand_macro (sym=0x33320) at macro.c:242
696 #3 0x7a88 in expand_token (obs=0x0, t=209696, td=0xf7fffa30)
698 #4 0x79dc in expand_input () at macro.c:40
699 #5 0x2930 in main (argc=0, argv=0xf7fffb20) at m4.c:195
703 We step through a few more lines to see what happens. The first two
704 times, we can use @samp{s}; the next two times we use @code{n} to avoid
705 falling into the @code{xstrdup} subroutine.
709 0x3b5c 532 if (rquote != def_rquote)
711 0x3b80 535 lquote = (lq == nil || *lq == '\0') ? \
712 def_lquote : xstrdup(lq);
714 536 rquote = (rq == nil || *rq == '\0') ? def_rquote\
717 538 len_lquote = strlen(rquote);
721 The last line displayed looks a little odd; we can examine the variables
722 @code{lquote} and @code{rquote} to see if they are in fact the new left
723 and right quotes we specified. We use the command @code{p}
724 (@code{print}) to see their values.
727 (@value{GDBP}) @b{p lquote}
728 $1 = 0x35d40 "<QUOTE>"
729 (@value{GDBP}) @b{p rquote}
730 $2 = 0x35d50 "<UNQUOTE>"
734 @code{lquote} and @code{rquote} are indeed the new left and right quotes.
735 To look at some context, we can display ten lines of source
736 surrounding the current line with the @code{l} (@code{list}) command.
742 535 lquote = (lq == nil || *lq == '\0') ? def_lquote\
744 536 rquote = (rq == nil || *rq == '\0') ? def_rquote\
747 538 len_lquote = strlen(rquote);
748 539 len_rquote = strlen(lquote);
755 Let us step past the two lines that set @code{len_lquote} and
756 @code{len_rquote}, and then examine the values of those variables.
760 539 len_rquote = strlen(lquote);
763 (@value{GDBP}) @b{p len_lquote}
765 (@value{GDBP}) @b{p len_rquote}
770 That certainly looks wrong, assuming @code{len_lquote} and
771 @code{len_rquote} are meant to be the lengths of @code{lquote} and
772 @code{rquote} respectively. We can set them to better values using
773 the @code{p} command, since it can print the value of
774 any expression---and that expression can include subroutine calls and
778 (@value{GDBP}) @b{p len_lquote=strlen(lquote)}
780 (@value{GDBP}) @b{p len_rquote=strlen(rquote)}
785 Is that enough to fix the problem of using the new quotes with the
786 @code{m4} built-in @code{defn}? We can allow @code{m4} to continue
787 executing with the @code{c} (@code{continue}) command, and then try the
788 example that caused trouble initially:
794 @b{define(baz,defn(<QUOTE>foo<UNQUOTE>))}
801 Success! The new quotes now work just as well as the default ones. The
802 problem seems to have been just the two typos defining the wrong
803 lengths. We allow @code{m4} exit by giving it an EOF as input:
807 Program exited normally.
811 The message @samp{Program exited normally.} is from @value{GDBN}; it
812 indicates @code{m4} has finished executing. We can end our @value{GDBN}
813 session with the @value{GDBN} @code{quit} command.
816 (@value{GDBP}) @b{quit}
820 @chapter Getting In and Out of @value{GDBN}
822 This chapter discusses how to start @value{GDBN}, and how to get out of it.
826 type @samp{@value{GDBP}} to start @value{GDBN}.
828 type @kbd{quit} or @kbd{Ctrl-d} to exit.
832 * Invoking GDB:: How to start @value{GDBN}
833 * Quitting GDB:: How to quit @value{GDBN}
834 * Shell Commands:: How to use shell commands inside @value{GDBN}
835 * Logging Output:: How to log @value{GDBN}'s output to a file
839 @section Invoking @value{GDBN}
841 Invoke @value{GDBN} by running the program @code{@value{GDBP}}. Once started,
842 @value{GDBN} reads commands from the terminal until you tell it to exit.
844 You can also run @code{@value{GDBP}} with a variety of arguments and options,
845 to specify more of your debugging environment at the outset.
847 The command-line options described here are designed
848 to cover a variety of situations; in some environments, some of these
849 options may effectively be unavailable.
851 The most usual way to start @value{GDBN} is with one argument,
852 specifying an executable program:
855 @value{GDBP} @var{program}
859 You can also start with both an executable program and a core file
863 @value{GDBP} @var{program} @var{core}
866 You can, instead, specify a process ID as a second argument, if you want
867 to debug a running process:
870 @value{GDBP} @var{program} 1234
874 would attach @value{GDBN} to process @code{1234} (unless you also have a file
875 named @file{1234}; @value{GDBN} does check for a core file first).
877 Taking advantage of the second command-line argument requires a fairly
878 complete operating system; when you use @value{GDBN} as a remote
879 debugger attached to a bare board, there may not be any notion of
880 ``process'', and there is often no way to get a core dump. @value{GDBN}
881 will warn you if it is unable to attach or to read core dumps.
883 You can optionally have @code{@value{GDBP}} pass any arguments after the
884 executable file to the inferior using @code{--args}. This option stops
887 @value{GDBP} --args gcc -O2 -c foo.c
889 This will cause @code{@value{GDBP}} to debug @code{gcc}, and to set
890 @code{gcc}'s command-line arguments (@pxref{Arguments}) to @samp{-O2 -c foo.c}.
892 You can run @code{@value{GDBP}} without printing the front material, which describes
893 @value{GDBN}'s non-warranty, by specifying @code{--silent}
894 (or @code{-q}/@code{--quiet}):
897 @value{GDBP} --silent
901 You can further control how @value{GDBN} starts up by using command-line
902 options. @value{GDBN} itself can remind you of the options available.
912 to display all available options and briefly describe their use
913 (@samp{@value{GDBP} -h} is a shorter equivalent).
915 All options and command line arguments you give are processed
916 in sequential order. The order makes a difference when the
917 @samp{-x} option is used.
921 * File Options:: Choosing files
922 * Mode Options:: Choosing modes
923 * Startup:: What @value{GDBN} does during startup
927 @subsection Choosing Files
929 When @value{GDBN} starts, it reads any arguments other than options as
930 specifying an executable file and core file (or process ID). This is
931 the same as if the arguments were specified by the @samp{-se} and
932 @samp{-c} (or @samp{-p}) options respectively. (@value{GDBN} reads the
933 first argument that does not have an associated option flag as
934 equivalent to the @samp{-se} option followed by that argument; and the
935 second argument that does not have an associated option flag, if any, as
936 equivalent to the @samp{-c}/@samp{-p} option followed by that argument.)
937 If the second argument begins with a decimal digit, @value{GDBN} will
938 first attempt to attach to it as a process, and if that fails, attempt
939 to open it as a corefile. If you have a corefile whose name begins with
940 a digit, you can prevent @value{GDBN} from treating it as a pid by
941 prefixing it with @file{./}, e.g.@: @file{./12345}.
943 If @value{GDBN} has not been configured to included core file support,
944 such as for most embedded targets, then it will complain about a second
945 argument and ignore it.
947 Many options have both long and short forms; both are shown in the
948 following list. @value{GDBN} also recognizes the long forms if you truncate
949 them, so long as enough of the option is present to be unambiguous.
950 (If you prefer, you can flag option arguments with @samp{--} rather
951 than @samp{-}, though we illustrate the more usual convention.)
953 @c NOTE: the @cindex entries here use double dashes ON PURPOSE. This
954 @c way, both those who look for -foo and --foo in the index, will find
958 @item -symbols @var{file}
960 @cindex @code{--symbols}
962 Read symbol table from file @var{file}.
964 @item -exec @var{file}
966 @cindex @code{--exec}
968 Use file @var{file} as the executable file to execute when appropriate,
969 and for examining pure data in conjunction with a core dump.
973 Read symbol table from file @var{file} and use it as the executable
976 @item -core @var{file}
978 @cindex @code{--core}
980 Use file @var{file} as a core dump to examine.
982 @item -pid @var{number}
983 @itemx -p @var{number}
986 Connect to process ID @var{number}, as with the @code{attach} command.
988 @item -command @var{file}
990 @cindex @code{--command}
992 Execute commands from file @var{file}. The contents of this file is
993 evaluated exactly as the @code{source} command would.
994 @xref{Command Files,, Command files}.
996 @item -eval-command @var{command}
997 @itemx -ex @var{command}
998 @cindex @code{--eval-command}
1000 Execute a single @value{GDBN} command.
1002 This option may be used multiple times to call multiple commands. It may
1003 also be interleaved with @samp{-command} as required.
1006 @value{GDBP} -ex 'target sim' -ex 'load' \
1007 -x setbreakpoints -ex 'run' a.out
1010 @item -init-command @var{file}
1011 @itemx -ix @var{file}
1012 @cindex @code{--init-command}
1014 Execute commands from file @var{file} before loading the inferior (but
1015 after loading gdbinit files).
1018 @item -init-eval-command @var{command}
1019 @itemx -iex @var{command}
1020 @cindex @code{--init-eval-command}
1022 Execute a single @value{GDBN} command before loading the inferior (but
1023 after loading gdbinit files).
1026 @item -directory @var{directory}
1027 @itemx -d @var{directory}
1028 @cindex @code{--directory}
1030 Add @var{directory} to the path to search for source and script files.
1034 @cindex @code{--readnow}
1036 Read each symbol file's entire symbol table immediately, rather than
1037 the default, which is to read it incrementally as it is needed.
1038 This makes startup slower, but makes future operations faster.
1043 @subsection Choosing Modes
1045 You can run @value{GDBN} in various alternative modes---for example, in
1046 batch mode or quiet mode.
1054 Do not execute commands found in any initialization file.
1055 There are three init files, loaded in the following order:
1058 @item @file{system.gdbinit}
1059 This is the system-wide init file.
1060 Its location is specified with the @code{--with-system-gdbinit}
1061 configure option (@pxref{System-wide configuration}).
1062 It is loaded first when @value{GDBN} starts, before command line options
1063 have been processed.
1064 @item @file{~/.gdbinit}
1065 This is the init file in your home directory.
1066 It is loaded next, after @file{system.gdbinit}, and before
1067 command options have been processed.
1068 @item @file{./.gdbinit}
1069 This is the init file in the current directory.
1070 It is loaded last, after command line options other than @code{-x} and
1071 @code{-ex} have been processed. Command line options @code{-x} and
1072 @code{-ex} are processed last, after @file{./.gdbinit} has been loaded.
1075 For further documentation on startup processing, @xref{Startup}.
1076 For documentation on how to write command files,
1077 @xref{Command Files,,Command Files}.
1082 Do not execute commands found in @file{~/.gdbinit}, the init file
1083 in your home directory.
1089 @cindex @code{--quiet}
1090 @cindex @code{--silent}
1092 ``Quiet''. Do not print the introductory and copyright messages. These
1093 messages are also suppressed in batch mode.
1096 @cindex @code{--batch}
1097 Run in batch mode. Exit with status @code{0} after processing all the
1098 command files specified with @samp{-x} (and all commands from
1099 initialization files, if not inhibited with @samp{-n}). Exit with
1100 nonzero status if an error occurs in executing the @value{GDBN} commands
1101 in the command files. Batch mode also disables pagination, sets unlimited
1102 terminal width and height @pxref{Screen Size}, and acts as if @kbd{set confirm
1103 off} were in effect (@pxref{Messages/Warnings}).
1105 Batch mode may be useful for running @value{GDBN} as a filter, for
1106 example to download and run a program on another computer; in order to
1107 make this more useful, the message
1110 Program exited normally.
1114 (which is ordinarily issued whenever a program running under
1115 @value{GDBN} control terminates) is not issued when running in batch
1119 @cindex @code{--batch-silent}
1120 Run in batch mode exactly like @samp{-batch}, but totally silently. All
1121 @value{GDBN} output to @code{stdout} is prevented (@code{stderr} is
1122 unaffected). This is much quieter than @samp{-silent} and would be useless
1123 for an interactive session.
1125 This is particularly useful when using targets that give @samp{Loading section}
1126 messages, for example.
1128 Note that targets that give their output via @value{GDBN}, as opposed to
1129 writing directly to @code{stdout}, will also be made silent.
1131 @item -return-child-result
1132 @cindex @code{--return-child-result}
1133 The return code from @value{GDBN} will be the return code from the child
1134 process (the process being debugged), with the following exceptions:
1138 @value{GDBN} exits abnormally. E.g., due to an incorrect argument or an
1139 internal error. In this case the exit code is the same as it would have been
1140 without @samp{-return-child-result}.
1142 The user quits with an explicit value. E.g., @samp{quit 1}.
1144 The child process never runs, or is not allowed to terminate, in which case
1145 the exit code will be -1.
1148 This option is useful in conjunction with @samp{-batch} or @samp{-batch-silent},
1149 when @value{GDBN} is being used as a remote program loader or simulator
1154 @cindex @code{--nowindows}
1156 ``No windows''. If @value{GDBN} comes with a graphical user interface
1157 (GUI) built in, then this option tells @value{GDBN} to only use the command-line
1158 interface. If no GUI is available, this option has no effect.
1162 @cindex @code{--windows}
1164 If @value{GDBN} includes a GUI, then this option requires it to be
1167 @item -cd @var{directory}
1169 Run @value{GDBN} using @var{directory} as its working directory,
1170 instead of the current directory.
1172 @item -data-directory @var{directory}
1173 @itemx -D @var{directory}
1174 @cindex @code{--data-directory}
1176 Run @value{GDBN} using @var{directory} as its data directory.
1177 The data directory is where @value{GDBN} searches for its
1178 auxiliary files. @xref{Data Files}.
1182 @cindex @code{--fullname}
1184 @sc{gnu} Emacs sets this option when it runs @value{GDBN} as a
1185 subprocess. It tells @value{GDBN} to output the full file name and line
1186 number in a standard, recognizable fashion each time a stack frame is
1187 displayed (which includes each time your program stops). This
1188 recognizable format looks like two @samp{\032} characters, followed by
1189 the file name, line number and character position separated by colons,
1190 and a newline. The Emacs-to-@value{GDBN} interface program uses the two
1191 @samp{\032} characters as a signal to display the source code for the
1194 @item -annotate @var{level}
1195 @cindex @code{--annotate}
1196 This option sets the @dfn{annotation level} inside @value{GDBN}. Its
1197 effect is identical to using @samp{set annotate @var{level}}
1198 (@pxref{Annotations}). The annotation @var{level} controls how much
1199 information @value{GDBN} prints together with its prompt, values of
1200 expressions, source lines, and other types of output. Level 0 is the
1201 normal, level 1 is for use when @value{GDBN} is run as a subprocess of
1202 @sc{gnu} Emacs, level 3 is the maximum annotation suitable for programs
1203 that control @value{GDBN}, and level 2 has been deprecated.
1205 The annotation mechanism has largely been superseded by @sc{gdb/mi}
1209 @cindex @code{--args}
1210 Change interpretation of command line so that arguments following the
1211 executable file are passed as command line arguments to the inferior.
1212 This option stops option processing.
1214 @item -baud @var{bps}
1216 @cindex @code{--baud}
1218 Set the line speed (baud rate or bits per second) of any serial
1219 interface used by @value{GDBN} for remote debugging.
1221 @item -l @var{timeout}
1223 Set the timeout (in seconds) of any communication used by @value{GDBN}
1224 for remote debugging.
1226 @item -tty @var{device}
1227 @itemx -t @var{device}
1228 @cindex @code{--tty}
1230 Run using @var{device} for your program's standard input and output.
1231 @c FIXME: kingdon thinks there is more to -tty. Investigate.
1233 @c resolve the situation of these eventually
1235 @cindex @code{--tui}
1236 Activate the @dfn{Text User Interface} when starting. The Text User
1237 Interface manages several text windows on the terminal, showing
1238 source, assembly, registers and @value{GDBN} command outputs
1239 (@pxref{TUI, ,@value{GDBN} Text User Interface}). Do not use this
1240 option if you run @value{GDBN} from Emacs (@pxref{Emacs, ,
1241 Using @value{GDBN} under @sc{gnu} Emacs}).
1243 @item -interpreter @var{interp}
1244 @cindex @code{--interpreter}
1245 Use the interpreter @var{interp} for interface with the controlling
1246 program or device. This option is meant to be set by programs which
1247 communicate with @value{GDBN} using it as a back end.
1248 @xref{Interpreters, , Command Interpreters}.
1250 @samp{--interpreter=mi} (or @samp{--interpreter=mi2}) causes
1251 @value{GDBN} to use the @dfn{@sc{gdb/mi} interface} (@pxref{GDB/MI, ,
1252 The @sc{gdb/mi} Interface}) included since @value{GDBN} version 6.0. The
1253 previous @sc{gdb/mi} interface, included in @value{GDBN} version 5.3 and
1254 selected with @samp{--interpreter=mi1}, is deprecated. Earlier
1255 @sc{gdb/mi} interfaces are no longer supported.
1258 @cindex @code{--write}
1259 Open the executable and core files for both reading and writing. This
1260 is equivalent to the @samp{set write on} command inside @value{GDBN}
1264 @cindex @code{--statistics}
1265 This option causes @value{GDBN} to print statistics about time and
1266 memory usage after it completes each command and returns to the prompt.
1269 @cindex @code{--version}
1270 This option causes @value{GDBN} to print its version number and
1271 no-warranty blurb, and exit.
1273 @item -configuration
1274 @cindex @code{--configuration}
1275 This option causes @value{GDBN} to print details about its build-time
1276 configuration parameters, and then exit. These details can be
1277 important when reporting @value{GDBN} bugs (@pxref{GDB Bugs}).
1282 @subsection What @value{GDBN} Does During Startup
1283 @cindex @value{GDBN} startup
1285 Here's the description of what @value{GDBN} does during session startup:
1289 Sets up the command interpreter as specified by the command line
1290 (@pxref{Mode Options, interpreter}).
1294 Reads the system-wide @dfn{init file} (if @option{--with-system-gdbinit} was
1295 used when building @value{GDBN}; @pxref{System-wide configuration,
1296 ,System-wide configuration and settings}) and executes all the commands in
1299 @anchor{Home Directory Init File}
1301 Reads the init file (if any) in your home directory@footnote{On
1302 DOS/Windows systems, the home directory is the one pointed to by the
1303 @code{HOME} environment variable.} and executes all the commands in
1306 @anchor{Option -init-eval-command}
1308 Executes commands and command files specified by the @samp{-iex} and
1309 @samp{-ix} options in their specified order. Usually you should use the
1310 @samp{-ex} and @samp{-x} options instead, but this way you can apply
1311 settings before @value{GDBN} init files get executed and before inferior
1315 Processes command line options and operands.
1317 @anchor{Init File in the Current Directory during Startup}
1319 Reads and executes the commands from init file (if any) in the current
1320 working directory as long as @samp{set auto-load local-gdbinit} is set to
1321 @samp{on} (@pxref{Init File in the Current Directory}).
1322 This is only done if the current directory is
1323 different from your home directory. Thus, you can have more than one
1324 init file, one generic in your home directory, and another, specific
1325 to the program you are debugging, in the directory where you invoke
1329 If the command line specified a program to debug, or a process to
1330 attach to, or a core file, @value{GDBN} loads any auto-loaded
1331 scripts provided for the program or for its loaded shared libraries.
1332 @xref{Auto-loading}.
1334 If you wish to disable the auto-loading during startup,
1335 you must do something like the following:
1338 $ gdb -iex "set auto-load python-scripts off" myprogram
1341 Option @samp{-ex} does not work because the auto-loading is then turned
1345 Executes commands and command files specified by the @samp{-ex} and
1346 @samp{-x} options in their specified order. @xref{Command Files}, for
1347 more details about @value{GDBN} command files.
1350 Reads the command history recorded in the @dfn{history file}.
1351 @xref{Command History}, for more details about the command history and the
1352 files where @value{GDBN} records it.
1355 Init files use the same syntax as @dfn{command files} (@pxref{Command
1356 Files}) and are processed by @value{GDBN} in the same way. The init
1357 file in your home directory can set options (such as @samp{set
1358 complaints}) that affect subsequent processing of command line options
1359 and operands. Init files are not executed if you use the @samp{-nx}
1360 option (@pxref{Mode Options, ,Choosing Modes}).
1362 To display the list of init files loaded by gdb at startup, you
1363 can use @kbd{gdb --help}.
1365 @cindex init file name
1366 @cindex @file{.gdbinit}
1367 @cindex @file{gdb.ini}
1368 The @value{GDBN} init files are normally called @file{.gdbinit}.
1369 The DJGPP port of @value{GDBN} uses the name @file{gdb.ini}, due to
1370 the limitations of file names imposed by DOS filesystems. The Windows
1371 port of @value{GDBN} uses the standard name, but if it finds a
1372 @file{gdb.ini} file in your home directory, it warns you about that
1373 and suggests to rename the file to the standard name.
1377 @section Quitting @value{GDBN}
1378 @cindex exiting @value{GDBN}
1379 @cindex leaving @value{GDBN}
1382 @kindex quit @r{[}@var{expression}@r{]}
1383 @kindex q @r{(@code{quit})}
1384 @item quit @r{[}@var{expression}@r{]}
1386 To exit @value{GDBN}, use the @code{quit} command (abbreviated
1387 @code{q}), or type an end-of-file character (usually @kbd{Ctrl-d}). If you
1388 do not supply @var{expression}, @value{GDBN} will terminate normally;
1389 otherwise it will terminate using the result of @var{expression} as the
1394 An interrupt (often @kbd{Ctrl-c}) does not exit from @value{GDBN}, but rather
1395 terminates the action of any @value{GDBN} command that is in progress and
1396 returns to @value{GDBN} command level. It is safe to type the interrupt
1397 character at any time because @value{GDBN} does not allow it to take effect
1398 until a time when it is safe.
1400 If you have been using @value{GDBN} to control an attached process or
1401 device, you can release it with the @code{detach} command
1402 (@pxref{Attach, ,Debugging an Already-running Process}).
1404 @node Shell Commands
1405 @section Shell Commands
1407 If you need to execute occasional shell commands during your
1408 debugging session, there is no need to leave or suspend @value{GDBN}; you can
1409 just use the @code{shell} command.
1414 @cindex shell escape
1415 @item shell @var{command-string}
1416 @itemx !@var{command-string}
1417 Invoke a standard shell to execute @var{command-string}.
1418 Note that no space is needed between @code{!} and @var{command-string}.
1419 If it exists, the environment variable @code{SHELL} determines which
1420 shell to run. Otherwise @value{GDBN} uses the default shell
1421 (@file{/bin/sh} on Unix systems, @file{COMMAND.COM} on MS-DOS, etc.).
1424 The utility @code{make} is often needed in development environments.
1425 You do not have to use the @code{shell} command for this purpose in
1430 @cindex calling make
1431 @item make @var{make-args}
1432 Execute the @code{make} program with the specified
1433 arguments. This is equivalent to @samp{shell make @var{make-args}}.
1436 @node Logging Output
1437 @section Logging Output
1438 @cindex logging @value{GDBN} output
1439 @cindex save @value{GDBN} output to a file
1441 You may want to save the output of @value{GDBN} commands to a file.
1442 There are several commands to control @value{GDBN}'s logging.
1446 @item set logging on
1448 @item set logging off
1450 @cindex logging file name
1451 @item set logging file @var{file}
1452 Change the name of the current logfile. The default logfile is @file{gdb.txt}.
1453 @item set logging overwrite [on|off]
1454 By default, @value{GDBN} will append to the logfile. Set @code{overwrite} if
1455 you want @code{set logging on} to overwrite the logfile instead.
1456 @item set logging redirect [on|off]
1457 By default, @value{GDBN} output will go to both the terminal and the logfile.
1458 Set @code{redirect} if you want output to go only to the log file.
1459 @kindex show logging
1461 Show the current values of the logging settings.
1465 @chapter @value{GDBN} Commands
1467 You can abbreviate a @value{GDBN} command to the first few letters of the command
1468 name, if that abbreviation is unambiguous; and you can repeat certain
1469 @value{GDBN} commands by typing just @key{RET}. You can also use the @key{TAB}
1470 key to get @value{GDBN} to fill out the rest of a word in a command (or to
1471 show you the alternatives available, if there is more than one possibility).
1474 * Command Syntax:: How to give commands to @value{GDBN}
1475 * Completion:: Command completion
1476 * Help:: How to ask @value{GDBN} for help
1479 @node Command Syntax
1480 @section Command Syntax
1482 A @value{GDBN} command is a single line of input. There is no limit on
1483 how long it can be. It starts with a command name, which is followed by
1484 arguments whose meaning depends on the command name. For example, the
1485 command @code{step} accepts an argument which is the number of times to
1486 step, as in @samp{step 5}. You can also use the @code{step} command
1487 with no arguments. Some commands do not allow any arguments.
1489 @cindex abbreviation
1490 @value{GDBN} command names may always be truncated if that abbreviation is
1491 unambiguous. Other possible command abbreviations are listed in the
1492 documentation for individual commands. In some cases, even ambiguous
1493 abbreviations are allowed; for example, @code{s} is specially defined as
1494 equivalent to @code{step} even though there are other commands whose
1495 names start with @code{s}. You can test abbreviations by using them as
1496 arguments to the @code{help} command.
1498 @cindex repeating commands
1499 @kindex RET @r{(repeat last command)}
1500 A blank line as input to @value{GDBN} (typing just @key{RET}) means to
1501 repeat the previous command. Certain commands (for example, @code{run})
1502 will not repeat this way; these are commands whose unintentional
1503 repetition might cause trouble and which you are unlikely to want to
1504 repeat. User-defined commands can disable this feature; see
1505 @ref{Define, dont-repeat}.
1507 The @code{list} and @code{x} commands, when you repeat them with
1508 @key{RET}, construct new arguments rather than repeating
1509 exactly as typed. This permits easy scanning of source or memory.
1511 @value{GDBN} can also use @key{RET} in another way: to partition lengthy
1512 output, in a way similar to the common utility @code{more}
1513 (@pxref{Screen Size,,Screen Size}). Since it is easy to press one
1514 @key{RET} too many in this situation, @value{GDBN} disables command
1515 repetition after any command that generates this sort of display.
1517 @kindex # @r{(a comment)}
1519 Any text from a @kbd{#} to the end of the line is a comment; it does
1520 nothing. This is useful mainly in command files (@pxref{Command
1521 Files,,Command Files}).
1523 @cindex repeating command sequences
1524 @kindex Ctrl-o @r{(operate-and-get-next)}
1525 The @kbd{Ctrl-o} binding is useful for repeating a complex sequence of
1526 commands. This command accepts the current line, like @key{RET}, and
1527 then fetches the next line relative to the current line from the history
1531 @section Command Completion
1534 @cindex word completion
1535 @value{GDBN} can fill in the rest of a word in a command for you, if there is
1536 only one possibility; it can also show you what the valid possibilities
1537 are for the next word in a command, at any time. This works for @value{GDBN}
1538 commands, @value{GDBN} subcommands, and the names of symbols in your program.
1540 Press the @key{TAB} key whenever you want @value{GDBN} to fill out the rest
1541 of a word. If there is only one possibility, @value{GDBN} fills in the
1542 word, and waits for you to finish the command (or press @key{RET} to
1543 enter it). For example, if you type
1545 @c FIXME "@key" does not distinguish its argument sufficiently to permit
1546 @c complete accuracy in these examples; space introduced for clarity.
1547 @c If texinfo enhancements make it unnecessary, it would be nice to
1548 @c replace " @key" by "@key" in the following...
1550 (@value{GDBP}) info bre @key{TAB}
1554 @value{GDBN} fills in the rest of the word @samp{breakpoints}, since that is
1555 the only @code{info} subcommand beginning with @samp{bre}:
1558 (@value{GDBP}) info breakpoints
1562 You can either press @key{RET} at this point, to run the @code{info
1563 breakpoints} command, or backspace and enter something else, if
1564 @samp{breakpoints} does not look like the command you expected. (If you
1565 were sure you wanted @code{info breakpoints} in the first place, you
1566 might as well just type @key{RET} immediately after @samp{info bre},
1567 to exploit command abbreviations rather than command completion).
1569 If there is more than one possibility for the next word when you press
1570 @key{TAB}, @value{GDBN} sounds a bell. You can either supply more
1571 characters and try again, or just press @key{TAB} a second time;
1572 @value{GDBN} displays all the possible completions for that word. For
1573 example, you might want to set a breakpoint on a subroutine whose name
1574 begins with @samp{make_}, but when you type @kbd{b make_@key{TAB}} @value{GDBN}
1575 just sounds the bell. Typing @key{TAB} again displays all the
1576 function names in your program that begin with those characters, for
1580 (@value{GDBP}) b make_ @key{TAB}
1581 @exdent @value{GDBN} sounds bell; press @key{TAB} again, to see:
1582 make_a_section_from_file make_environ
1583 make_abs_section make_function_type
1584 make_blockvector make_pointer_type
1585 make_cleanup make_reference_type
1586 make_command make_symbol_completion_list
1587 (@value{GDBP}) b make_
1591 After displaying the available possibilities, @value{GDBN} copies your
1592 partial input (@samp{b make_} in the example) so you can finish the
1595 If you just want to see the list of alternatives in the first place, you
1596 can press @kbd{M-?} rather than pressing @key{TAB} twice. @kbd{M-?}
1597 means @kbd{@key{META} ?}. You can type this either by holding down a
1598 key designated as the @key{META} shift on your keyboard (if there is
1599 one) while typing @kbd{?}, or as @key{ESC} followed by @kbd{?}.
1601 If the number of possible completions is large, @value{GDBN} will
1602 print as much of the list as it has collected, as well as a message
1603 indicating that the list may be truncated.
1606 (@value{GDBP}) b m@key{TAB}@key{TAB}
1608 <... the rest of the possible completions ...>
1609 *** List may be truncated, max-completions reached. ***
1614 This behavior can be controlled with the following commands:
1617 @kindex set max-completions
1618 @item set max-completions @var{limit}
1619 @itemx set max-completions unlimited
1620 Set the maximum number of completion candidates. @value{GDBN} will
1621 stop looking for more completions once it collects this many candidates.
1622 This is useful when completing on things like function names as collecting
1623 all the possible candidates can be time consuming.
1624 The default value is 200. A value of zero disables tab-completion.
1625 Note that setting either no limit or a very large limit can make
1627 @kindex show max-completions
1628 @item show max-completions
1629 Show the maximum number of candidates that @value{GDBN} will collect and show
1633 @cindex quotes in commands
1634 @cindex completion of quoted strings
1635 Sometimes the string you need, while logically a ``word'', may contain
1636 parentheses or other characters that @value{GDBN} normally excludes from
1637 its notion of a word. To permit word completion to work in this
1638 situation, you may enclose words in @code{'} (single quote marks) in
1639 @value{GDBN} commands.
1641 The most likely situation where you might need this is in typing the
1642 name of a C@t{++} function. This is because C@t{++} allows function
1643 overloading (multiple definitions of the same function, distinguished
1644 by argument type). For example, when you want to set a breakpoint you
1645 may need to distinguish whether you mean the version of @code{name}
1646 that takes an @code{int} parameter, @code{name(int)}, or the version
1647 that takes a @code{float} parameter, @code{name(float)}. To use the
1648 word-completion facilities in this situation, type a single quote
1649 @code{'} at the beginning of the function name. This alerts
1650 @value{GDBN} that it may need to consider more information than usual
1651 when you press @key{TAB} or @kbd{M-?} to request word completion:
1654 (@value{GDBP}) b 'bubble( @kbd{M-?}
1655 bubble(double,double) bubble(int,int)
1656 (@value{GDBP}) b 'bubble(
1659 In some cases, @value{GDBN} can tell that completing a name requires using
1660 quotes. When this happens, @value{GDBN} inserts the quote for you (while
1661 completing as much as it can) if you do not type the quote in the first
1665 (@value{GDBP}) b bub @key{TAB}
1666 @exdent @value{GDBN} alters your input line to the following, and rings a bell:
1667 (@value{GDBP}) b 'bubble(
1671 In general, @value{GDBN} can tell that a quote is needed (and inserts it) if
1672 you have not yet started typing the argument list when you ask for
1673 completion on an overloaded symbol.
1675 For more information about overloaded functions, see @ref{C Plus Plus
1676 Expressions, ,C@t{++} Expressions}. You can use the command @code{set
1677 overload-resolution off} to disable overload resolution;
1678 see @ref{Debugging C Plus Plus, ,@value{GDBN} Features for C@t{++}}.
1680 @cindex completion of structure field names
1681 @cindex structure field name completion
1682 @cindex completion of union field names
1683 @cindex union field name completion
1684 When completing in an expression which looks up a field in a
1685 structure, @value{GDBN} also tries@footnote{The completer can be
1686 confused by certain kinds of invalid expressions. Also, it only
1687 examines the static type of the expression, not the dynamic type.} to
1688 limit completions to the field names available in the type of the
1692 (@value{GDBP}) p gdb_stdout.@kbd{M-?}
1693 magic to_fputs to_rewind
1694 to_data to_isatty to_write
1695 to_delete to_put to_write_async_safe
1700 This is because the @code{gdb_stdout} is a variable of the type
1701 @code{struct ui_file} that is defined in @value{GDBN} sources as
1708 ui_file_flush_ftype *to_flush;
1709 ui_file_write_ftype *to_write;
1710 ui_file_write_async_safe_ftype *to_write_async_safe;
1711 ui_file_fputs_ftype *to_fputs;
1712 ui_file_read_ftype *to_read;
1713 ui_file_delete_ftype *to_delete;
1714 ui_file_isatty_ftype *to_isatty;
1715 ui_file_rewind_ftype *to_rewind;
1716 ui_file_put_ftype *to_put;
1723 @section Getting Help
1724 @cindex online documentation
1727 You can always ask @value{GDBN} itself for information on its commands,
1728 using the command @code{help}.
1731 @kindex h @r{(@code{help})}
1734 You can use @code{help} (abbreviated @code{h}) with no arguments to
1735 display a short list of named classes of commands:
1739 List of classes of commands:
1741 aliases -- Aliases of other commands
1742 breakpoints -- Making program stop at certain points
1743 data -- Examining data
1744 files -- Specifying and examining files
1745 internals -- Maintenance commands
1746 obscure -- Obscure features
1747 running -- Running the program
1748 stack -- Examining the stack
1749 status -- Status inquiries
1750 support -- Support facilities
1751 tracepoints -- Tracing of program execution without
1752 stopping the program
1753 user-defined -- User-defined commands
1755 Type "help" followed by a class name for a list of
1756 commands in that class.
1757 Type "help" followed by command name for full
1759 Command name abbreviations are allowed if unambiguous.
1762 @c the above line break eliminates huge line overfull...
1764 @item help @var{class}
1765 Using one of the general help classes as an argument, you can get a
1766 list of the individual commands in that class. For example, here is the
1767 help display for the class @code{status}:
1770 (@value{GDBP}) help status
1775 @c Line break in "show" line falsifies real output, but needed
1776 @c to fit in smallbook page size.
1777 info -- Generic command for showing things
1778 about the program being debugged
1779 show -- Generic command for showing things
1782 Type "help" followed by command name for full
1784 Command name abbreviations are allowed if unambiguous.
1788 @item help @var{command}
1789 With a command name as @code{help} argument, @value{GDBN} displays a
1790 short paragraph on how to use that command.
1793 @item apropos @var{args}
1794 The @code{apropos} command searches through all of the @value{GDBN}
1795 commands, and their documentation, for the regular expression specified in
1796 @var{args}. It prints out all matches found. For example:
1807 alias -- Define a new command that is an alias of an existing command
1808 aliases -- Aliases of other commands
1809 d -- Delete some breakpoints or auto-display expressions
1810 del -- Delete some breakpoints or auto-display expressions
1811 delete -- Delete some breakpoints or auto-display expressions
1816 @item complete @var{args}
1817 The @code{complete @var{args}} command lists all the possible completions
1818 for the beginning of a command. Use @var{args} to specify the beginning of the
1819 command you want completed. For example:
1825 @noindent results in:
1836 @noindent This is intended for use by @sc{gnu} Emacs.
1839 In addition to @code{help}, you can use the @value{GDBN} commands @code{info}
1840 and @code{show} to inquire about the state of your program, or the state
1841 of @value{GDBN} itself. Each command supports many topics of inquiry; this
1842 manual introduces each of them in the appropriate context. The listings
1843 under @code{info} and under @code{show} in the Command, Variable, and
1844 Function Index point to all the sub-commands. @xref{Command and Variable
1850 @kindex i @r{(@code{info})}
1852 This command (abbreviated @code{i}) is for describing the state of your
1853 program. For example, you can show the arguments passed to a function
1854 with @code{info args}, list the registers currently in use with @code{info
1855 registers}, or list the breakpoints you have set with @code{info breakpoints}.
1856 You can get a complete list of the @code{info} sub-commands with
1857 @w{@code{help info}}.
1861 You can assign the result of an expression to an environment variable with
1862 @code{set}. For example, you can set the @value{GDBN} prompt to a $-sign with
1863 @code{set prompt $}.
1867 In contrast to @code{info}, @code{show} is for describing the state of
1868 @value{GDBN} itself.
1869 You can change most of the things you can @code{show}, by using the
1870 related command @code{set}; for example, you can control what number
1871 system is used for displays with @code{set radix}, or simply inquire
1872 which is currently in use with @code{show radix}.
1875 To display all the settable parameters and their current
1876 values, you can use @code{show} with no arguments; you may also use
1877 @code{info set}. Both commands produce the same display.
1878 @c FIXME: "info set" violates the rule that "info" is for state of
1879 @c FIXME...program. Ck w/ GNU: "info set" to be called something else,
1880 @c FIXME...or change desc of rule---eg "state of prog and debugging session"?
1884 Here are several miscellaneous @code{show} subcommands, all of which are
1885 exceptional in lacking corresponding @code{set} commands:
1888 @kindex show version
1889 @cindex @value{GDBN} version number
1891 Show what version of @value{GDBN} is running. You should include this
1892 information in @value{GDBN} bug-reports. If multiple versions of
1893 @value{GDBN} are in use at your site, you may need to determine which
1894 version of @value{GDBN} you are running; as @value{GDBN} evolves, new
1895 commands are introduced, and old ones may wither away. Also, many
1896 system vendors ship variant versions of @value{GDBN}, and there are
1897 variant versions of @value{GDBN} in @sc{gnu}/Linux distributions as well.
1898 The version number is the same as the one announced when you start
1901 @kindex show copying
1902 @kindex info copying
1903 @cindex display @value{GDBN} copyright
1906 Display information about permission for copying @value{GDBN}.
1908 @kindex show warranty
1909 @kindex info warranty
1911 @itemx info warranty
1912 Display the @sc{gnu} ``NO WARRANTY'' statement, or a warranty,
1913 if your version of @value{GDBN} comes with one.
1915 @kindex show configuration
1916 @item show configuration
1917 Display detailed information about the way @value{GDBN} was configured
1918 when it was built. This displays the optional arguments passed to the
1919 @file{configure} script and also configuration parameters detected
1920 automatically by @command{configure}. When reporting a @value{GDBN}
1921 bug (@pxref{GDB Bugs}), it is important to include this information in
1927 @chapter Running Programs Under @value{GDBN}
1929 When you run a program under @value{GDBN}, you must first generate
1930 debugging information when you compile it.
1932 You may start @value{GDBN} with its arguments, if any, in an environment
1933 of your choice. If you are doing native debugging, you may redirect
1934 your program's input and output, debug an already running process, or
1935 kill a child process.
1938 * Compilation:: Compiling for debugging
1939 * Starting:: Starting your program
1940 * Arguments:: Your program's arguments
1941 * Environment:: Your program's environment
1943 * Working Directory:: Your program's working directory
1944 * Input/Output:: Your program's input and output
1945 * Attach:: Debugging an already-running process
1946 * Kill Process:: Killing the child process
1948 * Inferiors and Programs:: Debugging multiple inferiors and programs
1949 * Threads:: Debugging programs with multiple threads
1950 * Forks:: Debugging forks
1951 * Checkpoint/Restart:: Setting a @emph{bookmark} to return to later
1955 @section Compiling for Debugging
1957 In order to debug a program effectively, you need to generate
1958 debugging information when you compile it. This debugging information
1959 is stored in the object file; it describes the data type of each
1960 variable or function and the correspondence between source line numbers
1961 and addresses in the executable code.
1963 To request debugging information, specify the @samp{-g} option when you run
1966 Programs that are to be shipped to your customers are compiled with
1967 optimizations, using the @samp{-O} compiler option. However, some
1968 compilers are unable to handle the @samp{-g} and @samp{-O} options
1969 together. Using those compilers, you cannot generate optimized
1970 executables containing debugging information.
1972 @value{NGCC}, the @sc{gnu} C/C@t{++} compiler, supports @samp{-g} with or
1973 without @samp{-O}, making it possible to debug optimized code. We
1974 recommend that you @emph{always} use @samp{-g} whenever you compile a
1975 program. You may think your program is correct, but there is no sense
1976 in pushing your luck. For more information, see @ref{Optimized Code}.
1978 Older versions of the @sc{gnu} C compiler permitted a variant option
1979 @w{@samp{-gg}} for debugging information. @value{GDBN} no longer supports this
1980 format; if your @sc{gnu} C compiler has this option, do not use it.
1982 @value{GDBN} knows about preprocessor macros and can show you their
1983 expansion (@pxref{Macros}). Most compilers do not include information
1984 about preprocessor macros in the debugging information if you specify
1985 the @option{-g} flag alone. Version 3.1 and later of @value{NGCC},
1986 the @sc{gnu} C compiler, provides macro information if you are using
1987 the DWARF debugging format, and specify the option @option{-g3}.
1989 @xref{Debugging Options,,Options for Debugging Your Program or GCC,
1990 gcc.info, Using the @sc{gnu} Compiler Collection (GCC)}, for more
1991 information on @value{NGCC} options affecting debug information.
1993 You will have the best debugging experience if you use the latest
1994 version of the DWARF debugging format that your compiler supports.
1995 DWARF is currently the most expressive and best supported debugging
1996 format in @value{GDBN}.
2000 @section Starting your Program
2006 @kindex r @r{(@code{run})}
2009 Use the @code{run} command to start your program under @value{GDBN}.
2010 You must first specify the program name with an argument to
2011 @value{GDBN} (@pxref{Invocation, ,Getting In and Out of
2012 @value{GDBN}}), or by using the @code{file} or @code{exec-file}
2013 command (@pxref{Files, ,Commands to Specify Files}).
2017 If you are running your program in an execution environment that
2018 supports processes, @code{run} creates an inferior process and makes
2019 that process run your program. In some environments without processes,
2020 @code{run} jumps to the start of your program. Other targets,
2021 like @samp{remote}, are always running. If you get an error
2022 message like this one:
2025 The "remote" target does not support "run".
2026 Try "help target" or "continue".
2030 then use @code{continue} to run your program. You may need @code{load}
2031 first (@pxref{load}).
2033 The execution of a program is affected by certain information it
2034 receives from its superior. @value{GDBN} provides ways to specify this
2035 information, which you must do @emph{before} starting your program. (You
2036 can change it after starting your program, but such changes only affect
2037 your program the next time you start it.) This information may be
2038 divided into four categories:
2041 @item The @emph{arguments.}
2042 Specify the arguments to give your program as the arguments of the
2043 @code{run} command. If a shell is available on your target, the shell
2044 is used to pass the arguments, so that you may use normal conventions
2045 (such as wildcard expansion or variable substitution) in describing
2047 In Unix systems, you can control which shell is used with the
2048 @code{SHELL} environment variable. If you do not define @code{SHELL},
2049 @value{GDBN} uses the default shell (@file{/bin/sh}). You can disable
2050 use of any shell with the @code{set startup-with-shell} command (see
2053 @item The @emph{environment.}
2054 Your program normally inherits its environment from @value{GDBN}, but you can
2055 use the @value{GDBN} commands @code{set environment} and @code{unset
2056 environment} to change parts of the environment that affect
2057 your program. @xref{Environment, ,Your Program's Environment}.
2059 @item The @emph{working directory.}
2060 Your program inherits its working directory from @value{GDBN}. You can set
2061 the @value{GDBN} working directory with the @code{cd} command in @value{GDBN}.
2062 @xref{Working Directory, ,Your Program's Working Directory}.
2064 @item The @emph{standard input and output.}
2065 Your program normally uses the same device for standard input and
2066 standard output as @value{GDBN} is using. You can redirect input and output
2067 in the @code{run} command line, or you can use the @code{tty} command to
2068 set a different device for your program.
2069 @xref{Input/Output, ,Your Program's Input and Output}.
2072 @emph{Warning:} While input and output redirection work, you cannot use
2073 pipes to pass the output of the program you are debugging to another
2074 program; if you attempt this, @value{GDBN} is likely to wind up debugging the
2078 When you issue the @code{run} command, your program begins to execute
2079 immediately. @xref{Stopping, ,Stopping and Continuing}, for discussion
2080 of how to arrange for your program to stop. Once your program has
2081 stopped, you may call functions in your program, using the @code{print}
2082 or @code{call} commands. @xref{Data, ,Examining Data}.
2084 If the modification time of your symbol file has changed since the last
2085 time @value{GDBN} read its symbols, @value{GDBN} discards its symbol
2086 table, and reads it again. When it does this, @value{GDBN} tries to retain
2087 your current breakpoints.
2092 @cindex run to main procedure
2093 The name of the main procedure can vary from language to language.
2094 With C or C@t{++}, the main procedure name is always @code{main}, but
2095 other languages such as Ada do not require a specific name for their
2096 main procedure. The debugger provides a convenient way to start the
2097 execution of the program and to stop at the beginning of the main
2098 procedure, depending on the language used.
2100 The @samp{start} command does the equivalent of setting a temporary
2101 breakpoint at the beginning of the main procedure and then invoking
2102 the @samp{run} command.
2104 @cindex elaboration phase
2105 Some programs contain an @dfn{elaboration} phase where some startup code is
2106 executed before the main procedure is called. This depends on the
2107 languages used to write your program. In C@t{++}, for instance,
2108 constructors for static and global objects are executed before
2109 @code{main} is called. It is therefore possible that the debugger stops
2110 before reaching the main procedure. However, the temporary breakpoint
2111 will remain to halt execution.
2113 Specify the arguments to give to your program as arguments to the
2114 @samp{start} command. These arguments will be given verbatim to the
2115 underlying @samp{run} command. Note that the same arguments will be
2116 reused if no argument is provided during subsequent calls to
2117 @samp{start} or @samp{run}.
2119 It is sometimes necessary to debug the program during elaboration. In
2120 these cases, using the @code{start} command would stop the execution of
2121 your program too late, as the program would have already completed the
2122 elaboration phase. Under these circumstances, insert breakpoints in your
2123 elaboration code before running your program.
2125 @anchor{set exec-wrapper}
2126 @kindex set exec-wrapper
2127 @item set exec-wrapper @var{wrapper}
2128 @itemx show exec-wrapper
2129 @itemx unset exec-wrapper
2130 When @samp{exec-wrapper} is set, the specified wrapper is used to
2131 launch programs for debugging. @value{GDBN} starts your program
2132 with a shell command of the form @kbd{exec @var{wrapper}
2133 @var{program}}. Quoting is added to @var{program} and its
2134 arguments, but not to @var{wrapper}, so you should add quotes if
2135 appropriate for your shell. The wrapper runs until it executes
2136 your program, and then @value{GDBN} takes control.
2138 You can use any program that eventually calls @code{execve} with
2139 its arguments as a wrapper. Several standard Unix utilities do
2140 this, e.g.@: @code{env} and @code{nohup}. Any Unix shell script ending
2141 with @code{exec "$@@"} will also work.
2143 For example, you can use @code{env} to pass an environment variable to
2144 the debugged program, without setting the variable in your shell's
2148 (@value{GDBP}) set exec-wrapper env 'LD_PRELOAD=libtest.so'
2152 This command is available when debugging locally on most targets, excluding
2153 @sc{djgpp}, Cygwin, MS Windows, and QNX Neutrino.
2155 @kindex set startup-with-shell
2156 @anchor{set startup-with-shell}
2157 @item set startup-with-shell
2158 @itemx set startup-with-shell on
2159 @itemx set startup-with-shell off
2160 @itemx show set startup-with-shell
2161 On Unix systems, by default, if a shell is available on your target,
2162 @value{GDBN}) uses it to start your program. Arguments of the
2163 @code{run} command are passed to the shell, which does variable
2164 substitution, expands wildcard characters and performs redirection of
2165 I/O. In some circumstances, it may be useful to disable such use of a
2166 shell, for example, when debugging the shell itself or diagnosing
2167 startup failures such as:
2171 Starting program: ./a.out
2172 During startup program terminated with signal SIGSEGV, Segmentation fault.
2176 which indicates the shell or the wrapper specified with
2177 @samp{exec-wrapper} crashed, not your program. Most often, this is
2178 caused by something odd in your shell's non-interactive mode
2179 initialization file---such as @file{.cshrc} for C-shell,
2180 $@file{.zshenv} for the Z shell, or the file specified in the
2181 @samp{BASH_ENV} environment variable for BASH.
2183 @anchor{set auto-connect-native-target}
2184 @kindex set auto-connect-native-target
2185 @item set auto-connect-native-target
2186 @itemx set auto-connect-native-target on
2187 @itemx set auto-connect-native-target off
2188 @itemx show auto-connect-native-target
2190 By default, if not connected to any target yet (e.g., with
2191 @code{target remote}), the @code{run} command starts your program as a
2192 native process under @value{GDBN}, on your local machine. If you're
2193 sure you don't want to debug programs on your local machine, you can
2194 tell @value{GDBN} to not connect to the native target automatically
2195 with the @code{set auto-connect-native-target off} command.
2197 If @code{on}, which is the default, and if @value{GDBN} is not
2198 connected to a target already, the @code{run} command automaticaly
2199 connects to the native target, if one is available.
2201 If @code{off}, and if @value{GDBN} is not connected to a target
2202 already, the @code{run} command fails with an error:
2206 Don't know how to run. Try "help target".
2209 If @value{GDBN} is already connected to a target, @value{GDBN} always
2210 uses it with the @code{run} command.
2212 In any case, you can explicitly connect to the native target with the
2213 @code{target native} command. For example,
2216 (@value{GDBP}) set auto-connect-native-target off
2218 Don't know how to run. Try "help target".
2219 (@value{GDBP}) target native
2221 Starting program: ./a.out
2222 [Inferior 1 (process 10421) exited normally]
2225 In case you connected explicitly to the @code{native} target,
2226 @value{GDBN} remains connected even if all inferiors exit, ready for
2227 the next @code{run} command. Use the @code{disconnect} command to
2230 Examples of other commands that likewise respect the
2231 @code{auto-connect-native-target} setting: @code{attach}, @code{info
2232 proc}, @code{info os}.
2234 @kindex set disable-randomization
2235 @item set disable-randomization
2236 @itemx set disable-randomization on
2237 This option (enabled by default in @value{GDBN}) will turn off the native
2238 randomization of the virtual address space of the started program. This option
2239 is useful for multiple debugging sessions to make the execution better
2240 reproducible and memory addresses reusable across debugging sessions.
2242 This feature is implemented only on certain targets, including @sc{gnu}/Linux.
2243 On @sc{gnu}/Linux you can get the same behavior using
2246 (@value{GDBP}) set exec-wrapper setarch `uname -m` -R
2249 @item set disable-randomization off
2250 Leave the behavior of the started executable unchanged. Some bugs rear their
2251 ugly heads only when the program is loaded at certain addresses. If your bug
2252 disappears when you run the program under @value{GDBN}, that might be because
2253 @value{GDBN} by default disables the address randomization on platforms, such
2254 as @sc{gnu}/Linux, which do that for stand-alone programs. Use @kbd{set
2255 disable-randomization off} to try to reproduce such elusive bugs.
2257 On targets where it is available, virtual address space randomization
2258 protects the programs against certain kinds of security attacks. In these
2259 cases the attacker needs to know the exact location of a concrete executable
2260 code. Randomizing its location makes it impossible to inject jumps misusing
2261 a code at its expected addresses.
2263 Prelinking shared libraries provides a startup performance advantage but it
2264 makes addresses in these libraries predictable for privileged processes by
2265 having just unprivileged access at the target system. Reading the shared
2266 library binary gives enough information for assembling the malicious code
2267 misusing it. Still even a prelinked shared library can get loaded at a new
2268 random address just requiring the regular relocation process during the
2269 startup. Shared libraries not already prelinked are always loaded at
2270 a randomly chosen address.
2272 Position independent executables (PIE) contain position independent code
2273 similar to the shared libraries and therefore such executables get loaded at
2274 a randomly chosen address upon startup. PIE executables always load even
2275 already prelinked shared libraries at a random address. You can build such
2276 executable using @command{gcc -fPIE -pie}.
2278 Heap (malloc storage), stack and custom mmap areas are always placed randomly
2279 (as long as the randomization is enabled).
2281 @item show disable-randomization
2282 Show the current setting of the explicit disable of the native randomization of
2283 the virtual address space of the started program.
2288 @section Your Program's Arguments
2290 @cindex arguments (to your program)
2291 The arguments to your program can be specified by the arguments of the
2293 They are passed to a shell, which expands wildcard characters and
2294 performs redirection of I/O, and thence to your program. Your
2295 @code{SHELL} environment variable (if it exists) specifies what shell
2296 @value{GDBN} uses. If you do not define @code{SHELL}, @value{GDBN} uses
2297 the default shell (@file{/bin/sh} on Unix).
2299 On non-Unix systems, the program is usually invoked directly by
2300 @value{GDBN}, which emulates I/O redirection via the appropriate system
2301 calls, and the wildcard characters are expanded by the startup code of
2302 the program, not by the shell.
2304 @code{run} with no arguments uses the same arguments used by the previous
2305 @code{run}, or those set by the @code{set args} command.
2310 Specify the arguments to be used the next time your program is run. If
2311 @code{set args} has no arguments, @code{run} executes your program
2312 with no arguments. Once you have run your program with arguments,
2313 using @code{set args} before the next @code{run} is the only way to run
2314 it again without arguments.
2318 Show the arguments to give your program when it is started.
2322 @section Your Program's Environment
2324 @cindex environment (of your program)
2325 The @dfn{environment} consists of a set of environment variables and
2326 their values. Environment variables conventionally record such things as
2327 your user name, your home directory, your terminal type, and your search
2328 path for programs to run. Usually you set up environment variables with
2329 the shell and they are inherited by all the other programs you run. When
2330 debugging, it can be useful to try running your program with a modified
2331 environment without having to start @value{GDBN} over again.
2335 @item path @var{directory}
2336 Add @var{directory} to the front of the @code{PATH} environment variable
2337 (the search path for executables) that will be passed to your program.
2338 The value of @code{PATH} used by @value{GDBN} does not change.
2339 You may specify several directory names, separated by whitespace or by a
2340 system-dependent separator character (@samp{:} on Unix, @samp{;} on
2341 MS-DOS and MS-Windows). If @var{directory} is already in the path, it
2342 is moved to the front, so it is searched sooner.
2344 You can use the string @samp{$cwd} to refer to whatever is the current
2345 working directory at the time @value{GDBN} searches the path. If you
2346 use @samp{.} instead, it refers to the directory where you executed the
2347 @code{path} command. @value{GDBN} replaces @samp{.} in the
2348 @var{directory} argument (with the current path) before adding
2349 @var{directory} to the search path.
2350 @c 'path' is explicitly nonrepeatable, but RMS points out it is silly to
2351 @c document that, since repeating it would be a no-op.
2355 Display the list of search paths for executables (the @code{PATH}
2356 environment variable).
2358 @kindex show environment
2359 @item show environment @r{[}@var{varname}@r{]}
2360 Print the value of environment variable @var{varname} to be given to
2361 your program when it starts. If you do not supply @var{varname},
2362 print the names and values of all environment variables to be given to
2363 your program. You can abbreviate @code{environment} as @code{env}.
2365 @kindex set environment
2366 @item set environment @var{varname} @r{[}=@var{value}@r{]}
2367 Set environment variable @var{varname} to @var{value}. The value
2368 changes for your program (and the shell @value{GDBN} uses to launch
2369 it), not for @value{GDBN} itself. The @var{value} may be any string; the
2370 values of environment variables are just strings, and any
2371 interpretation is supplied by your program itself. The @var{value}
2372 parameter is optional; if it is eliminated, the variable is set to a
2374 @c "any string" here does not include leading, trailing
2375 @c blanks. Gnu asks: does anyone care?
2377 For example, this command:
2384 tells the debugged program, when subsequently run, that its user is named
2385 @samp{foo}. (The spaces around @samp{=} are used for clarity here; they
2386 are not actually required.)
2388 Note that on Unix systems, @value{GDBN} runs your program via a shell,
2389 which also inherits the environment set with @code{set environment}.
2390 If necessary, you can avoid that by using the @samp{env} program as a
2391 wrapper instead of using @code{set environment}. @xref{set
2392 exec-wrapper}, for an example doing just that.
2394 @kindex unset environment
2395 @item unset environment @var{varname}
2396 Remove variable @var{varname} from the environment to be passed to your
2397 program. This is different from @samp{set env @var{varname} =};
2398 @code{unset environment} removes the variable from the environment,
2399 rather than assigning it an empty value.
2402 @emph{Warning:} On Unix systems, @value{GDBN} runs your program using
2403 the shell indicated by your @code{SHELL} environment variable if it
2404 exists (or @code{/bin/sh} if not). If your @code{SHELL} variable
2405 names a shell that runs an initialization file when started
2406 non-interactively---such as @file{.cshrc} for C-shell, $@file{.zshenv}
2407 for the Z shell, or the file specified in the @samp{BASH_ENV}
2408 environment variable for BASH---any variables you set in that file
2409 affect your program. You may wish to move setting of environment
2410 variables to files that are only run when you sign on, such as
2411 @file{.login} or @file{.profile}.
2413 @node Working Directory
2414 @section Your Program's Working Directory
2416 @cindex working directory (of your program)
2417 Each time you start your program with @code{run}, it inherits its
2418 working directory from the current working directory of @value{GDBN}.
2419 The @value{GDBN} working directory is initially whatever it inherited
2420 from its parent process (typically the shell), but you can specify a new
2421 working directory in @value{GDBN} with the @code{cd} command.
2423 The @value{GDBN} working directory also serves as a default for the commands
2424 that specify files for @value{GDBN} to operate on. @xref{Files, ,Commands to
2429 @cindex change working directory
2430 @item cd @r{[}@var{directory}@r{]}
2431 Set the @value{GDBN} working directory to @var{directory}. If not
2432 given, @var{directory} uses @file{'~'}.
2436 Print the @value{GDBN} working directory.
2439 It is generally impossible to find the current working directory of
2440 the process being debugged (since a program can change its directory
2441 during its run). If you work on a system where @value{GDBN} is
2442 configured with the @file{/proc} support, you can use the @code{info
2443 proc} command (@pxref{SVR4 Process Information}) to find out the
2444 current working directory of the debuggee.
2447 @section Your Program's Input and Output
2452 By default, the program you run under @value{GDBN} does input and output to
2453 the same terminal that @value{GDBN} uses. @value{GDBN} switches the terminal
2454 to its own terminal modes to interact with you, but it records the terminal
2455 modes your program was using and switches back to them when you continue
2456 running your program.
2459 @kindex info terminal
2461 Displays information recorded by @value{GDBN} about the terminal modes your
2465 You can redirect your program's input and/or output using shell
2466 redirection with the @code{run} command. For example,
2473 starts your program, diverting its output to the file @file{outfile}.
2476 @cindex controlling terminal
2477 Another way to specify where your program should do input and output is
2478 with the @code{tty} command. This command accepts a file name as
2479 argument, and causes this file to be the default for future @code{run}
2480 commands. It also resets the controlling terminal for the child
2481 process, for future @code{run} commands. For example,
2488 directs that processes started with subsequent @code{run} commands
2489 default to do input and output on the terminal @file{/dev/ttyb} and have
2490 that as their controlling terminal.
2492 An explicit redirection in @code{run} overrides the @code{tty} command's
2493 effect on the input/output device, but not its effect on the controlling
2496 When you use the @code{tty} command or redirect input in the @code{run}
2497 command, only the input @emph{for your program} is affected. The input
2498 for @value{GDBN} still comes from your terminal. @code{tty} is an alias
2499 for @code{set inferior-tty}.
2501 @cindex inferior tty
2502 @cindex set inferior controlling terminal
2503 You can use the @code{show inferior-tty} command to tell @value{GDBN} to
2504 display the name of the terminal that will be used for future runs of your
2508 @item set inferior-tty [ @var{tty} ]
2509 @kindex set inferior-tty
2510 Set the tty for the program being debugged to @var{tty}. Omitting @var{tty}
2511 restores the default behavior, which is to use the same terminal as
2514 @item show inferior-tty
2515 @kindex show inferior-tty
2516 Show the current tty for the program being debugged.
2520 @section Debugging an Already-running Process
2525 @item attach @var{process-id}
2526 This command attaches to a running process---one that was started
2527 outside @value{GDBN}. (@code{info files} shows your active
2528 targets.) The command takes as argument a process ID. The usual way to
2529 find out the @var{process-id} of a Unix process is with the @code{ps} utility,
2530 or with the @samp{jobs -l} shell command.
2532 @code{attach} does not repeat if you press @key{RET} a second time after
2533 executing the command.
2536 To use @code{attach}, your program must be running in an environment
2537 which supports processes; for example, @code{attach} does not work for
2538 programs on bare-board targets that lack an operating system. You must
2539 also have permission to send the process a signal.
2541 When you use @code{attach}, the debugger finds the program running in
2542 the process first by looking in the current working directory, then (if
2543 the program is not found) by using the source file search path
2544 (@pxref{Source Path, ,Specifying Source Directories}). You can also use
2545 the @code{file} command to load the program. @xref{Files, ,Commands to
2548 The first thing @value{GDBN} does after arranging to debug the specified
2549 process is to stop it. You can examine and modify an attached process
2550 with all the @value{GDBN} commands that are ordinarily available when
2551 you start processes with @code{run}. You can insert breakpoints; you
2552 can step and continue; you can modify storage. If you would rather the
2553 process continue running, you may use the @code{continue} command after
2554 attaching @value{GDBN} to the process.
2559 When you have finished debugging the attached process, you can use the
2560 @code{detach} command to release it from @value{GDBN} control. Detaching
2561 the process continues its execution. After the @code{detach} command,
2562 that process and @value{GDBN} become completely independent once more, and you
2563 are ready to @code{attach} another process or start one with @code{run}.
2564 @code{detach} does not repeat if you press @key{RET} again after
2565 executing the command.
2568 If you exit @value{GDBN} while you have an attached process, you detach
2569 that process. If you use the @code{run} command, you kill that process.
2570 By default, @value{GDBN} asks for confirmation if you try to do either of these
2571 things; you can control whether or not you need to confirm by using the
2572 @code{set confirm} command (@pxref{Messages/Warnings, ,Optional Warnings and
2576 @section Killing the Child Process
2581 Kill the child process in which your program is running under @value{GDBN}.
2584 This command is useful if you wish to debug a core dump instead of a
2585 running process. @value{GDBN} ignores any core dump file while your program
2588 On some operating systems, a program cannot be executed outside @value{GDBN}
2589 while you have breakpoints set on it inside @value{GDBN}. You can use the
2590 @code{kill} command in this situation to permit running your program
2591 outside the debugger.
2593 The @code{kill} command is also useful if you wish to recompile and
2594 relink your program, since on many systems it is impossible to modify an
2595 executable file while it is running in a process. In this case, when you
2596 next type @code{run}, @value{GDBN} notices that the file has changed, and
2597 reads the symbol table again (while trying to preserve your current
2598 breakpoint settings).
2600 @node Inferiors and Programs
2601 @section Debugging Multiple Inferiors and Programs
2603 @value{GDBN} lets you run and debug multiple programs in a single
2604 session. In addition, @value{GDBN} on some systems may let you run
2605 several programs simultaneously (otherwise you have to exit from one
2606 before starting another). In the most general case, you can have
2607 multiple threads of execution in each of multiple processes, launched
2608 from multiple executables.
2611 @value{GDBN} represents the state of each program execution with an
2612 object called an @dfn{inferior}. An inferior typically corresponds to
2613 a process, but is more general and applies also to targets that do not
2614 have processes. Inferiors may be created before a process runs, and
2615 may be retained after a process exits. Inferiors have unique
2616 identifiers that are different from process ids. Usually each
2617 inferior will also have its own distinct address space, although some
2618 embedded targets may have several inferiors running in different parts
2619 of a single address space. Each inferior may in turn have multiple
2620 threads running in it.
2622 To find out what inferiors exist at any moment, use @w{@code{info
2626 @kindex info inferiors
2627 @item info inferiors
2628 Print a list of all inferiors currently being managed by @value{GDBN}.
2630 @value{GDBN} displays for each inferior (in this order):
2634 the inferior number assigned by @value{GDBN}
2637 the target system's inferior identifier
2640 the name of the executable the inferior is running.
2645 An asterisk @samp{*} preceding the @value{GDBN} inferior number
2646 indicates the current inferior.
2650 @c end table here to get a little more width for example
2653 (@value{GDBP}) info inferiors
2654 Num Description Executable
2655 2 process 2307 hello
2656 * 1 process 3401 goodbye
2659 To switch focus between inferiors, use the @code{inferior} command:
2662 @kindex inferior @var{infno}
2663 @item inferior @var{infno}
2664 Make inferior number @var{infno} the current inferior. The argument
2665 @var{infno} is the inferior number assigned by @value{GDBN}, as shown
2666 in the first field of the @samp{info inferiors} display.
2669 @vindex $_inferior@r{, convenience variable}
2670 The debugger convenience variable @samp{$_inferior} contains the
2671 number of the current inferior. You may find this useful in writing
2672 breakpoint conditional expressions, command scripts, and so forth.
2673 @xref{Convenience Vars,, Convenience Variables}, for general
2674 information on convenience variables.
2676 You can get multiple executables into a debugging session via the
2677 @code{add-inferior} and @w{@code{clone-inferior}} commands. On some
2678 systems @value{GDBN} can add inferiors to the debug session
2679 automatically by following calls to @code{fork} and @code{exec}. To
2680 remove inferiors from the debugging session use the
2681 @w{@code{remove-inferiors}} command.
2684 @kindex add-inferior
2685 @item add-inferior [ -copies @var{n} ] [ -exec @var{executable} ]
2686 Adds @var{n} inferiors to be run using @var{executable} as the
2687 executable; @var{n} defaults to 1. If no executable is specified,
2688 the inferiors begins empty, with no program. You can still assign or
2689 change the program assigned to the inferior at any time by using the
2690 @code{file} command with the executable name as its argument.
2692 @kindex clone-inferior
2693 @item clone-inferior [ -copies @var{n} ] [ @var{infno} ]
2694 Adds @var{n} inferiors ready to execute the same program as inferior
2695 @var{infno}; @var{n} defaults to 1, and @var{infno} defaults to the
2696 number of the current inferior. This is a convenient command when you
2697 want to run another instance of the inferior you are debugging.
2700 (@value{GDBP}) info inferiors
2701 Num Description Executable
2702 * 1 process 29964 helloworld
2703 (@value{GDBP}) clone-inferior
2706 (@value{GDBP}) info inferiors
2707 Num Description Executable
2709 * 1 process 29964 helloworld
2712 You can now simply switch focus to inferior 2 and run it.
2714 @kindex remove-inferiors
2715 @item remove-inferiors @var{infno}@dots{}
2716 Removes the inferior or inferiors @var{infno}@dots{}. It is not
2717 possible to remove an inferior that is running with this command. For
2718 those, use the @code{kill} or @code{detach} command first.
2722 To quit debugging one of the running inferiors that is not the current
2723 inferior, you can either detach from it by using the @w{@code{detach
2724 inferior}} command (allowing it to run independently), or kill it
2725 using the @w{@code{kill inferiors}} command:
2728 @kindex detach inferiors @var{infno}@dots{}
2729 @item detach inferior @var{infno}@dots{}
2730 Detach from the inferior or inferiors identified by @value{GDBN}
2731 inferior number(s) @var{infno}@dots{}. Note that the inferior's entry
2732 still stays on the list of inferiors shown by @code{info inferiors},
2733 but its Description will show @samp{<null>}.
2735 @kindex kill inferiors @var{infno}@dots{}
2736 @item kill inferiors @var{infno}@dots{}
2737 Kill the inferior or inferiors identified by @value{GDBN} inferior
2738 number(s) @var{infno}@dots{}. Note that the inferior's entry still
2739 stays on the list of inferiors shown by @code{info inferiors}, but its
2740 Description will show @samp{<null>}.
2743 After the successful completion of a command such as @code{detach},
2744 @code{detach inferiors}, @code{kill} or @code{kill inferiors}, or after
2745 a normal process exit, the inferior is still valid and listed with
2746 @code{info inferiors}, ready to be restarted.
2749 To be notified when inferiors are started or exit under @value{GDBN}'s
2750 control use @w{@code{set print inferior-events}}:
2753 @kindex set print inferior-events
2754 @cindex print messages on inferior start and exit
2755 @item set print inferior-events
2756 @itemx set print inferior-events on
2757 @itemx set print inferior-events off
2758 The @code{set print inferior-events} command allows you to enable or
2759 disable printing of messages when @value{GDBN} notices that new
2760 inferiors have started or that inferiors have exited or have been
2761 detached. By default, these messages will not be printed.
2763 @kindex show print inferior-events
2764 @item show print inferior-events
2765 Show whether messages will be printed when @value{GDBN} detects that
2766 inferiors have started, exited or have been detached.
2769 Many commands will work the same with multiple programs as with a
2770 single program: e.g., @code{print myglobal} will simply display the
2771 value of @code{myglobal} in the current inferior.
2774 Occasionaly, when debugging @value{GDBN} itself, it may be useful to
2775 get more info about the relationship of inferiors, programs, address
2776 spaces in a debug session. You can do that with the @w{@code{maint
2777 info program-spaces}} command.
2780 @kindex maint info program-spaces
2781 @item maint info program-spaces
2782 Print a list of all program spaces currently being managed by
2785 @value{GDBN} displays for each program space (in this order):
2789 the program space number assigned by @value{GDBN}
2792 the name of the executable loaded into the program space, with e.g.,
2793 the @code{file} command.
2798 An asterisk @samp{*} preceding the @value{GDBN} program space number
2799 indicates the current program space.
2801 In addition, below each program space line, @value{GDBN} prints extra
2802 information that isn't suitable to display in tabular form. For
2803 example, the list of inferiors bound to the program space.
2806 (@value{GDBP}) maint info program-spaces
2810 Bound inferiors: ID 1 (process 21561)
2813 Here we can see that no inferior is running the program @code{hello},
2814 while @code{process 21561} is running the program @code{goodbye}. On
2815 some targets, it is possible that multiple inferiors are bound to the
2816 same program space. The most common example is that of debugging both
2817 the parent and child processes of a @code{vfork} call. For example,
2820 (@value{GDBP}) maint info program-spaces
2823 Bound inferiors: ID 2 (process 18050), ID 1 (process 18045)
2826 Here, both inferior 2 and inferior 1 are running in the same program
2827 space as a result of inferior 1 having executed a @code{vfork} call.
2831 @section Debugging Programs with Multiple Threads
2833 @cindex threads of execution
2834 @cindex multiple threads
2835 @cindex switching threads
2836 In some operating systems, such as GNU/Linux and Solaris, a single program
2837 may have more than one @dfn{thread} of execution. The precise semantics
2838 of threads differ from one operating system to another, but in general
2839 the threads of a single program are akin to multiple processes---except
2840 that they share one address space (that is, they can all examine and
2841 modify the same variables). On the other hand, each thread has its own
2842 registers and execution stack, and perhaps private memory.
2844 @value{GDBN} provides these facilities for debugging multi-thread
2848 @item automatic notification of new threads
2849 @item @samp{thread @var{thread-id}}, a command to switch among threads
2850 @item @samp{info threads}, a command to inquire about existing threads
2851 @item @samp{thread apply [@var{thread-id-list}] [@var{all}] @var{args}},
2852 a command to apply a command to a list of threads
2853 @item thread-specific breakpoints
2854 @item @samp{set print thread-events}, which controls printing of
2855 messages on thread start and exit.
2856 @item @samp{set libthread-db-search-path @var{path}}, which lets
2857 the user specify which @code{libthread_db} to use if the default choice
2858 isn't compatible with the program.
2861 @cindex focus of debugging
2862 @cindex current thread
2863 The @value{GDBN} thread debugging facility allows you to observe all
2864 threads while your program runs---but whenever @value{GDBN} takes
2865 control, one thread in particular is always the focus of debugging.
2866 This thread is called the @dfn{current thread}. Debugging commands show
2867 program information from the perspective of the current thread.
2869 @cindex @code{New} @var{systag} message
2870 @cindex thread identifier (system)
2871 @c FIXME-implementors!! It would be more helpful if the [New...] message
2872 @c included GDB's numeric thread handle, so you could just go to that
2873 @c thread without first checking `info threads'.
2874 Whenever @value{GDBN} detects a new thread in your program, it displays
2875 the target system's identification for the thread with a message in the
2876 form @samp{[New @var{systag}]}, where @var{systag} is a thread identifier
2877 whose form varies depending on the particular system. For example, on
2878 @sc{gnu}/Linux, you might see
2881 [New Thread 0x41e02940 (LWP 25582)]
2885 when @value{GDBN} notices a new thread. In contrast, on other systems,
2886 the @var{systag} is simply something like @samp{process 368}, with no
2889 @c FIXME!! (1) Does the [New...] message appear even for the very first
2890 @c thread of a program, or does it only appear for the
2891 @c second---i.e.@: when it becomes obvious we have a multithread
2893 @c (2) *Is* there necessarily a first thread always? Or do some
2894 @c multithread systems permit starting a program with multiple
2895 @c threads ab initio?
2897 @anchor{thread numbers}
2898 @cindex thread number, per inferior
2899 @cindex thread identifier (GDB)
2900 For debugging purposes, @value{GDBN} associates its own thread number
2901 ---always a single integer---with each thread of an inferior. This
2902 number is unique between all threads of an inferior, but not unique
2903 between threads of different inferiors.
2905 @cindex qualified thread ID
2906 You can refer to a given thread in an inferior using the qualified
2907 @var{inferior-num}.@var{thread-num} syntax, also known as
2908 @dfn{qualified thread ID}, with @var{inferior-num} being the inferior
2909 number and @var{thread-num} being the thread number of the given
2910 inferior. For example, thread @code{2.3} refers to thread number 3 of
2911 inferior 2. If you omit @var{inferior-num} (e.g., @code{thread 3}),
2912 then @value{GDBN} infers you're referring to a thread of the current
2915 Until you create a second inferior, @value{GDBN} does not show the
2916 @var{inferior-num} part of thread IDs, even though you can always use
2917 the full @var{inferior-num}.@var{thread-num} form to refer to threads
2918 of inferior 1, the initial inferior.
2920 @anchor{thread ID lists}
2921 @cindex thread ID lists
2922 Some commands accept a space-separated @dfn{thread ID list} as
2923 argument. A list element can be:
2927 A thread ID as shown in the first field of the @samp{info threads}
2928 display, with or without an inferior qualifier. E.g., @samp{2.1} or
2932 A range of thread numbers, again with or without an inferior
2933 qualifier, as in @var{inf}.@var{thr1}-@var{thr2} or
2934 @var{thr1}-@var{thr2}. E.g., @samp{1.2-4} or @samp{2-4}.
2937 All threads of an inferior, specified with a star wildcard, with or
2938 without an inferior qualifier, as in @var{inf}.@code{*} (e.g.,
2939 @samp{1.*}) or @code{*}. The former refers to all threads of the
2940 given inferior, and the latter form without an inferior qualifier
2941 refers to all threads of the current inferior.
2945 For example, if the current inferior is 1, and inferior 7 has one
2946 thread with ID 7.1, the thread list @samp{1 2-3 4.5 6.7-9 7.*}
2947 includes threads 1 to 3 of inferior 1, thread 5 of inferior 4, threads
2948 7 to 9 of inferior 6 and all threads of inferior 7. That is, in
2949 expanded qualified form, the same as @samp{1.1 1.2 1.3 4.5 6.7 6.8 6.9
2953 @anchor{global thread numbers}
2954 @cindex global thread number
2955 @cindex global thread identifier (GDB)
2956 In addition to a @emph{per-inferior} number, each thread is also
2957 assigned a unique @emph{global} number, also known as @dfn{global
2958 thread ID}, a single integer. Unlike the thread number component of
2959 the thread ID, no two threads have the same global ID, even when
2960 you're debugging multiple inferiors.
2962 From @value{GDBN}'s perspective, a process always has at least one
2963 thread. In other words, @value{GDBN} assigns a thread number to the
2964 program's ``main thread'' even if the program is not multi-threaded.
2966 @vindex $_thread@r{, convenience variable}
2967 @vindex $_gthread@r{, convenience variable}
2968 The debugger convenience variables @samp{$_thread} and
2969 @samp{$_gthread} contain, respectively, the per-inferior thread number
2970 and the global thread number of the current thread. You may find this
2971 useful in writing breakpoint conditional expressions, command scripts,
2972 and so forth. @xref{Convenience Vars,, Convenience Variables}, for
2973 general information on convenience variables.
2975 If @value{GDBN} detects the program is multi-threaded, it augments the
2976 usual message about stopping at a breakpoint with the ID and name of
2977 the thread that hit the breakpoint.
2980 Thread 2 "client" hit Breakpoint 1, send_message () at client.c:68
2983 Likewise when the program receives a signal:
2986 Thread 1 "main" received signal SIGINT, Interrupt.
2990 @kindex info threads
2991 @item info threads @r{[}@var{thread-id-list}@r{]}
2993 Display information about one or more threads. With no arguments
2994 displays information about all threads. You can specify the list of
2995 threads that you want to display using the thread ID list syntax
2996 (@pxref{thread ID lists}).
2998 @value{GDBN} displays for each thread (in this order):
3002 the per-inferior thread number assigned by @value{GDBN}
3005 the global thread number assigned by @value{GDBN}, if the @samp{-gid}
3006 option was specified
3009 the target system's thread identifier (@var{systag})
3012 the thread's name, if one is known. A thread can either be named by
3013 the user (see @code{thread name}, below), or, in some cases, by the
3017 the current stack frame summary for that thread
3021 An asterisk @samp{*} to the left of the @value{GDBN} thread number
3022 indicates the current thread.
3026 @c end table here to get a little more width for example
3029 (@value{GDBP}) info threads
3031 * 1 process 35 thread 13 main (argc=1, argv=0x7ffffff8)
3032 2 process 35 thread 23 0x34e5 in sigpause ()
3033 3 process 35 thread 27 0x34e5 in sigpause ()
3037 If you're debugging multiple inferiors, @value{GDBN} displays thread
3038 IDs using the qualified @var{inferior-num}.@var{thread-num} format.
3039 Otherwise, only @var{thread-num} is shown.
3041 If you specify the @samp{-gid} option, @value{GDBN} displays a column
3042 indicating each thread's global thread ID:
3045 (@value{GDBP}) info threads
3046 Id GId Target Id Frame
3047 1.1 1 process 35 thread 13 main (argc=1, argv=0x7ffffff8)
3048 1.2 3 process 35 thread 23 0x34e5 in sigpause ()
3049 1.3 4 process 35 thread 27 0x34e5 in sigpause ()
3050 * 2.1 2 process 65 thread 1 main (argc=1, argv=0x7ffffff8)
3053 On Solaris, you can display more information about user threads with a
3054 Solaris-specific command:
3057 @item maint info sol-threads
3058 @kindex maint info sol-threads
3059 @cindex thread info (Solaris)
3060 Display info on Solaris user threads.
3064 @kindex thread @var{thread-id}
3065 @item thread @var{thread-id}
3066 Make thread ID @var{thread-id} the current thread. The command
3067 argument @var{thread-id} is the @value{GDBN} thread ID, as shown in
3068 the first field of the @samp{info threads} display, with or without an
3069 inferior qualifier (e.g., @samp{2.1} or @samp{1}).
3071 @value{GDBN} responds by displaying the system identifier of the
3072 thread you selected, and its current stack frame summary:
3075 (@value{GDBP}) thread 2
3076 [Switching to thread 2 (Thread 0xb7fdab70 (LWP 12747))]
3077 #0 some_function (ignore=0x0) at example.c:8
3078 8 printf ("hello\n");
3082 As with the @samp{[New @dots{}]} message, the form of the text after
3083 @samp{Switching to} depends on your system's conventions for identifying
3086 @kindex thread apply
3087 @cindex apply command to several threads
3088 @item thread apply [@var{thread-id-list} | all [-ascending]] @var{command}
3089 The @code{thread apply} command allows you to apply the named
3090 @var{command} to one or more threads. Specify the threads that you
3091 want affected using the thread ID list syntax (@pxref{thread ID
3092 lists}), or specify @code{all} to apply to all threads. To apply a
3093 command to all threads in descending order, type @kbd{thread apply all
3094 @var{command}}. To apply a command to all threads in ascending order,
3095 type @kbd{thread apply all -ascending @var{command}}.
3099 @cindex name a thread
3100 @item thread name [@var{name}]
3101 This command assigns a name to the current thread. If no argument is
3102 given, any existing user-specified name is removed. The thread name
3103 appears in the @samp{info threads} display.
3105 On some systems, such as @sc{gnu}/Linux, @value{GDBN} is able to
3106 determine the name of the thread as given by the OS. On these
3107 systems, a name specified with @samp{thread name} will override the
3108 system-give name, and removing the user-specified name will cause
3109 @value{GDBN} to once again display the system-specified name.
3112 @cindex search for a thread
3113 @item thread find [@var{regexp}]
3114 Search for and display thread ids whose name or @var{systag}
3115 matches the supplied regular expression.
3117 As well as being the complement to the @samp{thread name} command,
3118 this command also allows you to identify a thread by its target
3119 @var{systag}. For instance, on @sc{gnu}/Linux, the target @var{systag}
3123 (@value{GDBN}) thread find 26688
3124 Thread 4 has target id 'Thread 0x41e02940 (LWP 26688)'
3125 (@value{GDBN}) info thread 4
3127 4 Thread 0x41e02940 (LWP 26688) 0x00000031ca6cd372 in select ()
3130 @kindex set print thread-events
3131 @cindex print messages on thread start and exit
3132 @item set print thread-events
3133 @itemx set print thread-events on
3134 @itemx set print thread-events off
3135 The @code{set print thread-events} command allows you to enable or
3136 disable printing of messages when @value{GDBN} notices that new threads have
3137 started or that threads have exited. By default, these messages will
3138 be printed if detection of these events is supported by the target.
3139 Note that these messages cannot be disabled on all targets.
3141 @kindex show print thread-events
3142 @item show print thread-events
3143 Show whether messages will be printed when @value{GDBN} detects that threads
3144 have started and exited.
3147 @xref{Thread Stops,,Stopping and Starting Multi-thread Programs}, for
3148 more information about how @value{GDBN} behaves when you stop and start
3149 programs with multiple threads.
3151 @xref{Set Watchpoints,,Setting Watchpoints}, for information about
3152 watchpoints in programs with multiple threads.
3154 @anchor{set libthread-db-search-path}
3156 @kindex set libthread-db-search-path
3157 @cindex search path for @code{libthread_db}
3158 @item set libthread-db-search-path @r{[}@var{path}@r{]}
3159 If this variable is set, @var{path} is a colon-separated list of
3160 directories @value{GDBN} will use to search for @code{libthread_db}.
3161 If you omit @var{path}, @samp{libthread-db-search-path} will be reset to
3162 its default value (@code{$sdir:$pdir} on @sc{gnu}/Linux and Solaris systems).
3163 Internally, the default value comes from the @code{LIBTHREAD_DB_SEARCH_PATH}
3166 On @sc{gnu}/Linux and Solaris systems, @value{GDBN} uses a ``helper''
3167 @code{libthread_db} library to obtain information about threads in the
3168 inferior process. @value{GDBN} will use @samp{libthread-db-search-path}
3169 to find @code{libthread_db}. @value{GDBN} also consults first if inferior
3170 specific thread debugging library loading is enabled
3171 by @samp{set auto-load libthread-db} (@pxref{libthread_db.so.1 file}).
3173 A special entry @samp{$sdir} for @samp{libthread-db-search-path}
3174 refers to the default system directories that are
3175 normally searched for loading shared libraries. The @samp{$sdir} entry
3176 is the only kind not needing to be enabled by @samp{set auto-load libthread-db}
3177 (@pxref{libthread_db.so.1 file}).
3179 A special entry @samp{$pdir} for @samp{libthread-db-search-path}
3180 refers to the directory from which @code{libpthread}
3181 was loaded in the inferior process.
3183 For any @code{libthread_db} library @value{GDBN} finds in above directories,
3184 @value{GDBN} attempts to initialize it with the current inferior process.
3185 If this initialization fails (which could happen because of a version
3186 mismatch between @code{libthread_db} and @code{libpthread}), @value{GDBN}
3187 will unload @code{libthread_db}, and continue with the next directory.
3188 If none of @code{libthread_db} libraries initialize successfully,
3189 @value{GDBN} will issue a warning and thread debugging will be disabled.
3191 Setting @code{libthread-db-search-path} is currently implemented
3192 only on some platforms.
3194 @kindex show libthread-db-search-path
3195 @item show libthread-db-search-path
3196 Display current libthread_db search path.
3198 @kindex set debug libthread-db
3199 @kindex show debug libthread-db
3200 @cindex debugging @code{libthread_db}
3201 @item set debug libthread-db
3202 @itemx show debug libthread-db
3203 Turns on or off display of @code{libthread_db}-related events.
3204 Use @code{1} to enable, @code{0} to disable.
3208 @section Debugging Forks
3210 @cindex fork, debugging programs which call
3211 @cindex multiple processes
3212 @cindex processes, multiple
3213 On most systems, @value{GDBN} has no special support for debugging
3214 programs which create additional processes using the @code{fork}
3215 function. When a program forks, @value{GDBN} will continue to debug the
3216 parent process and the child process will run unimpeded. If you have
3217 set a breakpoint in any code which the child then executes, the child
3218 will get a @code{SIGTRAP} signal which (unless it catches the signal)
3219 will cause it to terminate.
3221 However, if you want to debug the child process there is a workaround
3222 which isn't too painful. Put a call to @code{sleep} in the code which
3223 the child process executes after the fork. It may be useful to sleep
3224 only if a certain environment variable is set, or a certain file exists,
3225 so that the delay need not occur when you don't want to run @value{GDBN}
3226 on the child. While the child is sleeping, use the @code{ps} program to
3227 get its process ID. Then tell @value{GDBN} (a new invocation of
3228 @value{GDBN} if you are also debugging the parent process) to attach to
3229 the child process (@pxref{Attach}). From that point on you can debug
3230 the child process just like any other process which you attached to.
3232 On some systems, @value{GDBN} provides support for debugging programs
3233 that create additional processes using the @code{fork} or @code{vfork}
3234 functions. On @sc{gnu}/Linux platforms, this feature is supported
3235 with kernel version 2.5.46 and later.
3237 The fork debugging commands are supported in native mode and when
3238 connected to @code{gdbserver} in either @code{target remote} mode or
3239 @code{target extended-remote} mode.
3241 By default, when a program forks, @value{GDBN} will continue to debug
3242 the parent process and the child process will run unimpeded.
3244 If you want to follow the child process instead of the parent process,
3245 use the command @w{@code{set follow-fork-mode}}.
3248 @kindex set follow-fork-mode
3249 @item set follow-fork-mode @var{mode}
3250 Set the debugger response to a program call of @code{fork} or
3251 @code{vfork}. A call to @code{fork} or @code{vfork} creates a new
3252 process. The @var{mode} argument can be:
3256 The original process is debugged after a fork. The child process runs
3257 unimpeded. This is the default.
3260 The new process is debugged after a fork. The parent process runs
3265 @kindex show follow-fork-mode
3266 @item show follow-fork-mode
3267 Display the current debugger response to a @code{fork} or @code{vfork} call.
3270 @cindex debugging multiple processes
3271 On Linux, if you want to debug both the parent and child processes, use the
3272 command @w{@code{set detach-on-fork}}.
3275 @kindex set detach-on-fork
3276 @item set detach-on-fork @var{mode}
3277 Tells gdb whether to detach one of the processes after a fork, or
3278 retain debugger control over them both.
3282 The child process (or parent process, depending on the value of
3283 @code{follow-fork-mode}) will be detached and allowed to run
3284 independently. This is the default.
3287 Both processes will be held under the control of @value{GDBN}.
3288 One process (child or parent, depending on the value of
3289 @code{follow-fork-mode}) is debugged as usual, while the other
3294 @kindex show detach-on-fork
3295 @item show detach-on-fork
3296 Show whether detach-on-fork mode is on/off.
3299 If you choose to set @samp{detach-on-fork} mode off, then @value{GDBN}
3300 will retain control of all forked processes (including nested forks).
3301 You can list the forked processes under the control of @value{GDBN} by
3302 using the @w{@code{info inferiors}} command, and switch from one fork
3303 to another by using the @code{inferior} command (@pxref{Inferiors and
3304 Programs, ,Debugging Multiple Inferiors and Programs}).
3306 To quit debugging one of the forked processes, you can either detach
3307 from it by using the @w{@code{detach inferiors}} command (allowing it
3308 to run independently), or kill it using the @w{@code{kill inferiors}}
3309 command. @xref{Inferiors and Programs, ,Debugging Multiple Inferiors
3312 If you ask to debug a child process and a @code{vfork} is followed by an
3313 @code{exec}, @value{GDBN} executes the new target up to the first
3314 breakpoint in the new target. If you have a breakpoint set on
3315 @code{main} in your original program, the breakpoint will also be set on
3316 the child process's @code{main}.
3318 On some systems, when a child process is spawned by @code{vfork}, you
3319 cannot debug the child or parent until an @code{exec} call completes.
3321 If you issue a @code{run} command to @value{GDBN} after an @code{exec}
3322 call executes, the new target restarts. To restart the parent
3323 process, use the @code{file} command with the parent executable name
3324 as its argument. By default, after an @code{exec} call executes,
3325 @value{GDBN} discards the symbols of the previous executable image.
3326 You can change this behaviour with the @w{@code{set follow-exec-mode}}
3330 @kindex set follow-exec-mode
3331 @item set follow-exec-mode @var{mode}
3333 Set debugger response to a program call of @code{exec}. An
3334 @code{exec} call replaces the program image of a process.
3336 @code{follow-exec-mode} can be:
3340 @value{GDBN} creates a new inferior and rebinds the process to this
3341 new inferior. The program the process was running before the
3342 @code{exec} call can be restarted afterwards by restarting the
3348 (@value{GDBP}) info inferiors
3350 Id Description Executable
3353 process 12020 is executing new program: prog2
3354 Program exited normally.
3355 (@value{GDBP}) info inferiors
3356 Id Description Executable
3362 @value{GDBN} keeps the process bound to the same inferior. The new
3363 executable image replaces the previous executable loaded in the
3364 inferior. Restarting the inferior after the @code{exec} call, with
3365 e.g., the @code{run} command, restarts the executable the process was
3366 running after the @code{exec} call. This is the default mode.
3371 (@value{GDBP}) info inferiors
3372 Id Description Executable
3375 process 12020 is executing new program: prog2
3376 Program exited normally.
3377 (@value{GDBP}) info inferiors
3378 Id Description Executable
3385 @code{follow-exec-mode} is supported in native mode and
3386 @code{target extended-remote} mode.
3388 You can use the @code{catch} command to make @value{GDBN} stop whenever
3389 a @code{fork}, @code{vfork}, or @code{exec} call is made. @xref{Set
3390 Catchpoints, ,Setting Catchpoints}.
3392 @node Checkpoint/Restart
3393 @section Setting a @emph{Bookmark} to Return to Later
3398 @cindex snapshot of a process
3399 @cindex rewind program state
3401 On certain operating systems@footnote{Currently, only
3402 @sc{gnu}/Linux.}, @value{GDBN} is able to save a @dfn{snapshot} of a
3403 program's state, called a @dfn{checkpoint}, and come back to it
3406 Returning to a checkpoint effectively undoes everything that has
3407 happened in the program since the @code{checkpoint} was saved. This
3408 includes changes in memory, registers, and even (within some limits)
3409 system state. Effectively, it is like going back in time to the
3410 moment when the checkpoint was saved.
3412 Thus, if you're stepping thru a program and you think you're
3413 getting close to the point where things go wrong, you can save
3414 a checkpoint. Then, if you accidentally go too far and miss
3415 the critical statement, instead of having to restart your program
3416 from the beginning, you can just go back to the checkpoint and
3417 start again from there.
3419 This can be especially useful if it takes a lot of time or
3420 steps to reach the point where you think the bug occurs.
3422 To use the @code{checkpoint}/@code{restart} method of debugging:
3427 Save a snapshot of the debugged program's current execution state.
3428 The @code{checkpoint} command takes no arguments, but each checkpoint
3429 is assigned a small integer id, similar to a breakpoint id.
3431 @kindex info checkpoints
3432 @item info checkpoints
3433 List the checkpoints that have been saved in the current debugging
3434 session. For each checkpoint, the following information will be
3441 @item Source line, or label
3444 @kindex restart @var{checkpoint-id}
3445 @item restart @var{checkpoint-id}
3446 Restore the program state that was saved as checkpoint number
3447 @var{checkpoint-id}. All program variables, registers, stack frames
3448 etc.@: will be returned to the values that they had when the checkpoint
3449 was saved. In essence, gdb will ``wind back the clock'' to the point
3450 in time when the checkpoint was saved.
3452 Note that breakpoints, @value{GDBN} variables, command history etc.
3453 are not affected by restoring a checkpoint. In general, a checkpoint
3454 only restores things that reside in the program being debugged, not in
3457 @kindex delete checkpoint @var{checkpoint-id}
3458 @item delete checkpoint @var{checkpoint-id}
3459 Delete the previously-saved checkpoint identified by @var{checkpoint-id}.
3463 Returning to a previously saved checkpoint will restore the user state
3464 of the program being debugged, plus a significant subset of the system
3465 (OS) state, including file pointers. It won't ``un-write'' data from
3466 a file, but it will rewind the file pointer to the previous location,
3467 so that the previously written data can be overwritten. For files
3468 opened in read mode, the pointer will also be restored so that the
3469 previously read data can be read again.
3471 Of course, characters that have been sent to a printer (or other
3472 external device) cannot be ``snatched back'', and characters received
3473 from eg.@: a serial device can be removed from internal program buffers,
3474 but they cannot be ``pushed back'' into the serial pipeline, ready to
3475 be received again. Similarly, the actual contents of files that have
3476 been changed cannot be restored (at this time).
3478 However, within those constraints, you actually can ``rewind'' your
3479 program to a previously saved point in time, and begin debugging it
3480 again --- and you can change the course of events so as to debug a
3481 different execution path this time.
3483 @cindex checkpoints and process id
3484 Finally, there is one bit of internal program state that will be
3485 different when you return to a checkpoint --- the program's process
3486 id. Each checkpoint will have a unique process id (or @var{pid}),
3487 and each will be different from the program's original @var{pid}.
3488 If your program has saved a local copy of its process id, this could
3489 potentially pose a problem.
3491 @subsection A Non-obvious Benefit of Using Checkpoints
3493 On some systems such as @sc{gnu}/Linux, address space randomization
3494 is performed on new processes for security reasons. This makes it
3495 difficult or impossible to set a breakpoint, or watchpoint, on an
3496 absolute address if you have to restart the program, since the
3497 absolute location of a symbol will change from one execution to the
3500 A checkpoint, however, is an @emph{identical} copy of a process.
3501 Therefore if you create a checkpoint at (eg.@:) the start of main,
3502 and simply return to that checkpoint instead of restarting the
3503 process, you can avoid the effects of address randomization and
3504 your symbols will all stay in the same place.
3507 @chapter Stopping and Continuing
3509 The principal purposes of using a debugger are so that you can stop your
3510 program before it terminates; or so that, if your program runs into
3511 trouble, you can investigate and find out why.
3513 Inside @value{GDBN}, your program may stop for any of several reasons,
3514 such as a signal, a breakpoint, or reaching a new line after a
3515 @value{GDBN} command such as @code{step}. You may then examine and
3516 change variables, set new breakpoints or remove old ones, and then
3517 continue execution. Usually, the messages shown by @value{GDBN} provide
3518 ample explanation of the status of your program---but you can also
3519 explicitly request this information at any time.
3522 @kindex info program
3524 Display information about the status of your program: whether it is
3525 running or not, what process it is, and why it stopped.
3529 * Breakpoints:: Breakpoints, watchpoints, and catchpoints
3530 * Continuing and Stepping:: Resuming execution
3531 * Skipping Over Functions and Files::
3532 Skipping over functions and files
3534 * Thread Stops:: Stopping and starting multi-thread programs
3538 @section Breakpoints, Watchpoints, and Catchpoints
3541 A @dfn{breakpoint} makes your program stop whenever a certain point in
3542 the program is reached. For each breakpoint, you can add conditions to
3543 control in finer detail whether your program stops. You can set
3544 breakpoints with the @code{break} command and its variants (@pxref{Set
3545 Breaks, ,Setting Breakpoints}), to specify the place where your program
3546 should stop by line number, function name or exact address in the
3549 On some systems, you can set breakpoints in shared libraries before
3550 the executable is run.
3553 @cindex data breakpoints
3554 @cindex memory tracing
3555 @cindex breakpoint on memory address
3556 @cindex breakpoint on variable modification
3557 A @dfn{watchpoint} is a special breakpoint that stops your program
3558 when the value of an expression changes. The expression may be a value
3559 of a variable, or it could involve values of one or more variables
3560 combined by operators, such as @samp{a + b}. This is sometimes called
3561 @dfn{data breakpoints}. You must use a different command to set
3562 watchpoints (@pxref{Set Watchpoints, ,Setting Watchpoints}), but aside
3563 from that, you can manage a watchpoint like any other breakpoint: you
3564 enable, disable, and delete both breakpoints and watchpoints using the
3567 You can arrange to have values from your program displayed automatically
3568 whenever @value{GDBN} stops at a breakpoint. @xref{Auto Display,,
3572 @cindex breakpoint on events
3573 A @dfn{catchpoint} is another special breakpoint that stops your program
3574 when a certain kind of event occurs, such as the throwing of a C@t{++}
3575 exception or the loading of a library. As with watchpoints, you use a
3576 different command to set a catchpoint (@pxref{Set Catchpoints, ,Setting
3577 Catchpoints}), but aside from that, you can manage a catchpoint like any
3578 other breakpoint. (To stop when your program receives a signal, use the
3579 @code{handle} command; see @ref{Signals, ,Signals}.)
3581 @cindex breakpoint numbers
3582 @cindex numbers for breakpoints
3583 @value{GDBN} assigns a number to each breakpoint, watchpoint, or
3584 catchpoint when you create it; these numbers are successive integers
3585 starting with one. In many of the commands for controlling various
3586 features of breakpoints you use the breakpoint number to say which
3587 breakpoint you want to change. Each breakpoint may be @dfn{enabled} or
3588 @dfn{disabled}; if disabled, it has no effect on your program until you
3591 @cindex breakpoint ranges
3592 @cindex breakpoint lists
3593 @cindex ranges of breakpoints
3594 @cindex lists of breakpoints
3595 Some @value{GDBN} commands accept a space-separated list of breakpoints
3596 on which to operate. A list element can be either a single breakpoint number,
3597 like @samp{5}, or a range of such numbers, like @samp{5-7}.
3598 When a breakpoint list is given to a command, all breakpoints in that list
3602 * Set Breaks:: Setting breakpoints
3603 * Set Watchpoints:: Setting watchpoints
3604 * Set Catchpoints:: Setting catchpoints
3605 * Delete Breaks:: Deleting breakpoints
3606 * Disabling:: Disabling breakpoints
3607 * Conditions:: Break conditions
3608 * Break Commands:: Breakpoint command lists
3609 * Dynamic Printf:: Dynamic printf
3610 * Save Breakpoints:: How to save breakpoints in a file
3611 * Static Probe Points:: Listing static probe points
3612 * Error in Breakpoints:: ``Cannot insert breakpoints''
3613 * Breakpoint-related Warnings:: ``Breakpoint address adjusted...''
3617 @subsection Setting Breakpoints
3619 @c FIXME LMB what does GDB do if no code on line of breakpt?
3620 @c consider in particular declaration with/without initialization.
3622 @c FIXME 2 is there stuff on this already? break at fun start, already init?
3625 @kindex b @r{(@code{break})}
3626 @vindex $bpnum@r{, convenience variable}
3627 @cindex latest breakpoint
3628 Breakpoints are set with the @code{break} command (abbreviated
3629 @code{b}). The debugger convenience variable @samp{$bpnum} records the
3630 number of the breakpoint you've set most recently; see @ref{Convenience
3631 Vars,, Convenience Variables}, for a discussion of what you can do with
3632 convenience variables.
3635 @item break @var{location}
3636 Set a breakpoint at the given @var{location}, which can specify a
3637 function name, a line number, or an address of an instruction.
3638 (@xref{Specify Location}, for a list of all the possible ways to
3639 specify a @var{location}.) The breakpoint will stop your program just
3640 before it executes any of the code in the specified @var{location}.
3642 When using source languages that permit overloading of symbols, such as
3643 C@t{++}, a function name may refer to more than one possible place to break.
3644 @xref{Ambiguous Expressions,,Ambiguous Expressions}, for a discussion of
3647 It is also possible to insert a breakpoint that will stop the program
3648 only if a specific thread (@pxref{Thread-Specific Breakpoints})
3649 or a specific task (@pxref{Ada Tasks}) hits that breakpoint.
3652 When called without any arguments, @code{break} sets a breakpoint at
3653 the next instruction to be executed in the selected stack frame
3654 (@pxref{Stack, ,Examining the Stack}). In any selected frame but the
3655 innermost, this makes your program stop as soon as control
3656 returns to that frame. This is similar to the effect of a
3657 @code{finish} command in the frame inside the selected frame---except
3658 that @code{finish} does not leave an active breakpoint. If you use
3659 @code{break} without an argument in the innermost frame, @value{GDBN} stops
3660 the next time it reaches the current location; this may be useful
3663 @value{GDBN} normally ignores breakpoints when it resumes execution, until at
3664 least one instruction has been executed. If it did not do this, you
3665 would be unable to proceed past a breakpoint without first disabling the
3666 breakpoint. This rule applies whether or not the breakpoint already
3667 existed when your program stopped.
3669 @item break @dots{} if @var{cond}
3670 Set a breakpoint with condition @var{cond}; evaluate the expression
3671 @var{cond} each time the breakpoint is reached, and stop only if the
3672 value is nonzero---that is, if @var{cond} evaluates as true.
3673 @samp{@dots{}} stands for one of the possible arguments described
3674 above (or no argument) specifying where to break. @xref{Conditions,
3675 ,Break Conditions}, for more information on breakpoint conditions.
3678 @item tbreak @var{args}
3679 Set a breakpoint enabled only for one stop. The @var{args} are the
3680 same as for the @code{break} command, and the breakpoint is set in the same
3681 way, but the breakpoint is automatically deleted after the first time your
3682 program stops there. @xref{Disabling, ,Disabling Breakpoints}.
3685 @cindex hardware breakpoints
3686 @item hbreak @var{args}
3687 Set a hardware-assisted breakpoint. The @var{args} are the same as for the
3688 @code{break} command and the breakpoint is set in the same way, but the
3689 breakpoint requires hardware support and some target hardware may not
3690 have this support. The main purpose of this is EPROM/ROM code
3691 debugging, so you can set a breakpoint at an instruction without
3692 changing the instruction. This can be used with the new trap-generation
3693 provided by SPARClite DSU and most x86-based targets. These targets
3694 will generate traps when a program accesses some data or instruction
3695 address that is assigned to the debug registers. However the hardware
3696 breakpoint registers can take a limited number of breakpoints. For
3697 example, on the DSU, only two data breakpoints can be set at a time, and
3698 @value{GDBN} will reject this command if more than two are used. Delete
3699 or disable unused hardware breakpoints before setting new ones
3700 (@pxref{Disabling, ,Disabling Breakpoints}).
3701 @xref{Conditions, ,Break Conditions}.
3702 For remote targets, you can restrict the number of hardware
3703 breakpoints @value{GDBN} will use, see @ref{set remote
3704 hardware-breakpoint-limit}.
3707 @item thbreak @var{args}
3708 Set a hardware-assisted breakpoint enabled only for one stop. The @var{args}
3709 are the same as for the @code{hbreak} command and the breakpoint is set in
3710 the same way. However, like the @code{tbreak} command,
3711 the breakpoint is automatically deleted after the
3712 first time your program stops there. Also, like the @code{hbreak}
3713 command, the breakpoint requires hardware support and some target hardware
3714 may not have this support. @xref{Disabling, ,Disabling Breakpoints}.
3715 See also @ref{Conditions, ,Break Conditions}.
3718 @cindex regular expression
3719 @cindex breakpoints at functions matching a regexp
3720 @cindex set breakpoints in many functions
3721 @item rbreak @var{regex}
3722 Set breakpoints on all functions matching the regular expression
3723 @var{regex}. This command sets an unconditional breakpoint on all
3724 matches, printing a list of all breakpoints it set. Once these
3725 breakpoints are set, they are treated just like the breakpoints set with
3726 the @code{break} command. You can delete them, disable them, or make
3727 them conditional the same way as any other breakpoint.
3729 The syntax of the regular expression is the standard one used with tools
3730 like @file{grep}. Note that this is different from the syntax used by
3731 shells, so for instance @code{foo*} matches all functions that include
3732 an @code{fo} followed by zero or more @code{o}s. There is an implicit
3733 @code{.*} leading and trailing the regular expression you supply, so to
3734 match only functions that begin with @code{foo}, use @code{^foo}.
3736 @cindex non-member C@t{++} functions, set breakpoint in
3737 When debugging C@t{++} programs, @code{rbreak} is useful for setting
3738 breakpoints on overloaded functions that are not members of any special
3741 @cindex set breakpoints on all functions
3742 The @code{rbreak} command can be used to set breakpoints in
3743 @strong{all} the functions in a program, like this:
3746 (@value{GDBP}) rbreak .
3749 @item rbreak @var{file}:@var{regex}
3750 If @code{rbreak} is called with a filename qualification, it limits
3751 the search for functions matching the given regular expression to the
3752 specified @var{file}. This can be used, for example, to set breakpoints on
3753 every function in a given file:
3756 (@value{GDBP}) rbreak file.c:.
3759 The colon separating the filename qualifier from the regex may
3760 optionally be surrounded by spaces.
3762 @kindex info breakpoints
3763 @cindex @code{$_} and @code{info breakpoints}
3764 @item info breakpoints @r{[}@var{list}@dots{}@r{]}
3765 @itemx info break @r{[}@var{list}@dots{}@r{]}
3766 Print a table of all breakpoints, watchpoints, and catchpoints set and
3767 not deleted. Optional argument @var{n} means print information only
3768 about the specified breakpoint(s) (or watchpoint(s) or catchpoint(s)).
3769 For each breakpoint, following columns are printed:
3772 @item Breakpoint Numbers
3774 Breakpoint, watchpoint, or catchpoint.
3776 Whether the breakpoint is marked to be disabled or deleted when hit.
3777 @item Enabled or Disabled
3778 Enabled breakpoints are marked with @samp{y}. @samp{n} marks breakpoints
3779 that are not enabled.
3781 Where the breakpoint is in your program, as a memory address. For a
3782 pending breakpoint whose address is not yet known, this field will
3783 contain @samp{<PENDING>}. Such breakpoint won't fire until a shared
3784 library that has the symbol or line referred by breakpoint is loaded.
3785 See below for details. A breakpoint with several locations will
3786 have @samp{<MULTIPLE>} in this field---see below for details.
3788 Where the breakpoint is in the source for your program, as a file and
3789 line number. For a pending breakpoint, the original string passed to
3790 the breakpoint command will be listed as it cannot be resolved until
3791 the appropriate shared library is loaded in the future.
3795 If a breakpoint is conditional, there are two evaluation modes: ``host'' and
3796 ``target''. If mode is ``host'', breakpoint condition evaluation is done by
3797 @value{GDBN} on the host's side. If it is ``target'', then the condition
3798 is evaluated by the target. The @code{info break} command shows
3799 the condition on the line following the affected breakpoint, together with
3800 its condition evaluation mode in between parentheses.
3802 Breakpoint commands, if any, are listed after that. A pending breakpoint is
3803 allowed to have a condition specified for it. The condition is not parsed for
3804 validity until a shared library is loaded that allows the pending
3805 breakpoint to resolve to a valid location.
3808 @code{info break} with a breakpoint
3809 number @var{n} as argument lists only that breakpoint. The
3810 convenience variable @code{$_} and the default examining-address for
3811 the @code{x} command are set to the address of the last breakpoint
3812 listed (@pxref{Memory, ,Examining Memory}).
3815 @code{info break} displays a count of the number of times the breakpoint
3816 has been hit. This is especially useful in conjunction with the
3817 @code{ignore} command. You can ignore a large number of breakpoint
3818 hits, look at the breakpoint info to see how many times the breakpoint
3819 was hit, and then run again, ignoring one less than that number. This
3820 will get you quickly to the last hit of that breakpoint.
3823 For a breakpoints with an enable count (xref) greater than 1,
3824 @code{info break} also displays that count.
3828 @value{GDBN} allows you to set any number of breakpoints at the same place in
3829 your program. There is nothing silly or meaningless about this. When
3830 the breakpoints are conditional, this is even useful
3831 (@pxref{Conditions, ,Break Conditions}).
3833 @cindex multiple locations, breakpoints
3834 @cindex breakpoints, multiple locations
3835 It is possible that a breakpoint corresponds to several locations
3836 in your program. Examples of this situation are:
3840 Multiple functions in the program may have the same name.
3843 For a C@t{++} constructor, the @value{NGCC} compiler generates several
3844 instances of the function body, used in different cases.
3847 For a C@t{++} template function, a given line in the function can
3848 correspond to any number of instantiations.
3851 For an inlined function, a given source line can correspond to
3852 several places where that function is inlined.
3855 In all those cases, @value{GDBN} will insert a breakpoint at all
3856 the relevant locations.
3858 A breakpoint with multiple locations is displayed in the breakpoint
3859 table using several rows---one header row, followed by one row for
3860 each breakpoint location. The header row has @samp{<MULTIPLE>} in the
3861 address column. The rows for individual locations contain the actual
3862 addresses for locations, and show the functions to which those
3863 locations belong. The number column for a location is of the form
3864 @var{breakpoint-number}.@var{location-number}.
3869 Num Type Disp Enb Address What
3870 1 breakpoint keep y <MULTIPLE>
3872 breakpoint already hit 1 time
3873 1.1 y 0x080486a2 in void foo<int>() at t.cc:8
3874 1.2 y 0x080486ca in void foo<double>() at t.cc:8
3877 Each location can be individually enabled or disabled by passing
3878 @var{breakpoint-number}.@var{location-number} as argument to the
3879 @code{enable} and @code{disable} commands. Note that you cannot
3880 delete the individual locations from the list, you can only delete the
3881 entire list of locations that belong to their parent breakpoint (with
3882 the @kbd{delete @var{num}} command, where @var{num} is the number of
3883 the parent breakpoint, 1 in the above example). Disabling or enabling
3884 the parent breakpoint (@pxref{Disabling}) affects all of the locations
3885 that belong to that breakpoint.
3887 @cindex pending breakpoints
3888 It's quite common to have a breakpoint inside a shared library.
3889 Shared libraries can be loaded and unloaded explicitly,
3890 and possibly repeatedly, as the program is executed. To support
3891 this use case, @value{GDBN} updates breakpoint locations whenever
3892 any shared library is loaded or unloaded. Typically, you would
3893 set a breakpoint in a shared library at the beginning of your
3894 debugging session, when the library is not loaded, and when the
3895 symbols from the library are not available. When you try to set
3896 breakpoint, @value{GDBN} will ask you if you want to set
3897 a so called @dfn{pending breakpoint}---breakpoint whose address
3898 is not yet resolved.
3900 After the program is run, whenever a new shared library is loaded,
3901 @value{GDBN} reevaluates all the breakpoints. When a newly loaded
3902 shared library contains the symbol or line referred to by some
3903 pending breakpoint, that breakpoint is resolved and becomes an
3904 ordinary breakpoint. When a library is unloaded, all breakpoints
3905 that refer to its symbols or source lines become pending again.
3907 This logic works for breakpoints with multiple locations, too. For
3908 example, if you have a breakpoint in a C@t{++} template function, and
3909 a newly loaded shared library has an instantiation of that template,
3910 a new location is added to the list of locations for the breakpoint.
3912 Except for having unresolved address, pending breakpoints do not
3913 differ from regular breakpoints. You can set conditions or commands,
3914 enable and disable them and perform other breakpoint operations.
3916 @value{GDBN} provides some additional commands for controlling what
3917 happens when the @samp{break} command cannot resolve breakpoint
3918 address specification to an address:
3920 @kindex set breakpoint pending
3921 @kindex show breakpoint pending
3923 @item set breakpoint pending auto
3924 This is the default behavior. When @value{GDBN} cannot find the breakpoint
3925 location, it queries you whether a pending breakpoint should be created.
3927 @item set breakpoint pending on
3928 This indicates that an unrecognized breakpoint location should automatically
3929 result in a pending breakpoint being created.
3931 @item set breakpoint pending off
3932 This indicates that pending breakpoints are not to be created. Any
3933 unrecognized breakpoint location results in an error. This setting does
3934 not affect any pending breakpoints previously created.
3936 @item show breakpoint pending
3937 Show the current behavior setting for creating pending breakpoints.
3940 The settings above only affect the @code{break} command and its
3941 variants. Once breakpoint is set, it will be automatically updated
3942 as shared libraries are loaded and unloaded.
3944 @cindex automatic hardware breakpoints
3945 For some targets, @value{GDBN} can automatically decide if hardware or
3946 software breakpoints should be used, depending on whether the
3947 breakpoint address is read-only or read-write. This applies to
3948 breakpoints set with the @code{break} command as well as to internal
3949 breakpoints set by commands like @code{next} and @code{finish}. For
3950 breakpoints set with @code{hbreak}, @value{GDBN} will always use hardware
3953 You can control this automatic behaviour with the following commands:
3955 @kindex set breakpoint auto-hw
3956 @kindex show breakpoint auto-hw
3958 @item set breakpoint auto-hw on
3959 This is the default behavior. When @value{GDBN} sets a breakpoint, it
3960 will try to use the target memory map to decide if software or hardware
3961 breakpoint must be used.
3963 @item set breakpoint auto-hw off
3964 This indicates @value{GDBN} should not automatically select breakpoint
3965 type. If the target provides a memory map, @value{GDBN} will warn when
3966 trying to set software breakpoint at a read-only address.
3969 @value{GDBN} normally implements breakpoints by replacing the program code
3970 at the breakpoint address with a special instruction, which, when
3971 executed, given control to the debugger. By default, the program
3972 code is so modified only when the program is resumed. As soon as
3973 the program stops, @value{GDBN} restores the original instructions. This
3974 behaviour guards against leaving breakpoints inserted in the
3975 target should gdb abrubptly disconnect. However, with slow remote
3976 targets, inserting and removing breakpoint can reduce the performance.
3977 This behavior can be controlled with the following commands::
3979 @kindex set breakpoint always-inserted
3980 @kindex show breakpoint always-inserted
3982 @item set breakpoint always-inserted off
3983 All breakpoints, including newly added by the user, are inserted in
3984 the target only when the target is resumed. All breakpoints are
3985 removed from the target when it stops. This is the default mode.
3987 @item set breakpoint always-inserted on
3988 Causes all breakpoints to be inserted in the target at all times. If
3989 the user adds a new breakpoint, or changes an existing breakpoint, the
3990 breakpoints in the target are updated immediately. A breakpoint is
3991 removed from the target only when breakpoint itself is deleted.
3994 @value{GDBN} handles conditional breakpoints by evaluating these conditions
3995 when a breakpoint breaks. If the condition is true, then the process being
3996 debugged stops, otherwise the process is resumed.
3998 If the target supports evaluating conditions on its end, @value{GDBN} may
3999 download the breakpoint, together with its conditions, to it.
4001 This feature can be controlled via the following commands:
4003 @kindex set breakpoint condition-evaluation
4004 @kindex show breakpoint condition-evaluation
4006 @item set breakpoint condition-evaluation host
4007 This option commands @value{GDBN} to evaluate the breakpoint
4008 conditions on the host's side. Unconditional breakpoints are sent to
4009 the target which in turn receives the triggers and reports them back to GDB
4010 for condition evaluation. This is the standard evaluation mode.
4012 @item set breakpoint condition-evaluation target
4013 This option commands @value{GDBN} to download breakpoint conditions
4014 to the target at the moment of their insertion. The target
4015 is responsible for evaluating the conditional expression and reporting
4016 breakpoint stop events back to @value{GDBN} whenever the condition
4017 is true. Due to limitations of target-side evaluation, some conditions
4018 cannot be evaluated there, e.g., conditions that depend on local data
4019 that is only known to the host. Examples include
4020 conditional expressions involving convenience variables, complex types
4021 that cannot be handled by the agent expression parser and expressions
4022 that are too long to be sent over to the target, specially when the
4023 target is a remote system. In these cases, the conditions will be
4024 evaluated by @value{GDBN}.
4026 @item set breakpoint condition-evaluation auto
4027 This is the default mode. If the target supports evaluating breakpoint
4028 conditions on its end, @value{GDBN} will download breakpoint conditions to
4029 the target (limitations mentioned previously apply). If the target does
4030 not support breakpoint condition evaluation, then @value{GDBN} will fallback
4031 to evaluating all these conditions on the host's side.
4035 @cindex negative breakpoint numbers
4036 @cindex internal @value{GDBN} breakpoints
4037 @value{GDBN} itself sometimes sets breakpoints in your program for
4038 special purposes, such as proper handling of @code{longjmp} (in C
4039 programs). These internal breakpoints are assigned negative numbers,
4040 starting with @code{-1}; @samp{info breakpoints} does not display them.
4041 You can see these breakpoints with the @value{GDBN} maintenance command
4042 @samp{maint info breakpoints} (@pxref{maint info breakpoints}).
4045 @node Set Watchpoints
4046 @subsection Setting Watchpoints
4048 @cindex setting watchpoints
4049 You can use a watchpoint to stop execution whenever the value of an
4050 expression changes, without having to predict a particular place where
4051 this may happen. (This is sometimes called a @dfn{data breakpoint}.)
4052 The expression may be as simple as the value of a single variable, or
4053 as complex as many variables combined by operators. Examples include:
4057 A reference to the value of a single variable.
4060 An address cast to an appropriate data type. For example,
4061 @samp{*(int *)0x12345678} will watch a 4-byte region at the specified
4062 address (assuming an @code{int} occupies 4 bytes).
4065 An arbitrarily complex expression, such as @samp{a*b + c/d}. The
4066 expression can use any operators valid in the program's native
4067 language (@pxref{Languages}).
4070 You can set a watchpoint on an expression even if the expression can
4071 not be evaluated yet. For instance, you can set a watchpoint on
4072 @samp{*global_ptr} before @samp{global_ptr} is initialized.
4073 @value{GDBN} will stop when your program sets @samp{global_ptr} and
4074 the expression produces a valid value. If the expression becomes
4075 valid in some other way than changing a variable (e.g.@: if the memory
4076 pointed to by @samp{*global_ptr} becomes readable as the result of a
4077 @code{malloc} call), @value{GDBN} may not stop until the next time
4078 the expression changes.
4080 @cindex software watchpoints
4081 @cindex hardware watchpoints
4082 Depending on your system, watchpoints may be implemented in software or
4083 hardware. @value{GDBN} does software watchpointing by single-stepping your
4084 program and testing the variable's value each time, which is hundreds of
4085 times slower than normal execution. (But this may still be worth it, to
4086 catch errors where you have no clue what part of your program is the
4089 On some systems, such as most PowerPC or x86-based targets,
4090 @value{GDBN} includes support for hardware watchpoints, which do not
4091 slow down the running of your program.
4095 @item watch @r{[}-l@r{|}-location@r{]} @var{expr} @r{[}thread @var{thread-id}@r{]} @r{[}mask @var{maskvalue}@r{]}
4096 Set a watchpoint for an expression. @value{GDBN} will break when the
4097 expression @var{expr} is written into by the program and its value
4098 changes. The simplest (and the most popular) use of this command is
4099 to watch the value of a single variable:
4102 (@value{GDBP}) watch foo
4105 If the command includes a @code{@r{[}thread @var{thread-id}@r{]}}
4106 argument, @value{GDBN} breaks only when the thread identified by
4107 @var{thread-id} changes the value of @var{expr}. If any other threads
4108 change the value of @var{expr}, @value{GDBN} will not break. Note
4109 that watchpoints restricted to a single thread in this way only work
4110 with Hardware Watchpoints.
4112 Ordinarily a watchpoint respects the scope of variables in @var{expr}
4113 (see below). The @code{-location} argument tells @value{GDBN} to
4114 instead watch the memory referred to by @var{expr}. In this case,
4115 @value{GDBN} will evaluate @var{expr}, take the address of the result,
4116 and watch the memory at that address. The type of the result is used
4117 to determine the size of the watched memory. If the expression's
4118 result does not have an address, then @value{GDBN} will print an
4121 The @code{@r{[}mask @var{maskvalue}@r{]}} argument allows creation
4122 of masked watchpoints, if the current architecture supports this
4123 feature (e.g., PowerPC Embedded architecture, see @ref{PowerPC
4124 Embedded}.) A @dfn{masked watchpoint} specifies a mask in addition
4125 to an address to watch. The mask specifies that some bits of an address
4126 (the bits which are reset in the mask) should be ignored when matching
4127 the address accessed by the inferior against the watchpoint address.
4128 Thus, a masked watchpoint watches many addresses simultaneously---those
4129 addresses whose unmasked bits are identical to the unmasked bits in the
4130 watchpoint address. The @code{mask} argument implies @code{-location}.
4134 (@value{GDBP}) watch foo mask 0xffff00ff
4135 (@value{GDBP}) watch *0xdeadbeef mask 0xffffff00
4139 @item rwatch @r{[}-l@r{|}-location@r{]} @var{expr} @r{[}thread @var{thread-id}@r{]} @r{[}mask @var{maskvalue}@r{]}
4140 Set a watchpoint that will break when the value of @var{expr} is read
4144 @item awatch @r{[}-l@r{|}-location@r{]} @var{expr} @r{[}thread @var{thread-id}@r{]} @r{[}mask @var{maskvalue}@r{]}
4145 Set a watchpoint that will break when @var{expr} is either read from
4146 or written into by the program.
4148 @kindex info watchpoints @r{[}@var{list}@dots{}@r{]}
4149 @item info watchpoints @r{[}@var{list}@dots{}@r{]}
4150 This command prints a list of watchpoints, using the same format as
4151 @code{info break} (@pxref{Set Breaks}).
4154 If you watch for a change in a numerically entered address you need to
4155 dereference it, as the address itself is just a constant number which will
4156 never change. @value{GDBN} refuses to create a watchpoint that watches
4157 a never-changing value:
4160 (@value{GDBP}) watch 0x600850
4161 Cannot watch constant value 0x600850.
4162 (@value{GDBP}) watch *(int *) 0x600850
4163 Watchpoint 1: *(int *) 6293584
4166 @value{GDBN} sets a @dfn{hardware watchpoint} if possible. Hardware
4167 watchpoints execute very quickly, and the debugger reports a change in
4168 value at the exact instruction where the change occurs. If @value{GDBN}
4169 cannot set a hardware watchpoint, it sets a software watchpoint, which
4170 executes more slowly and reports the change in value at the next
4171 @emph{statement}, not the instruction, after the change occurs.
4173 @cindex use only software watchpoints
4174 You can force @value{GDBN} to use only software watchpoints with the
4175 @kbd{set can-use-hw-watchpoints 0} command. With this variable set to
4176 zero, @value{GDBN} will never try to use hardware watchpoints, even if
4177 the underlying system supports them. (Note that hardware-assisted
4178 watchpoints that were set @emph{before} setting
4179 @code{can-use-hw-watchpoints} to zero will still use the hardware
4180 mechanism of watching expression values.)
4183 @item set can-use-hw-watchpoints
4184 @kindex set can-use-hw-watchpoints
4185 Set whether or not to use hardware watchpoints.
4187 @item show can-use-hw-watchpoints
4188 @kindex show can-use-hw-watchpoints
4189 Show the current mode of using hardware watchpoints.
4192 For remote targets, you can restrict the number of hardware
4193 watchpoints @value{GDBN} will use, see @ref{set remote
4194 hardware-breakpoint-limit}.
4196 When you issue the @code{watch} command, @value{GDBN} reports
4199 Hardware watchpoint @var{num}: @var{expr}
4203 if it was able to set a hardware watchpoint.
4205 Currently, the @code{awatch} and @code{rwatch} commands can only set
4206 hardware watchpoints, because accesses to data that don't change the
4207 value of the watched expression cannot be detected without examining
4208 every instruction as it is being executed, and @value{GDBN} does not do
4209 that currently. If @value{GDBN} finds that it is unable to set a
4210 hardware breakpoint with the @code{awatch} or @code{rwatch} command, it
4211 will print a message like this:
4214 Expression cannot be implemented with read/access watchpoint.
4217 Sometimes, @value{GDBN} cannot set a hardware watchpoint because the
4218 data type of the watched expression is wider than what a hardware
4219 watchpoint on the target machine can handle. For example, some systems
4220 can only watch regions that are up to 4 bytes wide; on such systems you
4221 cannot set hardware watchpoints for an expression that yields a
4222 double-precision floating-point number (which is typically 8 bytes
4223 wide). As a work-around, it might be possible to break the large region
4224 into a series of smaller ones and watch them with separate watchpoints.
4226 If you set too many hardware watchpoints, @value{GDBN} might be unable
4227 to insert all of them when you resume the execution of your program.
4228 Since the precise number of active watchpoints is unknown until such
4229 time as the program is about to be resumed, @value{GDBN} might not be
4230 able to warn you about this when you set the watchpoints, and the
4231 warning will be printed only when the program is resumed:
4234 Hardware watchpoint @var{num}: Could not insert watchpoint
4238 If this happens, delete or disable some of the watchpoints.
4240 Watching complex expressions that reference many variables can also
4241 exhaust the resources available for hardware-assisted watchpoints.
4242 That's because @value{GDBN} needs to watch every variable in the
4243 expression with separately allocated resources.
4245 If you call a function interactively using @code{print} or @code{call},
4246 any watchpoints you have set will be inactive until @value{GDBN} reaches another
4247 kind of breakpoint or the call completes.
4249 @value{GDBN} automatically deletes watchpoints that watch local
4250 (automatic) variables, or expressions that involve such variables, when
4251 they go out of scope, that is, when the execution leaves the block in
4252 which these variables were defined. In particular, when the program
4253 being debugged terminates, @emph{all} local variables go out of scope,
4254 and so only watchpoints that watch global variables remain set. If you
4255 rerun the program, you will need to set all such watchpoints again. One
4256 way of doing that would be to set a code breakpoint at the entry to the
4257 @code{main} function and when it breaks, set all the watchpoints.
4259 @cindex watchpoints and threads
4260 @cindex threads and watchpoints
4261 In multi-threaded programs, watchpoints will detect changes to the
4262 watched expression from every thread.
4265 @emph{Warning:} In multi-threaded programs, software watchpoints
4266 have only limited usefulness. If @value{GDBN} creates a software
4267 watchpoint, it can only watch the value of an expression @emph{in a
4268 single thread}. If you are confident that the expression can only
4269 change due to the current thread's activity (and if you are also
4270 confident that no other thread can become current), then you can use
4271 software watchpoints as usual. However, @value{GDBN} may not notice
4272 when a non-current thread's activity changes the expression. (Hardware
4273 watchpoints, in contrast, watch an expression in all threads.)
4276 @xref{set remote hardware-watchpoint-limit}.
4278 @node Set Catchpoints
4279 @subsection Setting Catchpoints
4280 @cindex catchpoints, setting
4281 @cindex exception handlers
4282 @cindex event handling
4284 You can use @dfn{catchpoints} to cause the debugger to stop for certain
4285 kinds of program events, such as C@t{++} exceptions or the loading of a
4286 shared library. Use the @code{catch} command to set a catchpoint.
4290 @item catch @var{event}
4291 Stop when @var{event} occurs. The @var{event} can be any of the following:
4294 @item throw @r{[}@var{regexp}@r{]}
4295 @itemx rethrow @r{[}@var{regexp}@r{]}
4296 @itemx catch @r{[}@var{regexp}@r{]}
4298 @kindex catch rethrow
4300 @cindex stop on C@t{++} exceptions
4301 The throwing, re-throwing, or catching of a C@t{++} exception.
4303 If @var{regexp} is given, then only exceptions whose type matches the
4304 regular expression will be caught.
4306 @vindex $_exception@r{, convenience variable}
4307 The convenience variable @code{$_exception} is available at an
4308 exception-related catchpoint, on some systems. This holds the
4309 exception being thrown.
4311 There are currently some limitations to C@t{++} exception handling in
4316 The support for these commands is system-dependent. Currently, only
4317 systems using the @samp{gnu-v3} C@t{++} ABI (@pxref{ABI}) are
4321 The regular expression feature and the @code{$_exception} convenience
4322 variable rely on the presence of some SDT probes in @code{libstdc++}.
4323 If these probes are not present, then these features cannot be used.
4324 These probes were first available in the GCC 4.8 release, but whether
4325 or not they are available in your GCC also depends on how it was
4329 The @code{$_exception} convenience variable is only valid at the
4330 instruction at which an exception-related catchpoint is set.
4333 When an exception-related catchpoint is hit, @value{GDBN} stops at a
4334 location in the system library which implements runtime exception
4335 support for C@t{++}, usually @code{libstdc++}. You can use @code{up}
4336 (@pxref{Selection}) to get to your code.
4339 If you call a function interactively, @value{GDBN} normally returns
4340 control to you when the function has finished executing. If the call
4341 raises an exception, however, the call may bypass the mechanism that
4342 returns control to you and cause your program either to abort or to
4343 simply continue running until it hits a breakpoint, catches a signal
4344 that @value{GDBN} is listening for, or exits. This is the case even if
4345 you set a catchpoint for the exception; catchpoints on exceptions are
4346 disabled within interactive calls. @xref{Calling}, for information on
4347 controlling this with @code{set unwind-on-terminating-exception}.
4350 You cannot raise an exception interactively.
4353 You cannot install an exception handler interactively.
4357 @kindex catch exception
4358 @cindex Ada exception catching
4359 @cindex catch Ada exceptions
4360 An Ada exception being raised. If an exception name is specified
4361 at the end of the command (eg @code{catch exception Program_Error}),
4362 the debugger will stop only when this specific exception is raised.
4363 Otherwise, the debugger stops execution when any Ada exception is raised.
4365 When inserting an exception catchpoint on a user-defined exception whose
4366 name is identical to one of the exceptions defined by the language, the
4367 fully qualified name must be used as the exception name. Otherwise,
4368 @value{GDBN} will assume that it should stop on the pre-defined exception
4369 rather than the user-defined one. For instance, assuming an exception
4370 called @code{Constraint_Error} is defined in package @code{Pck}, then
4371 the command to use to catch such exceptions is @kbd{catch exception
4372 Pck.Constraint_Error}.
4374 @item exception unhandled
4375 @kindex catch exception unhandled
4376 An exception that was raised but is not handled by the program.
4379 @kindex catch assert
4380 A failed Ada assertion.
4384 @cindex break on fork/exec
4385 A call to @code{exec}.
4388 @itemx syscall @r{[}@var{name} @r{|} @var{number} @r{|} @r{group:}@var{groupname} @r{|} @r{g:}@var{groupname}@r{]} @dots{}
4389 @kindex catch syscall
4390 @cindex break on a system call.
4391 A call to or return from a system call, a.k.a.@: @dfn{syscall}. A
4392 syscall is a mechanism for application programs to request a service
4393 from the operating system (OS) or one of the OS system services.
4394 @value{GDBN} can catch some or all of the syscalls issued by the
4395 debuggee, and show the related information for each syscall. If no
4396 argument is specified, calls to and returns from all system calls
4399 @var{name} can be any system call name that is valid for the
4400 underlying OS. Just what syscalls are valid depends on the OS. On
4401 GNU and Unix systems, you can find the full list of valid syscall
4402 names on @file{/usr/include/asm/unistd.h}.
4404 @c For MS-Windows, the syscall names and the corresponding numbers
4405 @c can be found, e.g., on this URL:
4406 @c http://www.metasploit.com/users/opcode/syscalls.html
4407 @c but we don't support Windows syscalls yet.
4409 Normally, @value{GDBN} knows in advance which syscalls are valid for
4410 each OS, so you can use the @value{GDBN} command-line completion
4411 facilities (@pxref{Completion,, command completion}) to list the
4414 You may also specify the system call numerically. A syscall's
4415 number is the value passed to the OS's syscall dispatcher to
4416 identify the requested service. When you specify the syscall by its
4417 name, @value{GDBN} uses its database of syscalls to convert the name
4418 into the corresponding numeric code, but using the number directly
4419 may be useful if @value{GDBN}'s database does not have the complete
4420 list of syscalls on your system (e.g., because @value{GDBN} lags
4421 behind the OS upgrades).
4423 You may specify a group of related syscalls to be caught at once using
4424 the @code{group:} syntax (@code{g:} is a shorter equivalent). For
4425 instance, on some platforms @value{GDBN} allows you to catch all
4426 network related syscalls, by passing the argument @code{group:network}
4427 to @code{catch syscall}. Note that not all syscall groups are
4428 available in every system. You can use the command completion
4429 facilities (@pxref{Completion,, command completion}) to list the
4430 syscall groups available on your environment.
4432 The example below illustrates how this command works if you don't provide
4436 (@value{GDBP}) catch syscall
4437 Catchpoint 1 (syscall)
4439 Starting program: /tmp/catch-syscall
4441 Catchpoint 1 (call to syscall 'close'), \
4442 0xffffe424 in __kernel_vsyscall ()
4446 Catchpoint 1 (returned from syscall 'close'), \
4447 0xffffe424 in __kernel_vsyscall ()
4451 Here is an example of catching a system call by name:
4454 (@value{GDBP}) catch syscall chroot
4455 Catchpoint 1 (syscall 'chroot' [61])
4457 Starting program: /tmp/catch-syscall
4459 Catchpoint 1 (call to syscall 'chroot'), \
4460 0xffffe424 in __kernel_vsyscall ()
4464 Catchpoint 1 (returned from syscall 'chroot'), \
4465 0xffffe424 in __kernel_vsyscall ()
4469 An example of specifying a system call numerically. In the case
4470 below, the syscall number has a corresponding entry in the XML
4471 file, so @value{GDBN} finds its name and prints it:
4474 (@value{GDBP}) catch syscall 252
4475 Catchpoint 1 (syscall(s) 'exit_group')
4477 Starting program: /tmp/catch-syscall
4479 Catchpoint 1 (call to syscall 'exit_group'), \
4480 0xffffe424 in __kernel_vsyscall ()
4484 Program exited normally.
4488 Here is an example of catching a syscall group:
4491 (@value{GDBP}) catch syscall group:process
4492 Catchpoint 1 (syscalls 'exit' [1] 'fork' [2] 'waitpid' [7]
4493 'execve' [11] 'wait4' [114] 'clone' [120] 'vfork' [190]
4494 'exit_group' [252] 'waitid' [284] 'unshare' [310])
4496 Starting program: /tmp/catch-syscall
4498 Catchpoint 1 (call to syscall fork), 0x00007ffff7df4e27 in open64 ()
4499 from /lib64/ld-linux-x86-64.so.2
4505 However, there can be situations when there is no corresponding name
4506 in XML file for that syscall number. In this case, @value{GDBN} prints
4507 a warning message saying that it was not able to find the syscall name,
4508 but the catchpoint will be set anyway. See the example below:
4511 (@value{GDBP}) catch syscall 764
4512 warning: The number '764' does not represent a known syscall.
4513 Catchpoint 2 (syscall 764)
4517 If you configure @value{GDBN} using the @samp{--without-expat} option,
4518 it will not be able to display syscall names. Also, if your
4519 architecture does not have an XML file describing its system calls,
4520 you will not be able to see the syscall names. It is important to
4521 notice that these two features are used for accessing the syscall
4522 name database. In either case, you will see a warning like this:
4525 (@value{GDBP}) catch syscall
4526 warning: Could not open "syscalls/i386-linux.xml"
4527 warning: Could not load the syscall XML file 'syscalls/i386-linux.xml'.
4528 GDB will not be able to display syscall names.
4529 Catchpoint 1 (syscall)
4533 Of course, the file name will change depending on your architecture and system.
4535 Still using the example above, you can also try to catch a syscall by its
4536 number. In this case, you would see something like:
4539 (@value{GDBP}) catch syscall 252
4540 Catchpoint 1 (syscall(s) 252)
4543 Again, in this case @value{GDBN} would not be able to display syscall's names.
4547 A call to @code{fork}.
4551 A call to @code{vfork}.
4553 @item load @r{[}regexp@r{]}
4554 @itemx unload @r{[}regexp@r{]}
4556 @kindex catch unload
4557 The loading or unloading of a shared library. If @var{regexp} is
4558 given, then the catchpoint will stop only if the regular expression
4559 matches one of the affected libraries.
4561 @item signal @r{[}@var{signal}@dots{} @r{|} @samp{all}@r{]}
4562 @kindex catch signal
4563 The delivery of a signal.
4565 With no arguments, this catchpoint will catch any signal that is not
4566 used internally by @value{GDBN}, specifically, all signals except
4567 @samp{SIGTRAP} and @samp{SIGINT}.
4569 With the argument @samp{all}, all signals, including those used by
4570 @value{GDBN}, will be caught. This argument cannot be used with other
4573 Otherwise, the arguments are a list of signal names as given to
4574 @code{handle} (@pxref{Signals}). Only signals specified in this list
4577 One reason that @code{catch signal} can be more useful than
4578 @code{handle} is that you can attach commands and conditions to the
4581 When a signal is caught by a catchpoint, the signal's @code{stop} and
4582 @code{print} settings, as specified by @code{handle}, are ignored.
4583 However, whether the signal is still delivered to the inferior depends
4584 on the @code{pass} setting; this can be changed in the catchpoint's
4589 @item tcatch @var{event}
4591 Set a catchpoint that is enabled only for one stop. The catchpoint is
4592 automatically deleted after the first time the event is caught.
4596 Use the @code{info break} command to list the current catchpoints.
4600 @subsection Deleting Breakpoints
4602 @cindex clearing breakpoints, watchpoints, catchpoints
4603 @cindex deleting breakpoints, watchpoints, catchpoints
4604 It is often necessary to eliminate a breakpoint, watchpoint, or
4605 catchpoint once it has done its job and you no longer want your program
4606 to stop there. This is called @dfn{deleting} the breakpoint. A
4607 breakpoint that has been deleted no longer exists; it is forgotten.
4609 With the @code{clear} command you can delete breakpoints according to
4610 where they are in your program. With the @code{delete} command you can
4611 delete individual breakpoints, watchpoints, or catchpoints by specifying
4612 their breakpoint numbers.
4614 It is not necessary to delete a breakpoint to proceed past it. @value{GDBN}
4615 automatically ignores breakpoints on the first instruction to be executed
4616 when you continue execution without changing the execution address.
4621 Delete any breakpoints at the next instruction to be executed in the
4622 selected stack frame (@pxref{Selection, ,Selecting a Frame}). When
4623 the innermost frame is selected, this is a good way to delete a
4624 breakpoint where your program just stopped.
4626 @item clear @var{location}
4627 Delete any breakpoints set at the specified @var{location}.
4628 @xref{Specify Location}, for the various forms of @var{location}; the
4629 most useful ones are listed below:
4632 @item clear @var{function}
4633 @itemx clear @var{filename}:@var{function}
4634 Delete any breakpoints set at entry to the named @var{function}.
4636 @item clear @var{linenum}
4637 @itemx clear @var{filename}:@var{linenum}
4638 Delete any breakpoints set at or within the code of the specified
4639 @var{linenum} of the specified @var{filename}.
4642 @cindex delete breakpoints
4644 @kindex d @r{(@code{delete})}
4645 @item delete @r{[}breakpoints@r{]} @r{[}@var{list}@dots{}@r{]}
4646 Delete the breakpoints, watchpoints, or catchpoints of the breakpoint
4647 list specified as argument. If no argument is specified, delete all
4648 breakpoints (@value{GDBN} asks confirmation, unless you have @code{set
4649 confirm off}). You can abbreviate this command as @code{d}.
4653 @subsection Disabling Breakpoints
4655 @cindex enable/disable a breakpoint
4656 Rather than deleting a breakpoint, watchpoint, or catchpoint, you might
4657 prefer to @dfn{disable} it. This makes the breakpoint inoperative as if
4658 it had been deleted, but remembers the information on the breakpoint so
4659 that you can @dfn{enable} it again later.
4661 You disable and enable breakpoints, watchpoints, and catchpoints with
4662 the @code{enable} and @code{disable} commands, optionally specifying
4663 one or more breakpoint numbers as arguments. Use @code{info break} to
4664 print a list of all breakpoints, watchpoints, and catchpoints if you
4665 do not know which numbers to use.
4667 Disabling and enabling a breakpoint that has multiple locations
4668 affects all of its locations.
4670 A breakpoint, watchpoint, or catchpoint can have any of several
4671 different states of enablement:
4675 Enabled. The breakpoint stops your program. A breakpoint set
4676 with the @code{break} command starts out in this state.
4678 Disabled. The breakpoint has no effect on your program.
4680 Enabled once. The breakpoint stops your program, but then becomes
4683 Enabled for a count. The breakpoint stops your program for the next
4684 N times, then becomes disabled.
4686 Enabled for deletion. The breakpoint stops your program, but
4687 immediately after it does so it is deleted permanently. A breakpoint
4688 set with the @code{tbreak} command starts out in this state.
4691 You can use the following commands to enable or disable breakpoints,
4692 watchpoints, and catchpoints:
4696 @kindex dis @r{(@code{disable})}
4697 @item disable @r{[}breakpoints@r{]} @r{[}@var{list}@dots{}@r{]}
4698 Disable the specified breakpoints---or all breakpoints, if none are
4699 listed. A disabled breakpoint has no effect but is not forgotten. All
4700 options such as ignore-counts, conditions and commands are remembered in
4701 case the breakpoint is enabled again later. You may abbreviate
4702 @code{disable} as @code{dis}.
4705 @item enable @r{[}breakpoints@r{]} @r{[}@var{list}@dots{}@r{]}
4706 Enable the specified breakpoints (or all defined breakpoints). They
4707 become effective once again in stopping your program.
4709 @item enable @r{[}breakpoints@r{]} once @var{list}@dots{}
4710 Enable the specified breakpoints temporarily. @value{GDBN} disables any
4711 of these breakpoints immediately after stopping your program.
4713 @item enable @r{[}breakpoints@r{]} count @var{count} @var{list}@dots{}
4714 Enable the specified breakpoints temporarily. @value{GDBN} records
4715 @var{count} with each of the specified breakpoints, and decrements a
4716 breakpoint's count when it is hit. When any count reaches 0,
4717 @value{GDBN} disables that breakpoint. If a breakpoint has an ignore
4718 count (@pxref{Conditions, ,Break Conditions}), that will be
4719 decremented to 0 before @var{count} is affected.
4721 @item enable @r{[}breakpoints@r{]} delete @var{list}@dots{}
4722 Enable the specified breakpoints to work once, then die. @value{GDBN}
4723 deletes any of these breakpoints as soon as your program stops there.
4724 Breakpoints set by the @code{tbreak} command start out in this state.
4727 @c FIXME: I think the following ``Except for [...] @code{tbreak}'' is
4728 @c confusing: tbreak is also initially enabled.
4729 Except for a breakpoint set with @code{tbreak} (@pxref{Set Breaks,
4730 ,Setting Breakpoints}), breakpoints that you set are initially enabled;
4731 subsequently, they become disabled or enabled only when you use one of
4732 the commands above. (The command @code{until} can set and delete a
4733 breakpoint of its own, but it does not change the state of your other
4734 breakpoints; see @ref{Continuing and Stepping, ,Continuing and
4738 @subsection Break Conditions
4739 @cindex conditional breakpoints
4740 @cindex breakpoint conditions
4742 @c FIXME what is scope of break condition expr? Context where wanted?
4743 @c in particular for a watchpoint?
4744 The simplest sort of breakpoint breaks every time your program reaches a
4745 specified place. You can also specify a @dfn{condition} for a
4746 breakpoint. A condition is just a Boolean expression in your
4747 programming language (@pxref{Expressions, ,Expressions}). A breakpoint with
4748 a condition evaluates the expression each time your program reaches it,
4749 and your program stops only if the condition is @emph{true}.
4751 This is the converse of using assertions for program validation; in that
4752 situation, you want to stop when the assertion is violated---that is,
4753 when the condition is false. In C, if you want to test an assertion expressed
4754 by the condition @var{assert}, you should set the condition
4755 @samp{! @var{assert}} on the appropriate breakpoint.
4757 Conditions are also accepted for watchpoints; you may not need them,
4758 since a watchpoint is inspecting the value of an expression anyhow---but
4759 it might be simpler, say, to just set a watchpoint on a variable name,
4760 and specify a condition that tests whether the new value is an interesting
4763 Break conditions can have side effects, and may even call functions in
4764 your program. This can be useful, for example, to activate functions
4765 that log program progress, or to use your own print functions to
4766 format special data structures. The effects are completely predictable
4767 unless there is another enabled breakpoint at the same address. (In
4768 that case, @value{GDBN} might see the other breakpoint first and stop your
4769 program without checking the condition of this one.) Note that
4770 breakpoint commands are usually more convenient and flexible than break
4772 purpose of performing side effects when a breakpoint is reached
4773 (@pxref{Break Commands, ,Breakpoint Command Lists}).
4775 Breakpoint conditions can also be evaluated on the target's side if
4776 the target supports it. Instead of evaluating the conditions locally,
4777 @value{GDBN} encodes the expression into an agent expression
4778 (@pxref{Agent Expressions}) suitable for execution on the target,
4779 independently of @value{GDBN}. Global variables become raw memory
4780 locations, locals become stack accesses, and so forth.
4782 In this case, @value{GDBN} will only be notified of a breakpoint trigger
4783 when its condition evaluates to true. This mechanism may provide faster
4784 response times depending on the performance characteristics of the target
4785 since it does not need to keep @value{GDBN} informed about
4786 every breakpoint trigger, even those with false conditions.
4788 Break conditions can be specified when a breakpoint is set, by using
4789 @samp{if} in the arguments to the @code{break} command. @xref{Set
4790 Breaks, ,Setting Breakpoints}. They can also be changed at any time
4791 with the @code{condition} command.
4793 You can also use the @code{if} keyword with the @code{watch} command.
4794 The @code{catch} command does not recognize the @code{if} keyword;
4795 @code{condition} is the only way to impose a further condition on a
4800 @item condition @var{bnum} @var{expression}
4801 Specify @var{expression} as the break condition for breakpoint,
4802 watchpoint, or catchpoint number @var{bnum}. After you set a condition,
4803 breakpoint @var{bnum} stops your program only if the value of
4804 @var{expression} is true (nonzero, in C). When you use
4805 @code{condition}, @value{GDBN} checks @var{expression} immediately for
4806 syntactic correctness, and to determine whether symbols in it have
4807 referents in the context of your breakpoint. If @var{expression} uses
4808 symbols not referenced in the context of the breakpoint, @value{GDBN}
4809 prints an error message:
4812 No symbol "foo" in current context.
4817 not actually evaluate @var{expression} at the time the @code{condition}
4818 command (or a command that sets a breakpoint with a condition, like
4819 @code{break if @dots{}}) is given, however. @xref{Expressions, ,Expressions}.
4821 @item condition @var{bnum}
4822 Remove the condition from breakpoint number @var{bnum}. It becomes
4823 an ordinary unconditional breakpoint.
4826 @cindex ignore count (of breakpoint)
4827 A special case of a breakpoint condition is to stop only when the
4828 breakpoint has been reached a certain number of times. This is so
4829 useful that there is a special way to do it, using the @dfn{ignore
4830 count} of the breakpoint. Every breakpoint has an ignore count, which
4831 is an integer. Most of the time, the ignore count is zero, and
4832 therefore has no effect. But if your program reaches a breakpoint whose
4833 ignore count is positive, then instead of stopping, it just decrements
4834 the ignore count by one and continues. As a result, if the ignore count
4835 value is @var{n}, the breakpoint does not stop the next @var{n} times
4836 your program reaches it.
4840 @item ignore @var{bnum} @var{count}
4841 Set the ignore count of breakpoint number @var{bnum} to @var{count}.
4842 The next @var{count} times the breakpoint is reached, your program's
4843 execution does not stop; other than to decrement the ignore count, @value{GDBN}
4846 To make the breakpoint stop the next time it is reached, specify
4849 When you use @code{continue} to resume execution of your program from a
4850 breakpoint, you can specify an ignore count directly as an argument to
4851 @code{continue}, rather than using @code{ignore}. @xref{Continuing and
4852 Stepping,,Continuing and Stepping}.
4854 If a breakpoint has a positive ignore count and a condition, the
4855 condition is not checked. Once the ignore count reaches zero,
4856 @value{GDBN} resumes checking the condition.
4858 You could achieve the effect of the ignore count with a condition such
4859 as @w{@samp{$foo-- <= 0}} using a debugger convenience variable that
4860 is decremented each time. @xref{Convenience Vars, ,Convenience
4864 Ignore counts apply to breakpoints, watchpoints, and catchpoints.
4867 @node Break Commands
4868 @subsection Breakpoint Command Lists
4870 @cindex breakpoint commands
4871 You can give any breakpoint (or watchpoint or catchpoint) a series of
4872 commands to execute when your program stops due to that breakpoint. For
4873 example, you might want to print the values of certain expressions, or
4874 enable other breakpoints.
4878 @kindex end@r{ (breakpoint commands)}
4879 @item commands @r{[}@var{list}@dots{}@r{]}
4880 @itemx @dots{} @var{command-list} @dots{}
4882 Specify a list of commands for the given breakpoints. The commands
4883 themselves appear on the following lines. Type a line containing just
4884 @code{end} to terminate the commands.
4886 To remove all commands from a breakpoint, type @code{commands} and
4887 follow it immediately with @code{end}; that is, give no commands.
4889 With no argument, @code{commands} refers to the last breakpoint,
4890 watchpoint, or catchpoint set (not to the breakpoint most recently
4891 encountered). If the most recent breakpoints were set with a single
4892 command, then the @code{commands} will apply to all the breakpoints
4893 set by that command. This applies to breakpoints set by
4894 @code{rbreak}, and also applies when a single @code{break} command
4895 creates multiple breakpoints (@pxref{Ambiguous Expressions,,Ambiguous
4899 Pressing @key{RET} as a means of repeating the last @value{GDBN} command is
4900 disabled within a @var{command-list}.
4902 You can use breakpoint commands to start your program up again. Simply
4903 use the @code{continue} command, or @code{step}, or any other command
4904 that resumes execution.
4906 Any other commands in the command list, after a command that resumes
4907 execution, are ignored. This is because any time you resume execution
4908 (even with a simple @code{next} or @code{step}), you may encounter
4909 another breakpoint---which could have its own command list, leading to
4910 ambiguities about which list to execute.
4913 If the first command you specify in a command list is @code{silent}, the
4914 usual message about stopping at a breakpoint is not printed. This may
4915 be desirable for breakpoints that are to print a specific message and
4916 then continue. If none of the remaining commands print anything, you
4917 see no sign that the breakpoint was reached. @code{silent} is
4918 meaningful only at the beginning of a breakpoint command list.
4920 The commands @code{echo}, @code{output}, and @code{printf} allow you to
4921 print precisely controlled output, and are often useful in silent
4922 breakpoints. @xref{Output, ,Commands for Controlled Output}.
4924 For example, here is how you could use breakpoint commands to print the
4925 value of @code{x} at entry to @code{foo} whenever @code{x} is positive.
4931 printf "x is %d\n",x
4936 One application for breakpoint commands is to compensate for one bug so
4937 you can test for another. Put a breakpoint just after the erroneous line
4938 of code, give it a condition to detect the case in which something
4939 erroneous has been done, and give it commands to assign correct values
4940 to any variables that need them. End with the @code{continue} command
4941 so that your program does not stop, and start with the @code{silent}
4942 command so that no output is produced. Here is an example:
4953 @node Dynamic Printf
4954 @subsection Dynamic Printf
4956 @cindex dynamic printf
4958 The dynamic printf command @code{dprintf} combines a breakpoint with
4959 formatted printing of your program's data to give you the effect of
4960 inserting @code{printf} calls into your program on-the-fly, without
4961 having to recompile it.
4963 In its most basic form, the output goes to the GDB console. However,
4964 you can set the variable @code{dprintf-style} for alternate handling.
4965 For instance, you can ask to format the output by calling your
4966 program's @code{printf} function. This has the advantage that the
4967 characters go to the program's output device, so they can recorded in
4968 redirects to files and so forth.
4970 If you are doing remote debugging with a stub or agent, you can also
4971 ask to have the printf handled by the remote agent. In addition to
4972 ensuring that the output goes to the remote program's device along
4973 with any other output the program might produce, you can also ask that
4974 the dprintf remain active even after disconnecting from the remote
4975 target. Using the stub/agent is also more efficient, as it can do
4976 everything without needing to communicate with @value{GDBN}.
4980 @item dprintf @var{location},@var{template},@var{expression}[,@var{expression}@dots{}]
4981 Whenever execution reaches @var{location}, print the values of one or
4982 more @var{expressions} under the control of the string @var{template}.
4983 To print several values, separate them with commas.
4985 @item set dprintf-style @var{style}
4986 Set the dprintf output to be handled in one of several different
4987 styles enumerated below. A change of style affects all existing
4988 dynamic printfs immediately. (If you need individual control over the
4989 print commands, simply define normal breakpoints with
4990 explicitly-supplied command lists.)
4994 @kindex dprintf-style gdb
4995 Handle the output using the @value{GDBN} @code{printf} command.
4998 @kindex dprintf-style call
4999 Handle the output by calling a function in your program (normally
5003 @kindex dprintf-style agent
5004 Have the remote debugging agent (such as @code{gdbserver}) handle
5005 the output itself. This style is only available for agents that
5006 support running commands on the target.
5009 @item set dprintf-function @var{function}
5010 Set the function to call if the dprintf style is @code{call}. By
5011 default its value is @code{printf}. You may set it to any expression.
5012 that @value{GDBN} can evaluate to a function, as per the @code{call}
5015 @item set dprintf-channel @var{channel}
5016 Set a ``channel'' for dprintf. If set to a non-empty value,
5017 @value{GDBN} will evaluate it as an expression and pass the result as
5018 a first argument to the @code{dprintf-function}, in the manner of
5019 @code{fprintf} and similar functions. Otherwise, the dprintf format
5020 string will be the first argument, in the manner of @code{printf}.
5022 As an example, if you wanted @code{dprintf} output to go to a logfile
5023 that is a standard I/O stream assigned to the variable @code{mylog},
5024 you could do the following:
5027 (gdb) set dprintf-style call
5028 (gdb) set dprintf-function fprintf
5029 (gdb) set dprintf-channel mylog
5030 (gdb) dprintf 25,"at line 25, glob=%d\n",glob
5031 Dprintf 1 at 0x123456: file main.c, line 25.
5033 1 dprintf keep y 0x00123456 in main at main.c:25
5034 call (void) fprintf (mylog,"at line 25, glob=%d\n",glob)
5039 Note that the @code{info break} displays the dynamic printf commands
5040 as normal breakpoint commands; you can thus easily see the effect of
5041 the variable settings.
5043 @item set disconnected-dprintf on
5044 @itemx set disconnected-dprintf off
5045 @kindex set disconnected-dprintf
5046 Choose whether @code{dprintf} commands should continue to run if
5047 @value{GDBN} has disconnected from the target. This only applies
5048 if the @code{dprintf-style} is @code{agent}.
5050 @item show disconnected-dprintf off
5051 @kindex show disconnected-dprintf
5052 Show the current choice for disconnected @code{dprintf}.
5056 @value{GDBN} does not check the validity of function and channel,
5057 relying on you to supply values that are meaningful for the contexts
5058 in which they are being used. For instance, the function and channel
5059 may be the values of local variables, but if that is the case, then
5060 all enabled dynamic prints must be at locations within the scope of
5061 those locals. If evaluation fails, @value{GDBN} will report an error.
5063 @node Save Breakpoints
5064 @subsection How to save breakpoints to a file
5066 To save breakpoint definitions to a file use the @w{@code{save
5067 breakpoints}} command.
5070 @kindex save breakpoints
5071 @cindex save breakpoints to a file for future sessions
5072 @item save breakpoints [@var{filename}]
5073 This command saves all current breakpoint definitions together with
5074 their commands and ignore counts, into a file @file{@var{filename}}
5075 suitable for use in a later debugging session. This includes all
5076 types of breakpoints (breakpoints, watchpoints, catchpoints,
5077 tracepoints). To read the saved breakpoint definitions, use the
5078 @code{source} command (@pxref{Command Files}). Note that watchpoints
5079 with expressions involving local variables may fail to be recreated
5080 because it may not be possible to access the context where the
5081 watchpoint is valid anymore. Because the saved breakpoint definitions
5082 are simply a sequence of @value{GDBN} commands that recreate the
5083 breakpoints, you can edit the file in your favorite editing program,
5084 and remove the breakpoint definitions you're not interested in, or
5085 that can no longer be recreated.
5088 @node Static Probe Points
5089 @subsection Static Probe Points
5091 @cindex static probe point, SystemTap
5092 @cindex static probe point, DTrace
5093 @value{GDBN} supports @dfn{SDT} probes in the code. @acronym{SDT} stands
5094 for Statically Defined Tracing, and the probes are designed to have a tiny
5095 runtime code and data footprint, and no dynamic relocations.
5097 Currently, the following types of probes are supported on
5098 ELF-compatible systems:
5102 @item @code{SystemTap} (@uref{http://sourceware.org/systemtap/})
5103 @acronym{SDT} probes@footnote{See
5104 @uref{http://sourceware.org/systemtap/wiki/AddingUserSpaceProbingToApps}
5105 for more information on how to add @code{SystemTap} @acronym{SDT}
5106 probes in your applications.}. @code{SystemTap} probes are usable
5107 from assembly, C and C@t{++} languages@footnote{See
5108 @uref{http://sourceware.org/systemtap/wiki/UserSpaceProbeImplementation}
5109 for a good reference on how the @acronym{SDT} probes are implemented.}.
5111 @item @code{DTrace} (@uref{http://oss.oracle.com/projects/DTrace})
5112 @acronym{USDT} probes. @code{DTrace} probes are usable from C and
5116 @cindex semaphores on static probe points
5117 Some @code{SystemTap} probes have an associated semaphore variable;
5118 for instance, this happens automatically if you defined your probe
5119 using a DTrace-style @file{.d} file. If your probe has a semaphore,
5120 @value{GDBN} will automatically enable it when you specify a
5121 breakpoint using the @samp{-probe-stap} notation. But, if you put a
5122 breakpoint at a probe's location by some other method (e.g.,
5123 @code{break file:line}), then @value{GDBN} will not automatically set
5124 the semaphore. @code{DTrace} probes do not support semaphores.
5126 You can examine the available static static probes using @code{info
5127 probes}, with optional arguments:
5131 @item info probes @r{[}@var{type}@r{]} @r{[}@var{provider} @r{[}@var{name} @r{[}@var{objfile}@r{]}@r{]}@r{]}
5132 If given, @var{type} is either @code{stap} for listing
5133 @code{SystemTap} probes or @code{dtrace} for listing @code{DTrace}
5134 probes. If omitted all probes are listed regardless of their types.
5136 If given, @var{provider} is a regular expression used to match against provider
5137 names when selecting which probes to list. If omitted, probes by all
5138 probes from all providers are listed.
5140 If given, @var{name} is a regular expression to match against probe names
5141 when selecting which probes to list. If omitted, probe names are not
5142 considered when deciding whether to display them.
5144 If given, @var{objfile} is a regular expression used to select which
5145 object files (executable or shared libraries) to examine. If not
5146 given, all object files are considered.
5148 @item info probes all
5149 List the available static probes, from all types.
5152 @cindex enabling and disabling probes
5153 Some probe points can be enabled and/or disabled. The effect of
5154 enabling or disabling a probe depends on the type of probe being
5155 handled. Some @code{DTrace} probes can be enabled or
5156 disabled, but @code{SystemTap} probes cannot be disabled.
5158 You can enable (or disable) one or more probes using the following
5159 commands, with optional arguments:
5162 @kindex enable probes
5163 @item enable probes @r{[}@var{provider} @r{[}@var{name} @r{[}@var{objfile}@r{]}@r{]}@r{]}
5164 If given, @var{provider} is a regular expression used to match against
5165 provider names when selecting which probes to enable. If omitted,
5166 all probes from all providers are enabled.
5168 If given, @var{name} is a regular expression to match against probe
5169 names when selecting which probes to enable. If omitted, probe names
5170 are not considered when deciding whether to enable them.
5172 If given, @var{objfile} is a regular expression used to select which
5173 object files (executable or shared libraries) to examine. If not
5174 given, all object files are considered.
5176 @kindex disable probes
5177 @item disable probes @r{[}@var{provider} @r{[}@var{name} @r{[}@var{objfile}@r{]}@r{]}@r{]}
5178 See the @code{enable probes} command above for a description of the
5179 optional arguments accepted by this command.
5182 @vindex $_probe_arg@r{, convenience variable}
5183 A probe may specify up to twelve arguments. These are available at the
5184 point at which the probe is defined---that is, when the current PC is
5185 at the probe's location. The arguments are available using the
5186 convenience variables (@pxref{Convenience Vars})
5187 @code{$_probe_arg0}@dots{}@code{$_probe_arg11}. In @code{SystemTap}
5188 probes each probe argument is an integer of the appropriate size;
5189 types are not preserved. In @code{DTrace} probes types are preserved
5190 provided that they are recognized as such by @value{GDBN}; otherwise
5191 the value of the probe argument will be a long integer. The
5192 convenience variable @code{$_probe_argc} holds the number of arguments
5193 at the current probe point.
5195 These variables are always available, but attempts to access them at
5196 any location other than a probe point will cause @value{GDBN} to give
5200 @c @ifclear BARETARGET
5201 @node Error in Breakpoints
5202 @subsection ``Cannot insert breakpoints''
5204 If you request too many active hardware-assisted breakpoints and
5205 watchpoints, you will see this error message:
5207 @c FIXME: the precise wording of this message may change; the relevant
5208 @c source change is not committed yet (Sep 3, 1999).
5210 Stopped; cannot insert breakpoints.
5211 You may have requested too many hardware breakpoints and watchpoints.
5215 This message is printed when you attempt to resume the program, since
5216 only then @value{GDBN} knows exactly how many hardware breakpoints and
5217 watchpoints it needs to insert.
5219 When this message is printed, you need to disable or remove some of the
5220 hardware-assisted breakpoints and watchpoints, and then continue.
5222 @node Breakpoint-related Warnings
5223 @subsection ``Breakpoint address adjusted...''
5224 @cindex breakpoint address adjusted
5226 Some processor architectures place constraints on the addresses at
5227 which breakpoints may be placed. For architectures thus constrained,
5228 @value{GDBN} will attempt to adjust the breakpoint's address to comply
5229 with the constraints dictated by the architecture.
5231 One example of such an architecture is the Fujitsu FR-V. The FR-V is
5232 a VLIW architecture in which a number of RISC-like instructions may be
5233 bundled together for parallel execution. The FR-V architecture
5234 constrains the location of a breakpoint instruction within such a
5235 bundle to the instruction with the lowest address. @value{GDBN}
5236 honors this constraint by adjusting a breakpoint's address to the
5237 first in the bundle.
5239 It is not uncommon for optimized code to have bundles which contain
5240 instructions from different source statements, thus it may happen that
5241 a breakpoint's address will be adjusted from one source statement to
5242 another. Since this adjustment may significantly alter @value{GDBN}'s
5243 breakpoint related behavior from what the user expects, a warning is
5244 printed when the breakpoint is first set and also when the breakpoint
5247 A warning like the one below is printed when setting a breakpoint
5248 that's been subject to address adjustment:
5251 warning: Breakpoint address adjusted from 0x00010414 to 0x00010410.
5254 Such warnings are printed both for user settable and @value{GDBN}'s
5255 internal breakpoints. If you see one of these warnings, you should
5256 verify that a breakpoint set at the adjusted address will have the
5257 desired affect. If not, the breakpoint in question may be removed and
5258 other breakpoints may be set which will have the desired behavior.
5259 E.g., it may be sufficient to place the breakpoint at a later
5260 instruction. A conditional breakpoint may also be useful in some
5261 cases to prevent the breakpoint from triggering too often.
5263 @value{GDBN} will also issue a warning when stopping at one of these
5264 adjusted breakpoints:
5267 warning: Breakpoint 1 address previously adjusted from 0x00010414
5271 When this warning is encountered, it may be too late to take remedial
5272 action except in cases where the breakpoint is hit earlier or more
5273 frequently than expected.
5275 @node Continuing and Stepping
5276 @section Continuing and Stepping
5280 @cindex resuming execution
5281 @dfn{Continuing} means resuming program execution until your program
5282 completes normally. In contrast, @dfn{stepping} means executing just
5283 one more ``step'' of your program, where ``step'' may mean either one
5284 line of source code, or one machine instruction (depending on what
5285 particular command you use). Either when continuing or when stepping,
5286 your program may stop even sooner, due to a breakpoint or a signal. (If
5287 it stops due to a signal, you may want to use @code{handle}, or use
5288 @samp{signal 0} to resume execution (@pxref{Signals, ,Signals}),
5289 or you may step into the signal's handler (@pxref{stepping and signal
5294 @kindex c @r{(@code{continue})}
5295 @kindex fg @r{(resume foreground execution)}
5296 @item continue @r{[}@var{ignore-count}@r{]}
5297 @itemx c @r{[}@var{ignore-count}@r{]}
5298 @itemx fg @r{[}@var{ignore-count}@r{]}
5299 Resume program execution, at the address where your program last stopped;
5300 any breakpoints set at that address are bypassed. The optional argument
5301 @var{ignore-count} allows you to specify a further number of times to
5302 ignore a breakpoint at this location; its effect is like that of
5303 @code{ignore} (@pxref{Conditions, ,Break Conditions}).
5305 The argument @var{ignore-count} is meaningful only when your program
5306 stopped due to a breakpoint. At other times, the argument to
5307 @code{continue} is ignored.
5309 The synonyms @code{c} and @code{fg} (for @dfn{foreground}, as the
5310 debugged program is deemed to be the foreground program) are provided
5311 purely for convenience, and have exactly the same behavior as
5315 To resume execution at a different place, you can use @code{return}
5316 (@pxref{Returning, ,Returning from a Function}) to go back to the
5317 calling function; or @code{jump} (@pxref{Jumping, ,Continuing at a
5318 Different Address}) to go to an arbitrary location in your program.
5320 A typical technique for using stepping is to set a breakpoint
5321 (@pxref{Breakpoints, ,Breakpoints; Watchpoints; and Catchpoints}) at the
5322 beginning of the function or the section of your program where a problem
5323 is believed to lie, run your program until it stops at that breakpoint,
5324 and then step through the suspect area, examining the variables that are
5325 interesting, until you see the problem happen.
5329 @kindex s @r{(@code{step})}
5331 Continue running your program until control reaches a different source
5332 line, then stop it and return control to @value{GDBN}. This command is
5333 abbreviated @code{s}.
5336 @c "without debugging information" is imprecise; actually "without line
5337 @c numbers in the debugging information". (gcc -g1 has debugging info but
5338 @c not line numbers). But it seems complex to try to make that
5339 @c distinction here.
5340 @emph{Warning:} If you use the @code{step} command while control is
5341 within a function that was compiled without debugging information,
5342 execution proceeds until control reaches a function that does have
5343 debugging information. Likewise, it will not step into a function which
5344 is compiled without debugging information. To step through functions
5345 without debugging information, use the @code{stepi} command, described
5349 The @code{step} command only stops at the first instruction of a source
5350 line. This prevents the multiple stops that could otherwise occur in
5351 @code{switch} statements, @code{for} loops, etc. @code{step} continues
5352 to stop if a function that has debugging information is called within
5353 the line. In other words, @code{step} @emph{steps inside} any functions
5354 called within the line.
5356 Also, the @code{step} command only enters a function if there is line
5357 number information for the function. Otherwise it acts like the
5358 @code{next} command. This avoids problems when using @code{cc -gl}
5359 on @acronym{MIPS} machines. Previously, @code{step} entered subroutines if there
5360 was any debugging information about the routine.
5362 @item step @var{count}
5363 Continue running as in @code{step}, but do so @var{count} times. If a
5364 breakpoint is reached, or a signal not related to stepping occurs before
5365 @var{count} steps, stepping stops right away.
5368 @kindex n @r{(@code{next})}
5369 @item next @r{[}@var{count}@r{]}
5370 Continue to the next source line in the current (innermost) stack frame.
5371 This is similar to @code{step}, but function calls that appear within
5372 the line of code are executed without stopping. Execution stops when
5373 control reaches a different line of code at the original stack level
5374 that was executing when you gave the @code{next} command. This command
5375 is abbreviated @code{n}.
5377 An argument @var{count} is a repeat count, as for @code{step}.
5380 @c FIX ME!! Do we delete this, or is there a way it fits in with
5381 @c the following paragraph? --- Vctoria
5383 @c @code{next} within a function that lacks debugging information acts like
5384 @c @code{step}, but any function calls appearing within the code of the
5385 @c function are executed without stopping.
5387 The @code{next} command only stops at the first instruction of a
5388 source line. This prevents multiple stops that could otherwise occur in
5389 @code{switch} statements, @code{for} loops, etc.
5391 @kindex set step-mode
5393 @cindex functions without line info, and stepping
5394 @cindex stepping into functions with no line info
5395 @itemx set step-mode on
5396 The @code{set step-mode on} command causes the @code{step} command to
5397 stop at the first instruction of a function which contains no debug line
5398 information rather than stepping over it.
5400 This is useful in cases where you may be interested in inspecting the
5401 machine instructions of a function which has no symbolic info and do not
5402 want @value{GDBN} to automatically skip over this function.
5404 @item set step-mode off
5405 Causes the @code{step} command to step over any functions which contains no
5406 debug information. This is the default.
5408 @item show step-mode
5409 Show whether @value{GDBN} will stop in or step over functions without
5410 source line debug information.
5413 @kindex fin @r{(@code{finish})}
5415 Continue running until just after function in the selected stack frame
5416 returns. Print the returned value (if any). This command can be
5417 abbreviated as @code{fin}.
5419 Contrast this with the @code{return} command (@pxref{Returning,
5420 ,Returning from a Function}).
5423 @kindex u @r{(@code{until})}
5424 @cindex run until specified location
5427 Continue running until a source line past the current line, in the
5428 current stack frame, is reached. This command is used to avoid single
5429 stepping through a loop more than once. It is like the @code{next}
5430 command, except that when @code{until} encounters a jump, it
5431 automatically continues execution until the program counter is greater
5432 than the address of the jump.
5434 This means that when you reach the end of a loop after single stepping
5435 though it, @code{until} makes your program continue execution until it
5436 exits the loop. In contrast, a @code{next} command at the end of a loop
5437 simply steps back to the beginning of the loop, which forces you to step
5438 through the next iteration.
5440 @code{until} always stops your program if it attempts to exit the current
5443 @code{until} may produce somewhat counterintuitive results if the order
5444 of machine code does not match the order of the source lines. For
5445 example, in the following excerpt from a debugging session, the @code{f}
5446 (@code{frame}) command shows that execution is stopped at line
5447 @code{206}; yet when we use @code{until}, we get to line @code{195}:
5451 #0 main (argc=4, argv=0xf7fffae8) at m4.c:206
5453 (@value{GDBP}) until
5454 195 for ( ; argc > 0; NEXTARG) @{
5457 This happened because, for execution efficiency, the compiler had
5458 generated code for the loop closure test at the end, rather than the
5459 start, of the loop---even though the test in a C @code{for}-loop is
5460 written before the body of the loop. The @code{until} command appeared
5461 to step back to the beginning of the loop when it advanced to this
5462 expression; however, it has not really gone to an earlier
5463 statement---not in terms of the actual machine code.
5465 @code{until} with no argument works by means of single
5466 instruction stepping, and hence is slower than @code{until} with an
5469 @item until @var{location}
5470 @itemx u @var{location}
5471 Continue running your program until either the specified @var{location} is
5472 reached, or the current stack frame returns. The location is any of
5473 the forms described in @ref{Specify Location}.
5474 This form of the command uses temporary breakpoints, and
5475 hence is quicker than @code{until} without an argument. The specified
5476 location is actually reached only if it is in the current frame. This
5477 implies that @code{until} can be used to skip over recursive function
5478 invocations. For instance in the code below, if the current location is
5479 line @code{96}, issuing @code{until 99} will execute the program up to
5480 line @code{99} in the same invocation of factorial, i.e., after the inner
5481 invocations have returned.
5484 94 int factorial (int value)
5486 96 if (value > 1) @{
5487 97 value *= factorial (value - 1);
5494 @kindex advance @var{location}
5495 @item advance @var{location}
5496 Continue running the program up to the given @var{location}. An argument is
5497 required, which should be of one of the forms described in
5498 @ref{Specify Location}.
5499 Execution will also stop upon exit from the current stack
5500 frame. This command is similar to @code{until}, but @code{advance} will
5501 not skip over recursive function calls, and the target location doesn't
5502 have to be in the same frame as the current one.
5506 @kindex si @r{(@code{stepi})}
5508 @itemx stepi @var{arg}
5510 Execute one machine instruction, then stop and return to the debugger.
5512 It is often useful to do @samp{display/i $pc} when stepping by machine
5513 instructions. This makes @value{GDBN} automatically display the next
5514 instruction to be executed, each time your program stops. @xref{Auto
5515 Display,, Automatic Display}.
5517 An argument is a repeat count, as in @code{step}.
5521 @kindex ni @r{(@code{nexti})}
5523 @itemx nexti @var{arg}
5525 Execute one machine instruction, but if it is a function call,
5526 proceed until the function returns.
5528 An argument is a repeat count, as in @code{next}.
5532 @anchor{range stepping}
5533 @cindex range stepping
5534 @cindex target-assisted range stepping
5535 By default, and if available, @value{GDBN} makes use of
5536 target-assisted @dfn{range stepping}. In other words, whenever you
5537 use a stepping command (e.g., @code{step}, @code{next}), @value{GDBN}
5538 tells the target to step the corresponding range of instruction
5539 addresses instead of issuing multiple single-steps. This speeds up
5540 line stepping, particularly for remote targets. Ideally, there should
5541 be no reason you would want to turn range stepping off. However, it's
5542 possible that a bug in the debug info, a bug in the remote stub (for
5543 remote targets), or even a bug in @value{GDBN} could make line
5544 stepping behave incorrectly when target-assisted range stepping is
5545 enabled. You can use the following command to turn off range stepping
5549 @kindex set range-stepping
5550 @kindex show range-stepping
5551 @item set range-stepping
5552 @itemx show range-stepping
5553 Control whether range stepping is enabled.
5555 If @code{on}, and the target supports it, @value{GDBN} tells the
5556 target to step a range of addresses itself, instead of issuing
5557 multiple single-steps. If @code{off}, @value{GDBN} always issues
5558 single-steps, even if range stepping is supported by the target. The
5559 default is @code{on}.
5563 @node Skipping Over Functions and Files
5564 @section Skipping Over Functions and Files
5565 @cindex skipping over functions and files
5567 The program you are debugging may contain some functions which are
5568 uninteresting to debug. The @code{skip} command lets you tell @value{GDBN} to
5569 skip a function, all functions in a file or a particular function in
5570 a particular file when stepping.
5572 For example, consider the following C function:
5583 Suppose you wish to step into the functions @code{foo} and @code{bar}, but you
5584 are not interested in stepping through @code{boring}. If you run @code{step}
5585 at line 103, you'll enter @code{boring()}, but if you run @code{next}, you'll
5586 step over both @code{foo} and @code{boring}!
5588 One solution is to @code{step} into @code{boring} and use the @code{finish}
5589 command to immediately exit it. But this can become tedious if @code{boring}
5590 is called from many places.
5592 A more flexible solution is to execute @kbd{skip boring}. This instructs
5593 @value{GDBN} never to step into @code{boring}. Now when you execute
5594 @code{step} at line 103, you'll step over @code{boring} and directly into
5597 Functions may be skipped by providing either a function name, linespec
5598 (@pxref{Specify Location}), regular expression that matches the function's
5599 name, file name or a @code{glob}-style pattern that matches the file name.
5601 On Posix systems the form of the regular expression is
5602 ``Extended Regular Expressions''. See for example @samp{man 7 regex}
5603 on @sc{gnu}/Linux systems. On non-Posix systems the form of the regular
5604 expression is whatever is provided by the @code{regcomp} function of
5605 the underlying system.
5606 See for example @samp{man 7 glob} on @sc{gnu}/Linux systems for a
5607 description of @code{glob}-style patterns.
5611 @item skip @r{[}@var{options}@r{]}
5612 The basic form of the @code{skip} command takes zero or more options
5613 that specify what to skip.
5614 The @var{options} argument is any useful combination of the following:
5617 @item -file @var{file}
5618 @itemx -fi @var{file}
5619 Functions in @var{file} will be skipped over when stepping.
5621 @item -gfile @var{file-glob-pattern}
5622 @itemx -gfi @var{file-glob-pattern}
5623 @cindex skipping over files via glob-style patterns
5624 Functions in files matching @var{file-glob-pattern} will be skipped
5628 (gdb) skip -gfi utils/*.c
5631 @item -function @var{linespec}
5632 @itemx -fu @var{linespec}
5633 Functions named by @var{linespec} or the function containing the line
5634 named by @var{linespec} will be skipped over when stepping.
5635 @xref{Specify Location}.
5637 @item -rfunction @var{regexp}
5638 @itemx -rfu @var{regexp}
5639 @cindex skipping over functions via regular expressions
5640 Functions whose name matches @var{regexp} will be skipped over when stepping.
5642 This form is useful for complex function names.
5643 For example, there is generally no need to step into C@t{++} @code{std::string}
5644 constructors or destructors. Plus with C@t{++} templates it can be hard to
5645 write out the full name of the function, and often it doesn't matter what
5646 the template arguments are. Specifying the function to be skipped as a
5647 regular expression makes this easier.
5650 (gdb) skip -rfu ^std::(allocator|basic_string)<.*>::~?\1 *\(
5653 If you want to skip every templated C@t{++} constructor and destructor
5654 in the @code{std} namespace you can do:
5657 (gdb) skip -rfu ^std::([a-zA-z0-9_]+)<.*>::~?\1 *\(
5661 If no options are specified, the function you're currently debugging
5664 @kindex skip function
5665 @item skip function @r{[}@var{linespec}@r{]}
5666 After running this command, the function named by @var{linespec} or the
5667 function containing the line named by @var{linespec} will be skipped over when
5668 stepping. @xref{Specify Location}.
5670 If you do not specify @var{linespec}, the function you're currently debugging
5673 (If you have a function called @code{file} that you want to skip, use
5674 @kbd{skip function file}.)
5677 @item skip file @r{[}@var{filename}@r{]}
5678 After running this command, any function whose source lives in @var{filename}
5679 will be skipped over when stepping.
5682 (gdb) skip file boring.c
5683 File boring.c will be skipped when stepping.
5686 If you do not specify @var{filename}, functions whose source lives in the file
5687 you're currently debugging will be skipped.
5690 Skips can be listed, deleted, disabled, and enabled, much like breakpoints.
5691 These are the commands for managing your list of skips:
5695 @item info skip @r{[}@var{range}@r{]}
5696 Print details about the specified skip(s). If @var{range} is not specified,
5697 print a table with details about all functions and files marked for skipping.
5698 @code{info skip} prints the following information about each skip:
5702 A number identifying this skip.
5703 @item Enabled or Disabled
5704 Enabled skips are marked with @samp{y}.
5705 Disabled skips are marked with @samp{n}.
5707 If the file name is a @samp{glob} pattern this is @samp{y}.
5708 Otherwise it is @samp{n}.
5710 The name or @samp{glob} pattern of the file to be skipped.
5711 If no file is specified this is @samp{<none>}.
5713 If the function name is a @samp{regular expression} this is @samp{y}.
5714 Otherwise it is @samp{n}.
5716 The name or regular expression of the function to skip.
5717 If no function is specified this is @samp{<none>}.
5721 @item skip delete @r{[}@var{range}@r{]}
5722 Delete the specified skip(s). If @var{range} is not specified, delete all
5726 @item skip enable @r{[}@var{range}@r{]}
5727 Enable the specified skip(s). If @var{range} is not specified, enable all
5730 @kindex skip disable
5731 @item skip disable @r{[}@var{range}@r{]}
5732 Disable the specified skip(s). If @var{range} is not specified, disable all
5741 A signal is an asynchronous event that can happen in a program. The
5742 operating system defines the possible kinds of signals, and gives each
5743 kind a name and a number. For example, in Unix @code{SIGINT} is the
5744 signal a program gets when you type an interrupt character (often @kbd{Ctrl-c});
5745 @code{SIGSEGV} is the signal a program gets from referencing a place in
5746 memory far away from all the areas in use; @code{SIGALRM} occurs when
5747 the alarm clock timer goes off (which happens only if your program has
5748 requested an alarm).
5750 @cindex fatal signals
5751 Some signals, including @code{SIGALRM}, are a normal part of the
5752 functioning of your program. Others, such as @code{SIGSEGV}, indicate
5753 errors; these signals are @dfn{fatal} (they kill your program immediately) if the
5754 program has not specified in advance some other way to handle the signal.
5755 @code{SIGINT} does not indicate an error in your program, but it is normally
5756 fatal so it can carry out the purpose of the interrupt: to kill the program.
5758 @value{GDBN} has the ability to detect any occurrence of a signal in your
5759 program. You can tell @value{GDBN} in advance what to do for each kind of
5762 @cindex handling signals
5763 Normally, @value{GDBN} is set up to let the non-erroneous signals like
5764 @code{SIGALRM} be silently passed to your program
5765 (so as not to interfere with their role in the program's functioning)
5766 but to stop your program immediately whenever an error signal happens.
5767 You can change these settings with the @code{handle} command.
5770 @kindex info signals
5774 Print a table of all the kinds of signals and how @value{GDBN} has been told to
5775 handle each one. You can use this to see the signal numbers of all
5776 the defined types of signals.
5778 @item info signals @var{sig}
5779 Similar, but print information only about the specified signal number.
5781 @code{info handle} is an alias for @code{info signals}.
5783 @item catch signal @r{[}@var{signal}@dots{} @r{|} @samp{all}@r{]}
5784 Set a catchpoint for the indicated signals. @xref{Set Catchpoints},
5785 for details about this command.
5788 @item handle @var{signal} @r{[}@var{keywords}@dots{}@r{]}
5789 Change the way @value{GDBN} handles signal @var{signal}. The @var{signal}
5790 can be the number of a signal or its name (with or without the
5791 @samp{SIG} at the beginning); a list of signal numbers of the form
5792 @samp{@var{low}-@var{high}}; or the word @samp{all}, meaning all the
5793 known signals. Optional arguments @var{keywords}, described below,
5794 say what change to make.
5798 The keywords allowed by the @code{handle} command can be abbreviated.
5799 Their full names are:
5803 @value{GDBN} should not stop your program when this signal happens. It may
5804 still print a message telling you that the signal has come in.
5807 @value{GDBN} should stop your program when this signal happens. This implies
5808 the @code{print} keyword as well.
5811 @value{GDBN} should print a message when this signal happens.
5814 @value{GDBN} should not mention the occurrence of the signal at all. This
5815 implies the @code{nostop} keyword as well.
5819 @value{GDBN} should allow your program to see this signal; your program
5820 can handle the signal, or else it may terminate if the signal is fatal
5821 and not handled. @code{pass} and @code{noignore} are synonyms.
5825 @value{GDBN} should not allow your program to see this signal.
5826 @code{nopass} and @code{ignore} are synonyms.
5830 When a signal stops your program, the signal is not visible to the
5832 continue. Your program sees the signal then, if @code{pass} is in
5833 effect for the signal in question @emph{at that time}. In other words,
5834 after @value{GDBN} reports a signal, you can use the @code{handle}
5835 command with @code{pass} or @code{nopass} to control whether your
5836 program sees that signal when you continue.
5838 The default is set to @code{nostop}, @code{noprint}, @code{pass} for
5839 non-erroneous signals such as @code{SIGALRM}, @code{SIGWINCH} and
5840 @code{SIGCHLD}, and to @code{stop}, @code{print}, @code{pass} for the
5843 You can also use the @code{signal} command to prevent your program from
5844 seeing a signal, or cause it to see a signal it normally would not see,
5845 or to give it any signal at any time. For example, if your program stopped
5846 due to some sort of memory reference error, you might store correct
5847 values into the erroneous variables and continue, hoping to see more
5848 execution; but your program would probably terminate immediately as
5849 a result of the fatal signal once it saw the signal. To prevent this,
5850 you can continue with @samp{signal 0}. @xref{Signaling, ,Giving your
5853 @cindex stepping and signal handlers
5854 @anchor{stepping and signal handlers}
5856 @value{GDBN} optimizes for stepping the mainline code. If a signal
5857 that has @code{handle nostop} and @code{handle pass} set arrives while
5858 a stepping command (e.g., @code{stepi}, @code{step}, @code{next}) is
5859 in progress, @value{GDBN} lets the signal handler run and then resumes
5860 stepping the mainline code once the signal handler returns. In other
5861 words, @value{GDBN} steps over the signal handler. This prevents
5862 signals that you've specified as not interesting (with @code{handle
5863 nostop}) from changing the focus of debugging unexpectedly. Note that
5864 the signal handler itself may still hit a breakpoint, stop for another
5865 signal that has @code{handle stop} in effect, or for any other event
5866 that normally results in stopping the stepping command sooner. Also
5867 note that @value{GDBN} still informs you that the program received a
5868 signal if @code{handle print} is set.
5870 @anchor{stepping into signal handlers}
5872 If you set @code{handle pass} for a signal, and your program sets up a
5873 handler for it, then issuing a stepping command, such as @code{step}
5874 or @code{stepi}, when your program is stopped due to the signal will
5875 step @emph{into} the signal handler (if the target supports that).
5877 Likewise, if you use the @code{queue-signal} command to queue a signal
5878 to be delivered to the current thread when execution of the thread
5879 resumes (@pxref{Signaling, ,Giving your Program a Signal}), then a
5880 stepping command will step into the signal handler.
5882 Here's an example, using @code{stepi} to step to the first instruction
5883 of @code{SIGUSR1}'s handler:
5886 (@value{GDBP}) handle SIGUSR1
5887 Signal Stop Print Pass to program Description
5888 SIGUSR1 Yes Yes Yes User defined signal 1
5892 Program received signal SIGUSR1, User defined signal 1.
5893 main () sigusr1.c:28
5896 sigusr1_handler () at sigusr1.c:9
5900 The same, but using @code{queue-signal} instead of waiting for the
5901 program to receive the signal first:
5906 (@value{GDBP}) queue-signal SIGUSR1
5908 sigusr1_handler () at sigusr1.c:9
5913 @cindex extra signal information
5914 @anchor{extra signal information}
5916 On some targets, @value{GDBN} can inspect extra signal information
5917 associated with the intercepted signal, before it is actually
5918 delivered to the program being debugged. This information is exported
5919 by the convenience variable @code{$_siginfo}, and consists of data
5920 that is passed by the kernel to the signal handler at the time of the
5921 receipt of a signal. The data type of the information itself is
5922 target dependent. You can see the data type using the @code{ptype
5923 $_siginfo} command. On Unix systems, it typically corresponds to the
5924 standard @code{siginfo_t} type, as defined in the @file{signal.h}
5927 Here's an example, on a @sc{gnu}/Linux system, printing the stray
5928 referenced address that raised a segmentation fault.
5932 (@value{GDBP}) continue
5933 Program received signal SIGSEGV, Segmentation fault.
5934 0x0000000000400766 in main ()
5936 (@value{GDBP}) ptype $_siginfo
5943 struct @{...@} _kill;
5944 struct @{...@} _timer;
5946 struct @{...@} _sigchld;
5947 struct @{...@} _sigfault;
5948 struct @{...@} _sigpoll;
5951 (@value{GDBP}) ptype $_siginfo._sifields._sigfault
5955 (@value{GDBP}) p $_siginfo._sifields._sigfault.si_addr
5956 $1 = (void *) 0x7ffff7ff7000
5960 Depending on target support, @code{$_siginfo} may also be writable.
5962 @cindex Intel MPX boundary violations
5963 @cindex boundary violations, Intel MPX
5964 On some targets, a @code{SIGSEGV} can be caused by a boundary
5965 violation, i.e., accessing an address outside of the allowed range.
5966 In those cases @value{GDBN} may displays additional information,
5967 depending on how @value{GDBN} has been told to handle the signal.
5968 With @code{handle stop SIGSEGV}, @value{GDBN} displays the violation
5969 kind: "Upper" or "Lower", the memory address accessed and the
5970 bounds, while with @code{handle nostop SIGSEGV} no additional
5971 information is displayed.
5973 The usual output of a segfault is:
5975 Program received signal SIGSEGV, Segmentation fault
5976 0x0000000000400d7c in upper () at i386-mpx-sigsegv.c:68
5977 68 value = *(p + len);
5980 While a bound violation is presented as:
5982 Program received signal SIGSEGV, Segmentation fault
5983 Upper bound violation while accessing address 0x7fffffffc3b3
5984 Bounds: [lower = 0x7fffffffc390, upper = 0x7fffffffc3a3]
5985 0x0000000000400d7c in upper () at i386-mpx-sigsegv.c:68
5986 68 value = *(p + len);
5990 @section Stopping and Starting Multi-thread Programs
5992 @cindex stopped threads
5993 @cindex threads, stopped
5995 @cindex continuing threads
5996 @cindex threads, continuing
5998 @value{GDBN} supports debugging programs with multiple threads
5999 (@pxref{Threads,, Debugging Programs with Multiple Threads}). There
6000 are two modes of controlling execution of your program within the
6001 debugger. In the default mode, referred to as @dfn{all-stop mode},
6002 when any thread in your program stops (for example, at a breakpoint
6003 or while being stepped), all other threads in the program are also stopped by
6004 @value{GDBN}. On some targets, @value{GDBN} also supports
6005 @dfn{non-stop mode}, in which other threads can continue to run freely while
6006 you examine the stopped thread in the debugger.
6009 * All-Stop Mode:: All threads stop when GDB takes control
6010 * Non-Stop Mode:: Other threads continue to execute
6011 * Background Execution:: Running your program asynchronously
6012 * Thread-Specific Breakpoints:: Controlling breakpoints
6013 * Interrupted System Calls:: GDB may interfere with system calls
6014 * Observer Mode:: GDB does not alter program behavior
6018 @subsection All-Stop Mode
6020 @cindex all-stop mode
6022 In all-stop mode, whenever your program stops under @value{GDBN} for any reason,
6023 @emph{all} threads of execution stop, not just the current thread. This
6024 allows you to examine the overall state of the program, including
6025 switching between threads, without worrying that things may change
6028 Conversely, whenever you restart the program, @emph{all} threads start
6029 executing. @emph{This is true even when single-stepping} with commands
6030 like @code{step} or @code{next}.
6032 In particular, @value{GDBN} cannot single-step all threads in lockstep.
6033 Since thread scheduling is up to your debugging target's operating
6034 system (not controlled by @value{GDBN}), other threads may
6035 execute more than one statement while the current thread completes a
6036 single step. Moreover, in general other threads stop in the middle of a
6037 statement, rather than at a clean statement boundary, when the program
6040 You might even find your program stopped in another thread after
6041 continuing or even single-stepping. This happens whenever some other
6042 thread runs into a breakpoint, a signal, or an exception before the
6043 first thread completes whatever you requested.
6045 @cindex automatic thread selection
6046 @cindex switching threads automatically
6047 @cindex threads, automatic switching
6048 Whenever @value{GDBN} stops your program, due to a breakpoint or a
6049 signal, it automatically selects the thread where that breakpoint or
6050 signal happened. @value{GDBN} alerts you to the context switch with a
6051 message such as @samp{[Switching to Thread @var{n}]} to identify the
6054 On some OSes, you can modify @value{GDBN}'s default behavior by
6055 locking the OS scheduler to allow only a single thread to run.
6058 @item set scheduler-locking @var{mode}
6059 @cindex scheduler locking mode
6060 @cindex lock scheduler
6061 Set the scheduler locking mode. It applies to normal execution,
6062 record mode, and replay mode. If it is @code{off}, then there is no
6063 locking and any thread may run at any time. If @code{on}, then only
6064 the current thread may run when the inferior is resumed. The
6065 @code{step} mode optimizes for single-stepping; it prevents other
6066 threads from preempting the current thread while you are stepping, so
6067 that the focus of debugging does not change unexpectedly. Other
6068 threads never get a chance to run when you step, and they are
6069 completely free to run when you use commands like @samp{continue},
6070 @samp{until}, or @samp{finish}. However, unless another thread hits a
6071 breakpoint during its timeslice, @value{GDBN} does not change the
6072 current thread away from the thread that you are debugging. The
6073 @code{replay} mode behaves like @code{off} in record mode and like
6074 @code{on} in replay mode.
6076 @item show scheduler-locking
6077 Display the current scheduler locking mode.
6080 @cindex resume threads of multiple processes simultaneously
6081 By default, when you issue one of the execution commands such as
6082 @code{continue}, @code{next} or @code{step}, @value{GDBN} allows only
6083 threads of the current inferior to run. For example, if @value{GDBN}
6084 is attached to two inferiors, each with two threads, the
6085 @code{continue} command resumes only the two threads of the current
6086 inferior. This is useful, for example, when you debug a program that
6087 forks and you want to hold the parent stopped (so that, for instance,
6088 it doesn't run to exit), while you debug the child. In other
6089 situations, you may not be interested in inspecting the current state
6090 of any of the processes @value{GDBN} is attached to, and you may want
6091 to resume them all until some breakpoint is hit. In the latter case,
6092 you can instruct @value{GDBN} to allow all threads of all the
6093 inferiors to run with the @w{@code{set schedule-multiple}} command.
6096 @kindex set schedule-multiple
6097 @item set schedule-multiple
6098 Set the mode for allowing threads of multiple processes to be resumed
6099 when an execution command is issued. When @code{on}, all threads of
6100 all processes are allowed to run. When @code{off}, only the threads
6101 of the current process are resumed. The default is @code{off}. The
6102 @code{scheduler-locking} mode takes precedence when set to @code{on},
6103 or while you are stepping and set to @code{step}.
6105 @item show schedule-multiple
6106 Display the current mode for resuming the execution of threads of
6111 @subsection Non-Stop Mode
6113 @cindex non-stop mode
6115 @c This section is really only a place-holder, and needs to be expanded
6116 @c with more details.
6118 For some multi-threaded targets, @value{GDBN} supports an optional
6119 mode of operation in which you can examine stopped program threads in
6120 the debugger while other threads continue to execute freely. This
6121 minimizes intrusion when debugging live systems, such as programs
6122 where some threads have real-time constraints or must continue to
6123 respond to external events. This is referred to as @dfn{non-stop} mode.
6125 In non-stop mode, when a thread stops to report a debugging event,
6126 @emph{only} that thread is stopped; @value{GDBN} does not stop other
6127 threads as well, in contrast to the all-stop mode behavior. Additionally,
6128 execution commands such as @code{continue} and @code{step} apply by default
6129 only to the current thread in non-stop mode, rather than all threads as
6130 in all-stop mode. This allows you to control threads explicitly in
6131 ways that are not possible in all-stop mode --- for example, stepping
6132 one thread while allowing others to run freely, stepping
6133 one thread while holding all others stopped, or stepping several threads
6134 independently and simultaneously.
6136 To enter non-stop mode, use this sequence of commands before you run
6137 or attach to your program:
6140 # If using the CLI, pagination breaks non-stop.
6143 # Finally, turn it on!
6147 You can use these commands to manipulate the non-stop mode setting:
6150 @kindex set non-stop
6151 @item set non-stop on
6152 Enable selection of non-stop mode.
6153 @item set non-stop off
6154 Disable selection of non-stop mode.
6155 @kindex show non-stop
6157 Show the current non-stop enablement setting.
6160 Note these commands only reflect whether non-stop mode is enabled,
6161 not whether the currently-executing program is being run in non-stop mode.
6162 In particular, the @code{set non-stop} preference is only consulted when
6163 @value{GDBN} starts or connects to the target program, and it is generally
6164 not possible to switch modes once debugging has started. Furthermore,
6165 since not all targets support non-stop mode, even when you have enabled
6166 non-stop mode, @value{GDBN} may still fall back to all-stop operation by
6169 In non-stop mode, all execution commands apply only to the current thread
6170 by default. That is, @code{continue} only continues one thread.
6171 To continue all threads, issue @code{continue -a} or @code{c -a}.
6173 You can use @value{GDBN}'s background execution commands
6174 (@pxref{Background Execution}) to run some threads in the background
6175 while you continue to examine or step others from @value{GDBN}.
6176 The MI execution commands (@pxref{GDB/MI Program Execution}) are
6177 always executed asynchronously in non-stop mode.
6179 Suspending execution is done with the @code{interrupt} command when
6180 running in the background, or @kbd{Ctrl-c} during foreground execution.
6181 In all-stop mode, this stops the whole process;
6182 but in non-stop mode the interrupt applies only to the current thread.
6183 To stop the whole program, use @code{interrupt -a}.
6185 Other execution commands do not currently support the @code{-a} option.
6187 In non-stop mode, when a thread stops, @value{GDBN} doesn't automatically make
6188 that thread current, as it does in all-stop mode. This is because the
6189 thread stop notifications are asynchronous with respect to @value{GDBN}'s
6190 command interpreter, and it would be confusing if @value{GDBN} unexpectedly
6191 changed to a different thread just as you entered a command to operate on the
6192 previously current thread.
6194 @node Background Execution
6195 @subsection Background Execution
6197 @cindex foreground execution
6198 @cindex background execution
6199 @cindex asynchronous execution
6200 @cindex execution, foreground, background and asynchronous
6202 @value{GDBN}'s execution commands have two variants: the normal
6203 foreground (synchronous) behavior, and a background
6204 (asynchronous) behavior. In foreground execution, @value{GDBN} waits for
6205 the program to report that some thread has stopped before prompting for
6206 another command. In background execution, @value{GDBN} immediately gives
6207 a command prompt so that you can issue other commands while your program runs.
6209 If the target doesn't support async mode, @value{GDBN} issues an error
6210 message if you attempt to use the background execution commands.
6212 To specify background execution, add a @code{&} to the command. For example,
6213 the background form of the @code{continue} command is @code{continue&}, or
6214 just @code{c&}. The execution commands that accept background execution
6220 @xref{Starting, , Starting your Program}.
6224 @xref{Attach, , Debugging an Already-running Process}.
6228 @xref{Continuing and Stepping, step}.
6232 @xref{Continuing and Stepping, stepi}.
6236 @xref{Continuing and Stepping, next}.
6240 @xref{Continuing and Stepping, nexti}.
6244 @xref{Continuing and Stepping, continue}.
6248 @xref{Continuing and Stepping, finish}.
6252 @xref{Continuing and Stepping, until}.
6256 Background execution is especially useful in conjunction with non-stop
6257 mode for debugging programs with multiple threads; see @ref{Non-Stop Mode}.
6258 However, you can also use these commands in the normal all-stop mode with
6259 the restriction that you cannot issue another execution command until the
6260 previous one finishes. Examples of commands that are valid in all-stop
6261 mode while the program is running include @code{help} and @code{info break}.
6263 You can interrupt your program while it is running in the background by
6264 using the @code{interrupt} command.
6271 Suspend execution of the running program. In all-stop mode,
6272 @code{interrupt} stops the whole process, but in non-stop mode, it stops
6273 only the current thread. To stop the whole program in non-stop mode,
6274 use @code{interrupt -a}.
6277 @node Thread-Specific Breakpoints
6278 @subsection Thread-Specific Breakpoints
6280 When your program has multiple threads (@pxref{Threads,, Debugging
6281 Programs with Multiple Threads}), you can choose whether to set
6282 breakpoints on all threads, or on a particular thread.
6285 @cindex breakpoints and threads
6286 @cindex thread breakpoints
6287 @kindex break @dots{} thread @var{thread-id}
6288 @item break @var{location} thread @var{thread-id}
6289 @itemx break @var{location} thread @var{thread-id} if @dots{}
6290 @var{location} specifies source lines; there are several ways of
6291 writing them (@pxref{Specify Location}), but the effect is always to
6292 specify some source line.
6294 Use the qualifier @samp{thread @var{thread-id}} with a breakpoint command
6295 to specify that you only want @value{GDBN} to stop the program when a
6296 particular thread reaches this breakpoint. The @var{thread-id} specifier
6297 is one of the thread identifiers assigned by @value{GDBN}, shown
6298 in the first column of the @samp{info threads} display.
6300 If you do not specify @samp{thread @var{thread-id}} when you set a
6301 breakpoint, the breakpoint applies to @emph{all} threads of your
6304 You can use the @code{thread} qualifier on conditional breakpoints as
6305 well; in this case, place @samp{thread @var{thread-id}} before or
6306 after the breakpoint condition, like this:
6309 (@value{GDBP}) break frik.c:13 thread 28 if bartab > lim
6314 Thread-specific breakpoints are automatically deleted when
6315 @value{GDBN} detects the corresponding thread is no longer in the
6316 thread list. For example:
6320 Thread-specific breakpoint 3 deleted - thread 28 no longer in the thread list.
6323 There are several ways for a thread to disappear, such as a regular
6324 thread exit, but also when you detach from the process with the
6325 @code{detach} command (@pxref{Attach, ,Debugging an Already-running
6326 Process}), or if @value{GDBN} loses the remote connection
6327 (@pxref{Remote Debugging}), etc. Note that with some targets,
6328 @value{GDBN} is only able to detect a thread has exited when the user
6329 explictly asks for the thread list with the @code{info threads}
6332 @node Interrupted System Calls
6333 @subsection Interrupted System Calls
6335 @cindex thread breakpoints and system calls
6336 @cindex system calls and thread breakpoints
6337 @cindex premature return from system calls
6338 There is an unfortunate side effect when using @value{GDBN} to debug
6339 multi-threaded programs. If one thread stops for a
6340 breakpoint, or for some other reason, and another thread is blocked in a
6341 system call, then the system call may return prematurely. This is a
6342 consequence of the interaction between multiple threads and the signals
6343 that @value{GDBN} uses to implement breakpoints and other events that
6346 To handle this problem, your program should check the return value of
6347 each system call and react appropriately. This is good programming
6350 For example, do not write code like this:
6356 The call to @code{sleep} will return early if a different thread stops
6357 at a breakpoint or for some other reason.
6359 Instead, write this:
6364 unslept = sleep (unslept);
6367 A system call is allowed to return early, so the system is still
6368 conforming to its specification. But @value{GDBN} does cause your
6369 multi-threaded program to behave differently than it would without
6372 Also, @value{GDBN} uses internal breakpoints in the thread library to
6373 monitor certain events such as thread creation and thread destruction.
6374 When such an event happens, a system call in another thread may return
6375 prematurely, even though your program does not appear to stop.
6378 @subsection Observer Mode
6380 If you want to build on non-stop mode and observe program behavior
6381 without any chance of disruption by @value{GDBN}, you can set
6382 variables to disable all of the debugger's attempts to modify state,
6383 whether by writing memory, inserting breakpoints, etc. These operate
6384 at a low level, intercepting operations from all commands.
6386 When all of these are set to @code{off}, then @value{GDBN} is said to
6387 be @dfn{observer mode}. As a convenience, the variable
6388 @code{observer} can be set to disable these, plus enable non-stop
6391 Note that @value{GDBN} will not prevent you from making nonsensical
6392 combinations of these settings. For instance, if you have enabled
6393 @code{may-insert-breakpoints} but disabled @code{may-write-memory},
6394 then breakpoints that work by writing trap instructions into the code
6395 stream will still not be able to be placed.
6400 @item set observer on
6401 @itemx set observer off
6402 When set to @code{on}, this disables all the permission variables
6403 below (except for @code{insert-fast-tracepoints}), plus enables
6404 non-stop debugging. Setting this to @code{off} switches back to
6405 normal debugging, though remaining in non-stop mode.
6408 Show whether observer mode is on or off.
6410 @kindex may-write-registers
6411 @item set may-write-registers on
6412 @itemx set may-write-registers off
6413 This controls whether @value{GDBN} will attempt to alter the values of
6414 registers, such as with assignment expressions in @code{print}, or the
6415 @code{jump} command. It defaults to @code{on}.
6417 @item show may-write-registers
6418 Show the current permission to write registers.
6420 @kindex may-write-memory
6421 @item set may-write-memory on
6422 @itemx set may-write-memory off
6423 This controls whether @value{GDBN} will attempt to alter the contents
6424 of memory, such as with assignment expressions in @code{print}. It
6425 defaults to @code{on}.
6427 @item show may-write-memory
6428 Show the current permission to write memory.
6430 @kindex may-insert-breakpoints
6431 @item set may-insert-breakpoints on
6432 @itemx set may-insert-breakpoints off
6433 This controls whether @value{GDBN} will attempt to insert breakpoints.
6434 This affects all breakpoints, including internal breakpoints defined
6435 by @value{GDBN}. It defaults to @code{on}.
6437 @item show may-insert-breakpoints
6438 Show the current permission to insert breakpoints.
6440 @kindex may-insert-tracepoints
6441 @item set may-insert-tracepoints on
6442 @itemx set may-insert-tracepoints off
6443 This controls whether @value{GDBN} will attempt to insert (regular)
6444 tracepoints at the beginning of a tracing experiment. It affects only
6445 non-fast tracepoints, fast tracepoints being under the control of
6446 @code{may-insert-fast-tracepoints}. It defaults to @code{on}.
6448 @item show may-insert-tracepoints
6449 Show the current permission to insert tracepoints.
6451 @kindex may-insert-fast-tracepoints
6452 @item set may-insert-fast-tracepoints on
6453 @itemx set may-insert-fast-tracepoints off
6454 This controls whether @value{GDBN} will attempt to insert fast
6455 tracepoints at the beginning of a tracing experiment. It affects only
6456 fast tracepoints, regular (non-fast) tracepoints being under the
6457 control of @code{may-insert-tracepoints}. It defaults to @code{on}.
6459 @item show may-insert-fast-tracepoints
6460 Show the current permission to insert fast tracepoints.
6462 @kindex may-interrupt
6463 @item set may-interrupt on
6464 @itemx set may-interrupt off
6465 This controls whether @value{GDBN} will attempt to interrupt or stop
6466 program execution. When this variable is @code{off}, the
6467 @code{interrupt} command will have no effect, nor will
6468 @kbd{Ctrl-c}. It defaults to @code{on}.
6470 @item show may-interrupt
6471 Show the current permission to interrupt or stop the program.
6475 @node Reverse Execution
6476 @chapter Running programs backward
6477 @cindex reverse execution
6478 @cindex running programs backward
6480 When you are debugging a program, it is not unusual to realize that
6481 you have gone too far, and some event of interest has already happened.
6482 If the target environment supports it, @value{GDBN} can allow you to
6483 ``rewind'' the program by running it backward.
6485 A target environment that supports reverse execution should be able
6486 to ``undo'' the changes in machine state that have taken place as the
6487 program was executing normally. Variables, registers etc.@: should
6488 revert to their previous values. Obviously this requires a great
6489 deal of sophistication on the part of the target environment; not
6490 all target environments can support reverse execution.
6492 When a program is executed in reverse, the instructions that
6493 have most recently been executed are ``un-executed'', in reverse
6494 order. The program counter runs backward, following the previous
6495 thread of execution in reverse. As each instruction is ``un-executed'',
6496 the values of memory and/or registers that were changed by that
6497 instruction are reverted to their previous states. After executing
6498 a piece of source code in reverse, all side effects of that code
6499 should be ``undone'', and all variables should be returned to their
6500 prior values@footnote{
6501 Note that some side effects are easier to undo than others. For instance,
6502 memory and registers are relatively easy, but device I/O is hard. Some
6503 targets may be able undo things like device I/O, and some may not.
6505 The contract between @value{GDBN} and the reverse executing target
6506 requires only that the target do something reasonable when
6507 @value{GDBN} tells it to execute backwards, and then report the
6508 results back to @value{GDBN}. Whatever the target reports back to
6509 @value{GDBN}, @value{GDBN} will report back to the user. @value{GDBN}
6510 assumes that the memory and registers that the target reports are in a
6511 consistant state, but @value{GDBN} accepts whatever it is given.
6514 If you are debugging in a target environment that supports
6515 reverse execution, @value{GDBN} provides the following commands.
6518 @kindex reverse-continue
6519 @kindex rc @r{(@code{reverse-continue})}
6520 @item reverse-continue @r{[}@var{ignore-count}@r{]}
6521 @itemx rc @r{[}@var{ignore-count}@r{]}
6522 Beginning at the point where your program last stopped, start executing
6523 in reverse. Reverse execution will stop for breakpoints and synchronous
6524 exceptions (signals), just like normal execution. Behavior of
6525 asynchronous signals depends on the target environment.
6527 @kindex reverse-step
6528 @kindex rs @r{(@code{step})}
6529 @item reverse-step @r{[}@var{count}@r{]}
6530 Run the program backward until control reaches the start of a
6531 different source line; then stop it, and return control to @value{GDBN}.
6533 Like the @code{step} command, @code{reverse-step} will only stop
6534 at the beginning of a source line. It ``un-executes'' the previously
6535 executed source line. If the previous source line included calls to
6536 debuggable functions, @code{reverse-step} will step (backward) into
6537 the called function, stopping at the beginning of the @emph{last}
6538 statement in the called function (typically a return statement).
6540 Also, as with the @code{step} command, if non-debuggable functions are
6541 called, @code{reverse-step} will run thru them backward without stopping.
6543 @kindex reverse-stepi
6544 @kindex rsi @r{(@code{reverse-stepi})}
6545 @item reverse-stepi @r{[}@var{count}@r{]}
6546 Reverse-execute one machine instruction. Note that the instruction
6547 to be reverse-executed is @emph{not} the one pointed to by the program
6548 counter, but the instruction executed prior to that one. For instance,
6549 if the last instruction was a jump, @code{reverse-stepi} will take you
6550 back from the destination of the jump to the jump instruction itself.
6552 @kindex reverse-next
6553 @kindex rn @r{(@code{reverse-next})}
6554 @item reverse-next @r{[}@var{count}@r{]}
6555 Run backward to the beginning of the previous line executed in
6556 the current (innermost) stack frame. If the line contains function
6557 calls, they will be ``un-executed'' without stopping. Starting from
6558 the first line of a function, @code{reverse-next} will take you back
6559 to the caller of that function, @emph{before} the function was called,
6560 just as the normal @code{next} command would take you from the last
6561 line of a function back to its return to its caller
6562 @footnote{Unless the code is too heavily optimized.}.
6564 @kindex reverse-nexti
6565 @kindex rni @r{(@code{reverse-nexti})}
6566 @item reverse-nexti @r{[}@var{count}@r{]}
6567 Like @code{nexti}, @code{reverse-nexti} executes a single instruction
6568 in reverse, except that called functions are ``un-executed'' atomically.
6569 That is, if the previously executed instruction was a return from
6570 another function, @code{reverse-nexti} will continue to execute
6571 in reverse until the call to that function (from the current stack
6574 @kindex reverse-finish
6575 @item reverse-finish
6576 Just as the @code{finish} command takes you to the point where the
6577 current function returns, @code{reverse-finish} takes you to the point
6578 where it was called. Instead of ending up at the end of the current
6579 function invocation, you end up at the beginning.
6581 @kindex set exec-direction
6582 @item set exec-direction
6583 Set the direction of target execution.
6584 @item set exec-direction reverse
6585 @cindex execute forward or backward in time
6586 @value{GDBN} will perform all execution commands in reverse, until the
6587 exec-direction mode is changed to ``forward''. Affected commands include
6588 @code{step, stepi, next, nexti, continue, and finish}. The @code{return}
6589 command cannot be used in reverse mode.
6590 @item set exec-direction forward
6591 @value{GDBN} will perform all execution commands in the normal fashion.
6592 This is the default.
6596 @node Process Record and Replay
6597 @chapter Recording Inferior's Execution and Replaying It
6598 @cindex process record and replay
6599 @cindex recording inferior's execution and replaying it
6601 On some platforms, @value{GDBN} provides a special @dfn{process record
6602 and replay} target that can record a log of the process execution, and
6603 replay it later with both forward and reverse execution commands.
6606 When this target is in use, if the execution log includes the record
6607 for the next instruction, @value{GDBN} will debug in @dfn{replay
6608 mode}. In the replay mode, the inferior does not really execute code
6609 instructions. Instead, all the events that normally happen during
6610 code execution are taken from the execution log. While code is not
6611 really executed in replay mode, the values of registers (including the
6612 program counter register) and the memory of the inferior are still
6613 changed as they normally would. Their contents are taken from the
6617 If the record for the next instruction is not in the execution log,
6618 @value{GDBN} will debug in @dfn{record mode}. In this mode, the
6619 inferior executes normally, and @value{GDBN} records the execution log
6622 The process record and replay target supports reverse execution
6623 (@pxref{Reverse Execution}), even if the platform on which the
6624 inferior runs does not. However, the reverse execution is limited in
6625 this case by the range of the instructions recorded in the execution
6626 log. In other words, reverse execution on platforms that don't
6627 support it directly can only be done in the replay mode.
6629 When debugging in the reverse direction, @value{GDBN} will work in
6630 replay mode as long as the execution log includes the record for the
6631 previous instruction; otherwise, it will work in record mode, if the
6632 platform supports reverse execution, or stop if not.
6634 For architecture environments that support process record and replay,
6635 @value{GDBN} provides the following commands:
6638 @kindex target record
6639 @kindex target record-full
6640 @kindex target record-btrace
6643 @kindex record btrace
6644 @kindex record btrace bts
6645 @kindex record btrace pt
6651 @kindex rec btrace bts
6652 @kindex rec btrace pt
6655 @item record @var{method}
6656 This command starts the process record and replay target. The
6657 recording method can be specified as parameter. Without a parameter
6658 the command uses the @code{full} recording method. The following
6659 recording methods are available:
6663 Full record/replay recording using @value{GDBN}'s software record and
6664 replay implementation. This method allows replaying and reverse
6667 @item btrace @var{format}
6668 Hardware-supported instruction recording. This method does not record
6669 data. Further, the data is collected in a ring buffer so old data will
6670 be overwritten when the buffer is full. It allows limited reverse
6671 execution. Variables and registers are not available during reverse
6672 execution. In remote debugging, recording continues on disconnect.
6673 Recorded data can be inspected after reconnecting. The recording may
6674 be stopped using @code{record stop}.
6676 The recording format can be specified as parameter. Without a parameter
6677 the command chooses the recording format. The following recording
6678 formats are available:
6682 @cindex branch trace store
6683 Use the @dfn{Branch Trace Store} (@acronym{BTS}) recording format. In
6684 this format, the processor stores a from/to record for each executed
6685 branch in the btrace ring buffer.
6688 @cindex Intel Processor Trace
6689 Use the @dfn{Intel Processor Trace} recording format. In this
6690 format, the processor stores the execution trace in a compressed form
6691 that is afterwards decoded by @value{GDBN}.
6693 The trace can be recorded with very low overhead. The compressed
6694 trace format also allows small trace buffers to already contain a big
6695 number of instructions compared to @acronym{BTS}.
6697 Decoding the recorded execution trace, on the other hand, is more
6698 expensive than decoding @acronym{BTS} trace. This is mostly due to the
6699 increased number of instructions to process. You should increase the
6700 buffer-size with care.
6703 Not all recording formats may be available on all processors.
6706 The process record and replay target can only debug a process that is
6707 already running. Therefore, you need first to start the process with
6708 the @kbd{run} or @kbd{start} commands, and then start the recording
6709 with the @kbd{record @var{method}} command.
6711 @cindex displaced stepping, and process record and replay
6712 Displaced stepping (@pxref{Maintenance Commands,, displaced stepping})
6713 will be automatically disabled when process record and replay target
6714 is started. That's because the process record and replay target
6715 doesn't support displaced stepping.
6717 @cindex non-stop mode, and process record and replay
6718 @cindex asynchronous execution, and process record and replay
6719 If the inferior is in the non-stop mode (@pxref{Non-Stop Mode}) or in
6720 the asynchronous execution mode (@pxref{Background Execution}), not
6721 all recording methods are available. The @code{full} recording method
6722 does not support these two modes.
6727 Stop the process record and replay target. When process record and
6728 replay target stops, the entire execution log will be deleted and the
6729 inferior will either be terminated, or will remain in its final state.
6731 When you stop the process record and replay target in record mode (at
6732 the end of the execution log), the inferior will be stopped at the
6733 next instruction that would have been recorded. In other words, if
6734 you record for a while and then stop recording, the inferior process
6735 will be left in the same state as if the recording never happened.
6737 On the other hand, if the process record and replay target is stopped
6738 while in replay mode (that is, not at the end of the execution log,
6739 but at some earlier point), the inferior process will become ``live''
6740 at that earlier state, and it will then be possible to continue the
6741 usual ``live'' debugging of the process from that state.
6743 When the inferior process exits, or @value{GDBN} detaches from it,
6744 process record and replay target will automatically stop itself.
6748 Go to a specific location in the execution log. There are several
6749 ways to specify the location to go to:
6752 @item record goto begin
6753 @itemx record goto start
6754 Go to the beginning of the execution log.
6756 @item record goto end
6757 Go to the end of the execution log.
6759 @item record goto @var{n}
6760 Go to instruction number @var{n} in the execution log.
6764 @item record save @var{filename}
6765 Save the execution log to a file @file{@var{filename}}.
6766 Default filename is @file{gdb_record.@var{process_id}}, where
6767 @var{process_id} is the process ID of the inferior.
6769 This command may not be available for all recording methods.
6771 @kindex record restore
6772 @item record restore @var{filename}
6773 Restore the execution log from a file @file{@var{filename}}.
6774 File must have been created with @code{record save}.
6776 @kindex set record full
6777 @item set record full insn-number-max @var{limit}
6778 @itemx set record full insn-number-max unlimited
6779 Set the limit of instructions to be recorded for the @code{full}
6780 recording method. Default value is 200000.
6782 If @var{limit} is a positive number, then @value{GDBN} will start
6783 deleting instructions from the log once the number of the record
6784 instructions becomes greater than @var{limit}. For every new recorded
6785 instruction, @value{GDBN} will delete the earliest recorded
6786 instruction to keep the number of recorded instructions at the limit.
6787 (Since deleting recorded instructions loses information, @value{GDBN}
6788 lets you control what happens when the limit is reached, by means of
6789 the @code{stop-at-limit} option, described below.)
6791 If @var{limit} is @code{unlimited} or zero, @value{GDBN} will never
6792 delete recorded instructions from the execution log. The number of
6793 recorded instructions is limited only by the available memory.
6795 @kindex show record full
6796 @item show record full insn-number-max
6797 Show the limit of instructions to be recorded with the @code{full}
6800 @item set record full stop-at-limit
6801 Control the behavior of the @code{full} recording method when the
6802 number of recorded instructions reaches the limit. If ON (the
6803 default), @value{GDBN} will stop when the limit is reached for the
6804 first time and ask you whether you want to stop the inferior or
6805 continue running it and recording the execution log. If you decide
6806 to continue recording, each new recorded instruction will cause the
6807 oldest one to be deleted.
6809 If this option is OFF, @value{GDBN} will automatically delete the
6810 oldest record to make room for each new one, without asking.
6812 @item show record full stop-at-limit
6813 Show the current setting of @code{stop-at-limit}.
6815 @item set record full memory-query
6816 Control the behavior when @value{GDBN} is unable to record memory
6817 changes caused by an instruction for the @code{full} recording method.
6818 If ON, @value{GDBN} will query whether to stop the inferior in that
6821 If this option is OFF (the default), @value{GDBN} will automatically
6822 ignore the effect of such instructions on memory. Later, when
6823 @value{GDBN} replays this execution log, it will mark the log of this
6824 instruction as not accessible, and it will not affect the replay
6827 @item show record full memory-query
6828 Show the current setting of @code{memory-query}.
6830 @kindex set record btrace
6831 The @code{btrace} record target does not trace data. As a
6832 convenience, when replaying, @value{GDBN} reads read-only memory off
6833 the live program directly, assuming that the addresses of the
6834 read-only areas don't change. This for example makes it possible to
6835 disassemble code while replaying, but not to print variables.
6836 In some cases, being able to inspect variables might be useful.
6837 You can use the following command for that:
6839 @item set record btrace replay-memory-access
6840 Control the behavior of the @code{btrace} recording method when
6841 accessing memory during replay. If @code{read-only} (the default),
6842 @value{GDBN} will only allow accesses to read-only memory.
6843 If @code{read-write}, @value{GDBN} will allow accesses to read-only
6844 and to read-write memory. Beware that the accessed memory corresponds
6845 to the live target and not necessarily to the current replay
6848 @kindex show record btrace
6849 @item show record btrace replay-memory-access
6850 Show the current setting of @code{replay-memory-access}.
6852 @kindex set record btrace bts
6853 @item set record btrace bts buffer-size @var{size}
6854 @itemx set record btrace bts buffer-size unlimited
6855 Set the requested ring buffer size for branch tracing in @acronym{BTS}
6856 format. Default is 64KB.
6858 If @var{size} is a positive number, then @value{GDBN} will try to
6859 allocate a buffer of at least @var{size} bytes for each new thread
6860 that uses the btrace recording method and the @acronym{BTS} format.
6861 The actually obtained buffer size may differ from the requested
6862 @var{size}. Use the @code{info record} command to see the actual
6863 buffer size for each thread that uses the btrace recording method and
6864 the @acronym{BTS} format.
6866 If @var{limit} is @code{unlimited} or zero, @value{GDBN} will try to
6867 allocate a buffer of 4MB.
6869 Bigger buffers mean longer traces. On the other hand, @value{GDBN} will
6870 also need longer to process the branch trace data before it can be used.
6872 @item show record btrace bts buffer-size @var{size}
6873 Show the current setting of the requested ring buffer size for branch
6874 tracing in @acronym{BTS} format.
6876 @kindex set record btrace pt
6877 @item set record btrace pt buffer-size @var{size}
6878 @itemx set record btrace pt buffer-size unlimited
6879 Set the requested ring buffer size for branch tracing in Intel
6880 Processor Trace format. Default is 16KB.
6882 If @var{size} is a positive number, then @value{GDBN} will try to
6883 allocate a buffer of at least @var{size} bytes for each new thread
6884 that uses the btrace recording method and the Intel Processor Trace
6885 format. The actually obtained buffer size may differ from the
6886 requested @var{size}. Use the @code{info record} command to see the
6887 actual buffer size for each thread.
6889 If @var{limit} is @code{unlimited} or zero, @value{GDBN} will try to
6890 allocate a buffer of 4MB.
6892 Bigger buffers mean longer traces. On the other hand, @value{GDBN} will
6893 also need longer to process the branch trace data before it can be used.
6895 @item show record btrace pt buffer-size @var{size}
6896 Show the current setting of the requested ring buffer size for branch
6897 tracing in Intel Processor Trace format.
6901 Show various statistics about the recording depending on the recording
6906 For the @code{full} recording method, it shows the state of process
6907 record and its in-memory execution log buffer, including:
6911 Whether in record mode or replay mode.
6913 Lowest recorded instruction number (counting from when the current execution log started recording instructions).
6915 Highest recorded instruction number.
6917 Current instruction about to be replayed (if in replay mode).
6919 Number of instructions contained in the execution log.
6921 Maximum number of instructions that may be contained in the execution log.
6925 For the @code{btrace} recording method, it shows:
6931 Number of instructions that have been recorded.
6933 Number of blocks of sequential control-flow formed by the recorded
6936 Whether in record mode or replay mode.
6939 For the @code{bts} recording format, it also shows:
6942 Size of the perf ring buffer.
6945 For the @code{pt} recording format, it also shows:
6948 Size of the perf ring buffer.
6952 @kindex record delete
6955 When record target runs in replay mode (``in the past''), delete the
6956 subsequent execution log and begin to record a new execution log starting
6957 from the current address. This means you will abandon the previously
6958 recorded ``future'' and begin recording a new ``future''.
6960 @kindex record instruction-history
6961 @kindex rec instruction-history
6962 @item record instruction-history
6963 Disassembles instructions from the recorded execution log. By
6964 default, ten instructions are disassembled. This can be changed using
6965 the @code{set record instruction-history-size} command. Instructions
6966 are printed in execution order.
6968 It can also print mixed source+disassembly if you specify the the
6969 @code{/m} or @code{/s} modifier, and print the raw instructions in hex
6970 as well as in symbolic form by specifying the @code{/r} modifier.
6972 The current position marker is printed for the instruction at the
6973 current program counter value. This instruction can appear multiple
6974 times in the trace and the current position marker will be printed
6975 every time. To omit the current position marker, specify the
6978 To better align the printed instructions when the trace contains
6979 instructions from more than one function, the function name may be
6980 omitted by specifying the @code{/f} modifier.
6982 Speculatively executed instructions are prefixed with @samp{?}. This
6983 feature is not available for all recording formats.
6985 There are several ways to specify what part of the execution log to
6989 @item record instruction-history @var{insn}
6990 Disassembles ten instructions starting from instruction number
6993 @item record instruction-history @var{insn}, +/-@var{n}
6994 Disassembles @var{n} instructions around instruction number
6995 @var{insn}. If @var{n} is preceded with @code{+}, disassembles
6996 @var{n} instructions after instruction number @var{insn}. If
6997 @var{n} is preceded with @code{-}, disassembles @var{n}
6998 instructions before instruction number @var{insn}.
7000 @item record instruction-history
7001 Disassembles ten more instructions after the last disassembly.
7003 @item record instruction-history -
7004 Disassembles ten more instructions before the last disassembly.
7006 @item record instruction-history @var{begin}, @var{end}
7007 Disassembles instructions beginning with instruction number
7008 @var{begin} until instruction number @var{end}. The instruction
7009 number @var{end} is included.
7012 This command may not be available for all recording methods.
7015 @item set record instruction-history-size @var{size}
7016 @itemx set record instruction-history-size unlimited
7017 Define how many instructions to disassemble in the @code{record
7018 instruction-history} command. The default value is 10.
7019 A @var{size} of @code{unlimited} means unlimited instructions.
7022 @item show record instruction-history-size
7023 Show how many instructions to disassemble in the @code{record
7024 instruction-history} command.
7026 @kindex record function-call-history
7027 @kindex rec function-call-history
7028 @item record function-call-history
7029 Prints the execution history at function granularity. It prints one
7030 line for each sequence of instructions that belong to the same
7031 function giving the name of that function, the source lines
7032 for this instruction sequence (if the @code{/l} modifier is
7033 specified), and the instructions numbers that form the sequence (if
7034 the @code{/i} modifier is specified). The function names are indented
7035 to reflect the call stack depth if the @code{/c} modifier is
7036 specified. The @code{/l}, @code{/i}, and @code{/c} modifiers can be
7040 (@value{GDBP}) @b{list 1, 10}
7051 (@value{GDBP}) @b{record function-call-history /ilc}
7052 1 bar inst 1,4 at foo.c:6,8
7053 2 foo inst 5,10 at foo.c:2,3
7054 3 bar inst 11,13 at foo.c:9,10
7057 By default, ten lines are printed. This can be changed using the
7058 @code{set record function-call-history-size} command. Functions are
7059 printed in execution order. There are several ways to specify what
7063 @item record function-call-history @var{func}
7064 Prints ten functions starting from function number @var{func}.
7066 @item record function-call-history @var{func}, +/-@var{n}
7067 Prints @var{n} functions around function number @var{func}. If
7068 @var{n} is preceded with @code{+}, prints @var{n} functions after
7069 function number @var{func}. If @var{n} is preceded with @code{-},
7070 prints @var{n} functions before function number @var{func}.
7072 @item record function-call-history
7073 Prints ten more functions after the last ten-line print.
7075 @item record function-call-history -
7076 Prints ten more functions before the last ten-line print.
7078 @item record function-call-history @var{begin}, @var{end}
7079 Prints functions beginning with function number @var{begin} until
7080 function number @var{end}. The function number @var{end} is included.
7083 This command may not be available for all recording methods.
7085 @item set record function-call-history-size @var{size}
7086 @itemx set record function-call-history-size unlimited
7087 Define how many lines to print in the
7088 @code{record function-call-history} command. The default value is 10.
7089 A size of @code{unlimited} means unlimited lines.
7091 @item show record function-call-history-size
7092 Show how many lines to print in the
7093 @code{record function-call-history} command.
7098 @chapter Examining the Stack
7100 When your program has stopped, the first thing you need to know is where it
7101 stopped and how it got there.
7104 Each time your program performs a function call, information about the call
7106 That information includes the location of the call in your program,
7107 the arguments of the call,
7108 and the local variables of the function being called.
7109 The information is saved in a block of data called a @dfn{stack frame}.
7110 The stack frames are allocated in a region of memory called the @dfn{call
7113 When your program stops, the @value{GDBN} commands for examining the
7114 stack allow you to see all of this information.
7116 @cindex selected frame
7117 One of the stack frames is @dfn{selected} by @value{GDBN} and many
7118 @value{GDBN} commands refer implicitly to the selected frame. In
7119 particular, whenever you ask @value{GDBN} for the value of a variable in
7120 your program, the value is found in the selected frame. There are
7121 special @value{GDBN} commands to select whichever frame you are
7122 interested in. @xref{Selection, ,Selecting a Frame}.
7124 When your program stops, @value{GDBN} automatically selects the
7125 currently executing frame and describes it briefly, similar to the
7126 @code{frame} command (@pxref{Frame Info, ,Information about a Frame}).
7129 * Frames:: Stack frames
7130 * Backtrace:: Backtraces
7131 * Selection:: Selecting a frame
7132 * Frame Info:: Information on a frame
7133 * Frame Filter Management:: Managing frame filters
7138 @section Stack Frames
7140 @cindex frame, definition
7142 The call stack is divided up into contiguous pieces called @dfn{stack
7143 frames}, or @dfn{frames} for short; each frame is the data associated
7144 with one call to one function. The frame contains the arguments given
7145 to the function, the function's local variables, and the address at
7146 which the function is executing.
7148 @cindex initial frame
7149 @cindex outermost frame
7150 @cindex innermost frame
7151 When your program is started, the stack has only one frame, that of the
7152 function @code{main}. This is called the @dfn{initial} frame or the
7153 @dfn{outermost} frame. Each time a function is called, a new frame is
7154 made. Each time a function returns, the frame for that function invocation
7155 is eliminated. If a function is recursive, there can be many frames for
7156 the same function. The frame for the function in which execution is
7157 actually occurring is called the @dfn{innermost} frame. This is the most
7158 recently created of all the stack frames that still exist.
7160 @cindex frame pointer
7161 Inside your program, stack frames are identified by their addresses. A
7162 stack frame consists of many bytes, each of which has its own address; each
7163 kind of computer has a convention for choosing one byte whose
7164 address serves as the address of the frame. Usually this address is kept
7165 in a register called the @dfn{frame pointer register}
7166 (@pxref{Registers, $fp}) while execution is going on in that frame.
7168 @cindex frame number
7169 @value{GDBN} assigns numbers to all existing stack frames, starting with
7170 zero for the innermost frame, one for the frame that called it,
7171 and so on upward. These numbers do not really exist in your program;
7172 they are assigned by @value{GDBN} to give you a way of designating stack
7173 frames in @value{GDBN} commands.
7175 @c The -fomit-frame-pointer below perennially causes hbox overflow
7176 @c underflow problems.
7177 @cindex frameless execution
7178 Some compilers provide a way to compile functions so that they operate
7179 without stack frames. (For example, the @value{NGCC} option
7181 @samp{-fomit-frame-pointer}
7183 generates functions without a frame.)
7184 This is occasionally done with heavily used library functions to save
7185 the frame setup time. @value{GDBN} has limited facilities for dealing
7186 with these function invocations. If the innermost function invocation
7187 has no stack frame, @value{GDBN} nevertheless regards it as though
7188 it had a separate frame, which is numbered zero as usual, allowing
7189 correct tracing of the function call chain. However, @value{GDBN} has
7190 no provision for frameless functions elsewhere in the stack.
7196 @cindex call stack traces
7197 A backtrace is a summary of how your program got where it is. It shows one
7198 line per frame, for many frames, starting with the currently executing
7199 frame (frame zero), followed by its caller (frame one), and on up the
7202 @anchor{backtrace-command}
7205 @kindex bt @r{(@code{backtrace})}
7208 Print a backtrace of the entire stack: one line per frame for all
7209 frames in the stack.
7211 You can stop the backtrace at any time by typing the system interrupt
7212 character, normally @kbd{Ctrl-c}.
7214 @item backtrace @var{n}
7216 Similar, but print only the innermost @var{n} frames.
7218 @item backtrace -@var{n}
7220 Similar, but print only the outermost @var{n} frames.
7222 @item backtrace full
7224 @itemx bt full @var{n}
7225 @itemx bt full -@var{n}
7226 Print the values of the local variables also. As described above,
7227 @var{n} specifies the number of frames to print.
7229 @item backtrace no-filters
7230 @itemx bt no-filters
7231 @itemx bt no-filters @var{n}
7232 @itemx bt no-filters -@var{n}
7233 @itemx bt no-filters full
7234 @itemx bt no-filters full @var{n}
7235 @itemx bt no-filters full -@var{n}
7236 Do not run Python frame filters on this backtrace. @xref{Frame
7237 Filter API}, for more information. Additionally use @ref{disable
7238 frame-filter all} to turn off all frame filters. This is only
7239 relevant when @value{GDBN} has been configured with @code{Python}
7245 The names @code{where} and @code{info stack} (abbreviated @code{info s})
7246 are additional aliases for @code{backtrace}.
7248 @cindex multiple threads, backtrace
7249 In a multi-threaded program, @value{GDBN} by default shows the
7250 backtrace only for the current thread. To display the backtrace for
7251 several or all of the threads, use the command @code{thread apply}
7252 (@pxref{Threads, thread apply}). For example, if you type @kbd{thread
7253 apply all backtrace}, @value{GDBN} will display the backtrace for all
7254 the threads; this is handy when you debug a core dump of a
7255 multi-threaded program.
7257 Each line in the backtrace shows the frame number and the function name.
7258 The program counter value is also shown---unless you use @code{set
7259 print address off}. The backtrace also shows the source file name and
7260 line number, as well as the arguments to the function. The program
7261 counter value is omitted if it is at the beginning of the code for that
7264 Here is an example of a backtrace. It was made with the command
7265 @samp{bt 3}, so it shows the innermost three frames.
7269 #0 m4_traceon (obs=0x24eb0, argc=1, argv=0x2b8c8)
7271 #1 0x6e38 in expand_macro (sym=0x2b600, data=...) at macro.c:242
7272 #2 0x6840 in expand_token (obs=0x0, t=177664, td=0xf7fffb08)
7274 (More stack frames follow...)
7279 The display for frame zero does not begin with a program counter
7280 value, indicating that your program has stopped at the beginning of the
7281 code for line @code{993} of @code{builtin.c}.
7284 The value of parameter @code{data} in frame 1 has been replaced by
7285 @code{@dots{}}. By default, @value{GDBN} prints the value of a parameter
7286 only if it is a scalar (integer, pointer, enumeration, etc). See command
7287 @kbd{set print frame-arguments} in @ref{Print Settings} for more details
7288 on how to configure the way function parameter values are printed.
7290 @cindex optimized out, in backtrace
7291 @cindex function call arguments, optimized out
7292 If your program was compiled with optimizations, some compilers will
7293 optimize away arguments passed to functions if those arguments are
7294 never used after the call. Such optimizations generate code that
7295 passes arguments through registers, but doesn't store those arguments
7296 in the stack frame. @value{GDBN} has no way of displaying such
7297 arguments in stack frames other than the innermost one. Here's what
7298 such a backtrace might look like:
7302 #0 m4_traceon (obs=0x24eb0, argc=1, argv=0x2b8c8)
7304 #1 0x6e38 in expand_macro (sym=<optimized out>) at macro.c:242
7305 #2 0x6840 in expand_token (obs=0x0, t=<optimized out>, td=0xf7fffb08)
7307 (More stack frames follow...)
7312 The values of arguments that were not saved in their stack frames are
7313 shown as @samp{<optimized out>}.
7315 If you need to display the values of such optimized-out arguments,
7316 either deduce that from other variables whose values depend on the one
7317 you are interested in, or recompile without optimizations.
7319 @cindex backtrace beyond @code{main} function
7320 @cindex program entry point
7321 @cindex startup code, and backtrace
7322 Most programs have a standard user entry point---a place where system
7323 libraries and startup code transition into user code. For C this is
7324 @code{main}@footnote{
7325 Note that embedded programs (the so-called ``free-standing''
7326 environment) are not required to have a @code{main} function as the
7327 entry point. They could even have multiple entry points.}.
7328 When @value{GDBN} finds the entry function in a backtrace
7329 it will terminate the backtrace, to avoid tracing into highly
7330 system-specific (and generally uninteresting) code.
7332 If you need to examine the startup code, or limit the number of levels
7333 in a backtrace, you can change this behavior:
7336 @item set backtrace past-main
7337 @itemx set backtrace past-main on
7338 @kindex set backtrace
7339 Backtraces will continue past the user entry point.
7341 @item set backtrace past-main off
7342 Backtraces will stop when they encounter the user entry point. This is the
7345 @item show backtrace past-main
7346 @kindex show backtrace
7347 Display the current user entry point backtrace policy.
7349 @item set backtrace past-entry
7350 @itemx set backtrace past-entry on
7351 Backtraces will continue past the internal entry point of an application.
7352 This entry point is encoded by the linker when the application is built,
7353 and is likely before the user entry point @code{main} (or equivalent) is called.
7355 @item set backtrace past-entry off
7356 Backtraces will stop when they encounter the internal entry point of an
7357 application. This is the default.
7359 @item show backtrace past-entry
7360 Display the current internal entry point backtrace policy.
7362 @item set backtrace limit @var{n}
7363 @itemx set backtrace limit 0
7364 @itemx set backtrace limit unlimited
7365 @cindex backtrace limit
7366 Limit the backtrace to @var{n} levels. A value of @code{unlimited}
7367 or zero means unlimited levels.
7369 @item show backtrace limit
7370 Display the current limit on backtrace levels.
7373 You can control how file names are displayed.
7376 @item set filename-display
7377 @itemx set filename-display relative
7378 @cindex filename-display
7379 Display file names relative to the compilation directory. This is the default.
7381 @item set filename-display basename
7382 Display only basename of a filename.
7384 @item set filename-display absolute
7385 Display an absolute filename.
7387 @item show filename-display
7388 Show the current way to display filenames.
7392 @section Selecting a Frame
7394 Most commands for examining the stack and other data in your program work on
7395 whichever stack frame is selected at the moment. Here are the commands for
7396 selecting a stack frame; all of them finish by printing a brief description
7397 of the stack frame just selected.
7400 @kindex frame@r{, selecting}
7401 @kindex f @r{(@code{frame})}
7404 Select frame number @var{n}. Recall that frame zero is the innermost
7405 (currently executing) frame, frame one is the frame that called the
7406 innermost one, and so on. The highest-numbered frame is the one for
7409 @item frame @var{stack-addr} [ @var{pc-addr} ]
7410 @itemx f @var{stack-addr} [ @var{pc-addr} ]
7411 Select the frame at address @var{stack-addr}. This is useful mainly if the
7412 chaining of stack frames has been damaged by a bug, making it
7413 impossible for @value{GDBN} to assign numbers properly to all frames. In
7414 addition, this can be useful when your program has multiple stacks and
7415 switches between them. The optional @var{pc-addr} can also be given to
7416 specify the value of PC for the stack frame.
7420 Move @var{n} frames up the stack; @var{n} defaults to 1. For positive
7421 numbers @var{n}, this advances toward the outermost frame, to higher
7422 frame numbers, to frames that have existed longer.
7425 @kindex do @r{(@code{down})}
7427 Move @var{n} frames down the stack; @var{n} defaults to 1. For
7428 positive numbers @var{n}, this advances toward the innermost frame, to
7429 lower frame numbers, to frames that were created more recently.
7430 You may abbreviate @code{down} as @code{do}.
7433 All of these commands end by printing two lines of output describing the
7434 frame. The first line shows the frame number, the function name, the
7435 arguments, and the source file and line number of execution in that
7436 frame. The second line shows the text of that source line.
7444 #1 0x22f0 in main (argc=1, argv=0xf7fffbf4, env=0xf7fffbfc)
7446 10 read_input_file (argv[i]);
7450 After such a printout, the @code{list} command with no arguments
7451 prints ten lines centered on the point of execution in the frame.
7452 You can also edit the program at the point of execution with your favorite
7453 editing program by typing @code{edit}.
7454 @xref{List, ,Printing Source Lines},
7458 @kindex select-frame
7460 The @code{select-frame} command is a variant of @code{frame} that does
7461 not display the new frame after selecting it. This command is
7462 intended primarily for use in @value{GDBN} command scripts, where the
7463 output might be unnecessary and distracting.
7465 @kindex down-silently
7467 @item up-silently @var{n}
7468 @itemx down-silently @var{n}
7469 These two commands are variants of @code{up} and @code{down},
7470 respectively; they differ in that they do their work silently, without
7471 causing display of the new frame. They are intended primarily for use
7472 in @value{GDBN} command scripts, where the output might be unnecessary and
7477 @section Information About a Frame
7479 There are several other commands to print information about the selected
7485 When used without any argument, this command does not change which
7486 frame is selected, but prints a brief description of the currently
7487 selected stack frame. It can be abbreviated @code{f}. With an
7488 argument, this command is used to select a stack frame.
7489 @xref{Selection, ,Selecting a Frame}.
7492 @kindex info f @r{(@code{info frame})}
7495 This command prints a verbose description of the selected stack frame,
7500 the address of the frame
7502 the address of the next frame down (called by this frame)
7504 the address of the next frame up (caller of this frame)
7506 the language in which the source code corresponding to this frame is written
7508 the address of the frame's arguments
7510 the address of the frame's local variables
7512 the program counter saved in it (the address of execution in the caller frame)
7514 which registers were saved in the frame
7517 @noindent The verbose description is useful when
7518 something has gone wrong that has made the stack format fail to fit
7519 the usual conventions.
7521 @item info frame @var{addr}
7522 @itemx info f @var{addr}
7523 Print a verbose description of the frame at address @var{addr}, without
7524 selecting that frame. The selected frame remains unchanged by this
7525 command. This requires the same kind of address (more than one for some
7526 architectures) that you specify in the @code{frame} command.
7527 @xref{Selection, ,Selecting a Frame}.
7531 Print the arguments of the selected frame, each on a separate line.
7535 Print the local variables of the selected frame, each on a separate
7536 line. These are all variables (declared either static or automatic)
7537 accessible at the point of execution of the selected frame.
7541 @node Frame Filter Management
7542 @section Management of Frame Filters.
7543 @cindex managing frame filters
7545 Frame filters are Python based utilities to manage and decorate the
7546 output of frames. @xref{Frame Filter API}, for further information.
7548 Managing frame filters is performed by several commands available
7549 within @value{GDBN}, detailed here.
7552 @kindex info frame-filter
7553 @item info frame-filter
7554 Print a list of installed frame filters from all dictionaries, showing
7555 their name, priority and enabled status.
7557 @kindex disable frame-filter
7558 @anchor{disable frame-filter all}
7559 @item disable frame-filter @var{filter-dictionary} @var{filter-name}
7560 Disable a frame filter in the dictionary matching
7561 @var{filter-dictionary} and @var{filter-name}. The
7562 @var{filter-dictionary} may be @code{all}, @code{global},
7563 @code{progspace}, or the name of the object file where the frame filter
7564 dictionary resides. When @code{all} is specified, all frame filters
7565 across all dictionaries are disabled. The @var{filter-name} is the name
7566 of the frame filter and is used when @code{all} is not the option for
7567 @var{filter-dictionary}. A disabled frame-filter is not deleted, it
7568 may be enabled again later.
7570 @kindex enable frame-filter
7571 @item enable frame-filter @var{filter-dictionary} @var{filter-name}
7572 Enable a frame filter in the dictionary matching
7573 @var{filter-dictionary} and @var{filter-name}. The
7574 @var{filter-dictionary} may be @code{all}, @code{global},
7575 @code{progspace} or the name of the object file where the frame filter
7576 dictionary resides. When @code{all} is specified, all frame filters across
7577 all dictionaries are enabled. The @var{filter-name} is the name of the frame
7578 filter and is used when @code{all} is not the option for
7579 @var{filter-dictionary}.
7584 (gdb) info frame-filter
7586 global frame-filters:
7587 Priority Enabled Name
7588 1000 No PrimaryFunctionFilter
7591 progspace /build/test frame-filters:
7592 Priority Enabled Name
7593 100 Yes ProgspaceFilter
7595 objfile /build/test frame-filters:
7596 Priority Enabled Name
7597 999 Yes BuildProgra Filter
7599 (gdb) disable frame-filter /build/test BuildProgramFilter
7600 (gdb) info frame-filter
7602 global frame-filters:
7603 Priority Enabled Name
7604 1000 No PrimaryFunctionFilter
7607 progspace /build/test frame-filters:
7608 Priority Enabled Name
7609 100 Yes ProgspaceFilter
7611 objfile /build/test frame-filters:
7612 Priority Enabled Name
7613 999 No BuildProgramFilter
7615 (gdb) enable frame-filter global PrimaryFunctionFilter
7616 (gdb) info frame-filter
7618 global frame-filters:
7619 Priority Enabled Name
7620 1000 Yes PrimaryFunctionFilter
7623 progspace /build/test frame-filters:
7624 Priority Enabled Name
7625 100 Yes ProgspaceFilter
7627 objfile /build/test frame-filters:
7628 Priority Enabled Name
7629 999 No BuildProgramFilter
7632 @kindex set frame-filter priority
7633 @item set frame-filter priority @var{filter-dictionary} @var{filter-name} @var{priority}
7634 Set the @var{priority} of a frame filter in the dictionary matching
7635 @var{filter-dictionary}, and the frame filter name matching
7636 @var{filter-name}. The @var{filter-dictionary} may be @code{global},
7637 @code{progspace} or the name of the object file where the frame filter
7638 dictionary resides. The @var{priority} is an integer.
7640 @kindex show frame-filter priority
7641 @item show frame-filter priority @var{filter-dictionary} @var{filter-name}
7642 Show the @var{priority} of a frame filter in the dictionary matching
7643 @var{filter-dictionary}, and the frame filter name matching
7644 @var{filter-name}. The @var{filter-dictionary} may be @code{global},
7645 @code{progspace} or the name of the object file where the frame filter
7651 (gdb) info frame-filter
7653 global frame-filters:
7654 Priority Enabled Name
7655 1000 Yes PrimaryFunctionFilter
7658 progspace /build/test frame-filters:
7659 Priority Enabled Name
7660 100 Yes ProgspaceFilter
7662 objfile /build/test frame-filters:
7663 Priority Enabled Name
7664 999 No BuildProgramFilter
7666 (gdb) set frame-filter priority global Reverse 50
7667 (gdb) info frame-filter
7669 global frame-filters:
7670 Priority Enabled Name
7671 1000 Yes PrimaryFunctionFilter
7674 progspace /build/test frame-filters:
7675 Priority Enabled Name
7676 100 Yes ProgspaceFilter
7678 objfile /build/test frame-filters:
7679 Priority Enabled Name
7680 999 No BuildProgramFilter
7685 @chapter Examining Source Files
7687 @value{GDBN} can print parts of your program's source, since the debugging
7688 information recorded in the program tells @value{GDBN} what source files were
7689 used to build it. When your program stops, @value{GDBN} spontaneously prints
7690 the line where it stopped. Likewise, when you select a stack frame
7691 (@pxref{Selection, ,Selecting a Frame}), @value{GDBN} prints the line where
7692 execution in that frame has stopped. You can print other portions of
7693 source files by explicit command.
7695 If you use @value{GDBN} through its @sc{gnu} Emacs interface, you may
7696 prefer to use Emacs facilities to view source; see @ref{Emacs, ,Using
7697 @value{GDBN} under @sc{gnu} Emacs}.
7700 * List:: Printing source lines
7701 * Specify Location:: How to specify code locations
7702 * Edit:: Editing source files
7703 * Search:: Searching source files
7704 * Source Path:: Specifying source directories
7705 * Machine Code:: Source and machine code
7709 @section Printing Source Lines
7712 @kindex l @r{(@code{list})}
7713 To print lines from a source file, use the @code{list} command
7714 (abbreviated @code{l}). By default, ten lines are printed.
7715 There are several ways to specify what part of the file you want to
7716 print; see @ref{Specify Location}, for the full list.
7718 Here are the forms of the @code{list} command most commonly used:
7721 @item list @var{linenum}
7722 Print lines centered around line number @var{linenum} in the
7723 current source file.
7725 @item list @var{function}
7726 Print lines centered around the beginning of function
7730 Print more lines. If the last lines printed were printed with a
7731 @code{list} command, this prints lines following the last lines
7732 printed; however, if the last line printed was a solitary line printed
7733 as part of displaying a stack frame (@pxref{Stack, ,Examining the
7734 Stack}), this prints lines centered around that line.
7737 Print lines just before the lines last printed.
7740 @cindex @code{list}, how many lines to display
7741 By default, @value{GDBN} prints ten source lines with any of these forms of
7742 the @code{list} command. You can change this using @code{set listsize}:
7745 @kindex set listsize
7746 @item set listsize @var{count}
7747 @itemx set listsize unlimited
7748 Make the @code{list} command display @var{count} source lines (unless
7749 the @code{list} argument explicitly specifies some other number).
7750 Setting @var{count} to @code{unlimited} or 0 means there's no limit.
7752 @kindex show listsize
7754 Display the number of lines that @code{list} prints.
7757 Repeating a @code{list} command with @key{RET} discards the argument,
7758 so it is equivalent to typing just @code{list}. This is more useful
7759 than listing the same lines again. An exception is made for an
7760 argument of @samp{-}; that argument is preserved in repetition so that
7761 each repetition moves up in the source file.
7763 In general, the @code{list} command expects you to supply zero, one or two
7764 @dfn{locations}. Locations specify source lines; there are several ways
7765 of writing them (@pxref{Specify Location}), but the effect is always
7766 to specify some source line.
7768 Here is a complete description of the possible arguments for @code{list}:
7771 @item list @var{location}
7772 Print lines centered around the line specified by @var{location}.
7774 @item list @var{first},@var{last}
7775 Print lines from @var{first} to @var{last}. Both arguments are
7776 locations. When a @code{list} command has two locations, and the
7777 source file of the second location is omitted, this refers to
7778 the same source file as the first location.
7780 @item list ,@var{last}
7781 Print lines ending with @var{last}.
7783 @item list @var{first},
7784 Print lines starting with @var{first}.
7787 Print lines just after the lines last printed.
7790 Print lines just before the lines last printed.
7793 As described in the preceding table.
7796 @node Specify Location
7797 @section Specifying a Location
7798 @cindex specifying location
7800 @cindex source location
7803 * Linespec Locations:: Linespec locations
7804 * Explicit Locations:: Explicit locations
7805 * Address Locations:: Address locations
7808 Several @value{GDBN} commands accept arguments that specify a location
7809 of your program's code. Since @value{GDBN} is a source-level
7810 debugger, a location usually specifies some line in the source code.
7811 Locations may be specified using three different formats:
7812 linespec locations, explicit locations, or address locations.
7814 @node Linespec Locations
7815 @subsection Linespec Locations
7816 @cindex linespec locations
7818 A @dfn{linespec} is a colon-separated list of source location parameters such
7819 as file name, function name, etc. Here are all the different ways of
7820 specifying a linespec:
7824 Specifies the line number @var{linenum} of the current source file.
7827 @itemx +@var{offset}
7828 Specifies the line @var{offset} lines before or after the @dfn{current
7829 line}. For the @code{list} command, the current line is the last one
7830 printed; for the breakpoint commands, this is the line at which
7831 execution stopped in the currently selected @dfn{stack frame}
7832 (@pxref{Frames, ,Frames}, for a description of stack frames.) When
7833 used as the second of the two linespecs in a @code{list} command,
7834 this specifies the line @var{offset} lines up or down from the first
7837 @item @var{filename}:@var{linenum}
7838 Specifies the line @var{linenum} in the source file @var{filename}.
7839 If @var{filename} is a relative file name, then it will match any
7840 source file name with the same trailing components. For example, if
7841 @var{filename} is @samp{gcc/expr.c}, then it will match source file
7842 name of @file{/build/trunk/gcc/expr.c}, but not
7843 @file{/build/trunk/libcpp/expr.c} or @file{/build/trunk/gcc/x-expr.c}.
7845 @item @var{function}
7846 Specifies the line that begins the body of the function @var{function}.
7847 For example, in C, this is the line with the open brace.
7849 @item @var{function}:@var{label}
7850 Specifies the line where @var{label} appears in @var{function}.
7852 @item @var{filename}:@var{function}
7853 Specifies the line that begins the body of the function @var{function}
7854 in the file @var{filename}. You only need the file name with a
7855 function name to avoid ambiguity when there are identically named
7856 functions in different source files.
7859 Specifies the line at which the label named @var{label} appears
7860 in the function corresponding to the currently selected stack frame.
7861 If there is no current selected stack frame (for instance, if the inferior
7862 is not running), then @value{GDBN} will not search for a label.
7864 @cindex breakpoint at static probe point
7865 @item -pstap|-probe-stap @r{[}@var{objfile}:@r{[}@var{provider}:@r{]}@r{]}@var{name}
7866 The @sc{gnu}/Linux tool @code{SystemTap} provides a way for
7867 applications to embed static probes. @xref{Static Probe Points}, for more
7868 information on finding and using static probes. This form of linespec
7869 specifies the location of such a static probe.
7871 If @var{objfile} is given, only probes coming from that shared library
7872 or executable matching @var{objfile} as a regular expression are considered.
7873 If @var{provider} is given, then only probes from that provider are considered.
7874 If several probes match the spec, @value{GDBN} will insert a breakpoint at
7875 each one of those probes.
7878 @node Explicit Locations
7879 @subsection Explicit Locations
7880 @cindex explicit locations
7882 @dfn{Explicit locations} allow the user to directly specify the source
7883 location's parameters using option-value pairs.
7885 Explicit locations are useful when several functions, labels, or
7886 file names have the same name (base name for files) in the program's
7887 sources. In these cases, explicit locations point to the source
7888 line you meant more accurately and unambiguously. Also, using
7889 explicit locations might be faster in large programs.
7891 For example, the linespec @samp{foo:bar} may refer to a function @code{bar}
7892 defined in the file named @file{foo} or the label @code{bar} in a function
7893 named @code{foo}. @value{GDBN} must search either the file system or
7894 the symbol table to know.
7896 The list of valid explicit location options is summarized in the
7900 @item -source @var{filename}
7901 The value specifies the source file name. To differentiate between
7902 files with the same base name, prepend as many directories as is necessary
7903 to uniquely identify the desired file, e.g., @file{foo/bar/baz.c}. Otherwise
7904 @value{GDBN} will use the first file it finds with the given base
7905 name. This option requires the use of either @code{-function} or @code{-line}.
7907 @item -function @var{function}
7908 The value specifies the name of a function. Operations
7909 on function locations unmodified by other options (such as @code{-label}
7910 or @code{-line}) refer to the line that begins the body of the function.
7911 In C, for example, this is the line with the open brace.
7913 @item -label @var{label}
7914 The value specifies the name of a label. When the function
7915 name is not specified, the label is searched in the function of the currently
7916 selected stack frame.
7918 @item -line @var{number}
7919 The value specifies a line offset for the location. The offset may either
7920 be absolute (@code{-line 3}) or relative (@code{-line +3}), depending on
7921 the command. When specified without any other options, the line offset is
7922 relative to the current line.
7925 Explicit location options may be abbreviated by omitting any non-unique
7926 trailing characters from the option name, e.g., @code{break -s main.c -li 3}.
7928 @node Address Locations
7929 @subsection Address Locations
7930 @cindex address locations
7932 @dfn{Address locations} indicate a specific program address. They have
7933 the generalized form *@var{address}.
7935 For line-oriented commands, such as @code{list} and @code{edit}, this
7936 specifies a source line that contains @var{address}. For @code{break} and
7937 other breakpoint-oriented commands, this can be used to set breakpoints in
7938 parts of your program which do not have debugging information or
7941 Here @var{address} may be any expression valid in the current working
7942 language (@pxref{Languages, working language}) that specifies a code
7943 address. In addition, as a convenience, @value{GDBN} extends the
7944 semantics of expressions used in locations to cover several situations
7945 that frequently occur during debugging. Here are the various forms
7949 @item @var{expression}
7950 Any expression valid in the current working language.
7952 @item @var{funcaddr}
7953 An address of a function or procedure derived from its name. In C,
7954 C@t{++}, Objective-C, Fortran, minimal, and assembly, this is
7955 simply the function's name @var{function} (and actually a special case
7956 of a valid expression). In Pascal and Modula-2, this is
7957 @code{&@var{function}}. In Ada, this is @code{@var{function}'Address}
7958 (although the Pascal form also works).
7960 This form specifies the address of the function's first instruction,
7961 before the stack frame and arguments have been set up.
7963 @item '@var{filename}':@var{funcaddr}
7964 Like @var{funcaddr} above, but also specifies the name of the source
7965 file explicitly. This is useful if the name of the function does not
7966 specify the function unambiguously, e.g., if there are several
7967 functions with identical names in different source files.
7971 @section Editing Source Files
7972 @cindex editing source files
7975 @kindex e @r{(@code{edit})}
7976 To edit the lines in a source file, use the @code{edit} command.
7977 The editing program of your choice
7978 is invoked with the current line set to
7979 the active line in the program.
7980 Alternatively, there are several ways to specify what part of the file you
7981 want to print if you want to see other parts of the program:
7984 @item edit @var{location}
7985 Edit the source file specified by @code{location}. Editing starts at
7986 that @var{location}, e.g., at the specified source line of the
7987 specified file. @xref{Specify Location}, for all the possible forms
7988 of the @var{location} argument; here are the forms of the @code{edit}
7989 command most commonly used:
7992 @item edit @var{number}
7993 Edit the current source file with @var{number} as the active line number.
7995 @item edit @var{function}
7996 Edit the file containing @var{function} at the beginning of its definition.
8001 @subsection Choosing your Editor
8002 You can customize @value{GDBN} to use any editor you want
8004 The only restriction is that your editor (say @code{ex}), recognizes the
8005 following command-line syntax:
8007 ex +@var{number} file
8009 The optional numeric value +@var{number} specifies the number of the line in
8010 the file where to start editing.}.
8011 By default, it is @file{@value{EDITOR}}, but you can change this
8012 by setting the environment variable @code{EDITOR} before using
8013 @value{GDBN}. For example, to configure @value{GDBN} to use the
8014 @code{vi} editor, you could use these commands with the @code{sh} shell:
8020 or in the @code{csh} shell,
8022 setenv EDITOR /usr/bin/vi
8027 @section Searching Source Files
8028 @cindex searching source files
8030 There are two commands for searching through the current source file for a
8035 @kindex forward-search
8036 @kindex fo @r{(@code{forward-search})}
8037 @item forward-search @var{regexp}
8038 @itemx search @var{regexp}
8039 The command @samp{forward-search @var{regexp}} checks each line,
8040 starting with the one following the last line listed, for a match for
8041 @var{regexp}. It lists the line that is found. You can use the
8042 synonym @samp{search @var{regexp}} or abbreviate the command name as
8045 @kindex reverse-search
8046 @item reverse-search @var{regexp}
8047 The command @samp{reverse-search @var{regexp}} checks each line, starting
8048 with the one before the last line listed and going backward, for a match
8049 for @var{regexp}. It lists the line that is found. You can abbreviate
8050 this command as @code{rev}.
8054 @section Specifying Source Directories
8057 @cindex directories for source files
8058 Executable programs sometimes do not record the directories of the source
8059 files from which they were compiled, just the names. Even when they do,
8060 the directories could be moved between the compilation and your debugging
8061 session. @value{GDBN} has a list of directories to search for source files;
8062 this is called the @dfn{source path}. Each time @value{GDBN} wants a source file,
8063 it tries all the directories in the list, in the order they are present
8064 in the list, until it finds a file with the desired name.
8066 For example, suppose an executable references the file
8067 @file{/usr/src/foo-1.0/lib/foo.c}, and our source path is
8068 @file{/mnt/cross}. The file is first looked up literally; if this
8069 fails, @file{/mnt/cross/usr/src/foo-1.0/lib/foo.c} is tried; if this
8070 fails, @file{/mnt/cross/foo.c} is opened; if this fails, an error
8071 message is printed. @value{GDBN} does not look up the parts of the
8072 source file name, such as @file{/mnt/cross/src/foo-1.0/lib/foo.c}.
8073 Likewise, the subdirectories of the source path are not searched: if
8074 the source path is @file{/mnt/cross}, and the binary refers to
8075 @file{foo.c}, @value{GDBN} would not find it under
8076 @file{/mnt/cross/usr/src/foo-1.0/lib}.
8078 Plain file names, relative file names with leading directories, file
8079 names containing dots, etc.@: are all treated as described above; for
8080 instance, if the source path is @file{/mnt/cross}, and the source file
8081 is recorded as @file{../lib/foo.c}, @value{GDBN} would first try
8082 @file{../lib/foo.c}, then @file{/mnt/cross/../lib/foo.c}, and after
8083 that---@file{/mnt/cross/foo.c}.
8085 Note that the executable search path is @emph{not} used to locate the
8088 Whenever you reset or rearrange the source path, @value{GDBN} clears out
8089 any information it has cached about where source files are found and where
8090 each line is in the file.
8094 When you start @value{GDBN}, its source path includes only @samp{cdir}
8095 and @samp{cwd}, in that order.
8096 To add other directories, use the @code{directory} command.
8098 The search path is used to find both program source files and @value{GDBN}
8099 script files (read using the @samp{-command} option and @samp{source} command).
8101 In addition to the source path, @value{GDBN} provides a set of commands
8102 that manage a list of source path substitution rules. A @dfn{substitution
8103 rule} specifies how to rewrite source directories stored in the program's
8104 debug information in case the sources were moved to a different
8105 directory between compilation and debugging. A rule is made of
8106 two strings, the first specifying what needs to be rewritten in
8107 the path, and the second specifying how it should be rewritten.
8108 In @ref{set substitute-path}, we name these two parts @var{from} and
8109 @var{to} respectively. @value{GDBN} does a simple string replacement
8110 of @var{from} with @var{to} at the start of the directory part of the
8111 source file name, and uses that result instead of the original file
8112 name to look up the sources.
8114 Using the previous example, suppose the @file{foo-1.0} tree has been
8115 moved from @file{/usr/src} to @file{/mnt/cross}, then you can tell
8116 @value{GDBN} to replace @file{/usr/src} in all source path names with
8117 @file{/mnt/cross}. The first lookup will then be
8118 @file{/mnt/cross/foo-1.0/lib/foo.c} in place of the original location
8119 of @file{/usr/src/foo-1.0/lib/foo.c}. To define a source path
8120 substitution rule, use the @code{set substitute-path} command
8121 (@pxref{set substitute-path}).
8123 To avoid unexpected substitution results, a rule is applied only if the
8124 @var{from} part of the directory name ends at a directory separator.
8125 For instance, a rule substituting @file{/usr/source} into
8126 @file{/mnt/cross} will be applied to @file{/usr/source/foo-1.0} but
8127 not to @file{/usr/sourceware/foo-2.0}. And because the substitution
8128 is applied only at the beginning of the directory name, this rule will
8129 not be applied to @file{/root/usr/source/baz.c} either.
8131 In many cases, you can achieve the same result using the @code{directory}
8132 command. However, @code{set substitute-path} can be more efficient in
8133 the case where the sources are organized in a complex tree with multiple
8134 subdirectories. With the @code{directory} command, you need to add each
8135 subdirectory of your project. If you moved the entire tree while
8136 preserving its internal organization, then @code{set substitute-path}
8137 allows you to direct the debugger to all the sources with one single
8140 @code{set substitute-path} is also more than just a shortcut command.
8141 The source path is only used if the file at the original location no
8142 longer exists. On the other hand, @code{set substitute-path} modifies
8143 the debugger behavior to look at the rewritten location instead. So, if
8144 for any reason a source file that is not relevant to your executable is
8145 located at the original location, a substitution rule is the only
8146 method available to point @value{GDBN} at the new location.
8148 @cindex @samp{--with-relocated-sources}
8149 @cindex default source path substitution
8150 You can configure a default source path substitution rule by
8151 configuring @value{GDBN} with the
8152 @samp{--with-relocated-sources=@var{dir}} option. The @var{dir}
8153 should be the name of a directory under @value{GDBN}'s configured
8154 prefix (set with @samp{--prefix} or @samp{--exec-prefix}), and
8155 directory names in debug information under @var{dir} will be adjusted
8156 automatically if the installed @value{GDBN} is moved to a new
8157 location. This is useful if @value{GDBN}, libraries or executables
8158 with debug information and corresponding source code are being moved
8162 @item directory @var{dirname} @dots{}
8163 @item dir @var{dirname} @dots{}
8164 Add directory @var{dirname} to the front of the source path. Several
8165 directory names may be given to this command, separated by @samp{:}
8166 (@samp{;} on MS-DOS and MS-Windows, where @samp{:} usually appears as
8167 part of absolute file names) or
8168 whitespace. You may specify a directory that is already in the source
8169 path; this moves it forward, so @value{GDBN} searches it sooner.
8173 @vindex $cdir@r{, convenience variable}
8174 @vindex $cwd@r{, convenience variable}
8175 @cindex compilation directory
8176 @cindex current directory
8177 @cindex working directory
8178 @cindex directory, current
8179 @cindex directory, compilation
8180 You can use the string @samp{$cdir} to refer to the compilation
8181 directory (if one is recorded), and @samp{$cwd} to refer to the current
8182 working directory. @samp{$cwd} is not the same as @samp{.}---the former
8183 tracks the current working directory as it changes during your @value{GDBN}
8184 session, while the latter is immediately expanded to the current
8185 directory at the time you add an entry to the source path.
8188 Reset the source path to its default value (@samp{$cdir:$cwd} on Unix systems). This requires confirmation.
8190 @c RET-repeat for @code{directory} is explicitly disabled, but since
8191 @c repeating it would be a no-op we do not say that. (thanks to RMS)
8193 @item set directories @var{path-list}
8194 @kindex set directories
8195 Set the source path to @var{path-list}.
8196 @samp{$cdir:$cwd} are added if missing.
8198 @item show directories
8199 @kindex show directories
8200 Print the source path: show which directories it contains.
8202 @anchor{set substitute-path}
8203 @item set substitute-path @var{from} @var{to}
8204 @kindex set substitute-path
8205 Define a source path substitution rule, and add it at the end of the
8206 current list of existing substitution rules. If a rule with the same
8207 @var{from} was already defined, then the old rule is also deleted.
8209 For example, if the file @file{/foo/bar/baz.c} was moved to
8210 @file{/mnt/cross/baz.c}, then the command
8213 (@value{GDBP}) set substitute-path /foo/bar /mnt/cross
8217 will tell @value{GDBN} to replace @samp{/foo/bar} with
8218 @samp{/mnt/cross}, which will allow @value{GDBN} to find the file
8219 @file{baz.c} even though it was moved.
8221 In the case when more than one substitution rule have been defined,
8222 the rules are evaluated one by one in the order where they have been
8223 defined. The first one matching, if any, is selected to perform
8226 For instance, if we had entered the following commands:
8229 (@value{GDBP}) set substitute-path /usr/src/include /mnt/include
8230 (@value{GDBP}) set substitute-path /usr/src /mnt/src
8234 @value{GDBN} would then rewrite @file{/usr/src/include/defs.h} into
8235 @file{/mnt/include/defs.h} by using the first rule. However, it would
8236 use the second rule to rewrite @file{/usr/src/lib/foo.c} into
8237 @file{/mnt/src/lib/foo.c}.
8240 @item unset substitute-path [path]
8241 @kindex unset substitute-path
8242 If a path is specified, search the current list of substitution rules
8243 for a rule that would rewrite that path. Delete that rule if found.
8244 A warning is emitted by the debugger if no rule could be found.
8246 If no path is specified, then all substitution rules are deleted.
8248 @item show substitute-path [path]
8249 @kindex show substitute-path
8250 If a path is specified, then print the source path substitution rule
8251 which would rewrite that path, if any.
8253 If no path is specified, then print all existing source path substitution
8258 If your source path is cluttered with directories that are no longer of
8259 interest, @value{GDBN} may sometimes cause confusion by finding the wrong
8260 versions of source. You can correct the situation as follows:
8264 Use @code{directory} with no argument to reset the source path to its default value.
8267 Use @code{directory} with suitable arguments to reinstall the
8268 directories you want in the source path. You can add all the
8269 directories in one command.
8273 @section Source and Machine Code
8274 @cindex source line and its code address
8276 You can use the command @code{info line} to map source lines to program
8277 addresses (and vice versa), and the command @code{disassemble} to display
8278 a range of addresses as machine instructions. You can use the command
8279 @code{set disassemble-next-line} to set whether to disassemble next
8280 source line when execution stops. When run under @sc{gnu} Emacs
8281 mode, the @code{info line} command causes the arrow to point to the
8282 line specified. Also, @code{info line} prints addresses in symbolic form as
8287 @item info line @var{location}
8288 Print the starting and ending addresses of the compiled code for
8289 source line @var{location}. You can specify source lines in any of
8290 the ways documented in @ref{Specify Location}.
8293 For example, we can use @code{info line} to discover the location of
8294 the object code for the first line of function
8295 @code{m4_changequote}:
8297 @c FIXME: I think this example should also show the addresses in
8298 @c symbolic form, as they usually would be displayed.
8300 (@value{GDBP}) info line m4_changequote
8301 Line 895 of "builtin.c" starts at pc 0x634c and ends at 0x6350.
8305 @cindex code address and its source line
8306 We can also inquire (using @code{*@var{addr}} as the form for
8307 @var{location}) what source line covers a particular address:
8309 (@value{GDBP}) info line *0x63ff
8310 Line 926 of "builtin.c" starts at pc 0x63e4 and ends at 0x6404.
8313 @cindex @code{$_} and @code{info line}
8314 @cindex @code{x} command, default address
8315 @kindex x@r{(examine), and} info line
8316 After @code{info line}, the default address for the @code{x} command
8317 is changed to the starting address of the line, so that @samp{x/i} is
8318 sufficient to begin examining the machine code (@pxref{Memory,
8319 ,Examining Memory}). Also, this address is saved as the value of the
8320 convenience variable @code{$_} (@pxref{Convenience Vars, ,Convenience
8325 @cindex assembly instructions
8326 @cindex instructions, assembly
8327 @cindex machine instructions
8328 @cindex listing machine instructions
8330 @itemx disassemble /m
8331 @itemx disassemble /s
8332 @itemx disassemble /r
8333 This specialized command dumps a range of memory as machine
8334 instructions. It can also print mixed source+disassembly by specifying
8335 the @code{/m} or @code{/s} modifier and print the raw instructions in hex
8336 as well as in symbolic form by specifying the @code{/r} modifier.
8337 The default memory range is the function surrounding the
8338 program counter of the selected frame. A single argument to this
8339 command is a program counter value; @value{GDBN} dumps the function
8340 surrounding this value. When two arguments are given, they should
8341 be separated by a comma, possibly surrounded by whitespace. The
8342 arguments specify a range of addresses to dump, in one of two forms:
8345 @item @var{start},@var{end}
8346 the addresses from @var{start} (inclusive) to @var{end} (exclusive)
8347 @item @var{start},+@var{length}
8348 the addresses from @var{start} (inclusive) to
8349 @code{@var{start}+@var{length}} (exclusive).
8353 When 2 arguments are specified, the name of the function is also
8354 printed (since there could be several functions in the given range).
8356 The argument(s) can be any expression yielding a numeric value, such as
8357 @samp{0x32c4}, @samp{&main+10} or @samp{$pc - 8}.
8359 If the range of memory being disassembled contains current program counter,
8360 the instruction at that location is shown with a @code{=>} marker.
8363 The following example shows the disassembly of a range of addresses of
8364 HP PA-RISC 2.0 code:
8367 (@value{GDBP}) disas 0x32c4, 0x32e4
8368 Dump of assembler code from 0x32c4 to 0x32e4:
8369 0x32c4 <main+204>: addil 0,dp
8370 0x32c8 <main+208>: ldw 0x22c(sr0,r1),r26
8371 0x32cc <main+212>: ldil 0x3000,r31
8372 0x32d0 <main+216>: ble 0x3f8(sr4,r31)
8373 0x32d4 <main+220>: ldo 0(r31),rp
8374 0x32d8 <main+224>: addil -0x800,dp
8375 0x32dc <main+228>: ldo 0x588(r1),r26
8376 0x32e0 <main+232>: ldil 0x3000,r31
8377 End of assembler dump.
8380 Here is an example showing mixed source+assembly for Intel x86
8381 with @code{/m} or @code{/s}, when the program is stopped just after
8382 function prologue in a non-optimized function with no inline code.
8385 (@value{GDBP}) disas /m main
8386 Dump of assembler code for function main:
8388 0x08048330 <+0>: push %ebp
8389 0x08048331 <+1>: mov %esp,%ebp
8390 0x08048333 <+3>: sub $0x8,%esp
8391 0x08048336 <+6>: and $0xfffffff0,%esp
8392 0x08048339 <+9>: sub $0x10,%esp
8394 6 printf ("Hello.\n");
8395 => 0x0804833c <+12>: movl $0x8048440,(%esp)
8396 0x08048343 <+19>: call 0x8048284 <puts@@plt>
8400 0x08048348 <+24>: mov $0x0,%eax
8401 0x0804834d <+29>: leave
8402 0x0804834e <+30>: ret
8404 End of assembler dump.
8407 The @code{/m} option is deprecated as its output is not useful when
8408 there is either inlined code or re-ordered code.
8409 The @code{/s} option is the preferred choice.
8410 Here is an example for AMD x86-64 showing the difference between
8411 @code{/m} output and @code{/s} output.
8412 This example has one inline function defined in a header file,
8413 and the code is compiled with @samp{-O2} optimization.
8414 Note how the @code{/m} output is missing the disassembly of
8415 several instructions that are present in the @code{/s} output.
8445 (@value{GDBP}) disas /m main
8446 Dump of assembler code for function main:
8450 0x0000000000400400 <+0>: mov 0x200c2e(%rip),%eax # 0x601034 <y>
8451 0x0000000000400417 <+23>: mov %eax,0x200c13(%rip) # 0x601030 <x>
8455 0x000000000040041d <+29>: xor %eax,%eax
8456 0x000000000040041f <+31>: retq
8457 0x0000000000400420 <+32>: add %eax,%eax
8458 0x0000000000400422 <+34>: jmp 0x400417 <main+23>
8460 End of assembler dump.
8461 (@value{GDBP}) disas /s main
8462 Dump of assembler code for function main:
8466 0x0000000000400400 <+0>: mov 0x200c2e(%rip),%eax # 0x601034 <y>
8470 0x0000000000400406 <+6>: test %eax,%eax
8471 0x0000000000400408 <+8>: js 0x400420 <main+32>
8476 0x000000000040040a <+10>: lea 0xa(%rax),%edx
8477 0x000000000040040d <+13>: test %eax,%eax
8478 0x000000000040040f <+15>: mov $0x1,%eax
8479 0x0000000000400414 <+20>: cmovne %edx,%eax
8483 0x0000000000400417 <+23>: mov %eax,0x200c13(%rip) # 0x601030 <x>
8487 0x000000000040041d <+29>: xor %eax,%eax
8488 0x000000000040041f <+31>: retq
8492 0x0000000000400420 <+32>: add %eax,%eax
8493 0x0000000000400422 <+34>: jmp 0x400417 <main+23>
8494 End of assembler dump.
8497 Here is another example showing raw instructions in hex for AMD x86-64,
8500 (gdb) disas /r 0x400281,+10
8501 Dump of assembler code from 0x400281 to 0x40028b:
8502 0x0000000000400281: 38 36 cmp %dh,(%rsi)
8503 0x0000000000400283: 2d 36 34 2e 73 sub $0x732e3436,%eax
8504 0x0000000000400288: 6f outsl %ds:(%rsi),(%dx)
8505 0x0000000000400289: 2e 32 00 xor %cs:(%rax),%al
8506 End of assembler dump.
8509 Addresses cannot be specified as a location (@pxref{Specify Location}).
8510 So, for example, if you want to disassemble function @code{bar}
8511 in file @file{foo.c}, you must type @samp{disassemble 'foo.c'::bar}
8512 and not @samp{disassemble foo.c:bar}.
8514 Some architectures have more than one commonly-used set of instruction
8515 mnemonics or other syntax.
8517 For programs that were dynamically linked and use shared libraries,
8518 instructions that call functions or branch to locations in the shared
8519 libraries might show a seemingly bogus location---it's actually a
8520 location of the relocation table. On some architectures, @value{GDBN}
8521 might be able to resolve these to actual function names.
8524 @kindex set disassembler-options
8525 @cindex disassembler options
8526 @item set disassembler-options @var{option1}[,@var{option2}@dots{}]
8527 This command controls the passing of target specific information to
8528 the disassembler. For a list of valid options, please refer to the
8529 @code{-M}/@code{--disassembler-options} section of the @samp{objdump}
8530 manual and/or the output of @kbd{objdump --help}
8531 (@pxref{objdump,,objdump,binutils.info,The GNU Binary Utilities}).
8532 The default value is the empty string.
8534 If it is necessary to specify more than one disassembler option, then
8535 multiple options can be placed together into a comma separated list.
8536 Currently this command is only supported on targets ARM, PowerPC
8539 @kindex show disassembler-options
8540 @item show disassembler-options
8541 Show the current setting of the disassembler options.
8545 @kindex set disassembly-flavor
8546 @cindex Intel disassembly flavor
8547 @cindex AT&T disassembly flavor
8548 @item set disassembly-flavor @var{instruction-set}
8549 Select the instruction set to use when disassembling the
8550 program via the @code{disassemble} or @code{x/i} commands.
8552 Currently this command is only defined for the Intel x86 family. You
8553 can set @var{instruction-set} to either @code{intel} or @code{att}.
8554 The default is @code{att}, the AT&T flavor used by default by Unix
8555 assemblers for x86-based targets.
8557 @kindex show disassembly-flavor
8558 @item show disassembly-flavor
8559 Show the current setting of the disassembly flavor.
8563 @kindex set disassemble-next-line
8564 @kindex show disassemble-next-line
8565 @item set disassemble-next-line
8566 @itemx show disassemble-next-line
8567 Control whether or not @value{GDBN} will disassemble the next source
8568 line or instruction when execution stops. If ON, @value{GDBN} will
8569 display disassembly of the next source line when execution of the
8570 program being debugged stops. This is @emph{in addition} to
8571 displaying the source line itself, which @value{GDBN} always does if
8572 possible. If the next source line cannot be displayed for some reason
8573 (e.g., if @value{GDBN} cannot find the source file, or there's no line
8574 info in the debug info), @value{GDBN} will display disassembly of the
8575 next @emph{instruction} instead of showing the next source line. If
8576 AUTO, @value{GDBN} will display disassembly of next instruction only
8577 if the source line cannot be displayed. This setting causes
8578 @value{GDBN} to display some feedback when you step through a function
8579 with no line info or whose source file is unavailable. The default is
8580 OFF, which means never display the disassembly of the next line or
8586 @chapter Examining Data
8588 @cindex printing data
8589 @cindex examining data
8592 The usual way to examine data in your program is with the @code{print}
8593 command (abbreviated @code{p}), or its synonym @code{inspect}. It
8594 evaluates and prints the value of an expression of the language your
8595 program is written in (@pxref{Languages, ,Using @value{GDBN} with
8596 Different Languages}). It may also print the expression using a
8597 Python-based pretty-printer (@pxref{Pretty Printing}).
8600 @item print @var{expr}
8601 @itemx print /@var{f} @var{expr}
8602 @var{expr} is an expression (in the source language). By default the
8603 value of @var{expr} is printed in a format appropriate to its data type;
8604 you can choose a different format by specifying @samp{/@var{f}}, where
8605 @var{f} is a letter specifying the format; see @ref{Output Formats,,Output
8609 @itemx print /@var{f}
8610 @cindex reprint the last value
8611 If you omit @var{expr}, @value{GDBN} displays the last value again (from the
8612 @dfn{value history}; @pxref{Value History, ,Value History}). This allows you to
8613 conveniently inspect the same value in an alternative format.
8616 A more low-level way of examining data is with the @code{x} command.
8617 It examines data in memory at a specified address and prints it in a
8618 specified format. @xref{Memory, ,Examining Memory}.
8620 If you are interested in information about types, or about how the
8621 fields of a struct or a class are declared, use the @code{ptype @var{exp}}
8622 command rather than @code{print}. @xref{Symbols, ,Examining the Symbol
8625 @cindex exploring hierarchical data structures
8627 Another way of examining values of expressions and type information is
8628 through the Python extension command @code{explore} (available only if
8629 the @value{GDBN} build is configured with @code{--with-python}). It
8630 offers an interactive way to start at the highest level (or, the most
8631 abstract level) of the data type of an expression (or, the data type
8632 itself) and explore all the way down to leaf scalar values/fields
8633 embedded in the higher level data types.
8636 @item explore @var{arg}
8637 @var{arg} is either an expression (in the source language), or a type
8638 visible in the current context of the program being debugged.
8641 The working of the @code{explore} command can be illustrated with an
8642 example. If a data type @code{struct ComplexStruct} is defined in your
8652 struct ComplexStruct
8654 struct SimpleStruct *ss_p;
8660 followed by variable declarations as
8663 struct SimpleStruct ss = @{ 10, 1.11 @};
8664 struct ComplexStruct cs = @{ &ss, @{ 0, 1, 2, 3, 4, 5, 6, 7, 8, 9 @} @};
8668 then, the value of the variable @code{cs} can be explored using the
8669 @code{explore} command as follows.
8673 The value of `cs' is a struct/class of type `struct ComplexStruct' with
8674 the following fields:
8676 ss_p = <Enter 0 to explore this field of type `struct SimpleStruct *'>
8677 arr = <Enter 1 to explore this field of type `int [10]'>
8679 Enter the field number of choice:
8683 Since the fields of @code{cs} are not scalar values, you are being
8684 prompted to chose the field you want to explore. Let's say you choose
8685 the field @code{ss_p} by entering @code{0}. Then, since this field is a
8686 pointer, you will be asked if it is pointing to a single value. From
8687 the declaration of @code{cs} above, it is indeed pointing to a single
8688 value, hence you enter @code{y}. If you enter @code{n}, then you will
8689 be asked if it were pointing to an array of values, in which case this
8690 field will be explored as if it were an array.
8693 `cs.ss_p' is a pointer to a value of type `struct SimpleStruct'
8694 Continue exploring it as a pointer to a single value [y/n]: y
8695 The value of `*(cs.ss_p)' is a struct/class of type `struct
8696 SimpleStruct' with the following fields:
8698 i = 10 .. (Value of type `int')
8699 d = 1.1100000000000001 .. (Value of type `double')
8701 Press enter to return to parent value:
8705 If the field @code{arr} of @code{cs} was chosen for exploration by
8706 entering @code{1} earlier, then since it is as array, you will be
8707 prompted to enter the index of the element in the array that you want
8711 `cs.arr' is an array of `int'.
8712 Enter the index of the element you want to explore in `cs.arr': 5
8714 `(cs.arr)[5]' is a scalar value of type `int'.
8718 Press enter to return to parent value:
8721 In general, at any stage of exploration, you can go deeper towards the
8722 leaf values by responding to the prompts appropriately, or hit the
8723 return key to return to the enclosing data structure (the @i{higher}
8724 level data structure).
8726 Similar to exploring values, you can use the @code{explore} command to
8727 explore types. Instead of specifying a value (which is typically a
8728 variable name or an expression valid in the current context of the
8729 program being debugged), you specify a type name. If you consider the
8730 same example as above, your can explore the type
8731 @code{struct ComplexStruct} by passing the argument
8732 @code{struct ComplexStruct} to the @code{explore} command.
8735 (gdb) explore struct ComplexStruct
8739 By responding to the prompts appropriately in the subsequent interactive
8740 session, you can explore the type @code{struct ComplexStruct} in a
8741 manner similar to how the value @code{cs} was explored in the above
8744 The @code{explore} command also has two sub-commands,
8745 @code{explore value} and @code{explore type}. The former sub-command is
8746 a way to explicitly specify that value exploration of the argument is
8747 being invoked, while the latter is a way to explicitly specify that type
8748 exploration of the argument is being invoked.
8751 @item explore value @var{expr}
8752 @cindex explore value
8753 This sub-command of @code{explore} explores the value of the
8754 expression @var{expr} (if @var{expr} is an expression valid in the
8755 current context of the program being debugged). The behavior of this
8756 command is identical to that of the behavior of the @code{explore}
8757 command being passed the argument @var{expr}.
8759 @item explore type @var{arg}
8760 @cindex explore type
8761 This sub-command of @code{explore} explores the type of @var{arg} (if
8762 @var{arg} is a type visible in the current context of program being
8763 debugged), or the type of the value/expression @var{arg} (if @var{arg}
8764 is an expression valid in the current context of the program being
8765 debugged). If @var{arg} is a type, then the behavior of this command is
8766 identical to that of the @code{explore} command being passed the
8767 argument @var{arg}. If @var{arg} is an expression, then the behavior of
8768 this command will be identical to that of the @code{explore} command
8769 being passed the type of @var{arg} as the argument.
8773 * Expressions:: Expressions
8774 * Ambiguous Expressions:: Ambiguous Expressions
8775 * Variables:: Program variables
8776 * Arrays:: Artificial arrays
8777 * Output Formats:: Output formats
8778 * Memory:: Examining memory
8779 * Auto Display:: Automatic display
8780 * Print Settings:: Print settings
8781 * Pretty Printing:: Python pretty printing
8782 * Value History:: Value history
8783 * Convenience Vars:: Convenience variables
8784 * Convenience Funs:: Convenience functions
8785 * Registers:: Registers
8786 * Floating Point Hardware:: Floating point hardware
8787 * Vector Unit:: Vector Unit
8788 * OS Information:: Auxiliary data provided by operating system
8789 * Memory Region Attributes:: Memory region attributes
8790 * Dump/Restore Files:: Copy between memory and a file
8791 * Core File Generation:: Cause a program dump its core
8792 * Character Sets:: Debugging programs that use a different
8793 character set than GDB does
8794 * Caching Target Data:: Data caching for targets
8795 * Searching Memory:: Searching memory for a sequence of bytes
8796 * Value Sizes:: Managing memory allocated for values
8800 @section Expressions
8803 @code{print} and many other @value{GDBN} commands accept an expression and
8804 compute its value. Any kind of constant, variable or operator defined
8805 by the programming language you are using is valid in an expression in
8806 @value{GDBN}. This includes conditional expressions, function calls,
8807 casts, and string constants. It also includes preprocessor macros, if
8808 you compiled your program to include this information; see
8811 @cindex arrays in expressions
8812 @value{GDBN} supports array constants in expressions input by
8813 the user. The syntax is @{@var{element}, @var{element}@dots{}@}. For example,
8814 you can use the command @code{print @{1, 2, 3@}} to create an array
8815 of three integers. If you pass an array to a function or assign it
8816 to a program variable, @value{GDBN} copies the array to memory that
8817 is @code{malloc}ed in the target program.
8819 Because C is so widespread, most of the expressions shown in examples in
8820 this manual are in C. @xref{Languages, , Using @value{GDBN} with Different
8821 Languages}, for information on how to use expressions in other
8824 In this section, we discuss operators that you can use in @value{GDBN}
8825 expressions regardless of your programming language.
8827 @cindex casts, in expressions
8828 Casts are supported in all languages, not just in C, because it is so
8829 useful to cast a number into a pointer in order to examine a structure
8830 at that address in memory.
8831 @c FIXME: casts supported---Mod2 true?
8833 @value{GDBN} supports these operators, in addition to those common
8834 to programming languages:
8838 @samp{@@} is a binary operator for treating parts of memory as arrays.
8839 @xref{Arrays, ,Artificial Arrays}, for more information.
8842 @samp{::} allows you to specify a variable in terms of the file or
8843 function where it is defined. @xref{Variables, ,Program Variables}.
8845 @cindex @{@var{type}@}
8846 @cindex type casting memory
8847 @cindex memory, viewing as typed object
8848 @cindex casts, to view memory
8849 @item @{@var{type}@} @var{addr}
8850 Refers to an object of type @var{type} stored at address @var{addr} in
8851 memory. The address @var{addr} may be any expression whose value is
8852 an integer or pointer (but parentheses are required around binary
8853 operators, just as in a cast). This construct is allowed regardless
8854 of what kind of data is normally supposed to reside at @var{addr}.
8857 @node Ambiguous Expressions
8858 @section Ambiguous Expressions
8859 @cindex ambiguous expressions
8861 Expressions can sometimes contain some ambiguous elements. For instance,
8862 some programming languages (notably Ada, C@t{++} and Objective-C) permit
8863 a single function name to be defined several times, for application in
8864 different contexts. This is called @dfn{overloading}. Another example
8865 involving Ada is generics. A @dfn{generic package} is similar to C@t{++}
8866 templates and is typically instantiated several times, resulting in
8867 the same function name being defined in different contexts.
8869 In some cases and depending on the language, it is possible to adjust
8870 the expression to remove the ambiguity. For instance in C@t{++}, you
8871 can specify the signature of the function you want to break on, as in
8872 @kbd{break @var{function}(@var{types})}. In Ada, using the fully
8873 qualified name of your function often makes the expression unambiguous
8876 When an ambiguity that needs to be resolved is detected, the debugger
8877 has the capability to display a menu of numbered choices for each
8878 possibility, and then waits for the selection with the prompt @samp{>}.
8879 The first option is always @samp{[0] cancel}, and typing @kbd{0 @key{RET}}
8880 aborts the current command. If the command in which the expression was
8881 used allows more than one choice to be selected, the next option in the
8882 menu is @samp{[1] all}, and typing @kbd{1 @key{RET}} selects all possible
8885 For example, the following session excerpt shows an attempt to set a
8886 breakpoint at the overloaded symbol @code{String::after}.
8887 We choose three particular definitions of that function name:
8889 @c FIXME! This is likely to change to show arg type lists, at least
8892 (@value{GDBP}) b String::after
8895 [2] file:String.cc; line number:867
8896 [3] file:String.cc; line number:860
8897 [4] file:String.cc; line number:875
8898 [5] file:String.cc; line number:853
8899 [6] file:String.cc; line number:846
8900 [7] file:String.cc; line number:735
8902 Breakpoint 1 at 0xb26c: file String.cc, line 867.
8903 Breakpoint 2 at 0xb344: file String.cc, line 875.
8904 Breakpoint 3 at 0xafcc: file String.cc, line 846.
8905 Multiple breakpoints were set.
8906 Use the "delete" command to delete unwanted
8913 @kindex set multiple-symbols
8914 @item set multiple-symbols @var{mode}
8915 @cindex multiple-symbols menu
8917 This option allows you to adjust the debugger behavior when an expression
8920 By default, @var{mode} is set to @code{all}. If the command with which
8921 the expression is used allows more than one choice, then @value{GDBN}
8922 automatically selects all possible choices. For instance, inserting
8923 a breakpoint on a function using an ambiguous name results in a breakpoint
8924 inserted on each possible match. However, if a unique choice must be made,
8925 then @value{GDBN} uses the menu to help you disambiguate the expression.
8926 For instance, printing the address of an overloaded function will result
8927 in the use of the menu.
8929 When @var{mode} is set to @code{ask}, the debugger always uses the menu
8930 when an ambiguity is detected.
8932 Finally, when @var{mode} is set to @code{cancel}, the debugger reports
8933 an error due to the ambiguity and the command is aborted.
8935 @kindex show multiple-symbols
8936 @item show multiple-symbols
8937 Show the current value of the @code{multiple-symbols} setting.
8941 @section Program Variables
8943 The most common kind of expression to use is the name of a variable
8946 Variables in expressions are understood in the selected stack frame
8947 (@pxref{Selection, ,Selecting a Frame}); they must be either:
8951 global (or file-static)
8958 visible according to the scope rules of the
8959 programming language from the point of execution in that frame
8962 @noindent This means that in the function
8977 you can examine and use the variable @code{a} whenever your program is
8978 executing within the function @code{foo}, but you can only use or
8979 examine the variable @code{b} while your program is executing inside
8980 the block where @code{b} is declared.
8982 @cindex variable name conflict
8983 There is an exception: you can refer to a variable or function whose
8984 scope is a single source file even if the current execution point is not
8985 in this file. But it is possible to have more than one such variable or
8986 function with the same name (in different source files). If that
8987 happens, referring to that name has unpredictable effects. If you wish,
8988 you can specify a static variable in a particular function or file by
8989 using the colon-colon (@code{::}) notation:
8991 @cindex colon-colon, context for variables/functions
8993 @c info cannot cope with a :: index entry, but why deprive hard copy readers?
8994 @cindex @code{::}, context for variables/functions
8997 @var{file}::@var{variable}
8998 @var{function}::@var{variable}
9002 Here @var{file} or @var{function} is the name of the context for the
9003 static @var{variable}. In the case of file names, you can use quotes to
9004 make sure @value{GDBN} parses the file name as a single word---for example,
9005 to print a global value of @code{x} defined in @file{f2.c}:
9008 (@value{GDBP}) p 'f2.c'::x
9011 The @code{::} notation is normally used for referring to
9012 static variables, since you typically disambiguate uses of local variables
9013 in functions by selecting the appropriate frame and using the
9014 simple name of the variable. However, you may also use this notation
9015 to refer to local variables in frames enclosing the selected frame:
9024 process (a); /* Stop here */
9035 For example, if there is a breakpoint at the commented line,
9036 here is what you might see
9037 when the program stops after executing the call @code{bar(0)}:
9042 (@value{GDBP}) p bar::a
9045 #2 0x080483d0 in foo (a=5) at foobar.c:12
9048 (@value{GDBP}) p bar::a
9052 @cindex C@t{++} scope resolution
9053 These uses of @samp{::} are very rarely in conflict with the very
9054 similar use of the same notation in C@t{++}. When they are in
9055 conflict, the C@t{++} meaning takes precedence; however, this can be
9056 overridden by quoting the file or function name with single quotes.
9058 For example, suppose the program is stopped in a method of a class
9059 that has a field named @code{includefile}, and there is also an
9060 include file named @file{includefile} that defines a variable,
9064 (@value{GDBP}) p includefile
9066 (@value{GDBP}) p includefile::some_global
9067 A syntax error in expression, near `'.
9068 (@value{GDBP}) p 'includefile'::some_global
9072 @cindex wrong values
9073 @cindex variable values, wrong
9074 @cindex function entry/exit, wrong values of variables
9075 @cindex optimized code, wrong values of variables
9077 @emph{Warning:} Occasionally, a local variable may appear to have the
9078 wrong value at certain points in a function---just after entry to a new
9079 scope, and just before exit.
9081 You may see this problem when you are stepping by machine instructions.
9082 This is because, on most machines, it takes more than one instruction to
9083 set up a stack frame (including local variable definitions); if you are
9084 stepping by machine instructions, variables may appear to have the wrong
9085 values until the stack frame is completely built. On exit, it usually
9086 also takes more than one machine instruction to destroy a stack frame;
9087 after you begin stepping through that group of instructions, local
9088 variable definitions may be gone.
9090 This may also happen when the compiler does significant optimizations.
9091 To be sure of always seeing accurate values, turn off all optimization
9094 @cindex ``No symbol "foo" in current context''
9095 Another possible effect of compiler optimizations is to optimize
9096 unused variables out of existence, or assign variables to registers (as
9097 opposed to memory addresses). Depending on the support for such cases
9098 offered by the debug info format used by the compiler, @value{GDBN}
9099 might not be able to display values for such local variables. If that
9100 happens, @value{GDBN} will print a message like this:
9103 No symbol "foo" in current context.
9106 To solve such problems, either recompile without optimizations, or use a
9107 different debug info format, if the compiler supports several such
9108 formats. @xref{Compilation}, for more information on choosing compiler
9109 options. @xref{C, ,C and C@t{++}}, for more information about debug
9110 info formats that are best suited to C@t{++} programs.
9112 If you ask to print an object whose contents are unknown to
9113 @value{GDBN}, e.g., because its data type is not completely specified
9114 by the debug information, @value{GDBN} will say @samp{<incomplete
9115 type>}. @xref{Symbols, incomplete type}, for more about this.
9117 If you append @kbd{@@entry} string to a function parameter name you get its
9118 value at the time the function got called. If the value is not available an
9119 error message is printed. Entry values are available only with some compilers.
9120 Entry values are normally also printed at the function parameter list according
9121 to @ref{set print entry-values}.
9124 Breakpoint 1, d (i=30) at gdb.base/entry-value.c:29
9130 (gdb) print i@@entry
9134 Strings are identified as arrays of @code{char} values without specified
9135 signedness. Arrays of either @code{signed char} or @code{unsigned char} get
9136 printed as arrays of 1 byte sized integers. @code{-fsigned-char} or
9137 @code{-funsigned-char} @value{NGCC} options have no effect as @value{GDBN}
9138 defines literal string type @code{"char"} as @code{char} without a sign.
9143 signed char var1[] = "A";
9146 You get during debugging
9151 $2 = @{65 'A', 0 '\0'@}
9155 @section Artificial Arrays
9157 @cindex artificial array
9159 @kindex @@@r{, referencing memory as an array}
9160 It is often useful to print out several successive objects of the
9161 same type in memory; a section of an array, or an array of
9162 dynamically determined size for which only a pointer exists in the
9165 You can do this by referring to a contiguous span of memory as an
9166 @dfn{artificial array}, using the binary operator @samp{@@}. The left
9167 operand of @samp{@@} should be the first element of the desired array
9168 and be an individual object. The right operand should be the desired length
9169 of the array. The result is an array value whose elements are all of
9170 the type of the left argument. The first element is actually the left
9171 argument; the second element comes from bytes of memory immediately
9172 following those that hold the first element, and so on. Here is an
9173 example. If a program says
9176 int *array = (int *) malloc (len * sizeof (int));
9180 you can print the contents of @code{array} with
9186 The left operand of @samp{@@} must reside in memory. Array values made
9187 with @samp{@@} in this way behave just like other arrays in terms of
9188 subscripting, and are coerced to pointers when used in expressions.
9189 Artificial arrays most often appear in expressions via the value history
9190 (@pxref{Value History, ,Value History}), after printing one out.
9192 Another way to create an artificial array is to use a cast.
9193 This re-interprets a value as if it were an array.
9194 The value need not be in memory:
9196 (@value{GDBP}) p/x (short[2])0x12345678
9197 $1 = @{0x1234, 0x5678@}
9200 As a convenience, if you leave the array length out (as in
9201 @samp{(@var{type}[])@var{value}}) @value{GDBN} calculates the size to fill
9202 the value (as @samp{sizeof(@var{value})/sizeof(@var{type})}:
9204 (@value{GDBP}) p/x (short[])0x12345678
9205 $2 = @{0x1234, 0x5678@}
9208 Sometimes the artificial array mechanism is not quite enough; in
9209 moderately complex data structures, the elements of interest may not
9210 actually be adjacent---for example, if you are interested in the values
9211 of pointers in an array. One useful work-around in this situation is
9212 to use a convenience variable (@pxref{Convenience Vars, ,Convenience
9213 Variables}) as a counter in an expression that prints the first
9214 interesting value, and then repeat that expression via @key{RET}. For
9215 instance, suppose you have an array @code{dtab} of pointers to
9216 structures, and you are interested in the values of a field @code{fv}
9217 in each structure. Here is an example of what you might type:
9227 @node Output Formats
9228 @section Output Formats
9230 @cindex formatted output
9231 @cindex output formats
9232 By default, @value{GDBN} prints a value according to its data type. Sometimes
9233 this is not what you want. For example, you might want to print a number
9234 in hex, or a pointer in decimal. Or you might want to view data in memory
9235 at a certain address as a character string or as an instruction. To do
9236 these things, specify an @dfn{output format} when you print a value.
9238 The simplest use of output formats is to say how to print a value
9239 already computed. This is done by starting the arguments of the
9240 @code{print} command with a slash and a format letter. The format
9241 letters supported are:
9245 Regard the bits of the value as an integer, and print the integer in
9249 Print as integer in signed decimal.
9252 Print as integer in unsigned decimal.
9255 Print as integer in octal.
9258 Print as integer in binary. The letter @samp{t} stands for ``two''.
9259 @footnote{@samp{b} cannot be used because these format letters are also
9260 used with the @code{x} command, where @samp{b} stands for ``byte'';
9261 see @ref{Memory,,Examining Memory}.}
9264 @cindex unknown address, locating
9265 @cindex locate address
9266 Print as an address, both absolute in hexadecimal and as an offset from
9267 the nearest preceding symbol. You can use this format used to discover
9268 where (in what function) an unknown address is located:
9271 (@value{GDBP}) p/a 0x54320
9272 $3 = 0x54320 <_initialize_vx+396>
9276 The command @code{info symbol 0x54320} yields similar results.
9277 @xref{Symbols, info symbol}.
9280 Regard as an integer and print it as a character constant. This
9281 prints both the numerical value and its character representation. The
9282 character representation is replaced with the octal escape @samp{\nnn}
9283 for characters outside the 7-bit @sc{ascii} range.
9285 Without this format, @value{GDBN} displays @code{char},
9286 @w{@code{unsigned char}}, and @w{@code{signed char}} data as character
9287 constants. Single-byte members of vectors are displayed as integer
9291 Regard the bits of the value as a floating point number and print
9292 using typical floating point syntax.
9295 @cindex printing strings
9296 @cindex printing byte arrays
9297 Regard as a string, if possible. With this format, pointers to single-byte
9298 data are displayed as null-terminated strings and arrays of single-byte data
9299 are displayed as fixed-length strings. Other values are displayed in their
9302 Without this format, @value{GDBN} displays pointers to and arrays of
9303 @code{char}, @w{@code{unsigned char}}, and @w{@code{signed char}} as
9304 strings. Single-byte members of a vector are displayed as an integer
9308 Like @samp{x} formatting, the value is treated as an integer and
9309 printed as hexadecimal, but leading zeros are printed to pad the value
9310 to the size of the integer type.
9313 @cindex raw printing
9314 Print using the @samp{raw} formatting. By default, @value{GDBN} will
9315 use a Python-based pretty-printer, if one is available (@pxref{Pretty
9316 Printing}). This typically results in a higher-level display of the
9317 value's contents. The @samp{r} format bypasses any Python
9318 pretty-printer which might exist.
9321 For example, to print the program counter in hex (@pxref{Registers}), type
9328 Note that no space is required before the slash; this is because command
9329 names in @value{GDBN} cannot contain a slash.
9331 To reprint the last value in the value history with a different format,
9332 you can use the @code{print} command with just a format and no
9333 expression. For example, @samp{p/x} reprints the last value in hex.
9336 @section Examining Memory
9338 You can use the command @code{x} (for ``examine'') to examine memory in
9339 any of several formats, independently of your program's data types.
9341 @cindex examining memory
9343 @kindex x @r{(examine memory)}
9344 @item x/@var{nfu} @var{addr}
9347 Use the @code{x} command to examine memory.
9350 @var{n}, @var{f}, and @var{u} are all optional parameters that specify how
9351 much memory to display and how to format it; @var{addr} is an
9352 expression giving the address where you want to start displaying memory.
9353 If you use defaults for @var{nfu}, you need not type the slash @samp{/}.
9354 Several commands set convenient defaults for @var{addr}.
9357 @item @var{n}, the repeat count
9358 The repeat count is a decimal integer; the default is 1. It specifies
9359 how much memory (counting by units @var{u}) to display. If a negative
9360 number is specified, memory is examined backward from @var{addr}.
9361 @c This really is **decimal**; unaffected by 'set radix' as of GDB
9364 @item @var{f}, the display format
9365 The display format is one of the formats used by @code{print}
9366 (@samp{x}, @samp{d}, @samp{u}, @samp{o}, @samp{t}, @samp{a}, @samp{c},
9367 @samp{f}, @samp{s}), and in addition @samp{i} (for machine instructions).
9368 The default is @samp{x} (hexadecimal) initially. The default changes
9369 each time you use either @code{x} or @code{print}.
9371 @item @var{u}, the unit size
9372 The unit size is any of
9378 Halfwords (two bytes).
9380 Words (four bytes). This is the initial default.
9382 Giant words (eight bytes).
9385 Each time you specify a unit size with @code{x}, that size becomes the
9386 default unit the next time you use @code{x}. For the @samp{i} format,
9387 the unit size is ignored and is normally not written. For the @samp{s} format,
9388 the unit size defaults to @samp{b}, unless it is explicitly given.
9389 Use @kbd{x /hs} to display 16-bit char strings and @kbd{x /ws} to display
9390 32-bit strings. The next use of @kbd{x /s} will again display 8-bit strings.
9391 Note that the results depend on the programming language of the
9392 current compilation unit. If the language is C, the @samp{s}
9393 modifier will use the UTF-16 encoding while @samp{w} will use
9394 UTF-32. The encoding is set by the programming language and cannot
9397 @item @var{addr}, starting display address
9398 @var{addr} is the address where you want @value{GDBN} to begin displaying
9399 memory. The expression need not have a pointer value (though it may);
9400 it is always interpreted as an integer address of a byte of memory.
9401 @xref{Expressions, ,Expressions}, for more information on expressions. The default for
9402 @var{addr} is usually just after the last address examined---but several
9403 other commands also set the default address: @code{info breakpoints} (to
9404 the address of the last breakpoint listed), @code{info line} (to the
9405 starting address of a line), and @code{print} (if you use it to display
9406 a value from memory).
9409 For example, @samp{x/3uh 0x54320} is a request to display three halfwords
9410 (@code{h}) of memory, formatted as unsigned decimal integers (@samp{u}),
9411 starting at address @code{0x54320}. @samp{x/4xw $sp} prints the four
9412 words (@samp{w}) of memory above the stack pointer (here, @samp{$sp};
9413 @pxref{Registers, ,Registers}) in hexadecimal (@samp{x}).
9415 You can also specify a negative repeat count to examine memory backward
9416 from the given address. For example, @samp{x/-3uh 0x54320} prints three
9417 halfwords (@code{h}) at @code{0x54314}, @code{0x54328}, and @code{0x5431c}.
9419 Since the letters indicating unit sizes are all distinct from the
9420 letters specifying output formats, you do not have to remember whether
9421 unit size or format comes first; either order works. The output
9422 specifications @samp{4xw} and @samp{4wx} mean exactly the same thing.
9423 (However, the count @var{n} must come first; @samp{wx4} does not work.)
9425 Even though the unit size @var{u} is ignored for the formats @samp{s}
9426 and @samp{i}, you might still want to use a count @var{n}; for example,
9427 @samp{3i} specifies that you want to see three machine instructions,
9428 including any operands. For convenience, especially when used with
9429 the @code{display} command, the @samp{i} format also prints branch delay
9430 slot instructions, if any, beyond the count specified, which immediately
9431 follow the last instruction that is within the count. The command
9432 @code{disassemble} gives an alternative way of inspecting machine
9433 instructions; see @ref{Machine Code,,Source and Machine Code}.
9435 If a negative repeat count is specified for the formats @samp{s} or @samp{i},
9436 the command displays null-terminated strings or instructions before the given
9437 address as many as the absolute value of the given number. For the @samp{i}
9438 format, we use line number information in the debug info to accurately locate
9439 instruction boundaries while disassembling backward. If line info is not
9440 available, the command stops examining memory with an error message.
9442 All the defaults for the arguments to @code{x} are designed to make it
9443 easy to continue scanning memory with minimal specifications each time
9444 you use @code{x}. For example, after you have inspected three machine
9445 instructions with @samp{x/3i @var{addr}}, you can inspect the next seven
9446 with just @samp{x/7}. If you use @key{RET} to repeat the @code{x} command,
9447 the repeat count @var{n} is used again; the other arguments default as
9448 for successive uses of @code{x}.
9450 When examining machine instructions, the instruction at current program
9451 counter is shown with a @code{=>} marker. For example:
9454 (@value{GDBP}) x/5i $pc-6
9455 0x804837f <main+11>: mov %esp,%ebp
9456 0x8048381 <main+13>: push %ecx
9457 0x8048382 <main+14>: sub $0x4,%esp
9458 => 0x8048385 <main+17>: movl $0x8048460,(%esp)
9459 0x804838c <main+24>: call 0x80482d4 <puts@@plt>
9462 @cindex @code{$_}, @code{$__}, and value history
9463 The addresses and contents printed by the @code{x} command are not saved
9464 in the value history because there is often too much of them and they
9465 would get in the way. Instead, @value{GDBN} makes these values available for
9466 subsequent use in expressions as values of the convenience variables
9467 @code{$_} and @code{$__}. After an @code{x} command, the last address
9468 examined is available for use in expressions in the convenience variable
9469 @code{$_}. The contents of that address, as examined, are available in
9470 the convenience variable @code{$__}.
9472 If the @code{x} command has a repeat count, the address and contents saved
9473 are from the last memory unit printed; this is not the same as the last
9474 address printed if several units were printed on the last line of output.
9476 @anchor{addressable memory unit}
9477 @cindex addressable memory unit
9478 Most targets have an addressable memory unit size of 8 bits. This means
9479 that to each memory address are associated 8 bits of data. Some
9480 targets, however, have other addressable memory unit sizes.
9481 Within @value{GDBN} and this document, the term
9482 @dfn{addressable memory unit} (or @dfn{memory unit} for short) is used
9483 when explicitly referring to a chunk of data of that size. The word
9484 @dfn{byte} is used to refer to a chunk of data of 8 bits, regardless of
9485 the addressable memory unit size of the target. For most systems,
9486 addressable memory unit is a synonym of byte.
9488 @cindex remote memory comparison
9489 @cindex target memory comparison
9490 @cindex verify remote memory image
9491 @cindex verify target memory image
9492 When you are debugging a program running on a remote target machine
9493 (@pxref{Remote Debugging}), you may wish to verify the program's image
9494 in the remote machine's memory against the executable file you
9495 downloaded to the target. Or, on any target, you may want to check
9496 whether the program has corrupted its own read-only sections. The
9497 @code{compare-sections} command is provided for such situations.
9500 @kindex compare-sections
9501 @item compare-sections @r{[}@var{section-name}@r{|}@code{-r}@r{]}
9502 Compare the data of a loadable section @var{section-name} in the
9503 executable file of the program being debugged with the same section in
9504 the target machine's memory, and report any mismatches. With no
9505 arguments, compares all loadable sections. With an argument of
9506 @code{-r}, compares all loadable read-only sections.
9508 Note: for remote targets, this command can be accelerated if the
9509 target supports computing the CRC checksum of a block of memory
9510 (@pxref{qCRC packet}).
9514 @section Automatic Display
9515 @cindex automatic display
9516 @cindex display of expressions
9518 If you find that you want to print the value of an expression frequently
9519 (to see how it changes), you might want to add it to the @dfn{automatic
9520 display list} so that @value{GDBN} prints its value each time your program stops.
9521 Each expression added to the list is given a number to identify it;
9522 to remove an expression from the list, you specify that number.
9523 The automatic display looks like this:
9527 3: bar[5] = (struct hack *) 0x3804
9531 This display shows item numbers, expressions and their current values. As with
9532 displays you request manually using @code{x} or @code{print}, you can
9533 specify the output format you prefer; in fact, @code{display} decides
9534 whether to use @code{print} or @code{x} depending your format
9535 specification---it uses @code{x} if you specify either the @samp{i}
9536 or @samp{s} format, or a unit size; otherwise it uses @code{print}.
9540 @item display @var{expr}
9541 Add the expression @var{expr} to the list of expressions to display
9542 each time your program stops. @xref{Expressions, ,Expressions}.
9544 @code{display} does not repeat if you press @key{RET} again after using it.
9546 @item display/@var{fmt} @var{expr}
9547 For @var{fmt} specifying only a display format and not a size or
9548 count, add the expression @var{expr} to the auto-display list but
9549 arrange to display it each time in the specified format @var{fmt}.
9550 @xref{Output Formats,,Output Formats}.
9552 @item display/@var{fmt} @var{addr}
9553 For @var{fmt} @samp{i} or @samp{s}, or including a unit-size or a
9554 number of units, add the expression @var{addr} as a memory address to
9555 be examined each time your program stops. Examining means in effect
9556 doing @samp{x/@var{fmt} @var{addr}}. @xref{Memory, ,Examining Memory}.
9559 For example, @samp{display/i $pc} can be helpful, to see the machine
9560 instruction about to be executed each time execution stops (@samp{$pc}
9561 is a common name for the program counter; @pxref{Registers, ,Registers}).
9564 @kindex delete display
9566 @item undisplay @var{dnums}@dots{}
9567 @itemx delete display @var{dnums}@dots{}
9568 Remove items from the list of expressions to display. Specify the
9569 numbers of the displays that you want affected with the command
9570 argument @var{dnums}. It can be a single display number, one of the
9571 numbers shown in the first field of the @samp{info display} display;
9572 or it could be a range of display numbers, as in @code{2-4}.
9574 @code{undisplay} does not repeat if you press @key{RET} after using it.
9575 (Otherwise you would just get the error @samp{No display number @dots{}}.)
9577 @kindex disable display
9578 @item disable display @var{dnums}@dots{}
9579 Disable the display of item numbers @var{dnums}. A disabled display
9580 item is not printed automatically, but is not forgotten. It may be
9581 enabled again later. Specify the numbers of the displays that you
9582 want affected with the command argument @var{dnums}. It can be a
9583 single display number, one of the numbers shown in the first field of
9584 the @samp{info display} display; or it could be a range of display
9585 numbers, as in @code{2-4}.
9587 @kindex enable display
9588 @item enable display @var{dnums}@dots{}
9589 Enable display of item numbers @var{dnums}. It becomes effective once
9590 again in auto display of its expression, until you specify otherwise.
9591 Specify the numbers of the displays that you want affected with the
9592 command argument @var{dnums}. It can be a single display number, one
9593 of the numbers shown in the first field of the @samp{info display}
9594 display; or it could be a range of display numbers, as in @code{2-4}.
9597 Display the current values of the expressions on the list, just as is
9598 done when your program stops.
9600 @kindex info display
9602 Print the list of expressions previously set up to display
9603 automatically, each one with its item number, but without showing the
9604 values. This includes disabled expressions, which are marked as such.
9605 It also includes expressions which would not be displayed right now
9606 because they refer to automatic variables not currently available.
9609 @cindex display disabled out of scope
9610 If a display expression refers to local variables, then it does not make
9611 sense outside the lexical context for which it was set up. Such an
9612 expression is disabled when execution enters a context where one of its
9613 variables is not defined. For example, if you give the command
9614 @code{display last_char} while inside a function with an argument
9615 @code{last_char}, @value{GDBN} displays this argument while your program
9616 continues to stop inside that function. When it stops elsewhere---where
9617 there is no variable @code{last_char}---the display is disabled
9618 automatically. The next time your program stops where @code{last_char}
9619 is meaningful, you can enable the display expression once again.
9621 @node Print Settings
9622 @section Print Settings
9624 @cindex format options
9625 @cindex print settings
9626 @value{GDBN} provides the following ways to control how arrays, structures,
9627 and symbols are printed.
9630 These settings are useful for debugging programs in any language:
9634 @item set print address
9635 @itemx set print address on
9636 @cindex print/don't print memory addresses
9637 @value{GDBN} prints memory addresses showing the location of stack
9638 traces, structure values, pointer values, breakpoints, and so forth,
9639 even when it also displays the contents of those addresses. The default
9640 is @code{on}. For example, this is what a stack frame display looks like with
9641 @code{set print address on}:
9646 #0 set_quotes (lq=0x34c78 "<<", rq=0x34c88 ">>")
9648 530 if (lquote != def_lquote)
9652 @item set print address off
9653 Do not print addresses when displaying their contents. For example,
9654 this is the same stack frame displayed with @code{set print address off}:
9658 (@value{GDBP}) set print addr off
9660 #0 set_quotes (lq="<<", rq=">>") at input.c:530
9661 530 if (lquote != def_lquote)
9665 You can use @samp{set print address off} to eliminate all machine
9666 dependent displays from the @value{GDBN} interface. For example, with
9667 @code{print address off}, you should get the same text for backtraces on
9668 all machines---whether or not they involve pointer arguments.
9671 @item show print address
9672 Show whether or not addresses are to be printed.
9675 When @value{GDBN} prints a symbolic address, it normally prints the
9676 closest earlier symbol plus an offset. If that symbol does not uniquely
9677 identify the address (for example, it is a name whose scope is a single
9678 source file), you may need to clarify. One way to do this is with
9679 @code{info line}, for example @samp{info line *0x4537}. Alternately,
9680 you can set @value{GDBN} to print the source file and line number when
9681 it prints a symbolic address:
9684 @item set print symbol-filename on
9685 @cindex source file and line of a symbol
9686 @cindex symbol, source file and line
9687 Tell @value{GDBN} to print the source file name and line number of a
9688 symbol in the symbolic form of an address.
9690 @item set print symbol-filename off
9691 Do not print source file name and line number of a symbol. This is the
9694 @item show print symbol-filename
9695 Show whether or not @value{GDBN} will print the source file name and
9696 line number of a symbol in the symbolic form of an address.
9699 Another situation where it is helpful to show symbol filenames and line
9700 numbers is when disassembling code; @value{GDBN} shows you the line
9701 number and source file that corresponds to each instruction.
9703 Also, you may wish to see the symbolic form only if the address being
9704 printed is reasonably close to the closest earlier symbol:
9707 @item set print max-symbolic-offset @var{max-offset}
9708 @itemx set print max-symbolic-offset unlimited
9709 @cindex maximum value for offset of closest symbol
9710 Tell @value{GDBN} to only display the symbolic form of an address if the
9711 offset between the closest earlier symbol and the address is less than
9712 @var{max-offset}. The default is @code{unlimited}, which tells @value{GDBN}
9713 to always print the symbolic form of an address if any symbol precedes
9714 it. Zero is equivalent to @code{unlimited}.
9716 @item show print max-symbolic-offset
9717 Ask how large the maximum offset is that @value{GDBN} prints in a
9721 @cindex wild pointer, interpreting
9722 @cindex pointer, finding referent
9723 If you have a pointer and you are not sure where it points, try
9724 @samp{set print symbol-filename on}. Then you can determine the name
9725 and source file location of the variable where it points, using
9726 @samp{p/a @var{pointer}}. This interprets the address in symbolic form.
9727 For example, here @value{GDBN} shows that a variable @code{ptt} points
9728 at another variable @code{t}, defined in @file{hi2.c}:
9731 (@value{GDBP}) set print symbol-filename on
9732 (@value{GDBP}) p/a ptt
9733 $4 = 0xe008 <t in hi2.c>
9737 @emph{Warning:} For pointers that point to a local variable, @samp{p/a}
9738 does not show the symbol name and filename of the referent, even with
9739 the appropriate @code{set print} options turned on.
9742 You can also enable @samp{/a}-like formatting all the time using
9743 @samp{set print symbol on}:
9746 @item set print symbol on
9747 Tell @value{GDBN} to print the symbol corresponding to an address, if
9750 @item set print symbol off
9751 Tell @value{GDBN} not to print the symbol corresponding to an
9752 address. In this mode, @value{GDBN} will still print the symbol
9753 corresponding to pointers to functions. This is the default.
9755 @item show print symbol
9756 Show whether @value{GDBN} will display the symbol corresponding to an
9760 Other settings control how different kinds of objects are printed:
9763 @item set print array
9764 @itemx set print array on
9765 @cindex pretty print arrays
9766 Pretty print arrays. This format is more convenient to read,
9767 but uses more space. The default is off.
9769 @item set print array off
9770 Return to compressed format for arrays.
9772 @item show print array
9773 Show whether compressed or pretty format is selected for displaying
9776 @cindex print array indexes
9777 @item set print array-indexes
9778 @itemx set print array-indexes on
9779 Print the index of each element when displaying arrays. May be more
9780 convenient to locate a given element in the array or quickly find the
9781 index of a given element in that printed array. The default is off.
9783 @item set print array-indexes off
9784 Stop printing element indexes when displaying arrays.
9786 @item show print array-indexes
9787 Show whether the index of each element is printed when displaying
9790 @item set print elements @var{number-of-elements}
9791 @itemx set print elements unlimited
9792 @cindex number of array elements to print
9793 @cindex limit on number of printed array elements
9794 Set a limit on how many elements of an array @value{GDBN} will print.
9795 If @value{GDBN} is printing a large array, it stops printing after it has
9796 printed the number of elements set by the @code{set print elements} command.
9797 This limit also applies to the display of strings.
9798 When @value{GDBN} starts, this limit is set to 200.
9799 Setting @var{number-of-elements} to @code{unlimited} or zero means
9800 that the number of elements to print is unlimited.
9802 @item show print elements
9803 Display the number of elements of a large array that @value{GDBN} will print.
9804 If the number is 0, then the printing is unlimited.
9806 @item set print frame-arguments @var{value}
9807 @kindex set print frame-arguments
9808 @cindex printing frame argument values
9809 @cindex print all frame argument values
9810 @cindex print frame argument values for scalars only
9811 @cindex do not print frame argument values
9812 This command allows to control how the values of arguments are printed
9813 when the debugger prints a frame (@pxref{Frames}). The possible
9818 The values of all arguments are printed.
9821 Print the value of an argument only if it is a scalar. The value of more
9822 complex arguments such as arrays, structures, unions, etc, is replaced
9823 by @code{@dots{}}. This is the default. Here is an example where
9824 only scalar arguments are shown:
9827 #1 0x08048361 in call_me (i=3, s=@dots{}, ss=0xbf8d508c, u=@dots{}, e=green)
9832 None of the argument values are printed. Instead, the value of each argument
9833 is replaced by @code{@dots{}}. In this case, the example above now becomes:
9836 #1 0x08048361 in call_me (i=@dots{}, s=@dots{}, ss=@dots{}, u=@dots{}, e=@dots{})
9841 By default, only scalar arguments are printed. This command can be used
9842 to configure the debugger to print the value of all arguments, regardless
9843 of their type. However, it is often advantageous to not print the value
9844 of more complex parameters. For instance, it reduces the amount of
9845 information printed in each frame, making the backtrace more readable.
9846 Also, it improves performance when displaying Ada frames, because
9847 the computation of large arguments can sometimes be CPU-intensive,
9848 especially in large applications. Setting @code{print frame-arguments}
9849 to @code{scalars} (the default) or @code{none} avoids this computation,
9850 thus speeding up the display of each Ada frame.
9852 @item show print frame-arguments
9853 Show how the value of arguments should be displayed when printing a frame.
9855 @item set print raw frame-arguments on
9856 Print frame arguments in raw, non pretty-printed, form.
9858 @item set print raw frame-arguments off
9859 Print frame arguments in pretty-printed form, if there is a pretty-printer
9860 for the value (@pxref{Pretty Printing}),
9861 otherwise print the value in raw form.
9862 This is the default.
9864 @item show print raw frame-arguments
9865 Show whether to print frame arguments in raw form.
9867 @anchor{set print entry-values}
9868 @item set print entry-values @var{value}
9869 @kindex set print entry-values
9870 Set printing of frame argument values at function entry. In some cases
9871 @value{GDBN} can determine the value of function argument which was passed by
9872 the function caller, even if the value was modified inside the called function
9873 and therefore is different. With optimized code, the current value could be
9874 unavailable, but the entry value may still be known.
9876 The default value is @code{default} (see below for its description). Older
9877 @value{GDBN} behaved as with the setting @code{no}. Compilers not supporting
9878 this feature will behave in the @code{default} setting the same way as with the
9881 This functionality is currently supported only by DWARF 2 debugging format and
9882 the compiler has to produce @samp{DW_TAG_call_site} tags. With
9883 @value{NGCC}, you need to specify @option{-O -g} during compilation, to get
9886 The @var{value} parameter can be one of the following:
9890 Print only actual parameter values, never print values from function entry
9894 #0 different (val=6)
9895 #0 lost (val=<optimized out>)
9897 #0 invalid (val=<optimized out>)
9901 Print only parameter values from function entry point. The actual parameter
9902 values are never printed.
9904 #0 equal (val@@entry=5)
9905 #0 different (val@@entry=5)
9906 #0 lost (val@@entry=5)
9907 #0 born (val@@entry=<optimized out>)
9908 #0 invalid (val@@entry=<optimized out>)
9912 Print only parameter values from function entry point. If value from function
9913 entry point is not known while the actual value is known, print the actual
9914 value for such parameter.
9916 #0 equal (val@@entry=5)
9917 #0 different (val@@entry=5)
9918 #0 lost (val@@entry=5)
9920 #0 invalid (val@@entry=<optimized out>)
9924 Print actual parameter values. If actual parameter value is not known while
9925 value from function entry point is known, print the entry point value for such
9929 #0 different (val=6)
9930 #0 lost (val@@entry=5)
9932 #0 invalid (val=<optimized out>)
9936 Always print both the actual parameter value and its value from function entry
9937 point, even if values of one or both are not available due to compiler
9940 #0 equal (val=5, val@@entry=5)
9941 #0 different (val=6, val@@entry=5)
9942 #0 lost (val=<optimized out>, val@@entry=5)
9943 #0 born (val=10, val@@entry=<optimized out>)
9944 #0 invalid (val=<optimized out>, val@@entry=<optimized out>)
9948 Print the actual parameter value if it is known and also its value from
9949 function entry point if it is known. If neither is known, print for the actual
9950 value @code{<optimized out>}. If not in MI mode (@pxref{GDB/MI}) and if both
9951 values are known and identical, print the shortened
9952 @code{param=param@@entry=VALUE} notation.
9954 #0 equal (val=val@@entry=5)
9955 #0 different (val=6, val@@entry=5)
9956 #0 lost (val@@entry=5)
9958 #0 invalid (val=<optimized out>)
9962 Always print the actual parameter value. Print also its value from function
9963 entry point, but only if it is known. If not in MI mode (@pxref{GDB/MI}) and
9964 if both values are known and identical, print the shortened
9965 @code{param=param@@entry=VALUE} notation.
9967 #0 equal (val=val@@entry=5)
9968 #0 different (val=6, val@@entry=5)
9969 #0 lost (val=<optimized out>, val@@entry=5)
9971 #0 invalid (val=<optimized out>)
9975 For analysis messages on possible failures of frame argument values at function
9976 entry resolution see @ref{set debug entry-values}.
9978 @item show print entry-values
9979 Show the method being used for printing of frame argument values at function
9982 @item set print repeats @var{number-of-repeats}
9983 @itemx set print repeats unlimited
9984 @cindex repeated array elements
9985 Set the threshold for suppressing display of repeated array
9986 elements. When the number of consecutive identical elements of an
9987 array exceeds the threshold, @value{GDBN} prints the string
9988 @code{"<repeats @var{n} times>"}, where @var{n} is the number of
9989 identical repetitions, instead of displaying the identical elements
9990 themselves. Setting the threshold to @code{unlimited} or zero will
9991 cause all elements to be individually printed. The default threshold
9994 @item show print repeats
9995 Display the current threshold for printing repeated identical
9998 @item set print null-stop
9999 @cindex @sc{null} elements in arrays
10000 Cause @value{GDBN} to stop printing the characters of an array when the first
10001 @sc{null} is encountered. This is useful when large arrays actually
10002 contain only short strings.
10003 The default is off.
10005 @item show print null-stop
10006 Show whether @value{GDBN} stops printing an array on the first
10007 @sc{null} character.
10009 @item set print pretty on
10010 @cindex print structures in indented form
10011 @cindex indentation in structure display
10012 Cause @value{GDBN} to print structures in an indented format with one member
10013 per line, like this:
10028 @item set print pretty off
10029 Cause @value{GDBN} to print structures in a compact format, like this:
10033 $1 = @{next = 0x0, flags = @{sweet = 1, sour = 1@}, \
10034 meat = 0x54 "Pork"@}
10039 This is the default format.
10041 @item show print pretty
10042 Show which format @value{GDBN} is using to print structures.
10044 @item set print sevenbit-strings on
10045 @cindex eight-bit characters in strings
10046 @cindex octal escapes in strings
10047 Print using only seven-bit characters; if this option is set,
10048 @value{GDBN} displays any eight-bit characters (in strings or
10049 character values) using the notation @code{\}@var{nnn}. This setting is
10050 best if you are working in English (@sc{ascii}) and you use the
10051 high-order bit of characters as a marker or ``meta'' bit.
10053 @item set print sevenbit-strings off
10054 Print full eight-bit characters. This allows the use of more
10055 international character sets, and is the default.
10057 @item show print sevenbit-strings
10058 Show whether or not @value{GDBN} is printing only seven-bit characters.
10060 @item set print union on
10061 @cindex unions in structures, printing
10062 Tell @value{GDBN} to print unions which are contained in structures
10063 and other unions. This is the default setting.
10065 @item set print union off
10066 Tell @value{GDBN} not to print unions which are contained in
10067 structures and other unions. @value{GDBN} will print @code{"@{...@}"}
10070 @item show print union
10071 Ask @value{GDBN} whether or not it will print unions which are contained in
10072 structures and other unions.
10074 For example, given the declarations
10077 typedef enum @{Tree, Bug@} Species;
10078 typedef enum @{Big_tree, Acorn, Seedling@} Tree_forms;
10079 typedef enum @{Caterpillar, Cocoon, Butterfly@}
10090 struct thing foo = @{Tree, @{Acorn@}@};
10094 with @code{set print union on} in effect @samp{p foo} would print
10097 $1 = @{it = Tree, form = @{tree = Acorn, bug = Cocoon@}@}
10101 and with @code{set print union off} in effect it would print
10104 $1 = @{it = Tree, form = @{...@}@}
10108 @code{set print union} affects programs written in C-like languages
10114 These settings are of interest when debugging C@t{++} programs:
10117 @cindex demangling C@t{++} names
10118 @item set print demangle
10119 @itemx set print demangle on
10120 Print C@t{++} names in their source form rather than in the encoded
10121 (``mangled'') form passed to the assembler and linker for type-safe
10122 linkage. The default is on.
10124 @item show print demangle
10125 Show whether C@t{++} names are printed in mangled or demangled form.
10127 @item set print asm-demangle
10128 @itemx set print asm-demangle on
10129 Print C@t{++} names in their source form rather than their mangled form, even
10130 in assembler code printouts such as instruction disassemblies.
10131 The default is off.
10133 @item show print asm-demangle
10134 Show whether C@t{++} names in assembly listings are printed in mangled
10137 @cindex C@t{++} symbol decoding style
10138 @cindex symbol decoding style, C@t{++}
10139 @kindex set demangle-style
10140 @item set demangle-style @var{style}
10141 Choose among several encoding schemes used by different compilers to
10142 represent C@t{++} names. The choices for @var{style} are currently:
10146 Allow @value{GDBN} to choose a decoding style by inspecting your program.
10147 This is the default.
10150 Decode based on the @sc{gnu} C@t{++} compiler (@code{g++}) encoding algorithm.
10153 Decode based on the HP ANSI C@t{++} (@code{aCC}) encoding algorithm.
10156 Decode based on the Lucid C@t{++} compiler (@code{lcc}) encoding algorithm.
10159 Decode using the algorithm in the @cite{C@t{++} Annotated Reference Manual}.
10160 @strong{Warning:} this setting alone is not sufficient to allow
10161 debugging @code{cfront}-generated executables. @value{GDBN} would
10162 require further enhancement to permit that.
10165 If you omit @var{style}, you will see a list of possible formats.
10167 @item show demangle-style
10168 Display the encoding style currently in use for decoding C@t{++} symbols.
10170 @item set print object
10171 @itemx set print object on
10172 @cindex derived type of an object, printing
10173 @cindex display derived types
10174 When displaying a pointer to an object, identify the @emph{actual}
10175 (derived) type of the object rather than the @emph{declared} type, using
10176 the virtual function table. Note that the virtual function table is
10177 required---this feature can only work for objects that have run-time
10178 type identification; a single virtual method in the object's declared
10179 type is sufficient. Note that this setting is also taken into account when
10180 working with variable objects via MI (@pxref{GDB/MI}).
10182 @item set print object off
10183 Display only the declared type of objects, without reference to the
10184 virtual function table. This is the default setting.
10186 @item show print object
10187 Show whether actual, or declared, object types are displayed.
10189 @item set print static-members
10190 @itemx set print static-members on
10191 @cindex static members of C@t{++} objects
10192 Print static members when displaying a C@t{++} object. The default is on.
10194 @item set print static-members off
10195 Do not print static members when displaying a C@t{++} object.
10197 @item show print static-members
10198 Show whether C@t{++} static members are printed or not.
10200 @item set print pascal_static-members
10201 @itemx set print pascal_static-members on
10202 @cindex static members of Pascal objects
10203 @cindex Pascal objects, static members display
10204 Print static members when displaying a Pascal object. The default is on.
10206 @item set print pascal_static-members off
10207 Do not print static members when displaying a Pascal object.
10209 @item show print pascal_static-members
10210 Show whether Pascal static members are printed or not.
10212 @c These don't work with HP ANSI C++ yet.
10213 @item set print vtbl
10214 @itemx set print vtbl on
10215 @cindex pretty print C@t{++} virtual function tables
10216 @cindex virtual functions (C@t{++}) display
10217 @cindex VTBL display
10218 Pretty print C@t{++} virtual function tables. The default is off.
10219 (The @code{vtbl} commands do not work on programs compiled with the HP
10220 ANSI C@t{++} compiler (@code{aCC}).)
10222 @item set print vtbl off
10223 Do not pretty print C@t{++} virtual function tables.
10225 @item show print vtbl
10226 Show whether C@t{++} virtual function tables are pretty printed, or not.
10229 @node Pretty Printing
10230 @section Pretty Printing
10232 @value{GDBN} provides a mechanism to allow pretty-printing of values using
10233 Python code. It greatly simplifies the display of complex objects. This
10234 mechanism works for both MI and the CLI.
10237 * Pretty-Printer Introduction:: Introduction to pretty-printers
10238 * Pretty-Printer Example:: An example pretty-printer
10239 * Pretty-Printer Commands:: Pretty-printer commands
10242 @node Pretty-Printer Introduction
10243 @subsection Pretty-Printer Introduction
10245 When @value{GDBN} prints a value, it first sees if there is a pretty-printer
10246 registered for the value. If there is then @value{GDBN} invokes the
10247 pretty-printer to print the value. Otherwise the value is printed normally.
10249 Pretty-printers are normally named. This makes them easy to manage.
10250 The @samp{info pretty-printer} command will list all the installed
10251 pretty-printers with their names.
10252 If a pretty-printer can handle multiple data types, then its
10253 @dfn{subprinters} are the printers for the individual data types.
10254 Each such subprinter has its own name.
10255 The format of the name is @var{printer-name};@var{subprinter-name}.
10257 Pretty-printers are installed by @dfn{registering} them with @value{GDBN}.
10258 Typically they are automatically loaded and registered when the corresponding
10259 debug information is loaded, thus making them available without having to
10260 do anything special.
10262 There are three places where a pretty-printer can be registered.
10266 Pretty-printers registered globally are available when debugging
10270 Pretty-printers registered with a program space are available only
10271 when debugging that program.
10272 @xref{Progspaces In Python}, for more details on program spaces in Python.
10275 Pretty-printers registered with an objfile are loaded and unloaded
10276 with the corresponding objfile (e.g., shared library).
10277 @xref{Objfiles In Python}, for more details on objfiles in Python.
10280 @xref{Selecting Pretty-Printers}, for further information on how
10281 pretty-printers are selected,
10283 @xref{Writing a Pretty-Printer}, for implementing pretty printers
10286 @node Pretty-Printer Example
10287 @subsection Pretty-Printer Example
10289 Here is how a C@t{++} @code{std::string} looks without a pretty-printer:
10292 (@value{GDBP}) print s
10294 static npos = 4294967295,
10296 <std::allocator<char>> = @{
10297 <__gnu_cxx::new_allocator<char>> = @{
10298 <No data fields>@}, <No data fields>
10300 members of std::basic_string<char, std::char_traits<char>,
10301 std::allocator<char> >::_Alloc_hider:
10302 _M_p = 0x804a014 "abcd"
10307 With a pretty-printer for @code{std::string} only the contents are printed:
10310 (@value{GDBP}) print s
10314 @node Pretty-Printer Commands
10315 @subsection Pretty-Printer Commands
10316 @cindex pretty-printer commands
10319 @kindex info pretty-printer
10320 @item info pretty-printer [@var{object-regexp} [@var{name-regexp}]]
10321 Print the list of installed pretty-printers.
10322 This includes disabled pretty-printers, which are marked as such.
10324 @var{object-regexp} is a regular expression matching the objects
10325 whose pretty-printers to list.
10326 Objects can be @code{global}, the program space's file
10327 (@pxref{Progspaces In Python}),
10328 and the object files within that program space (@pxref{Objfiles In Python}).
10329 @xref{Selecting Pretty-Printers}, for details on how @value{GDBN}
10330 looks up a printer from these three objects.
10332 @var{name-regexp} is a regular expression matching the name of the printers
10335 @kindex disable pretty-printer
10336 @item disable pretty-printer [@var{object-regexp} [@var{name-regexp}]]
10337 Disable pretty-printers matching @var{object-regexp} and @var{name-regexp}.
10338 A disabled pretty-printer is not forgotten, it may be enabled again later.
10340 @kindex enable pretty-printer
10341 @item enable pretty-printer [@var{object-regexp} [@var{name-regexp}]]
10342 Enable pretty-printers matching @var{object-regexp} and @var{name-regexp}.
10347 Suppose we have three pretty-printers installed: one from library1.so
10348 named @code{foo} that prints objects of type @code{foo}, and
10349 another from library2.so named @code{bar} that prints two types of objects,
10350 @code{bar1} and @code{bar2}.
10353 (gdb) info pretty-printer
10360 (gdb) info pretty-printer library2
10365 (gdb) disable pretty-printer library1
10367 2 of 3 printers enabled
10368 (gdb) info pretty-printer
10375 (gdb) disable pretty-printer library2 bar:bar1
10377 1 of 3 printers enabled
10378 (gdb) info pretty-printer library2
10385 (gdb) disable pretty-printer library2 bar
10387 0 of 3 printers enabled
10388 (gdb) info pretty-printer library2
10397 Note that for @code{bar} the entire printer can be disabled,
10398 as can each individual subprinter.
10400 @node Value History
10401 @section Value History
10403 @cindex value history
10404 @cindex history of values printed by @value{GDBN}
10405 Values printed by the @code{print} command are saved in the @value{GDBN}
10406 @dfn{value history}. This allows you to refer to them in other expressions.
10407 Values are kept until the symbol table is re-read or discarded
10408 (for example with the @code{file} or @code{symbol-file} commands).
10409 When the symbol table changes, the value history is discarded,
10410 since the values may contain pointers back to the types defined in the
10415 @cindex history number
10416 The values printed are given @dfn{history numbers} by which you can
10417 refer to them. These are successive integers starting with one.
10418 @code{print} shows you the history number assigned to a value by
10419 printing @samp{$@var{num} = } before the value; here @var{num} is the
10422 To refer to any previous value, use @samp{$} followed by the value's
10423 history number. The way @code{print} labels its output is designed to
10424 remind you of this. Just @code{$} refers to the most recent value in
10425 the history, and @code{$$} refers to the value before that.
10426 @code{$$@var{n}} refers to the @var{n}th value from the end; @code{$$2}
10427 is the value just prior to @code{$$}, @code{$$1} is equivalent to
10428 @code{$$}, and @code{$$0} is equivalent to @code{$}.
10430 For example, suppose you have just printed a pointer to a structure and
10431 want to see the contents of the structure. It suffices to type
10437 If you have a chain of structures where the component @code{next} points
10438 to the next one, you can print the contents of the next one with this:
10445 You can print successive links in the chain by repeating this
10446 command---which you can do by just typing @key{RET}.
10448 Note that the history records values, not expressions. If the value of
10449 @code{x} is 4 and you type these commands:
10457 then the value recorded in the value history by the @code{print} command
10458 remains 4 even though the value of @code{x} has changed.
10461 @kindex show values
10463 Print the last ten values in the value history, with their item numbers.
10464 This is like @samp{p@ $$9} repeated ten times, except that @code{show
10465 values} does not change the history.
10467 @item show values @var{n}
10468 Print ten history values centered on history item number @var{n}.
10470 @item show values +
10471 Print ten history values just after the values last printed. If no more
10472 values are available, @code{show values +} produces no display.
10475 Pressing @key{RET} to repeat @code{show values @var{n}} has exactly the
10476 same effect as @samp{show values +}.
10478 @node Convenience Vars
10479 @section Convenience Variables
10481 @cindex convenience variables
10482 @cindex user-defined variables
10483 @value{GDBN} provides @dfn{convenience variables} that you can use within
10484 @value{GDBN} to hold on to a value and refer to it later. These variables
10485 exist entirely within @value{GDBN}; they are not part of your program, and
10486 setting a convenience variable has no direct effect on further execution
10487 of your program. That is why you can use them freely.
10489 Convenience variables are prefixed with @samp{$}. Any name preceded by
10490 @samp{$} can be used for a convenience variable, unless it is one of
10491 the predefined machine-specific register names (@pxref{Registers, ,Registers}).
10492 (Value history references, in contrast, are @emph{numbers} preceded
10493 by @samp{$}. @xref{Value History, ,Value History}.)
10495 You can save a value in a convenience variable with an assignment
10496 expression, just as you would set a variable in your program.
10500 set $foo = *object_ptr
10504 would save in @code{$foo} the value contained in the object pointed to by
10507 Using a convenience variable for the first time creates it, but its
10508 value is @code{void} until you assign a new value. You can alter the
10509 value with another assignment at any time.
10511 Convenience variables have no fixed types. You can assign a convenience
10512 variable any type of value, including structures and arrays, even if
10513 that variable already has a value of a different type. The convenience
10514 variable, when used as an expression, has the type of its current value.
10517 @kindex show convenience
10518 @cindex show all user variables and functions
10519 @item show convenience
10520 Print a list of convenience variables used so far, and their values,
10521 as well as a list of the convenience functions.
10522 Abbreviated @code{show conv}.
10524 @kindex init-if-undefined
10525 @cindex convenience variables, initializing
10526 @item init-if-undefined $@var{variable} = @var{expression}
10527 Set a convenience variable if it has not already been set. This is useful
10528 for user-defined commands that keep some state. It is similar, in concept,
10529 to using local static variables with initializers in C (except that
10530 convenience variables are global). It can also be used to allow users to
10531 override default values used in a command script.
10533 If the variable is already defined then the expression is not evaluated so
10534 any side-effects do not occur.
10537 One of the ways to use a convenience variable is as a counter to be
10538 incremented or a pointer to be advanced. For example, to print
10539 a field from successive elements of an array of structures:
10543 print bar[$i++]->contents
10547 Repeat that command by typing @key{RET}.
10549 Some convenience variables are created automatically by @value{GDBN} and given
10550 values likely to be useful.
10553 @vindex $_@r{, convenience variable}
10555 The variable @code{$_} is automatically set by the @code{x} command to
10556 the last address examined (@pxref{Memory, ,Examining Memory}). Other
10557 commands which provide a default address for @code{x} to examine also
10558 set @code{$_} to that address; these commands include @code{info line}
10559 and @code{info breakpoint}. The type of @code{$_} is @code{void *}
10560 except when set by the @code{x} command, in which case it is a pointer
10561 to the type of @code{$__}.
10563 @vindex $__@r{, convenience variable}
10565 The variable @code{$__} is automatically set by the @code{x} command
10566 to the value found in the last address examined. Its type is chosen
10567 to match the format in which the data was printed.
10570 @vindex $_exitcode@r{, convenience variable}
10571 When the program being debugged terminates normally, @value{GDBN}
10572 automatically sets this variable to the exit code of the program, and
10573 resets @code{$_exitsignal} to @code{void}.
10576 @vindex $_exitsignal@r{, convenience variable}
10577 When the program being debugged dies due to an uncaught signal,
10578 @value{GDBN} automatically sets this variable to that signal's number,
10579 and resets @code{$_exitcode} to @code{void}.
10581 To distinguish between whether the program being debugged has exited
10582 (i.e., @code{$_exitcode} is not @code{void}) or signalled (i.e.,
10583 @code{$_exitsignal} is not @code{void}), the convenience function
10584 @code{$_isvoid} can be used (@pxref{Convenience Funs,, Convenience
10585 Functions}). For example, considering the following source code:
10588 #include <signal.h>
10591 main (int argc, char *argv[])
10598 A valid way of telling whether the program being debugged has exited
10599 or signalled would be:
10602 (@value{GDBP}) define has_exited_or_signalled
10603 Type commands for definition of ``has_exited_or_signalled''.
10604 End with a line saying just ``end''.
10605 >if $_isvoid ($_exitsignal)
10606 >echo The program has exited\n
10608 >echo The program has signalled\n
10614 Program terminated with signal SIGALRM, Alarm clock.
10615 The program no longer exists.
10616 (@value{GDBP}) has_exited_or_signalled
10617 The program has signalled
10620 As can be seen, @value{GDBN} correctly informs that the program being
10621 debugged has signalled, since it calls @code{raise} and raises a
10622 @code{SIGALRM} signal. If the program being debugged had not called
10623 @code{raise}, then @value{GDBN} would report a normal exit:
10626 (@value{GDBP}) has_exited_or_signalled
10627 The program has exited
10631 The variable @code{$_exception} is set to the exception object being
10632 thrown at an exception-related catchpoint. @xref{Set Catchpoints}.
10635 @itemx $_probe_arg0@dots{}$_probe_arg11
10636 Arguments to a static probe. @xref{Static Probe Points}.
10639 @vindex $_sdata@r{, inspect, convenience variable}
10640 The variable @code{$_sdata} contains extra collected static tracepoint
10641 data. @xref{Tracepoint Actions,,Tracepoint Action Lists}. Note that
10642 @code{$_sdata} could be empty, if not inspecting a trace buffer, or
10643 if extra static tracepoint data has not been collected.
10646 @vindex $_siginfo@r{, convenience variable}
10647 The variable @code{$_siginfo} contains extra signal information
10648 (@pxref{extra signal information}). Note that @code{$_siginfo}
10649 could be empty, if the application has not yet received any signals.
10650 For example, it will be empty before you execute the @code{run} command.
10653 @vindex $_tlb@r{, convenience variable}
10654 The variable @code{$_tlb} is automatically set when debugging
10655 applications running on MS-Windows in native mode or connected to
10656 gdbserver that supports the @code{qGetTIBAddr} request.
10657 @xref{General Query Packets}.
10658 This variable contains the address of the thread information block.
10661 The number of the current inferior. @xref{Inferiors and
10662 Programs, ,Debugging Multiple Inferiors and Programs}.
10665 The thread number of the current thread. @xref{thread numbers}.
10668 The global number of the current thread. @xref{global thread numbers}.
10672 @node Convenience Funs
10673 @section Convenience Functions
10675 @cindex convenience functions
10676 @value{GDBN} also supplies some @dfn{convenience functions}. These
10677 have a syntax similar to convenience variables. A convenience
10678 function can be used in an expression just like an ordinary function;
10679 however, a convenience function is implemented internally to
10682 These functions do not require @value{GDBN} to be configured with
10683 @code{Python} support, which means that they are always available.
10687 @item $_isvoid (@var{expr})
10688 @findex $_isvoid@r{, convenience function}
10689 Return one if the expression @var{expr} is @code{void}. Otherwise it
10692 A @code{void} expression is an expression where the type of the result
10693 is @code{void}. For example, you can examine a convenience variable
10694 (see @ref{Convenience Vars,, Convenience Variables}) to check whether
10698 (@value{GDBP}) print $_exitcode
10700 (@value{GDBP}) print $_isvoid ($_exitcode)
10703 Starting program: ./a.out
10704 [Inferior 1 (process 29572) exited normally]
10705 (@value{GDBP}) print $_exitcode
10707 (@value{GDBP}) print $_isvoid ($_exitcode)
10711 In the example above, we used @code{$_isvoid} to check whether
10712 @code{$_exitcode} is @code{void} before and after the execution of the
10713 program being debugged. Before the execution there is no exit code to
10714 be examined, therefore @code{$_exitcode} is @code{void}. After the
10715 execution the program being debugged returned zero, therefore
10716 @code{$_exitcode} is zero, which means that it is not @code{void}
10719 The @code{void} expression can also be a call of a function from the
10720 program being debugged. For example, given the following function:
10729 The result of calling it inside @value{GDBN} is @code{void}:
10732 (@value{GDBP}) print foo ()
10734 (@value{GDBP}) print $_isvoid (foo ())
10736 (@value{GDBP}) set $v = foo ()
10737 (@value{GDBP}) print $v
10739 (@value{GDBP}) print $_isvoid ($v)
10745 These functions require @value{GDBN} to be configured with
10746 @code{Python} support.
10750 @item $_memeq(@var{buf1}, @var{buf2}, @var{length})
10751 @findex $_memeq@r{, convenience function}
10752 Returns one if the @var{length} bytes at the addresses given by
10753 @var{buf1} and @var{buf2} are equal.
10754 Otherwise it returns zero.
10756 @item $_regex(@var{str}, @var{regex})
10757 @findex $_regex@r{, convenience function}
10758 Returns one if the string @var{str} matches the regular expression
10759 @var{regex}. Otherwise it returns zero.
10760 The syntax of the regular expression is that specified by @code{Python}'s
10761 regular expression support.
10763 @item $_streq(@var{str1}, @var{str2})
10764 @findex $_streq@r{, convenience function}
10765 Returns one if the strings @var{str1} and @var{str2} are equal.
10766 Otherwise it returns zero.
10768 @item $_strlen(@var{str})
10769 @findex $_strlen@r{, convenience function}
10770 Returns the length of string @var{str}.
10772 @item $_caller_is(@var{name}@r{[}, @var{number_of_frames}@r{]})
10773 @findex $_caller_is@r{, convenience function}
10774 Returns one if the calling function's name is equal to @var{name}.
10775 Otherwise it returns zero.
10777 If the optional argument @var{number_of_frames} is provided,
10778 it is the number of frames up in the stack to look.
10786 at testsuite/gdb.python/py-caller-is.c:21
10787 #1 0x00000000004005a0 in middle_func ()
10788 at testsuite/gdb.python/py-caller-is.c:27
10789 #2 0x00000000004005ab in top_func ()
10790 at testsuite/gdb.python/py-caller-is.c:33
10791 #3 0x00000000004005b6 in main ()
10792 at testsuite/gdb.python/py-caller-is.c:39
10793 (gdb) print $_caller_is ("middle_func")
10795 (gdb) print $_caller_is ("top_func", 2)
10799 @item $_caller_matches(@var{regexp}@r{[}, @var{number_of_frames}@r{]})
10800 @findex $_caller_matches@r{, convenience function}
10801 Returns one if the calling function's name matches the regular expression
10802 @var{regexp}. Otherwise it returns zero.
10804 If the optional argument @var{number_of_frames} is provided,
10805 it is the number of frames up in the stack to look.
10808 @item $_any_caller_is(@var{name}@r{[}, @var{number_of_frames}@r{]})
10809 @findex $_any_caller_is@r{, convenience function}
10810 Returns one if any calling function's name is equal to @var{name}.
10811 Otherwise it returns zero.
10813 If the optional argument @var{number_of_frames} is provided,
10814 it is the number of frames up in the stack to look.
10817 This function differs from @code{$_caller_is} in that this function
10818 checks all stack frames from the immediate caller to the frame specified
10819 by @var{number_of_frames}, whereas @code{$_caller_is} only checks the
10820 frame specified by @var{number_of_frames}.
10822 @item $_any_caller_matches(@var{regexp}@r{[}, @var{number_of_frames}@r{]})
10823 @findex $_any_caller_matches@r{, convenience function}
10824 Returns one if any calling function's name matches the regular expression
10825 @var{regexp}. Otherwise it returns zero.
10827 If the optional argument @var{number_of_frames} is provided,
10828 it is the number of frames up in the stack to look.
10831 This function differs from @code{$_caller_matches} in that this function
10832 checks all stack frames from the immediate caller to the frame specified
10833 by @var{number_of_frames}, whereas @code{$_caller_matches} only checks the
10834 frame specified by @var{number_of_frames}.
10836 @item $_as_string(@var{value})
10837 @findex $_as_string@r{, convenience function}
10838 Return the string representation of @var{value}.
10840 This function is useful to obtain the textual label (enumerator) of an
10841 enumeration value. For example, assuming the variable @var{node} is of
10842 an enumerated type:
10845 (gdb) printf "Visiting node of type %s\n", $_as_string(node)
10846 Visiting node of type NODE_INTEGER
10851 @value{GDBN} provides the ability to list and get help on
10852 convenience functions.
10855 @item help function
10856 @kindex help function
10857 @cindex show all convenience functions
10858 Print a list of all convenience functions.
10865 You can refer to machine register contents, in expressions, as variables
10866 with names starting with @samp{$}. The names of registers are different
10867 for each machine; use @code{info registers} to see the names used on
10871 @kindex info registers
10872 @item info registers
10873 Print the names and values of all registers except floating-point
10874 and vector registers (in the selected stack frame).
10876 @kindex info all-registers
10877 @cindex floating point registers
10878 @item info all-registers
10879 Print the names and values of all registers, including floating-point
10880 and vector registers (in the selected stack frame).
10882 @item info registers @var{regname} @dots{}
10883 Print the @dfn{relativized} value of each specified register @var{regname}.
10884 As discussed in detail below, register values are normally relative to
10885 the selected stack frame. The @var{regname} may be any register name valid on
10886 the machine you are using, with or without the initial @samp{$}.
10889 @anchor{standard registers}
10890 @cindex stack pointer register
10891 @cindex program counter register
10892 @cindex process status register
10893 @cindex frame pointer register
10894 @cindex standard registers
10895 @value{GDBN} has four ``standard'' register names that are available (in
10896 expressions) on most machines---whenever they do not conflict with an
10897 architecture's canonical mnemonics for registers. The register names
10898 @code{$pc} and @code{$sp} are used for the program counter register and
10899 the stack pointer. @code{$fp} is used for a register that contains a
10900 pointer to the current stack frame, and @code{$ps} is used for a
10901 register that contains the processor status. For example,
10902 you could print the program counter in hex with
10909 or print the instruction to be executed next with
10916 or add four to the stack pointer@footnote{This is a way of removing
10917 one word from the stack, on machines where stacks grow downward in
10918 memory (most machines, nowadays). This assumes that the innermost
10919 stack frame is selected; setting @code{$sp} is not allowed when other
10920 stack frames are selected. To pop entire frames off the stack,
10921 regardless of machine architecture, use @code{return};
10922 see @ref{Returning, ,Returning from a Function}.} with
10928 Whenever possible, these four standard register names are available on
10929 your machine even though the machine has different canonical mnemonics,
10930 so long as there is no conflict. The @code{info registers} command
10931 shows the canonical names. For example, on the SPARC, @code{info
10932 registers} displays the processor status register as @code{$psr} but you
10933 can also refer to it as @code{$ps}; and on x86-based machines @code{$ps}
10934 is an alias for the @sc{eflags} register.
10936 @value{GDBN} always considers the contents of an ordinary register as an
10937 integer when the register is examined in this way. Some machines have
10938 special registers which can hold nothing but floating point; these
10939 registers are considered to have floating point values. There is no way
10940 to refer to the contents of an ordinary register as floating point value
10941 (although you can @emph{print} it as a floating point value with
10942 @samp{print/f $@var{regname}}).
10944 Some registers have distinct ``raw'' and ``virtual'' data formats. This
10945 means that the data format in which the register contents are saved by
10946 the operating system is not the same one that your program normally
10947 sees. For example, the registers of the 68881 floating point
10948 coprocessor are always saved in ``extended'' (raw) format, but all C
10949 programs expect to work with ``double'' (virtual) format. In such
10950 cases, @value{GDBN} normally works with the virtual format only (the format
10951 that makes sense for your program), but the @code{info registers} command
10952 prints the data in both formats.
10954 @cindex SSE registers (x86)
10955 @cindex MMX registers (x86)
10956 Some machines have special registers whose contents can be interpreted
10957 in several different ways. For example, modern x86-based machines
10958 have SSE and MMX registers that can hold several values packed
10959 together in several different formats. @value{GDBN} refers to such
10960 registers in @code{struct} notation:
10963 (@value{GDBP}) print $xmm1
10965 v4_float = @{0, 3.43859137e-038, 1.54142831e-044, 1.821688e-044@},
10966 v2_double = @{9.92129282474342e-303, 2.7585945287983262e-313@},
10967 v16_int8 = "\000\000\000\000\3706;\001\v\000\000\000\r\000\000",
10968 v8_int16 = @{0, 0, 14072, 315, 11, 0, 13, 0@},
10969 v4_int32 = @{0, 20657912, 11, 13@},
10970 v2_int64 = @{88725056443645952, 55834574859@},
10971 uint128 = 0x0000000d0000000b013b36f800000000
10976 To set values of such registers, you need to tell @value{GDBN} which
10977 view of the register you wish to change, as if you were assigning
10978 value to a @code{struct} member:
10981 (@value{GDBP}) set $xmm1.uint128 = 0x000000000000000000000000FFFFFFFF
10984 Normally, register values are relative to the selected stack frame
10985 (@pxref{Selection, ,Selecting a Frame}). This means that you get the
10986 value that the register would contain if all stack frames farther in
10987 were exited and their saved registers restored. In order to see the
10988 true contents of hardware registers, you must select the innermost
10989 frame (with @samp{frame 0}).
10991 @cindex caller-saved registers
10992 @cindex call-clobbered registers
10993 @cindex volatile registers
10994 @cindex <not saved> values
10995 Usually ABIs reserve some registers as not needed to be saved by the
10996 callee (a.k.a.: ``caller-saved'', ``call-clobbered'' or ``volatile''
10997 registers). It may therefore not be possible for @value{GDBN} to know
10998 the value a register had before the call (in other words, in the outer
10999 frame), if the register value has since been changed by the callee.
11000 @value{GDBN} tries to deduce where the inner frame saved
11001 (``callee-saved'') registers, from the debug info, unwind info, or the
11002 machine code generated by your compiler. If some register is not
11003 saved, and @value{GDBN} knows the register is ``caller-saved'' (via
11004 its own knowledge of the ABI, or because the debug/unwind info
11005 explicitly says the register's value is undefined), @value{GDBN}
11006 displays @w{@samp{<not saved>}} as the register's value. With targets
11007 that @value{GDBN} has no knowledge of the register saving convention,
11008 if a register was not saved by the callee, then its value and location
11009 in the outer frame are assumed to be the same of the inner frame.
11010 This is usually harmless, because if the register is call-clobbered,
11011 the caller either does not care what is in the register after the
11012 call, or has code to restore the value that it does care about. Note,
11013 however, that if you change such a register in the outer frame, you
11014 may also be affecting the inner frame. Also, the more ``outer'' the
11015 frame is you're looking at, the more likely a call-clobbered
11016 register's value is to be wrong, in the sense that it doesn't actually
11017 represent the value the register had just before the call.
11019 @node Floating Point Hardware
11020 @section Floating Point Hardware
11021 @cindex floating point
11023 Depending on the configuration, @value{GDBN} may be able to give
11024 you more information about the status of the floating point hardware.
11029 Display hardware-dependent information about the floating
11030 point unit. The exact contents and layout vary depending on the
11031 floating point chip. Currently, @samp{info float} is supported on
11032 the ARM and x86 machines.
11036 @section Vector Unit
11037 @cindex vector unit
11039 Depending on the configuration, @value{GDBN} may be able to give you
11040 more information about the status of the vector unit.
11043 @kindex info vector
11045 Display information about the vector unit. The exact contents and
11046 layout vary depending on the hardware.
11049 @node OS Information
11050 @section Operating System Auxiliary Information
11051 @cindex OS information
11053 @value{GDBN} provides interfaces to useful OS facilities that can help
11054 you debug your program.
11056 @cindex auxiliary vector
11057 @cindex vector, auxiliary
11058 Some operating systems supply an @dfn{auxiliary vector} to programs at
11059 startup. This is akin to the arguments and environment that you
11060 specify for a program, but contains a system-dependent variety of
11061 binary values that tell system libraries important details about the
11062 hardware, operating system, and process. Each value's purpose is
11063 identified by an integer tag; the meanings are well-known but system-specific.
11064 Depending on the configuration and operating system facilities,
11065 @value{GDBN} may be able to show you this information. For remote
11066 targets, this functionality may further depend on the remote stub's
11067 support of the @samp{qXfer:auxv:read} packet, see
11068 @ref{qXfer auxiliary vector read}.
11073 Display the auxiliary vector of the inferior, which can be either a
11074 live process or a core dump file. @value{GDBN} prints each tag value
11075 numerically, and also shows names and text descriptions for recognized
11076 tags. Some values in the vector are numbers, some bit masks, and some
11077 pointers to strings or other data. @value{GDBN} displays each value in the
11078 most appropriate form for a recognized tag, and in hexadecimal for
11079 an unrecognized tag.
11082 On some targets, @value{GDBN} can access operating system-specific
11083 information and show it to you. The types of information available
11084 will differ depending on the type of operating system running on the
11085 target. The mechanism used to fetch the data is described in
11086 @ref{Operating System Information}. For remote targets, this
11087 functionality depends on the remote stub's support of the
11088 @samp{qXfer:osdata:read} packet, see @ref{qXfer osdata read}.
11092 @item info os @var{infotype}
11094 Display OS information of the requested type.
11096 On @sc{gnu}/Linux, the following values of @var{infotype} are valid:
11098 @anchor{linux info os infotypes}
11100 @kindex info os cpus
11102 Display the list of all CPUs/cores. For each CPU/core, @value{GDBN} prints
11103 the available fields from /proc/cpuinfo. For each supported architecture
11104 different fields are available. Two common entries are processor which gives
11105 CPU number and bogomips; a system constant that is calculated during
11106 kernel initialization.
11108 @kindex info os files
11110 Display the list of open file descriptors on the target. For each
11111 file descriptor, @value{GDBN} prints the identifier of the process
11112 owning the descriptor, the command of the owning process, the value
11113 of the descriptor, and the target of the descriptor.
11115 @kindex info os modules
11117 Display the list of all loaded kernel modules on the target. For each
11118 module, @value{GDBN} prints the module name, the size of the module in
11119 bytes, the number of times the module is used, the dependencies of the
11120 module, the status of the module, and the address of the loaded module
11123 @kindex info os msg
11125 Display the list of all System V message queues on the target. For each
11126 message queue, @value{GDBN} prints the message queue key, the message
11127 queue identifier, the access permissions, the current number of bytes
11128 on the queue, the current number of messages on the queue, the processes
11129 that last sent and received a message on the queue, the user and group
11130 of the owner and creator of the message queue, the times at which a
11131 message was last sent and received on the queue, and the time at which
11132 the message queue was last changed.
11134 @kindex info os processes
11136 Display the list of processes on the target. For each process,
11137 @value{GDBN} prints the process identifier, the name of the user, the
11138 command corresponding to the process, and the list of processor cores
11139 that the process is currently running on. (To understand what these
11140 properties mean, for this and the following info types, please consult
11141 the general @sc{gnu}/Linux documentation.)
11143 @kindex info os procgroups
11145 Display the list of process groups on the target. For each process,
11146 @value{GDBN} prints the identifier of the process group that it belongs
11147 to, the command corresponding to the process group leader, the process
11148 identifier, and the command line of the process. The list is sorted
11149 first by the process group identifier, then by the process identifier,
11150 so that processes belonging to the same process group are grouped together
11151 and the process group leader is listed first.
11153 @kindex info os semaphores
11155 Display the list of all System V semaphore sets on the target. For each
11156 semaphore set, @value{GDBN} prints the semaphore set key, the semaphore
11157 set identifier, the access permissions, the number of semaphores in the
11158 set, the user and group of the owner and creator of the semaphore set,
11159 and the times at which the semaphore set was operated upon and changed.
11161 @kindex info os shm
11163 Display the list of all System V shared-memory regions on the target.
11164 For each shared-memory region, @value{GDBN} prints the region key,
11165 the shared-memory identifier, the access permissions, the size of the
11166 region, the process that created the region, the process that last
11167 attached to or detached from the region, the current number of live
11168 attaches to the region, and the times at which the region was last
11169 attached to, detach from, and changed.
11171 @kindex info os sockets
11173 Display the list of Internet-domain sockets on the target. For each
11174 socket, @value{GDBN} prints the address and port of the local and
11175 remote endpoints, the current state of the connection, the creator of
11176 the socket, the IP address family of the socket, and the type of the
11179 @kindex info os threads
11181 Display the list of threads running on the target. For each thread,
11182 @value{GDBN} prints the identifier of the process that the thread
11183 belongs to, the command of the process, the thread identifier, and the
11184 processor core that it is currently running on. The main thread of a
11185 process is not listed.
11189 If @var{infotype} is omitted, then list the possible values for
11190 @var{infotype} and the kind of OS information available for each
11191 @var{infotype}. If the target does not return a list of possible
11192 types, this command will report an error.
11195 @node Memory Region Attributes
11196 @section Memory Region Attributes
11197 @cindex memory region attributes
11199 @dfn{Memory region attributes} allow you to describe special handling
11200 required by regions of your target's memory. @value{GDBN} uses
11201 attributes to determine whether to allow certain types of memory
11202 accesses; whether to use specific width accesses; and whether to cache
11203 target memory. By default the description of memory regions is
11204 fetched from the target (if the current target supports this), but the
11205 user can override the fetched regions.
11207 Defined memory regions can be individually enabled and disabled. When a
11208 memory region is disabled, @value{GDBN} uses the default attributes when
11209 accessing memory in that region. Similarly, if no memory regions have
11210 been defined, @value{GDBN} uses the default attributes when accessing
11213 When a memory region is defined, it is given a number to identify it;
11214 to enable, disable, or remove a memory region, you specify that number.
11218 @item mem @var{lower} @var{upper} @var{attributes}@dots{}
11219 Define a memory region bounded by @var{lower} and @var{upper} with
11220 attributes @var{attributes}@dots{}, and add it to the list of regions
11221 monitored by @value{GDBN}. Note that @var{upper} == 0 is a special
11222 case: it is treated as the target's maximum memory address.
11223 (0xffff on 16 bit targets, 0xffffffff on 32 bit targets, etc.)
11226 Discard any user changes to the memory regions and use target-supplied
11227 regions, if available, or no regions if the target does not support.
11230 @item delete mem @var{nums}@dots{}
11231 Remove memory regions @var{nums}@dots{} from the list of regions
11232 monitored by @value{GDBN}.
11234 @kindex disable mem
11235 @item disable mem @var{nums}@dots{}
11236 Disable monitoring of memory regions @var{nums}@dots{}.
11237 A disabled memory region is not forgotten.
11238 It may be enabled again later.
11241 @item enable mem @var{nums}@dots{}
11242 Enable monitoring of memory regions @var{nums}@dots{}.
11246 Print a table of all defined memory regions, with the following columns
11250 @item Memory Region Number
11251 @item Enabled or Disabled.
11252 Enabled memory regions are marked with @samp{y}.
11253 Disabled memory regions are marked with @samp{n}.
11256 The address defining the inclusive lower bound of the memory region.
11259 The address defining the exclusive upper bound of the memory region.
11262 The list of attributes set for this memory region.
11267 @subsection Attributes
11269 @subsubsection Memory Access Mode
11270 The access mode attributes set whether @value{GDBN} may make read or
11271 write accesses to a memory region.
11273 While these attributes prevent @value{GDBN} from performing invalid
11274 memory accesses, they do nothing to prevent the target system, I/O DMA,
11275 etc.@: from accessing memory.
11279 Memory is read only.
11281 Memory is write only.
11283 Memory is read/write. This is the default.
11286 @subsubsection Memory Access Size
11287 The access size attribute tells @value{GDBN} to use specific sized
11288 accesses in the memory region. Often memory mapped device registers
11289 require specific sized accesses. If no access size attribute is
11290 specified, @value{GDBN} may use accesses of any size.
11294 Use 8 bit memory accesses.
11296 Use 16 bit memory accesses.
11298 Use 32 bit memory accesses.
11300 Use 64 bit memory accesses.
11303 @c @subsubsection Hardware/Software Breakpoints
11304 @c The hardware/software breakpoint attributes set whether @value{GDBN}
11305 @c will use hardware or software breakpoints for the internal breakpoints
11306 @c used by the step, next, finish, until, etc. commands.
11310 @c Always use hardware breakpoints
11311 @c @item swbreak (default)
11314 @subsubsection Data Cache
11315 The data cache attributes set whether @value{GDBN} will cache target
11316 memory. While this generally improves performance by reducing debug
11317 protocol overhead, it can lead to incorrect results because @value{GDBN}
11318 does not know about volatile variables or memory mapped device
11323 Enable @value{GDBN} to cache target memory.
11325 Disable @value{GDBN} from caching target memory. This is the default.
11328 @subsection Memory Access Checking
11329 @value{GDBN} can be instructed to refuse accesses to memory that is
11330 not explicitly described. This can be useful if accessing such
11331 regions has undesired effects for a specific target, or to provide
11332 better error checking. The following commands control this behaviour.
11335 @kindex set mem inaccessible-by-default
11336 @item set mem inaccessible-by-default [on|off]
11337 If @code{on} is specified, make @value{GDBN} treat memory not
11338 explicitly described by the memory ranges as non-existent and refuse accesses
11339 to such memory. The checks are only performed if there's at least one
11340 memory range defined. If @code{off} is specified, make @value{GDBN}
11341 treat the memory not explicitly described by the memory ranges as RAM.
11342 The default value is @code{on}.
11343 @kindex show mem inaccessible-by-default
11344 @item show mem inaccessible-by-default
11345 Show the current handling of accesses to unknown memory.
11349 @c @subsubsection Memory Write Verification
11350 @c The memory write verification attributes set whether @value{GDBN}
11351 @c will re-reads data after each write to verify the write was successful.
11355 @c @item noverify (default)
11358 @node Dump/Restore Files
11359 @section Copy Between Memory and a File
11360 @cindex dump/restore files
11361 @cindex append data to a file
11362 @cindex dump data to a file
11363 @cindex restore data from a file
11365 You can use the commands @code{dump}, @code{append}, and
11366 @code{restore} to copy data between target memory and a file. The
11367 @code{dump} and @code{append} commands write data to a file, and the
11368 @code{restore} command reads data from a file back into the inferior's
11369 memory. Files may be in binary, Motorola S-record, Intel hex,
11370 Tektronix Hex, or Verilog Hex format; however, @value{GDBN} can only
11371 append to binary files, and cannot read from Verilog Hex files.
11376 @item dump @r{[}@var{format}@r{]} memory @var{filename} @var{start_addr} @var{end_addr}
11377 @itemx dump @r{[}@var{format}@r{]} value @var{filename} @var{expr}
11378 Dump the contents of memory from @var{start_addr} to @var{end_addr},
11379 or the value of @var{expr}, to @var{filename} in the given format.
11381 The @var{format} parameter may be any one of:
11388 Motorola S-record format.
11390 Tektronix Hex format.
11392 Verilog Hex format.
11395 @value{GDBN} uses the same definitions of these formats as the
11396 @sc{gnu} binary utilities, like @samp{objdump} and @samp{objcopy}. If
11397 @var{format} is omitted, @value{GDBN} dumps the data in raw binary
11401 @item append @r{[}binary@r{]} memory @var{filename} @var{start_addr} @var{end_addr}
11402 @itemx append @r{[}binary@r{]} value @var{filename} @var{expr}
11403 Append the contents of memory from @var{start_addr} to @var{end_addr},
11404 or the value of @var{expr}, to the file @var{filename}, in raw binary form.
11405 (@value{GDBN} can only append data to files in raw binary form.)
11408 @item restore @var{filename} @r{[}binary@r{]} @var{bias} @var{start} @var{end}
11409 Restore the contents of file @var{filename} into memory. The
11410 @code{restore} command can automatically recognize any known @sc{bfd}
11411 file format, except for raw binary. To restore a raw binary file you
11412 must specify the optional keyword @code{binary} after the filename.
11414 If @var{bias} is non-zero, its value will be added to the addresses
11415 contained in the file. Binary files always start at address zero, so
11416 they will be restored at address @var{bias}. Other bfd files have
11417 a built-in location; they will be restored at offset @var{bias}
11418 from that location.
11420 If @var{start} and/or @var{end} are non-zero, then only data between
11421 file offset @var{start} and file offset @var{end} will be restored.
11422 These offsets are relative to the addresses in the file, before
11423 the @var{bias} argument is applied.
11427 @node Core File Generation
11428 @section How to Produce a Core File from Your Program
11429 @cindex dump core from inferior
11431 A @dfn{core file} or @dfn{core dump} is a file that records the memory
11432 image of a running process and its process status (register values
11433 etc.). Its primary use is post-mortem debugging of a program that
11434 crashed while it ran outside a debugger. A program that crashes
11435 automatically produces a core file, unless this feature is disabled by
11436 the user. @xref{Files}, for information on invoking @value{GDBN} in
11437 the post-mortem debugging mode.
11439 Occasionally, you may wish to produce a core file of the program you
11440 are debugging in order to preserve a snapshot of its state.
11441 @value{GDBN} has a special command for that.
11445 @kindex generate-core-file
11446 @item generate-core-file [@var{file}]
11447 @itemx gcore [@var{file}]
11448 Produce a core dump of the inferior process. The optional argument
11449 @var{file} specifies the file name where to put the core dump. If not
11450 specified, the file name defaults to @file{core.@var{pid}}, where
11451 @var{pid} is the inferior process ID.
11453 Note that this command is implemented only for some systems (as of
11454 this writing, @sc{gnu}/Linux, FreeBSD, Solaris, and S390).
11456 On @sc{gnu}/Linux, this command can take into account the value of the
11457 file @file{/proc/@var{pid}/coredump_filter} when generating the core
11458 dump (@pxref{set use-coredump-filter}).
11460 @kindex set use-coredump-filter
11461 @anchor{set use-coredump-filter}
11462 @item set use-coredump-filter on
11463 @itemx set use-coredump-filter off
11464 Enable or disable the use of the file
11465 @file{/proc/@var{pid}/coredump_filter} when generating core dump
11466 files. This file is used by the Linux kernel to decide what types of
11467 memory mappings will be dumped or ignored when generating a core dump
11468 file. @var{pid} is the process ID of a currently running process.
11470 To make use of this feature, you have to write in the
11471 @file{/proc/@var{pid}/coredump_filter} file a value, in hexadecimal,
11472 which is a bit mask representing the memory mapping types. If a bit
11473 is set in the bit mask, then the memory mappings of the corresponding
11474 types will be dumped; otherwise, they will be ignored. This
11475 configuration is inherited by child processes. For more information
11476 about the bits that can be set in the
11477 @file{/proc/@var{pid}/coredump_filter} file, please refer to the
11478 manpage of @code{core(5)}.
11480 By default, this option is @code{on}. If this option is turned
11481 @code{off}, @value{GDBN} does not read the @file{coredump_filter} file
11482 and instead uses the same default value as the Linux kernel in order
11483 to decide which pages will be dumped in the core dump file. This
11484 value is currently @code{0x33}, which means that bits @code{0}
11485 (anonymous private mappings), @code{1} (anonymous shared mappings),
11486 @code{4} (ELF headers) and @code{5} (private huge pages) are active.
11487 This will cause these memory mappings to be dumped automatically.
11490 @node Character Sets
11491 @section Character Sets
11492 @cindex character sets
11494 @cindex translating between character sets
11495 @cindex host character set
11496 @cindex target character set
11498 If the program you are debugging uses a different character set to
11499 represent characters and strings than the one @value{GDBN} uses itself,
11500 @value{GDBN} can automatically translate between the character sets for
11501 you. The character set @value{GDBN} uses we call the @dfn{host
11502 character set}; the one the inferior program uses we call the
11503 @dfn{target character set}.
11505 For example, if you are running @value{GDBN} on a @sc{gnu}/Linux system, which
11506 uses the ISO Latin 1 character set, but you are using @value{GDBN}'s
11507 remote protocol (@pxref{Remote Debugging}) to debug a program
11508 running on an IBM mainframe, which uses the @sc{ebcdic} character set,
11509 then the host character set is Latin-1, and the target character set is
11510 @sc{ebcdic}. If you give @value{GDBN} the command @code{set
11511 target-charset EBCDIC-US}, then @value{GDBN} translates between
11512 @sc{ebcdic} and Latin 1 as you print character or string values, or use
11513 character and string literals in expressions.
11515 @value{GDBN} has no way to automatically recognize which character set
11516 the inferior program uses; you must tell it, using the @code{set
11517 target-charset} command, described below.
11519 Here are the commands for controlling @value{GDBN}'s character set
11523 @item set target-charset @var{charset}
11524 @kindex set target-charset
11525 Set the current target character set to @var{charset}. To display the
11526 list of supported target character sets, type
11527 @kbd{@w{set target-charset @key{TAB}@key{TAB}}}.
11529 @item set host-charset @var{charset}
11530 @kindex set host-charset
11531 Set the current host character set to @var{charset}.
11533 By default, @value{GDBN} uses a host character set appropriate to the
11534 system it is running on; you can override that default using the
11535 @code{set host-charset} command. On some systems, @value{GDBN} cannot
11536 automatically determine the appropriate host character set. In this
11537 case, @value{GDBN} uses @samp{UTF-8}.
11539 @value{GDBN} can only use certain character sets as its host character
11540 set. If you type @kbd{@w{set host-charset @key{TAB}@key{TAB}}},
11541 @value{GDBN} will list the host character sets it supports.
11543 @item set charset @var{charset}
11544 @kindex set charset
11545 Set the current host and target character sets to @var{charset}. As
11546 above, if you type @kbd{@w{set charset @key{TAB}@key{TAB}}},
11547 @value{GDBN} will list the names of the character sets that can be used
11548 for both host and target.
11551 @kindex show charset
11552 Show the names of the current host and target character sets.
11554 @item show host-charset
11555 @kindex show host-charset
11556 Show the name of the current host character set.
11558 @item show target-charset
11559 @kindex show target-charset
11560 Show the name of the current target character set.
11562 @item set target-wide-charset @var{charset}
11563 @kindex set target-wide-charset
11564 Set the current target's wide character set to @var{charset}. This is
11565 the character set used by the target's @code{wchar_t} type. To
11566 display the list of supported wide character sets, type
11567 @kbd{@w{set target-wide-charset @key{TAB}@key{TAB}}}.
11569 @item show target-wide-charset
11570 @kindex show target-wide-charset
11571 Show the name of the current target's wide character set.
11574 Here is an example of @value{GDBN}'s character set support in action.
11575 Assume that the following source code has been placed in the file
11576 @file{charset-test.c}:
11582 = @{72, 101, 108, 108, 111, 44, 32, 119,
11583 111, 114, 108, 100, 33, 10, 0@};
11584 char ibm1047_hello[]
11585 = @{200, 133, 147, 147, 150, 107, 64, 166,
11586 150, 153, 147, 132, 90, 37, 0@};
11590 printf ("Hello, world!\n");
11594 In this program, @code{ascii_hello} and @code{ibm1047_hello} are arrays
11595 containing the string @samp{Hello, world!} followed by a newline,
11596 encoded in the @sc{ascii} and @sc{ibm1047} character sets.
11598 We compile the program, and invoke the debugger on it:
11601 $ gcc -g charset-test.c -o charset-test
11602 $ gdb -nw charset-test
11603 GNU gdb 2001-12-19-cvs
11604 Copyright 2001 Free Software Foundation, Inc.
11609 We can use the @code{show charset} command to see what character sets
11610 @value{GDBN} is currently using to interpret and display characters and
11614 (@value{GDBP}) show charset
11615 The current host and target character set is `ISO-8859-1'.
11619 For the sake of printing this manual, let's use @sc{ascii} as our
11620 initial character set:
11622 (@value{GDBP}) set charset ASCII
11623 (@value{GDBP}) show charset
11624 The current host and target character set is `ASCII'.
11628 Let's assume that @sc{ascii} is indeed the correct character set for our
11629 host system --- in other words, let's assume that if @value{GDBN} prints
11630 characters using the @sc{ascii} character set, our terminal will display
11631 them properly. Since our current target character set is also
11632 @sc{ascii}, the contents of @code{ascii_hello} print legibly:
11635 (@value{GDBP}) print ascii_hello
11636 $1 = 0x401698 "Hello, world!\n"
11637 (@value{GDBP}) print ascii_hello[0]
11642 @value{GDBN} uses the target character set for character and string
11643 literals you use in expressions:
11646 (@value{GDBP}) print '+'
11651 The @sc{ascii} character set uses the number 43 to encode the @samp{+}
11654 @value{GDBN} relies on the user to tell it which character set the
11655 target program uses. If we print @code{ibm1047_hello} while our target
11656 character set is still @sc{ascii}, we get jibberish:
11659 (@value{GDBP}) print ibm1047_hello
11660 $4 = 0x4016a8 "\310\205\223\223\226k@@\246\226\231\223\204Z%"
11661 (@value{GDBP}) print ibm1047_hello[0]
11666 If we invoke the @code{set target-charset} followed by @key{TAB}@key{TAB},
11667 @value{GDBN} tells us the character sets it supports:
11670 (@value{GDBP}) set target-charset
11671 ASCII EBCDIC-US IBM1047 ISO-8859-1
11672 (@value{GDBP}) set target-charset
11675 We can select @sc{ibm1047} as our target character set, and examine the
11676 program's strings again. Now the @sc{ascii} string is wrong, but
11677 @value{GDBN} translates the contents of @code{ibm1047_hello} from the
11678 target character set, @sc{ibm1047}, to the host character set,
11679 @sc{ascii}, and they display correctly:
11682 (@value{GDBP}) set target-charset IBM1047
11683 (@value{GDBP}) show charset
11684 The current host character set is `ASCII'.
11685 The current target character set is `IBM1047'.
11686 (@value{GDBP}) print ascii_hello
11687 $6 = 0x401698 "\110\145%%?\054\040\167?\162%\144\041\012"
11688 (@value{GDBP}) print ascii_hello[0]
11690 (@value{GDBP}) print ibm1047_hello
11691 $8 = 0x4016a8 "Hello, world!\n"
11692 (@value{GDBP}) print ibm1047_hello[0]
11697 As above, @value{GDBN} uses the target character set for character and
11698 string literals you use in expressions:
11701 (@value{GDBP}) print '+'
11706 The @sc{ibm1047} character set uses the number 78 to encode the @samp{+}
11709 @node Caching Target Data
11710 @section Caching Data of Targets
11711 @cindex caching data of targets
11713 @value{GDBN} caches data exchanged between the debugger and a target.
11714 Each cache is associated with the address space of the inferior.
11715 @xref{Inferiors and Programs}, about inferior and address space.
11716 Such caching generally improves performance in remote debugging
11717 (@pxref{Remote Debugging}), because it reduces the overhead of the
11718 remote protocol by bundling memory reads and writes into large chunks.
11719 Unfortunately, simply caching everything would lead to incorrect results,
11720 since @value{GDBN} does not necessarily know anything about volatile
11721 values, memory-mapped I/O addresses, etc. Furthermore, in non-stop mode
11722 (@pxref{Non-Stop Mode}) memory can be changed @emph{while} a gdb command
11724 Therefore, by default, @value{GDBN} only caches data
11725 known to be on the stack@footnote{In non-stop mode, it is moderately
11726 rare for a running thread to modify the stack of a stopped thread
11727 in a way that would interfere with a backtrace, and caching of
11728 stack reads provides a significant speed up of remote backtraces.} or
11729 in the code segment.
11730 Other regions of memory can be explicitly marked as
11731 cacheable; @pxref{Memory Region Attributes}.
11734 @kindex set remotecache
11735 @item set remotecache on
11736 @itemx set remotecache off
11737 This option no longer does anything; it exists for compatibility
11740 @kindex show remotecache
11741 @item show remotecache
11742 Show the current state of the obsolete remotecache flag.
11744 @kindex set stack-cache
11745 @item set stack-cache on
11746 @itemx set stack-cache off
11747 Enable or disable caching of stack accesses. When @code{on}, use
11748 caching. By default, this option is @code{on}.
11750 @kindex show stack-cache
11751 @item show stack-cache
11752 Show the current state of data caching for memory accesses.
11754 @kindex set code-cache
11755 @item set code-cache on
11756 @itemx set code-cache off
11757 Enable or disable caching of code segment accesses. When @code{on},
11758 use caching. By default, this option is @code{on}. This improves
11759 performance of disassembly in remote debugging.
11761 @kindex show code-cache
11762 @item show code-cache
11763 Show the current state of target memory cache for code segment
11766 @kindex info dcache
11767 @item info dcache @r{[}line@r{]}
11768 Print the information about the performance of data cache of the
11769 current inferior's address space. The information displayed
11770 includes the dcache width and depth, and for each cache line, its
11771 number, address, and how many times it was referenced. This
11772 command is useful for debugging the data cache operation.
11774 If a line number is specified, the contents of that line will be
11777 @item set dcache size @var{size}
11778 @cindex dcache size
11779 @kindex set dcache size
11780 Set maximum number of entries in dcache (dcache depth above).
11782 @item set dcache line-size @var{line-size}
11783 @cindex dcache line-size
11784 @kindex set dcache line-size
11785 Set number of bytes each dcache entry caches (dcache width above).
11786 Must be a power of 2.
11788 @item show dcache size
11789 @kindex show dcache size
11790 Show maximum number of dcache entries. @xref{Caching Target Data, info dcache}.
11792 @item show dcache line-size
11793 @kindex show dcache line-size
11794 Show default size of dcache lines.
11798 @node Searching Memory
11799 @section Search Memory
11800 @cindex searching memory
11802 Memory can be searched for a particular sequence of bytes with the
11803 @code{find} command.
11807 @item find @r{[}/@var{sn}@r{]} @var{start_addr}, +@var{len}, @var{val1} @r{[}, @var{val2}, @dots{}@r{]}
11808 @itemx find @r{[}/@var{sn}@r{]} @var{start_addr}, @var{end_addr}, @var{val1} @r{[}, @var{val2}, @dots{}@r{]}
11809 Search memory for the sequence of bytes specified by @var{val1}, @var{val2},
11810 etc. The search begins at address @var{start_addr} and continues for either
11811 @var{len} bytes or through to @var{end_addr} inclusive.
11814 @var{s} and @var{n} are optional parameters.
11815 They may be specified in either order, apart or together.
11818 @item @var{s}, search query size
11819 The size of each search query value.
11825 halfwords (two bytes)
11829 giant words (eight bytes)
11832 All values are interpreted in the current language.
11833 This means, for example, that if the current source language is C/C@t{++}
11834 then searching for the string ``hello'' includes the trailing '\0'.
11836 If the value size is not specified, it is taken from the
11837 value's type in the current language.
11838 This is useful when one wants to specify the search
11839 pattern as a mixture of types.
11840 Note that this means, for example, that in the case of C-like languages
11841 a search for an untyped 0x42 will search for @samp{(int) 0x42}
11842 which is typically four bytes.
11844 @item @var{n}, maximum number of finds
11845 The maximum number of matches to print. The default is to print all finds.
11848 You can use strings as search values. Quote them with double-quotes
11850 The string value is copied into the search pattern byte by byte,
11851 regardless of the endianness of the target and the size specification.
11853 The address of each match found is printed as well as a count of the
11854 number of matches found.
11856 The address of the last value found is stored in convenience variable
11858 A count of the number of matches is stored in @samp{$numfound}.
11860 For example, if stopped at the @code{printf} in this function:
11866 static char hello[] = "hello-hello";
11867 static struct @{ char c; short s; int i; @}
11868 __attribute__ ((packed)) mixed
11869 = @{ 'c', 0x1234, 0x87654321 @};
11870 printf ("%s\n", hello);
11875 you get during debugging:
11878 (gdb) find &hello[0], +sizeof(hello), "hello"
11879 0x804956d <hello.1620+6>
11881 (gdb) find &hello[0], +sizeof(hello), 'h', 'e', 'l', 'l', 'o'
11882 0x8049567 <hello.1620>
11883 0x804956d <hello.1620+6>
11885 (gdb) find /b1 &hello[0], +sizeof(hello), 'h', 0x65, 'l'
11886 0x8049567 <hello.1620>
11888 (gdb) find &mixed, +sizeof(mixed), (char) 'c', (short) 0x1234, (int) 0x87654321
11889 0x8049560 <mixed.1625>
11891 (gdb) print $numfound
11894 $2 = (void *) 0x8049560
11898 @section Value Sizes
11900 Whenever @value{GDBN} prints a value memory will be allocated within
11901 @value{GDBN} to hold the contents of the value. It is possible in
11902 some languages with dynamic typing systems, that an invalid program
11903 may indicate a value that is incorrectly large, this in turn may cause
11904 @value{GDBN} to try and allocate an overly large ammount of memory.
11907 @kindex set max-value-size
11908 @item set max-value-size @var{bytes}
11909 @itemx set max-value-size unlimited
11910 Set the maximum size of memory that @value{GDBN} will allocate for the
11911 contents of a value to @var{bytes}, trying to display a value that
11912 requires more memory than that will result in an error.
11914 Setting this variable does not effect values that have already been
11915 allocated within @value{GDBN}, only future allocations.
11917 There's a minimum size that @code{max-value-size} can be set to in
11918 order that @value{GDBN} can still operate correctly, this minimum is
11919 currently 16 bytes.
11921 The limit applies to the results of some subexpressions as well as to
11922 complete expressions. For example, an expression denoting a simple
11923 integer component, such as @code{x.y.z}, may fail if the size of
11924 @var{x.y} is dynamic and exceeds @var{bytes}. On the other hand,
11925 @value{GDBN} is sometimes clever; the expression @code{A[i]}, where
11926 @var{A} is an array variable with non-constant size, will generally
11927 succeed regardless of the bounds on @var{A}, as long as the component
11928 size is less than @var{bytes}.
11930 The default value of @code{max-value-size} is currently 64k.
11932 @kindex show max-value-size
11933 @item show max-value-size
11934 Show the maximum size of memory, in bytes, that @value{GDBN} will
11935 allocate for the contents of a value.
11938 @node Optimized Code
11939 @chapter Debugging Optimized Code
11940 @cindex optimized code, debugging
11941 @cindex debugging optimized code
11943 Almost all compilers support optimization. With optimization
11944 disabled, the compiler generates assembly code that corresponds
11945 directly to your source code, in a simplistic way. As the compiler
11946 applies more powerful optimizations, the generated assembly code
11947 diverges from your original source code. With help from debugging
11948 information generated by the compiler, @value{GDBN} can map from
11949 the running program back to constructs from your original source.
11951 @value{GDBN} is more accurate with optimization disabled. If you
11952 can recompile without optimization, it is easier to follow the
11953 progress of your program during debugging. But, there are many cases
11954 where you may need to debug an optimized version.
11956 When you debug a program compiled with @samp{-g -O}, remember that the
11957 optimizer has rearranged your code; the debugger shows you what is
11958 really there. Do not be too surprised when the execution path does not
11959 exactly match your source file! An extreme example: if you define a
11960 variable, but never use it, @value{GDBN} never sees that
11961 variable---because the compiler optimizes it out of existence.
11963 Some things do not work as well with @samp{-g -O} as with just
11964 @samp{-g}, particularly on machines with instruction scheduling. If in
11965 doubt, recompile with @samp{-g} alone, and if this fixes the problem,
11966 please report it to us as a bug (including a test case!).
11967 @xref{Variables}, for more information about debugging optimized code.
11970 * Inline Functions:: How @value{GDBN} presents inlining
11971 * Tail Call Frames:: @value{GDBN} analysis of jumps to functions
11974 @node Inline Functions
11975 @section Inline Functions
11976 @cindex inline functions, debugging
11978 @dfn{Inlining} is an optimization that inserts a copy of the function
11979 body directly at each call site, instead of jumping to a shared
11980 routine. @value{GDBN} displays inlined functions just like
11981 non-inlined functions. They appear in backtraces. You can view their
11982 arguments and local variables, step into them with @code{step}, skip
11983 them with @code{next}, and escape from them with @code{finish}.
11984 You can check whether a function was inlined by using the
11985 @code{info frame} command.
11987 For @value{GDBN} to support inlined functions, the compiler must
11988 record information about inlining in the debug information ---
11989 @value{NGCC} using the @sc{dwarf 2} format does this, and several
11990 other compilers do also. @value{GDBN} only supports inlined functions
11991 when using @sc{dwarf 2}. Versions of @value{NGCC} before 4.1
11992 do not emit two required attributes (@samp{DW_AT_call_file} and
11993 @samp{DW_AT_call_line}); @value{GDBN} does not display inlined
11994 function calls with earlier versions of @value{NGCC}. It instead
11995 displays the arguments and local variables of inlined functions as
11996 local variables in the caller.
11998 The body of an inlined function is directly included at its call site;
11999 unlike a non-inlined function, there are no instructions devoted to
12000 the call. @value{GDBN} still pretends that the call site and the
12001 start of the inlined function are different instructions. Stepping to
12002 the call site shows the call site, and then stepping again shows
12003 the first line of the inlined function, even though no additional
12004 instructions are executed.
12006 This makes source-level debugging much clearer; you can see both the
12007 context of the call and then the effect of the call. Only stepping by
12008 a single instruction using @code{stepi} or @code{nexti} does not do
12009 this; single instruction steps always show the inlined body.
12011 There are some ways that @value{GDBN} does not pretend that inlined
12012 function calls are the same as normal calls:
12016 Setting breakpoints at the call site of an inlined function may not
12017 work, because the call site does not contain any code. @value{GDBN}
12018 may incorrectly move the breakpoint to the next line of the enclosing
12019 function, after the call. This limitation will be removed in a future
12020 version of @value{GDBN}; until then, set a breakpoint on an earlier line
12021 or inside the inlined function instead.
12024 @value{GDBN} cannot locate the return value of inlined calls after
12025 using the @code{finish} command. This is a limitation of compiler-generated
12026 debugging information; after @code{finish}, you can step to the next line
12027 and print a variable where your program stored the return value.
12031 @node Tail Call Frames
12032 @section Tail Call Frames
12033 @cindex tail call frames, debugging
12035 Function @code{B} can call function @code{C} in its very last statement. In
12036 unoptimized compilation the call of @code{C} is immediately followed by return
12037 instruction at the end of @code{B} code. Optimizing compiler may replace the
12038 call and return in function @code{B} into one jump to function @code{C}
12039 instead. Such use of a jump instruction is called @dfn{tail call}.
12041 During execution of function @code{C}, there will be no indication in the
12042 function call stack frames that it was tail-called from @code{B}. If function
12043 @code{A} regularly calls function @code{B} which tail-calls function @code{C},
12044 then @value{GDBN} will see @code{A} as the caller of @code{C}. However, in
12045 some cases @value{GDBN} can determine that @code{C} was tail-called from
12046 @code{B}, and it will then create fictitious call frame for that, with the
12047 return address set up as if @code{B} called @code{C} normally.
12049 This functionality is currently supported only by DWARF 2 debugging format and
12050 the compiler has to produce @samp{DW_TAG_call_site} tags. With
12051 @value{NGCC}, you need to specify @option{-O -g} during compilation, to get
12054 @kbd{info frame} command (@pxref{Frame Info}) will indicate the tail call frame
12055 kind by text @code{tail call frame} such as in this sample @value{GDBN} output:
12059 0x40066b <b(int, double)+11>: jmp 0x400640 <c(int, double)>
12061 Stack level 1, frame at 0x7fffffffda30:
12062 rip = 0x40066d in b (amd64-entry-value.cc:59); saved rip 0x4004c5
12063 tail call frame, caller of frame at 0x7fffffffda30
12064 source language c++.
12065 Arglist at unknown address.
12066 Locals at unknown address, Previous frame's sp is 0x7fffffffda30
12069 The detection of all the possible code path executions can find them ambiguous.
12070 There is no execution history stored (possible @ref{Reverse Execution} is never
12071 used for this purpose) and the last known caller could have reached the known
12072 callee by multiple different jump sequences. In such case @value{GDBN} still
12073 tries to show at least all the unambiguous top tail callers and all the
12074 unambiguous bottom tail calees, if any.
12077 @anchor{set debug entry-values}
12078 @item set debug entry-values
12079 @kindex set debug entry-values
12080 When set to on, enables printing of analysis messages for both frame argument
12081 values at function entry and tail calls. It will show all the possible valid
12082 tail calls code paths it has considered. It will also print the intersection
12083 of them with the final unambiguous (possibly partial or even empty) code path
12086 @item show debug entry-values
12087 @kindex show debug entry-values
12088 Show the current state of analysis messages printing for both frame argument
12089 values at function entry and tail calls.
12092 The analysis messages for tail calls can for example show why the virtual tail
12093 call frame for function @code{c} has not been recognized (due to the indirect
12094 reference by variable @code{x}):
12097 static void __attribute__((noinline, noclone)) c (void);
12098 void (*x) (void) = c;
12099 static void __attribute__((noinline, noclone)) a (void) @{ x++; @}
12100 static void __attribute__((noinline, noclone)) c (void) @{ a (); @}
12101 int main (void) @{ x (); return 0; @}
12103 Breakpoint 1, DW_OP_entry_value resolving cannot find
12104 DW_TAG_call_site 0x40039a in main
12106 3 static void __attribute__((noinline, noclone)) a (void) @{ x++; @}
12109 #1 0x000000000040039a in main () at t.c:5
12112 Another possibility is an ambiguous virtual tail call frames resolution:
12116 static void __attribute__((noinline, noclone)) f (void) @{ i++; @}
12117 static void __attribute__((noinline, noclone)) e (void) @{ f (); @}
12118 static void __attribute__((noinline, noclone)) d (void) @{ f (); @}
12119 static void __attribute__((noinline, noclone)) c (void) @{ d (); @}
12120 static void __attribute__((noinline, noclone)) b (void)
12121 @{ if (i) c (); else e (); @}
12122 static void __attribute__((noinline, noclone)) a (void) @{ b (); @}
12123 int main (void) @{ a (); return 0; @}
12125 tailcall: initial: 0x4004d2(a) 0x4004ce(b) 0x4004b2(c) 0x4004a2(d)
12126 tailcall: compare: 0x4004d2(a) 0x4004cc(b) 0x400492(e)
12127 tailcall: reduced: 0x4004d2(a) |
12130 #1 0x00000000004004d2 in a () at t.c:8
12131 #2 0x0000000000400395 in main () at t.c:9
12134 @set CALLSEQ1A @code{main@value{ARROW}a@value{ARROW}b@value{ARROW}c@value{ARROW}d@value{ARROW}f}
12135 @set CALLSEQ2A @code{main@value{ARROW}a@value{ARROW}b@value{ARROW}e@value{ARROW}f}
12137 @c Convert CALLSEQ#A to CALLSEQ#B depending on HAVE_MAKEINFO_CLICK.
12138 @ifset HAVE_MAKEINFO_CLICK
12139 @set ARROW @click{}
12140 @set CALLSEQ1B @clicksequence{@value{CALLSEQ1A}}
12141 @set CALLSEQ2B @clicksequence{@value{CALLSEQ2A}}
12143 @ifclear HAVE_MAKEINFO_CLICK
12145 @set CALLSEQ1B @value{CALLSEQ1A}
12146 @set CALLSEQ2B @value{CALLSEQ2A}
12149 Frames #0 and #2 are real, #1 is a virtual tail call frame.
12150 The code can have possible execution paths @value{CALLSEQ1B} or
12151 @value{CALLSEQ2B}, @value{GDBN} cannot find which one from the inferior state.
12153 @code{initial:} state shows some random possible calling sequence @value{GDBN}
12154 has found. It then finds another possible calling sequcen - that one is
12155 prefixed by @code{compare:}. The non-ambiguous intersection of these two is
12156 printed as the @code{reduced:} calling sequence. That one could have many
12157 futher @code{compare:} and @code{reduced:} statements as long as there remain
12158 any non-ambiguous sequence entries.
12160 For the frame of function @code{b} in both cases there are different possible
12161 @code{$pc} values (@code{0x4004cc} or @code{0x4004ce}), therefore this frame is
12162 also ambigous. The only non-ambiguous frame is the one for function @code{a},
12163 therefore this one is displayed to the user while the ambiguous frames are
12166 There can be also reasons why printing of frame argument values at function
12171 static void __attribute__((noinline, noclone)) c (int i) @{ v++; @}
12172 static void __attribute__((noinline, noclone)) a (int i);
12173 static void __attribute__((noinline, noclone)) b (int i) @{ a (i); @}
12174 static void __attribute__((noinline, noclone)) a (int i)
12175 @{ if (i) b (i - 1); else c (0); @}
12176 int main (void) @{ a (5); return 0; @}
12179 #0 c (i=i@@entry=0) at t.c:2
12180 #1 0x0000000000400428 in a (DW_OP_entry_value resolving has found
12181 function "a" at 0x400420 can call itself via tail calls
12182 i=<optimized out>) at t.c:6
12183 #2 0x000000000040036e in main () at t.c:7
12186 @value{GDBN} cannot find out from the inferior state if and how many times did
12187 function @code{a} call itself (via function @code{b}) as these calls would be
12188 tail calls. Such tail calls would modify thue @code{i} variable, therefore
12189 @value{GDBN} cannot be sure the value it knows would be right - @value{GDBN}
12190 prints @code{<optimized out>} instead.
12193 @chapter C Preprocessor Macros
12195 Some languages, such as C and C@t{++}, provide a way to define and invoke
12196 ``preprocessor macros'' which expand into strings of tokens.
12197 @value{GDBN} can evaluate expressions containing macro invocations, show
12198 the result of macro expansion, and show a macro's definition, including
12199 where it was defined.
12201 You may need to compile your program specially to provide @value{GDBN}
12202 with information about preprocessor macros. Most compilers do not
12203 include macros in their debugging information, even when you compile
12204 with the @option{-g} flag. @xref{Compilation}.
12206 A program may define a macro at one point, remove that definition later,
12207 and then provide a different definition after that. Thus, at different
12208 points in the program, a macro may have different definitions, or have
12209 no definition at all. If there is a current stack frame, @value{GDBN}
12210 uses the macros in scope at that frame's source code line. Otherwise,
12211 @value{GDBN} uses the macros in scope at the current listing location;
12214 Whenever @value{GDBN} evaluates an expression, it always expands any
12215 macro invocations present in the expression. @value{GDBN} also provides
12216 the following commands for working with macros explicitly.
12220 @kindex macro expand
12221 @cindex macro expansion, showing the results of preprocessor
12222 @cindex preprocessor macro expansion, showing the results of
12223 @cindex expanding preprocessor macros
12224 @item macro expand @var{expression}
12225 @itemx macro exp @var{expression}
12226 Show the results of expanding all preprocessor macro invocations in
12227 @var{expression}. Since @value{GDBN} simply expands macros, but does
12228 not parse the result, @var{expression} need not be a valid expression;
12229 it can be any string of tokens.
12232 @item macro expand-once @var{expression}
12233 @itemx macro exp1 @var{expression}
12234 @cindex expand macro once
12235 @i{(This command is not yet implemented.)} Show the results of
12236 expanding those preprocessor macro invocations that appear explicitly in
12237 @var{expression}. Macro invocations appearing in that expansion are
12238 left unchanged. This command allows you to see the effect of a
12239 particular macro more clearly, without being confused by further
12240 expansions. Since @value{GDBN} simply expands macros, but does not
12241 parse the result, @var{expression} need not be a valid expression; it
12242 can be any string of tokens.
12245 @cindex macro definition, showing
12246 @cindex definition of a macro, showing
12247 @cindex macros, from debug info
12248 @item info macro [-a|-all] [--] @var{macro}
12249 Show the current definition or all definitions of the named @var{macro},
12250 and describe the source location or compiler command-line where that
12251 definition was established. The optional double dash is to signify the end of
12252 argument processing and the beginning of @var{macro} for non C-like macros where
12253 the macro may begin with a hyphen.
12255 @kindex info macros
12256 @item info macros @var{location}
12257 Show all macro definitions that are in effect at the location specified
12258 by @var{location}, and describe the source location or compiler
12259 command-line where those definitions were established.
12261 @kindex macro define
12262 @cindex user-defined macros
12263 @cindex defining macros interactively
12264 @cindex macros, user-defined
12265 @item macro define @var{macro} @var{replacement-list}
12266 @itemx macro define @var{macro}(@var{arglist}) @var{replacement-list}
12267 Introduce a definition for a preprocessor macro named @var{macro},
12268 invocations of which are replaced by the tokens given in
12269 @var{replacement-list}. The first form of this command defines an
12270 ``object-like'' macro, which takes no arguments; the second form
12271 defines a ``function-like'' macro, which takes the arguments given in
12274 A definition introduced by this command is in scope in every
12275 expression evaluated in @value{GDBN}, until it is removed with the
12276 @code{macro undef} command, described below. The definition overrides
12277 all definitions for @var{macro} present in the program being debugged,
12278 as well as any previous user-supplied definition.
12280 @kindex macro undef
12281 @item macro undef @var{macro}
12282 Remove any user-supplied definition for the macro named @var{macro}.
12283 This command only affects definitions provided with the @code{macro
12284 define} command, described above; it cannot remove definitions present
12285 in the program being debugged.
12289 List all the macros defined using the @code{macro define} command.
12292 @cindex macros, example of debugging with
12293 Here is a transcript showing the above commands in action. First, we
12294 show our source files:
12299 #include "sample.h"
12302 #define ADD(x) (M + x)
12307 printf ("Hello, world!\n");
12309 printf ("We're so creative.\n");
12311 printf ("Goodbye, world!\n");
12318 Now, we compile the program using the @sc{gnu} C compiler,
12319 @value{NGCC}. We pass the @option{-gdwarf-2}@footnote{This is the
12320 minimum. Recent versions of @value{NGCC} support @option{-gdwarf-3}
12321 and @option{-gdwarf-4}; we recommend always choosing the most recent
12322 version of DWARF.} @emph{and} @option{-g3} flags to ensure the compiler
12323 includes information about preprocessor macros in the debugging
12327 $ gcc -gdwarf-2 -g3 sample.c -o sample
12331 Now, we start @value{GDBN} on our sample program:
12335 GNU gdb 2002-05-06-cvs
12336 Copyright 2002 Free Software Foundation, Inc.
12337 GDB is free software, @dots{}
12341 We can expand macros and examine their definitions, even when the
12342 program is not running. @value{GDBN} uses the current listing position
12343 to decide which macro definitions are in scope:
12346 (@value{GDBP}) list main
12349 5 #define ADD(x) (M + x)
12354 10 printf ("Hello, world!\n");
12356 12 printf ("We're so creative.\n");
12357 (@value{GDBP}) info macro ADD
12358 Defined at /home/jimb/gdb/macros/play/sample.c:5
12359 #define ADD(x) (M + x)
12360 (@value{GDBP}) info macro Q
12361 Defined at /home/jimb/gdb/macros/play/sample.h:1
12362 included at /home/jimb/gdb/macros/play/sample.c:2
12364 (@value{GDBP}) macro expand ADD(1)
12365 expands to: (42 + 1)
12366 (@value{GDBP}) macro expand-once ADD(1)
12367 expands to: once (M + 1)
12371 In the example above, note that @code{macro expand-once} expands only
12372 the macro invocation explicit in the original text --- the invocation of
12373 @code{ADD} --- but does not expand the invocation of the macro @code{M},
12374 which was introduced by @code{ADD}.
12376 Once the program is running, @value{GDBN} uses the macro definitions in
12377 force at the source line of the current stack frame:
12380 (@value{GDBP}) break main
12381 Breakpoint 1 at 0x8048370: file sample.c, line 10.
12383 Starting program: /home/jimb/gdb/macros/play/sample
12385 Breakpoint 1, main () at sample.c:10
12386 10 printf ("Hello, world!\n");
12390 At line 10, the definition of the macro @code{N} at line 9 is in force:
12393 (@value{GDBP}) info macro N
12394 Defined at /home/jimb/gdb/macros/play/sample.c:9
12396 (@value{GDBP}) macro expand N Q M
12397 expands to: 28 < 42
12398 (@value{GDBP}) print N Q M
12403 As we step over directives that remove @code{N}'s definition, and then
12404 give it a new definition, @value{GDBN} finds the definition (or lack
12405 thereof) in force at each point:
12408 (@value{GDBP}) next
12410 12 printf ("We're so creative.\n");
12411 (@value{GDBP}) info macro N
12412 The symbol `N' has no definition as a C/C++ preprocessor macro
12413 at /home/jimb/gdb/macros/play/sample.c:12
12414 (@value{GDBP}) next
12416 14 printf ("Goodbye, world!\n");
12417 (@value{GDBP}) info macro N
12418 Defined at /home/jimb/gdb/macros/play/sample.c:13
12420 (@value{GDBP}) macro expand N Q M
12421 expands to: 1729 < 42
12422 (@value{GDBP}) print N Q M
12427 In addition to source files, macros can be defined on the compilation command
12428 line using the @option{-D@var{name}=@var{value}} syntax. For macros defined in
12429 such a way, @value{GDBN} displays the location of their definition as line zero
12430 of the source file submitted to the compiler.
12433 (@value{GDBP}) info macro __STDC__
12434 Defined at /home/jimb/gdb/macros/play/sample.c:0
12441 @chapter Tracepoints
12442 @c This chapter is based on the documentation written by Michael
12443 @c Snyder, David Taylor, Jim Blandy, and Elena Zannoni.
12445 @cindex tracepoints
12446 In some applications, it is not feasible for the debugger to interrupt
12447 the program's execution long enough for the developer to learn
12448 anything helpful about its behavior. If the program's correctness
12449 depends on its real-time behavior, delays introduced by a debugger
12450 might cause the program to change its behavior drastically, or perhaps
12451 fail, even when the code itself is correct. It is useful to be able
12452 to observe the program's behavior without interrupting it.
12454 Using @value{GDBN}'s @code{trace} and @code{collect} commands, you can
12455 specify locations in the program, called @dfn{tracepoints}, and
12456 arbitrary expressions to evaluate when those tracepoints are reached.
12457 Later, using the @code{tfind} command, you can examine the values
12458 those expressions had when the program hit the tracepoints. The
12459 expressions may also denote objects in memory---structures or arrays,
12460 for example---whose values @value{GDBN} should record; while visiting
12461 a particular tracepoint, you may inspect those objects as if they were
12462 in memory at that moment. However, because @value{GDBN} records these
12463 values without interacting with you, it can do so quickly and
12464 unobtrusively, hopefully not disturbing the program's behavior.
12466 The tracepoint facility is currently available only for remote
12467 targets. @xref{Targets}. In addition, your remote target must know
12468 how to collect trace data. This functionality is implemented in the
12469 remote stub; however, none of the stubs distributed with @value{GDBN}
12470 support tracepoints as of this writing. The format of the remote
12471 packets used to implement tracepoints are described in @ref{Tracepoint
12474 It is also possible to get trace data from a file, in a manner reminiscent
12475 of corefiles; you specify the filename, and use @code{tfind} to search
12476 through the file. @xref{Trace Files}, for more details.
12478 This chapter describes the tracepoint commands and features.
12481 * Set Tracepoints::
12482 * Analyze Collected Data::
12483 * Tracepoint Variables::
12487 @node Set Tracepoints
12488 @section Commands to Set Tracepoints
12490 Before running such a @dfn{trace experiment}, an arbitrary number of
12491 tracepoints can be set. A tracepoint is actually a special type of
12492 breakpoint (@pxref{Set Breaks}), so you can manipulate it using
12493 standard breakpoint commands. For instance, as with breakpoints,
12494 tracepoint numbers are successive integers starting from one, and many
12495 of the commands associated with tracepoints take the tracepoint number
12496 as their argument, to identify which tracepoint to work on.
12498 For each tracepoint, you can specify, in advance, some arbitrary set
12499 of data that you want the target to collect in the trace buffer when
12500 it hits that tracepoint. The collected data can include registers,
12501 local variables, or global data. Later, you can use @value{GDBN}
12502 commands to examine the values these data had at the time the
12503 tracepoint was hit.
12505 Tracepoints do not support every breakpoint feature. Ignore counts on
12506 tracepoints have no effect, and tracepoints cannot run @value{GDBN}
12507 commands when they are hit. Tracepoints may not be thread-specific
12510 @cindex fast tracepoints
12511 Some targets may support @dfn{fast tracepoints}, which are inserted in
12512 a different way (such as with a jump instead of a trap), that is
12513 faster but possibly restricted in where they may be installed.
12515 @cindex static tracepoints
12516 @cindex markers, static tracepoints
12517 @cindex probing markers, static tracepoints
12518 Regular and fast tracepoints are dynamic tracing facilities, meaning
12519 that they can be used to insert tracepoints at (almost) any location
12520 in the target. Some targets may also support controlling @dfn{static
12521 tracepoints} from @value{GDBN}. With static tracing, a set of
12522 instrumentation points, also known as @dfn{markers}, are embedded in
12523 the target program, and can be activated or deactivated by name or
12524 address. These are usually placed at locations which facilitate
12525 investigating what the target is actually doing. @value{GDBN}'s
12526 support for static tracing includes being able to list instrumentation
12527 points, and attach them with @value{GDBN} defined high level
12528 tracepoints that expose the whole range of convenience of
12529 @value{GDBN}'s tracepoints support. Namely, support for collecting
12530 registers values and values of global or local (to the instrumentation
12531 point) variables; tracepoint conditions and trace state variables.
12532 The act of installing a @value{GDBN} static tracepoint on an
12533 instrumentation point, or marker, is referred to as @dfn{probing} a
12534 static tracepoint marker.
12536 @code{gdbserver} supports tracepoints on some target systems.
12537 @xref{Server,,Tracepoints support in @code{gdbserver}}.
12539 This section describes commands to set tracepoints and associated
12540 conditions and actions.
12543 * Create and Delete Tracepoints::
12544 * Enable and Disable Tracepoints::
12545 * Tracepoint Passcounts::
12546 * Tracepoint Conditions::
12547 * Trace State Variables::
12548 * Tracepoint Actions::
12549 * Listing Tracepoints::
12550 * Listing Static Tracepoint Markers::
12551 * Starting and Stopping Trace Experiments::
12552 * Tracepoint Restrictions::
12555 @node Create and Delete Tracepoints
12556 @subsection Create and Delete Tracepoints
12559 @cindex set tracepoint
12561 @item trace @var{location}
12562 The @code{trace} command is very similar to the @code{break} command.
12563 Its argument @var{location} can be any valid location.
12564 @xref{Specify Location}. The @code{trace} command defines a tracepoint,
12565 which is a point in the target program where the debugger will briefly stop,
12566 collect some data, and then allow the program to continue. Setting a tracepoint
12567 or changing its actions takes effect immediately if the remote stub
12568 supports the @samp{InstallInTrace} feature (@pxref{install tracepoint
12570 If remote stub doesn't support the @samp{InstallInTrace} feature, all
12571 these changes don't take effect until the next @code{tstart}
12572 command, and once a trace experiment is running, further changes will
12573 not have any effect until the next trace experiment starts. In addition,
12574 @value{GDBN} supports @dfn{pending tracepoints}---tracepoints whose
12575 address is not yet resolved. (This is similar to pending breakpoints.)
12576 Pending tracepoints are not downloaded to the target and not installed
12577 until they are resolved. The resolution of pending tracepoints requires
12578 @value{GDBN} support---when debugging with the remote target, and
12579 @value{GDBN} disconnects from the remote stub (@pxref{disconnected
12580 tracing}), pending tracepoints can not be resolved (and downloaded to
12581 the remote stub) while @value{GDBN} is disconnected.
12583 Here are some examples of using the @code{trace} command:
12586 (@value{GDBP}) @b{trace foo.c:121} // a source file and line number
12588 (@value{GDBP}) @b{trace +2} // 2 lines forward
12590 (@value{GDBP}) @b{trace my_function} // first source line of function
12592 (@value{GDBP}) @b{trace *my_function} // EXACT start address of function
12594 (@value{GDBP}) @b{trace *0x2117c4} // an address
12598 You can abbreviate @code{trace} as @code{tr}.
12600 @item trace @var{location} if @var{cond}
12601 Set a tracepoint with condition @var{cond}; evaluate the expression
12602 @var{cond} each time the tracepoint is reached, and collect data only
12603 if the value is nonzero---that is, if @var{cond} evaluates as true.
12604 @xref{Tracepoint Conditions, ,Tracepoint Conditions}, for more
12605 information on tracepoint conditions.
12607 @item ftrace @var{location} [ if @var{cond} ]
12608 @cindex set fast tracepoint
12609 @cindex fast tracepoints, setting
12611 The @code{ftrace} command sets a fast tracepoint. For targets that
12612 support them, fast tracepoints will use a more efficient but possibly
12613 less general technique to trigger data collection, such as a jump
12614 instruction instead of a trap, or some sort of hardware support. It
12615 may not be possible to create a fast tracepoint at the desired
12616 location, in which case the command will exit with an explanatory
12619 @value{GDBN} handles arguments to @code{ftrace} exactly as for
12622 On 32-bit x86-architecture systems, fast tracepoints normally need to
12623 be placed at an instruction that is 5 bytes or longer, but can be
12624 placed at 4-byte instructions if the low 64K of memory of the target
12625 program is available to install trampolines. Some Unix-type systems,
12626 such as @sc{gnu}/Linux, exclude low addresses from the program's
12627 address space; but for instance with the Linux kernel it is possible
12628 to let @value{GDBN} use this area by doing a @command{sysctl} command
12629 to set the @code{mmap_min_addr} kernel parameter, as in
12632 sudo sysctl -w vm.mmap_min_addr=32768
12636 which sets the low address to 32K, which leaves plenty of room for
12637 trampolines. The minimum address should be set to a page boundary.
12639 @item strace @var{location} [ if @var{cond} ]
12640 @cindex set static tracepoint
12641 @cindex static tracepoints, setting
12642 @cindex probe static tracepoint marker
12644 The @code{strace} command sets a static tracepoint. For targets that
12645 support it, setting a static tracepoint probes a static
12646 instrumentation point, or marker, found at @var{location}. It may not
12647 be possible to set a static tracepoint at the desired location, in
12648 which case the command will exit with an explanatory message.
12650 @value{GDBN} handles arguments to @code{strace} exactly as for
12651 @code{trace}, with the addition that the user can also specify
12652 @code{-m @var{marker}} as @var{location}. This probes the marker
12653 identified by the @var{marker} string identifier. This identifier
12654 depends on the static tracepoint backend library your program is
12655 using. You can find all the marker identifiers in the @samp{ID} field
12656 of the @code{info static-tracepoint-markers} command output.
12657 @xref{Listing Static Tracepoint Markers,,Listing Static Tracepoint
12658 Markers}. For example, in the following small program using the UST
12664 trace_mark(ust, bar33, "str %s", "FOOBAZ");
12669 the marker id is composed of joining the first two arguments to the
12670 @code{trace_mark} call with a slash, which translates to:
12673 (@value{GDBP}) info static-tracepoint-markers
12674 Cnt Enb ID Address What
12675 1 n ust/bar33 0x0000000000400ddc in main at stexample.c:22
12681 so you may probe the marker above with:
12684 (@value{GDBP}) strace -m ust/bar33
12687 Static tracepoints accept an extra collect action --- @code{collect
12688 $_sdata}. This collects arbitrary user data passed in the probe point
12689 call to the tracing library. In the UST example above, you'll see
12690 that the third argument to @code{trace_mark} is a printf-like format
12691 string. The user data is then the result of running that formating
12692 string against the following arguments. Note that @code{info
12693 static-tracepoint-markers} command output lists that format string in
12694 the @samp{Data:} field.
12696 You can inspect this data when analyzing the trace buffer, by printing
12697 the $_sdata variable like any other variable available to
12698 @value{GDBN}. @xref{Tracepoint Actions,,Tracepoint Action Lists}.
12701 @cindex last tracepoint number
12702 @cindex recent tracepoint number
12703 @cindex tracepoint number
12704 The convenience variable @code{$tpnum} records the tracepoint number
12705 of the most recently set tracepoint.
12707 @kindex delete tracepoint
12708 @cindex tracepoint deletion
12709 @item delete tracepoint @r{[}@var{num}@r{]}
12710 Permanently delete one or more tracepoints. With no argument, the
12711 default is to delete all tracepoints. Note that the regular
12712 @code{delete} command can remove tracepoints also.
12717 (@value{GDBP}) @b{delete trace 1 2 3} // remove three tracepoints
12719 (@value{GDBP}) @b{delete trace} // remove all tracepoints
12723 You can abbreviate this command as @code{del tr}.
12726 @node Enable and Disable Tracepoints
12727 @subsection Enable and Disable Tracepoints
12729 These commands are deprecated; they are equivalent to plain @code{disable} and @code{enable}.
12732 @kindex disable tracepoint
12733 @item disable tracepoint @r{[}@var{num}@r{]}
12734 Disable tracepoint @var{num}, or all tracepoints if no argument
12735 @var{num} is given. A disabled tracepoint will have no effect during
12736 a trace experiment, but it is not forgotten. You can re-enable
12737 a disabled tracepoint using the @code{enable tracepoint} command.
12738 If the command is issued during a trace experiment and the debug target
12739 has support for disabling tracepoints during a trace experiment, then the
12740 change will be effective immediately. Otherwise, it will be applied to the
12741 next trace experiment.
12743 @kindex enable tracepoint
12744 @item enable tracepoint @r{[}@var{num}@r{]}
12745 Enable tracepoint @var{num}, or all tracepoints. If this command is
12746 issued during a trace experiment and the debug target supports enabling
12747 tracepoints during a trace experiment, then the enabled tracepoints will
12748 become effective immediately. Otherwise, they will become effective the
12749 next time a trace experiment is run.
12752 @node Tracepoint Passcounts
12753 @subsection Tracepoint Passcounts
12757 @cindex tracepoint pass count
12758 @item passcount @r{[}@var{n} @r{[}@var{num}@r{]]}
12759 Set the @dfn{passcount} of a tracepoint. The passcount is a way to
12760 automatically stop a trace experiment. If a tracepoint's passcount is
12761 @var{n}, then the trace experiment will be automatically stopped on
12762 the @var{n}'th time that tracepoint is hit. If the tracepoint number
12763 @var{num} is not specified, the @code{passcount} command sets the
12764 passcount of the most recently defined tracepoint. If no passcount is
12765 given, the trace experiment will run until stopped explicitly by the
12771 (@value{GDBP}) @b{passcount 5 2} // Stop on the 5th execution of
12772 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// tracepoint 2}
12774 (@value{GDBP}) @b{passcount 12} // Stop on the 12th execution of the
12775 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// most recently defined tracepoint.}
12776 (@value{GDBP}) @b{trace foo}
12777 (@value{GDBP}) @b{pass 3}
12778 (@value{GDBP}) @b{trace bar}
12779 (@value{GDBP}) @b{pass 2}
12780 (@value{GDBP}) @b{trace baz}
12781 (@value{GDBP}) @b{pass 1} // Stop tracing when foo has been
12782 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// executed 3 times OR when bar has}
12783 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// been executed 2 times}
12784 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// OR when baz has been executed 1 time.}
12788 @node Tracepoint Conditions
12789 @subsection Tracepoint Conditions
12790 @cindex conditional tracepoints
12791 @cindex tracepoint conditions
12793 The simplest sort of tracepoint collects data every time your program
12794 reaches a specified place. You can also specify a @dfn{condition} for
12795 a tracepoint. A condition is just a Boolean expression in your
12796 programming language (@pxref{Expressions, ,Expressions}). A
12797 tracepoint with a condition evaluates the expression each time your
12798 program reaches it, and data collection happens only if the condition
12801 Tracepoint conditions can be specified when a tracepoint is set, by
12802 using @samp{if} in the arguments to the @code{trace} command.
12803 @xref{Create and Delete Tracepoints, ,Setting Tracepoints}. They can
12804 also be set or changed at any time with the @code{condition} command,
12805 just as with breakpoints.
12807 Unlike breakpoint conditions, @value{GDBN} does not actually evaluate
12808 the conditional expression itself. Instead, @value{GDBN} encodes the
12809 expression into an agent expression (@pxref{Agent Expressions})
12810 suitable for execution on the target, independently of @value{GDBN}.
12811 Global variables become raw memory locations, locals become stack
12812 accesses, and so forth.
12814 For instance, suppose you have a function that is usually called
12815 frequently, but should not be called after an error has occurred. You
12816 could use the following tracepoint command to collect data about calls
12817 of that function that happen while the error code is propagating
12818 through the program; an unconditional tracepoint could end up
12819 collecting thousands of useless trace frames that you would have to
12823 (@value{GDBP}) @kbd{trace normal_operation if errcode > 0}
12826 @node Trace State Variables
12827 @subsection Trace State Variables
12828 @cindex trace state variables
12830 A @dfn{trace state variable} is a special type of variable that is
12831 created and managed by target-side code. The syntax is the same as
12832 that for GDB's convenience variables (a string prefixed with ``$''),
12833 but they are stored on the target. They must be created explicitly,
12834 using a @code{tvariable} command. They are always 64-bit signed
12837 Trace state variables are remembered by @value{GDBN}, and downloaded
12838 to the target along with tracepoint information when the trace
12839 experiment starts. There are no intrinsic limits on the number of
12840 trace state variables, beyond memory limitations of the target.
12842 @cindex convenience variables, and trace state variables
12843 Although trace state variables are managed by the target, you can use
12844 them in print commands and expressions as if they were convenience
12845 variables; @value{GDBN} will get the current value from the target
12846 while the trace experiment is running. Trace state variables share
12847 the same namespace as other ``$'' variables, which means that you
12848 cannot have trace state variables with names like @code{$23} or
12849 @code{$pc}, nor can you have a trace state variable and a convenience
12850 variable with the same name.
12854 @item tvariable $@var{name} [ = @var{expression} ]
12856 The @code{tvariable} command creates a new trace state variable named
12857 @code{$@var{name}}, and optionally gives it an initial value of
12858 @var{expression}. The @var{expression} is evaluated when this command is
12859 entered; the result will be converted to an integer if possible,
12860 otherwise @value{GDBN} will report an error. A subsequent
12861 @code{tvariable} command specifying the same name does not create a
12862 variable, but instead assigns the supplied initial value to the
12863 existing variable of that name, overwriting any previous initial
12864 value. The default initial value is 0.
12866 @item info tvariables
12867 @kindex info tvariables
12868 List all the trace state variables along with their initial values.
12869 Their current values may also be displayed, if the trace experiment is
12872 @item delete tvariable @r{[} $@var{name} @dots{} @r{]}
12873 @kindex delete tvariable
12874 Delete the given trace state variables, or all of them if no arguments
12879 @node Tracepoint Actions
12880 @subsection Tracepoint Action Lists
12884 @cindex tracepoint actions
12885 @item actions @r{[}@var{num}@r{]}
12886 This command will prompt for a list of actions to be taken when the
12887 tracepoint is hit. If the tracepoint number @var{num} is not
12888 specified, this command sets the actions for the one that was most
12889 recently defined (so that you can define a tracepoint and then say
12890 @code{actions} without bothering about its number). You specify the
12891 actions themselves on the following lines, one action at a time, and
12892 terminate the actions list with a line containing just @code{end}. So
12893 far, the only defined actions are @code{collect}, @code{teval}, and
12894 @code{while-stepping}.
12896 @code{actions} is actually equivalent to @code{commands} (@pxref{Break
12897 Commands, ,Breakpoint Command Lists}), except that only the defined
12898 actions are allowed; any other @value{GDBN} command is rejected.
12900 @cindex remove actions from a tracepoint
12901 To remove all actions from a tracepoint, type @samp{actions @var{num}}
12902 and follow it immediately with @samp{end}.
12905 (@value{GDBP}) @b{collect @var{data}} // collect some data
12907 (@value{GDBP}) @b{while-stepping 5} // single-step 5 times, collect data
12909 (@value{GDBP}) @b{end} // signals the end of actions.
12912 In the following example, the action list begins with @code{collect}
12913 commands indicating the things to be collected when the tracepoint is
12914 hit. Then, in order to single-step and collect additional data
12915 following the tracepoint, a @code{while-stepping} command is used,
12916 followed by the list of things to be collected after each step in a
12917 sequence of single steps. The @code{while-stepping} command is
12918 terminated by its own separate @code{end} command. Lastly, the action
12919 list is terminated by an @code{end} command.
12922 (@value{GDBP}) @b{trace foo}
12923 (@value{GDBP}) @b{actions}
12924 Enter actions for tracepoint 1, one per line:
12927 > while-stepping 12
12928 > collect $pc, arr[i]
12933 @kindex collect @r{(tracepoints)}
12934 @item collect@r{[}/@var{mods}@r{]} @var{expr1}, @var{expr2}, @dots{}
12935 Collect values of the given expressions when the tracepoint is hit.
12936 This command accepts a comma-separated list of any valid expressions.
12937 In addition to global, static, or local variables, the following
12938 special arguments are supported:
12942 Collect all registers.
12945 Collect all function arguments.
12948 Collect all local variables.
12951 Collect the return address. This is helpful if you want to see more
12954 @emph{Note:} The return address location can not always be reliably
12955 determined up front, and the wrong address / registers may end up
12956 collected instead. On some architectures the reliability is higher
12957 for tracepoints at function entry, while on others it's the opposite.
12958 When this happens, backtracing will stop because the return address is
12959 found unavailable (unless another collect rule happened to match it).
12962 Collects the number of arguments from the static probe at which the
12963 tracepoint is located.
12964 @xref{Static Probe Points}.
12966 @item $_probe_arg@var{n}
12967 @var{n} is an integer between 0 and 11. Collects the @var{n}th argument
12968 from the static probe at which the tracepoint is located.
12969 @xref{Static Probe Points}.
12972 @vindex $_sdata@r{, collect}
12973 Collect static tracepoint marker specific data. Only available for
12974 static tracepoints. @xref{Tracepoint Actions,,Tracepoint Action
12975 Lists}. On the UST static tracepoints library backend, an
12976 instrumentation point resembles a @code{printf} function call. The
12977 tracing library is able to collect user specified data formatted to a
12978 character string using the format provided by the programmer that
12979 instrumented the program. Other backends have similar mechanisms.
12980 Here's an example of a UST marker call:
12983 const char master_name[] = "$your_name";
12984 trace_mark(channel1, marker1, "hello %s", master_name)
12987 In this case, collecting @code{$_sdata} collects the string
12988 @samp{hello $yourname}. When analyzing the trace buffer, you can
12989 inspect @samp{$_sdata} like any other variable available to
12993 You can give several consecutive @code{collect} commands, each one
12994 with a single argument, or one @code{collect} command with several
12995 arguments separated by commas; the effect is the same.
12997 The optional @var{mods} changes the usual handling of the arguments.
12998 @code{s} requests that pointers to chars be handled as strings, in
12999 particular collecting the contents of the memory being pointed at, up
13000 to the first zero. The upper bound is by default the value of the
13001 @code{print elements} variable; if @code{s} is followed by a decimal
13002 number, that is the upper bound instead. So for instance
13003 @samp{collect/s25 mystr} collects as many as 25 characters at
13006 The command @code{info scope} (@pxref{Symbols, info scope}) is
13007 particularly useful for figuring out what data to collect.
13009 @kindex teval @r{(tracepoints)}
13010 @item teval @var{expr1}, @var{expr2}, @dots{}
13011 Evaluate the given expressions when the tracepoint is hit. This
13012 command accepts a comma-separated list of expressions. The results
13013 are discarded, so this is mainly useful for assigning values to trace
13014 state variables (@pxref{Trace State Variables}) without adding those
13015 values to the trace buffer, as would be the case if the @code{collect}
13018 @kindex while-stepping @r{(tracepoints)}
13019 @item while-stepping @var{n}
13020 Perform @var{n} single-step instruction traces after the tracepoint,
13021 collecting new data after each step. The @code{while-stepping}
13022 command is followed by the list of what to collect while stepping
13023 (followed by its own @code{end} command):
13026 > while-stepping 12
13027 > collect $regs, myglobal
13033 Note that @code{$pc} is not automatically collected by
13034 @code{while-stepping}; you need to explicitly collect that register if
13035 you need it. You may abbreviate @code{while-stepping} as @code{ws} or
13038 @item set default-collect @var{expr1}, @var{expr2}, @dots{}
13039 @kindex set default-collect
13040 @cindex default collection action
13041 This variable is a list of expressions to collect at each tracepoint
13042 hit. It is effectively an additional @code{collect} action prepended
13043 to every tracepoint action list. The expressions are parsed
13044 individually for each tracepoint, so for instance a variable named
13045 @code{xyz} may be interpreted as a global for one tracepoint, and a
13046 local for another, as appropriate to the tracepoint's location.
13048 @item show default-collect
13049 @kindex show default-collect
13050 Show the list of expressions that are collected by default at each
13055 @node Listing Tracepoints
13056 @subsection Listing Tracepoints
13059 @kindex info tracepoints @r{[}@var{n}@dots{}@r{]}
13060 @kindex info tp @r{[}@var{n}@dots{}@r{]}
13061 @cindex information about tracepoints
13062 @item info tracepoints @r{[}@var{num}@dots{}@r{]}
13063 Display information about the tracepoint @var{num}. If you don't
13064 specify a tracepoint number, displays information about all the
13065 tracepoints defined so far. The format is similar to that used for
13066 @code{info breakpoints}; in fact, @code{info tracepoints} is the same
13067 command, simply restricting itself to tracepoints.
13069 A tracepoint's listing may include additional information specific to
13074 its passcount as given by the @code{passcount @var{n}} command
13077 the state about installed on target of each location
13081 (@value{GDBP}) @b{info trace}
13082 Num Type Disp Enb Address What
13083 1 tracepoint keep y 0x0804ab57 in foo() at main.cxx:7
13085 collect globfoo, $regs
13090 2 tracepoint keep y <MULTIPLE>
13092 2.1 y 0x0804859c in func4 at change-loc.h:35
13093 installed on target
13094 2.2 y 0xb7ffc480 in func4 at change-loc.h:35
13095 installed on target
13096 2.3 y <PENDING> set_tracepoint
13097 3 tracepoint keep y 0x080485b1 in foo at change-loc.c:29
13098 not installed on target
13103 This command can be abbreviated @code{info tp}.
13106 @node Listing Static Tracepoint Markers
13107 @subsection Listing Static Tracepoint Markers
13110 @kindex info static-tracepoint-markers
13111 @cindex information about static tracepoint markers
13112 @item info static-tracepoint-markers
13113 Display information about all static tracepoint markers defined in the
13116 For each marker, the following columns are printed:
13120 An incrementing counter, output to help readability. This is not a
13123 The marker ID, as reported by the target.
13124 @item Enabled or Disabled
13125 Probed markers are tagged with @samp{y}. @samp{n} identifies marks
13126 that are not enabled.
13128 Where the marker is in your program, as a memory address.
13130 Where the marker is in the source for your program, as a file and line
13131 number. If the debug information included in the program does not
13132 allow @value{GDBN} to locate the source of the marker, this column
13133 will be left blank.
13137 In addition, the following information may be printed for each marker:
13141 User data passed to the tracing library by the marker call. In the
13142 UST backend, this is the format string passed as argument to the
13144 @item Static tracepoints probing the marker
13145 The list of static tracepoints attached to the marker.
13149 (@value{GDBP}) info static-tracepoint-markers
13150 Cnt ID Enb Address What
13151 1 ust/bar2 y 0x0000000000400e1a in main at stexample.c:25
13152 Data: number1 %d number2 %d
13153 Probed by static tracepoints: #2
13154 2 ust/bar33 n 0x0000000000400c87 in main at stexample.c:24
13160 @node Starting and Stopping Trace Experiments
13161 @subsection Starting and Stopping Trace Experiments
13164 @kindex tstart [ @var{notes} ]
13165 @cindex start a new trace experiment
13166 @cindex collected data discarded
13168 This command starts the trace experiment, and begins collecting data.
13169 It has the side effect of discarding all the data collected in the
13170 trace buffer during the previous trace experiment. If any arguments
13171 are supplied, they are taken as a note and stored with the trace
13172 experiment's state. The notes may be arbitrary text, and are
13173 especially useful with disconnected tracing in a multi-user context;
13174 the notes can explain what the trace is doing, supply user contact
13175 information, and so forth.
13177 @kindex tstop [ @var{notes} ]
13178 @cindex stop a running trace experiment
13180 This command stops the trace experiment. If any arguments are
13181 supplied, they are recorded with the experiment as a note. This is
13182 useful if you are stopping a trace started by someone else, for
13183 instance if the trace is interfering with the system's behavior and
13184 needs to be stopped quickly.
13186 @strong{Note}: a trace experiment and data collection may stop
13187 automatically if any tracepoint's passcount is reached
13188 (@pxref{Tracepoint Passcounts}), or if the trace buffer becomes full.
13191 @cindex status of trace data collection
13192 @cindex trace experiment, status of
13194 This command displays the status of the current trace data
13198 Here is an example of the commands we described so far:
13201 (@value{GDBP}) @b{trace gdb_c_test}
13202 (@value{GDBP}) @b{actions}
13203 Enter actions for tracepoint #1, one per line.
13204 > collect $regs,$locals,$args
13205 > while-stepping 11
13209 (@value{GDBP}) @b{tstart}
13210 [time passes @dots{}]
13211 (@value{GDBP}) @b{tstop}
13214 @anchor{disconnected tracing}
13215 @cindex disconnected tracing
13216 You can choose to continue running the trace experiment even if
13217 @value{GDBN} disconnects from the target, voluntarily or
13218 involuntarily. For commands such as @code{detach}, the debugger will
13219 ask what you want to do with the trace. But for unexpected
13220 terminations (@value{GDBN} crash, network outage), it would be
13221 unfortunate to lose hard-won trace data, so the variable
13222 @code{disconnected-tracing} lets you decide whether the trace should
13223 continue running without @value{GDBN}.
13226 @item set disconnected-tracing on
13227 @itemx set disconnected-tracing off
13228 @kindex set disconnected-tracing
13229 Choose whether a tracing run should continue to run if @value{GDBN}
13230 has disconnected from the target. Note that @code{detach} or
13231 @code{quit} will ask you directly what to do about a running trace no
13232 matter what this variable's setting, so the variable is mainly useful
13233 for handling unexpected situations, such as loss of the network.
13235 @item show disconnected-tracing
13236 @kindex show disconnected-tracing
13237 Show the current choice for disconnected tracing.
13241 When you reconnect to the target, the trace experiment may or may not
13242 still be running; it might have filled the trace buffer in the
13243 meantime, or stopped for one of the other reasons. If it is running,
13244 it will continue after reconnection.
13246 Upon reconnection, the target will upload information about the
13247 tracepoints in effect. @value{GDBN} will then compare that
13248 information to the set of tracepoints currently defined, and attempt
13249 to match them up, allowing for the possibility that the numbers may
13250 have changed due to creation and deletion in the meantime. If one of
13251 the target's tracepoints does not match any in @value{GDBN}, the
13252 debugger will create a new tracepoint, so that you have a number with
13253 which to specify that tracepoint. This matching-up process is
13254 necessarily heuristic, and it may result in useless tracepoints being
13255 created; you may simply delete them if they are of no use.
13257 @cindex circular trace buffer
13258 If your target agent supports a @dfn{circular trace buffer}, then you
13259 can run a trace experiment indefinitely without filling the trace
13260 buffer; when space runs out, the agent deletes already-collected trace
13261 frames, oldest first, until there is enough room to continue
13262 collecting. This is especially useful if your tracepoints are being
13263 hit too often, and your trace gets terminated prematurely because the
13264 buffer is full. To ask for a circular trace buffer, simply set
13265 @samp{circular-trace-buffer} to on. You can set this at any time,
13266 including during tracing; if the agent can do it, it will change
13267 buffer handling on the fly, otherwise it will not take effect until
13271 @item set circular-trace-buffer on
13272 @itemx set circular-trace-buffer off
13273 @kindex set circular-trace-buffer
13274 Choose whether a tracing run should use a linear or circular buffer
13275 for trace data. A linear buffer will not lose any trace data, but may
13276 fill up prematurely, while a circular buffer will discard old trace
13277 data, but it will have always room for the latest tracepoint hits.
13279 @item show circular-trace-buffer
13280 @kindex show circular-trace-buffer
13281 Show the current choice for the trace buffer. Note that this may not
13282 match the agent's current buffer handling, nor is it guaranteed to
13283 match the setting that might have been in effect during a past run,
13284 for instance if you are looking at frames from a trace file.
13289 @item set trace-buffer-size @var{n}
13290 @itemx set trace-buffer-size unlimited
13291 @kindex set trace-buffer-size
13292 Request that the target use a trace buffer of @var{n} bytes. Not all
13293 targets will honor the request; they may have a compiled-in size for
13294 the trace buffer, or some other limitation. Set to a value of
13295 @code{unlimited} or @code{-1} to let the target use whatever size it
13296 likes. This is also the default.
13298 @item show trace-buffer-size
13299 @kindex show trace-buffer-size
13300 Show the current requested size for the trace buffer. Note that this
13301 will only match the actual size if the target supports size-setting,
13302 and was able to handle the requested size. For instance, if the
13303 target can only change buffer size between runs, this variable will
13304 not reflect the change until the next run starts. Use @code{tstatus}
13305 to get a report of the actual buffer size.
13309 @item set trace-user @var{text}
13310 @kindex set trace-user
13312 @item show trace-user
13313 @kindex show trace-user
13315 @item set trace-notes @var{text}
13316 @kindex set trace-notes
13317 Set the trace run's notes.
13319 @item show trace-notes
13320 @kindex show trace-notes
13321 Show the trace run's notes.
13323 @item set trace-stop-notes @var{text}
13324 @kindex set trace-stop-notes
13325 Set the trace run's stop notes. The handling of the note is as for
13326 @code{tstop} arguments; the set command is convenient way to fix a
13327 stop note that is mistaken or incomplete.
13329 @item show trace-stop-notes
13330 @kindex show trace-stop-notes
13331 Show the trace run's stop notes.
13335 @node Tracepoint Restrictions
13336 @subsection Tracepoint Restrictions
13338 @cindex tracepoint restrictions
13339 There are a number of restrictions on the use of tracepoints. As
13340 described above, tracepoint data gathering occurs on the target
13341 without interaction from @value{GDBN}. Thus the full capabilities of
13342 the debugger are not available during data gathering, and then at data
13343 examination time, you will be limited by only having what was
13344 collected. The following items describe some common problems, but it
13345 is not exhaustive, and you may run into additional difficulties not
13351 Tracepoint expressions are intended to gather objects (lvalues). Thus
13352 the full flexibility of GDB's expression evaluator is not available.
13353 You cannot call functions, cast objects to aggregate types, access
13354 convenience variables or modify values (except by assignment to trace
13355 state variables). Some language features may implicitly call
13356 functions (for instance Objective-C fields with accessors), and therefore
13357 cannot be collected either.
13360 Collection of local variables, either individually or in bulk with
13361 @code{$locals} or @code{$args}, during @code{while-stepping} may
13362 behave erratically. The stepping action may enter a new scope (for
13363 instance by stepping into a function), or the location of the variable
13364 may change (for instance it is loaded into a register). The
13365 tracepoint data recorded uses the location information for the
13366 variables that is correct for the tracepoint location. When the
13367 tracepoint is created, it is not possible, in general, to determine
13368 where the steps of a @code{while-stepping} sequence will advance the
13369 program---particularly if a conditional branch is stepped.
13372 Collection of an incompletely-initialized or partially-destroyed object
13373 may result in something that @value{GDBN} cannot display, or displays
13374 in a misleading way.
13377 When @value{GDBN} displays a pointer to character it automatically
13378 dereferences the pointer to also display characters of the string
13379 being pointed to. However, collecting the pointer during tracing does
13380 not automatically collect the string. You need to explicitly
13381 dereference the pointer and provide size information if you want to
13382 collect not only the pointer, but the memory pointed to. For example,
13383 @code{*ptr@@50} can be used to collect the 50 element array pointed to
13387 It is not possible to collect a complete stack backtrace at a
13388 tracepoint. Instead, you may collect the registers and a few hundred
13389 bytes from the stack pointer with something like @code{*(unsigned char *)$esp@@300}
13390 (adjust to use the name of the actual stack pointer register on your
13391 target architecture, and the amount of stack you wish to capture).
13392 Then the @code{backtrace} command will show a partial backtrace when
13393 using a trace frame. The number of stack frames that can be examined
13394 depends on the sizes of the frames in the collected stack. Note that
13395 if you ask for a block so large that it goes past the bottom of the
13396 stack, the target agent may report an error trying to read from an
13400 If you do not collect registers at a tracepoint, @value{GDBN} can
13401 infer that the value of @code{$pc} must be the same as the address of
13402 the tracepoint and use that when you are looking at a trace frame
13403 for that tracepoint. However, this cannot work if the tracepoint has
13404 multiple locations (for instance if it was set in a function that was
13405 inlined), or if it has a @code{while-stepping} loop. In those cases
13406 @value{GDBN} will warn you that it can't infer @code{$pc}, and default
13411 @node Analyze Collected Data
13412 @section Using the Collected Data
13414 After the tracepoint experiment ends, you use @value{GDBN} commands
13415 for examining the trace data. The basic idea is that each tracepoint
13416 collects a trace @dfn{snapshot} every time it is hit and another
13417 snapshot every time it single-steps. All these snapshots are
13418 consecutively numbered from zero and go into a buffer, and you can
13419 examine them later. The way you examine them is to @dfn{focus} on a
13420 specific trace snapshot. When the remote stub is focused on a trace
13421 snapshot, it will respond to all @value{GDBN} requests for memory and
13422 registers by reading from the buffer which belongs to that snapshot,
13423 rather than from @emph{real} memory or registers of the program being
13424 debugged. This means that @strong{all} @value{GDBN} commands
13425 (@code{print}, @code{info registers}, @code{backtrace}, etc.) will
13426 behave as if we were currently debugging the program state as it was
13427 when the tracepoint occurred. Any requests for data that are not in
13428 the buffer will fail.
13431 * tfind:: How to select a trace snapshot
13432 * tdump:: How to display all data for a snapshot
13433 * save tracepoints:: How to save tracepoints for a future run
13437 @subsection @code{tfind @var{n}}
13440 @cindex select trace snapshot
13441 @cindex find trace snapshot
13442 The basic command for selecting a trace snapshot from the buffer is
13443 @code{tfind @var{n}}, which finds trace snapshot number @var{n},
13444 counting from zero. If no argument @var{n} is given, the next
13445 snapshot is selected.
13447 Here are the various forms of using the @code{tfind} command.
13451 Find the first snapshot in the buffer. This is a synonym for
13452 @code{tfind 0} (since 0 is the number of the first snapshot).
13455 Stop debugging trace snapshots, resume @emph{live} debugging.
13458 Same as @samp{tfind none}.
13461 No argument means find the next trace snapshot or find the first
13462 one if no trace snapshot is selected.
13465 Find the previous trace snapshot before the current one. This permits
13466 retracing earlier steps.
13468 @item tfind tracepoint @var{num}
13469 Find the next snapshot associated with tracepoint @var{num}. Search
13470 proceeds forward from the last examined trace snapshot. If no
13471 argument @var{num} is given, it means find the next snapshot collected
13472 for the same tracepoint as the current snapshot.
13474 @item tfind pc @var{addr}
13475 Find the next snapshot associated with the value @var{addr} of the
13476 program counter. Search proceeds forward from the last examined trace
13477 snapshot. If no argument @var{addr} is given, it means find the next
13478 snapshot with the same value of PC as the current snapshot.
13480 @item tfind outside @var{addr1}, @var{addr2}
13481 Find the next snapshot whose PC is outside the given range of
13482 addresses (exclusive).
13484 @item tfind range @var{addr1}, @var{addr2}
13485 Find the next snapshot whose PC is between @var{addr1} and
13486 @var{addr2} (inclusive).
13488 @item tfind line @r{[}@var{file}:@r{]}@var{n}
13489 Find the next snapshot associated with the source line @var{n}. If
13490 the optional argument @var{file} is given, refer to line @var{n} in
13491 that source file. Search proceeds forward from the last examined
13492 trace snapshot. If no argument @var{n} is given, it means find the
13493 next line other than the one currently being examined; thus saying
13494 @code{tfind line} repeatedly can appear to have the same effect as
13495 stepping from line to line in a @emph{live} debugging session.
13498 The default arguments for the @code{tfind} commands are specifically
13499 designed to make it easy to scan through the trace buffer. For
13500 instance, @code{tfind} with no argument selects the next trace
13501 snapshot, and @code{tfind -} with no argument selects the previous
13502 trace snapshot. So, by giving one @code{tfind} command, and then
13503 simply hitting @key{RET} repeatedly you can examine all the trace
13504 snapshots in order. Or, by saying @code{tfind -} and then hitting
13505 @key{RET} repeatedly you can examine the snapshots in reverse order.
13506 The @code{tfind line} command with no argument selects the snapshot
13507 for the next source line executed. The @code{tfind pc} command with
13508 no argument selects the next snapshot with the same program counter
13509 (PC) as the current frame. The @code{tfind tracepoint} command with
13510 no argument selects the next trace snapshot collected by the same
13511 tracepoint as the current one.
13513 In addition to letting you scan through the trace buffer manually,
13514 these commands make it easy to construct @value{GDBN} scripts that
13515 scan through the trace buffer and print out whatever collected data
13516 you are interested in. Thus, if we want to examine the PC, FP, and SP
13517 registers from each trace frame in the buffer, we can say this:
13520 (@value{GDBP}) @b{tfind start}
13521 (@value{GDBP}) @b{while ($trace_frame != -1)}
13522 > printf "Frame %d, PC = %08X, SP = %08X, FP = %08X\n", \
13523 $trace_frame, $pc, $sp, $fp
13527 Frame 0, PC = 0020DC64, SP = 0030BF3C, FP = 0030BF44
13528 Frame 1, PC = 0020DC6C, SP = 0030BF38, FP = 0030BF44
13529 Frame 2, PC = 0020DC70, SP = 0030BF34, FP = 0030BF44
13530 Frame 3, PC = 0020DC74, SP = 0030BF30, FP = 0030BF44
13531 Frame 4, PC = 0020DC78, SP = 0030BF2C, FP = 0030BF44
13532 Frame 5, PC = 0020DC7C, SP = 0030BF28, FP = 0030BF44
13533 Frame 6, PC = 0020DC80, SP = 0030BF24, FP = 0030BF44
13534 Frame 7, PC = 0020DC84, SP = 0030BF20, FP = 0030BF44
13535 Frame 8, PC = 0020DC88, SP = 0030BF1C, FP = 0030BF44
13536 Frame 9, PC = 0020DC8E, SP = 0030BF18, FP = 0030BF44
13537 Frame 10, PC = 00203F6C, SP = 0030BE3C, FP = 0030BF14
13540 Or, if we want to examine the variable @code{X} at each source line in
13544 (@value{GDBP}) @b{tfind start}
13545 (@value{GDBP}) @b{while ($trace_frame != -1)}
13546 > printf "Frame %d, X == %d\n", $trace_frame, X
13556 @subsection @code{tdump}
13558 @cindex dump all data collected at tracepoint
13559 @cindex tracepoint data, display
13561 This command takes no arguments. It prints all the data collected at
13562 the current trace snapshot.
13565 (@value{GDBP}) @b{trace 444}
13566 (@value{GDBP}) @b{actions}
13567 Enter actions for tracepoint #2, one per line:
13568 > collect $regs, $locals, $args, gdb_long_test
13571 (@value{GDBP}) @b{tstart}
13573 (@value{GDBP}) @b{tfind line 444}
13574 #0 gdb_test (p1=0x11, p2=0x22, p3=0x33, p4=0x44, p5=0x55, p6=0x66)
13576 444 printp( "%s: arguments = 0x%X 0x%X 0x%X 0x%X 0x%X 0x%X\n", )
13578 (@value{GDBP}) @b{tdump}
13579 Data collected at tracepoint 2, trace frame 1:
13580 d0 0xc4aa0085 -995491707
13584 d4 0x71aea3d 119204413
13587 d7 0x380035 3670069
13588 a0 0x19e24a 1696330
13589 a1 0x3000668 50333288
13591 a3 0x322000 3284992
13592 a4 0x3000698 50333336
13593 a5 0x1ad3cc 1758156
13594 fp 0x30bf3c 0x30bf3c
13595 sp 0x30bf34 0x30bf34
13597 pc 0x20b2c8 0x20b2c8
13601 p = 0x20e5b4 "gdb-test"
13608 gdb_long_test = 17 '\021'
13613 @code{tdump} works by scanning the tracepoint's current collection
13614 actions and printing the value of each expression listed. So
13615 @code{tdump} can fail, if after a run, you change the tracepoint's
13616 actions to mention variables that were not collected during the run.
13618 Also, for tracepoints with @code{while-stepping} loops, @code{tdump}
13619 uses the collected value of @code{$pc} to distinguish between trace
13620 frames that were collected at the tracepoint hit, and frames that were
13621 collected while stepping. This allows it to correctly choose whether
13622 to display the basic list of collections, or the collections from the
13623 body of the while-stepping loop. However, if @code{$pc} was not collected,
13624 then @code{tdump} will always attempt to dump using the basic collection
13625 list, and may fail if a while-stepping frame does not include all the
13626 same data that is collected at the tracepoint hit.
13627 @c This is getting pretty arcane, example would be good.
13629 @node save tracepoints
13630 @subsection @code{save tracepoints @var{filename}}
13631 @kindex save tracepoints
13632 @kindex save-tracepoints
13633 @cindex save tracepoints for future sessions
13635 This command saves all current tracepoint definitions together with
13636 their actions and passcounts, into a file @file{@var{filename}}
13637 suitable for use in a later debugging session. To read the saved
13638 tracepoint definitions, use the @code{source} command (@pxref{Command
13639 Files}). The @w{@code{save-tracepoints}} command is a deprecated
13640 alias for @w{@code{save tracepoints}}
13642 @node Tracepoint Variables
13643 @section Convenience Variables for Tracepoints
13644 @cindex tracepoint variables
13645 @cindex convenience variables for tracepoints
13648 @vindex $trace_frame
13649 @item (int) $trace_frame
13650 The current trace snapshot (a.k.a.@: @dfn{frame}) number, or -1 if no
13651 snapshot is selected.
13653 @vindex $tracepoint
13654 @item (int) $tracepoint
13655 The tracepoint for the current trace snapshot.
13657 @vindex $trace_line
13658 @item (int) $trace_line
13659 The line number for the current trace snapshot.
13661 @vindex $trace_file
13662 @item (char []) $trace_file
13663 The source file for the current trace snapshot.
13665 @vindex $trace_func
13666 @item (char []) $trace_func
13667 The name of the function containing @code{$tracepoint}.
13670 Note: @code{$trace_file} is not suitable for use in @code{printf},
13671 use @code{output} instead.
13673 Here's a simple example of using these convenience variables for
13674 stepping through all the trace snapshots and printing some of their
13675 data. Note that these are not the same as trace state variables,
13676 which are managed by the target.
13679 (@value{GDBP}) @b{tfind start}
13681 (@value{GDBP}) @b{while $trace_frame != -1}
13682 > output $trace_file
13683 > printf ", line %d (tracepoint #%d)\n", $trace_line, $tracepoint
13689 @section Using Trace Files
13690 @cindex trace files
13692 In some situations, the target running a trace experiment may no
13693 longer be available; perhaps it crashed, or the hardware was needed
13694 for a different activity. To handle these cases, you can arrange to
13695 dump the trace data into a file, and later use that file as a source
13696 of trace data, via the @code{target tfile} command.
13701 @item tsave [ -r ] @var{filename}
13702 @itemx tsave [-ctf] @var{dirname}
13703 Save the trace data to @var{filename}. By default, this command
13704 assumes that @var{filename} refers to the host filesystem, so if
13705 necessary @value{GDBN} will copy raw trace data up from the target and
13706 then save it. If the target supports it, you can also supply the
13707 optional argument @code{-r} (``remote'') to direct the target to save
13708 the data directly into @var{filename} in its own filesystem, which may be
13709 more efficient if the trace buffer is very large. (Note, however, that
13710 @code{target tfile} can only read from files accessible to the host.)
13711 By default, this command will save trace frame in tfile format.
13712 You can supply the optional argument @code{-ctf} to save data in CTF
13713 format. The @dfn{Common Trace Format} (CTF) is proposed as a trace format
13714 that can be shared by multiple debugging and tracing tools. Please go to
13715 @indicateurl{http://www.efficios.com/ctf} to get more information.
13717 @kindex target tfile
13721 @item target tfile @var{filename}
13722 @itemx target ctf @var{dirname}
13723 Use the file named @var{filename} or directory named @var{dirname} as
13724 a source of trace data. Commands that examine data work as they do with
13725 a live target, but it is not possible to run any new trace experiments.
13726 @code{tstatus} will report the state of the trace run at the moment
13727 the data was saved, as well as the current trace frame you are examining.
13728 Both @var{filename} and @var{dirname} must be on a filesystem accessible to
13732 (@value{GDBP}) target ctf ctf.ctf
13733 (@value{GDBP}) tfind
13734 Found trace frame 0, tracepoint 2
13735 39 ++a; /* set tracepoint 1 here */
13736 (@value{GDBP}) tdump
13737 Data collected at tracepoint 2, trace frame 0:
13741 c = @{"123", "456", "789", "123", "456", "789"@}
13742 d = @{@{@{a = 1, b = 2@}, @{a = 3, b = 4@}@}, @{@{a = 5, b = 6@}, @{a = 7, b = 8@}@}@}
13750 @chapter Debugging Programs That Use Overlays
13753 If your program is too large to fit completely in your target system's
13754 memory, you can sometimes use @dfn{overlays} to work around this
13755 problem. @value{GDBN} provides some support for debugging programs that
13759 * How Overlays Work:: A general explanation of overlays.
13760 * Overlay Commands:: Managing overlays in @value{GDBN}.
13761 * Automatic Overlay Debugging:: @value{GDBN} can find out which overlays are
13762 mapped by asking the inferior.
13763 * Overlay Sample Program:: A sample program using overlays.
13766 @node How Overlays Work
13767 @section How Overlays Work
13768 @cindex mapped overlays
13769 @cindex unmapped overlays
13770 @cindex load address, overlay's
13771 @cindex mapped address
13772 @cindex overlay area
13774 Suppose you have a computer whose instruction address space is only 64
13775 kilobytes long, but which has much more memory which can be accessed by
13776 other means: special instructions, segment registers, or memory
13777 management hardware, for example. Suppose further that you want to
13778 adapt a program which is larger than 64 kilobytes to run on this system.
13780 One solution is to identify modules of your program which are relatively
13781 independent, and need not call each other directly; call these modules
13782 @dfn{overlays}. Separate the overlays from the main program, and place
13783 their machine code in the larger memory. Place your main program in
13784 instruction memory, but leave at least enough space there to hold the
13785 largest overlay as well.
13787 Now, to call a function located in an overlay, you must first copy that
13788 overlay's machine code from the large memory into the space set aside
13789 for it in the instruction memory, and then jump to its entry point
13792 @c NB: In the below the mapped area's size is greater or equal to the
13793 @c size of all overlays. This is intentional to remind the developer
13794 @c that overlays don't necessarily need to be the same size.
13798 Data Instruction Larger
13799 Address Space Address Space Address Space
13800 +-----------+ +-----------+ +-----------+
13802 +-----------+ +-----------+ +-----------+<-- overlay 1
13803 | program | | main | .----| overlay 1 | load address
13804 | variables | | program | | +-----------+
13805 | and heap | | | | | |
13806 +-----------+ | | | +-----------+<-- overlay 2
13807 | | +-----------+ | | | load address
13808 +-----------+ | | | .-| overlay 2 |
13810 mapped --->+-----------+ | | +-----------+
13811 address | | | | | |
13812 | overlay | <-' | | |
13813 | area | <---' +-----------+<-- overlay 3
13814 | | <---. | | load address
13815 +-----------+ `--| overlay 3 |
13822 @anchor{A code overlay}A code overlay
13826 The diagram (@pxref{A code overlay}) shows a system with separate data
13827 and instruction address spaces. To map an overlay, the program copies
13828 its code from the larger address space to the instruction address space.
13829 Since the overlays shown here all use the same mapped address, only one
13830 may be mapped at a time. For a system with a single address space for
13831 data and instructions, the diagram would be similar, except that the
13832 program variables and heap would share an address space with the main
13833 program and the overlay area.
13835 An overlay loaded into instruction memory and ready for use is called a
13836 @dfn{mapped} overlay; its @dfn{mapped address} is its address in the
13837 instruction memory. An overlay not present (or only partially present)
13838 in instruction memory is called @dfn{unmapped}; its @dfn{load address}
13839 is its address in the larger memory. The mapped address is also called
13840 the @dfn{virtual memory address}, or @dfn{VMA}; the load address is also
13841 called the @dfn{load memory address}, or @dfn{LMA}.
13843 Unfortunately, overlays are not a completely transparent way to adapt a
13844 program to limited instruction memory. They introduce a new set of
13845 global constraints you must keep in mind as you design your program:
13850 Before calling or returning to a function in an overlay, your program
13851 must make sure that overlay is actually mapped. Otherwise, the call or
13852 return will transfer control to the right address, but in the wrong
13853 overlay, and your program will probably crash.
13856 If the process of mapping an overlay is expensive on your system, you
13857 will need to choose your overlays carefully to minimize their effect on
13858 your program's performance.
13861 The executable file you load onto your system must contain each
13862 overlay's instructions, appearing at the overlay's load address, not its
13863 mapped address. However, each overlay's instructions must be relocated
13864 and its symbols defined as if the overlay were at its mapped address.
13865 You can use GNU linker scripts to specify different load and relocation
13866 addresses for pieces of your program; see @ref{Overlay Description,,,
13867 ld.info, Using ld: the GNU linker}.
13870 The procedure for loading executable files onto your system must be able
13871 to load their contents into the larger address space as well as the
13872 instruction and data spaces.
13876 The overlay system described above is rather simple, and could be
13877 improved in many ways:
13882 If your system has suitable bank switch registers or memory management
13883 hardware, you could use those facilities to make an overlay's load area
13884 contents simply appear at their mapped address in instruction space.
13885 This would probably be faster than copying the overlay to its mapped
13886 area in the usual way.
13889 If your overlays are small enough, you could set aside more than one
13890 overlay area, and have more than one overlay mapped at a time.
13893 You can use overlays to manage data, as well as instructions. In
13894 general, data overlays are even less transparent to your design than
13895 code overlays: whereas code overlays only require care when you call or
13896 return to functions, data overlays require care every time you access
13897 the data. Also, if you change the contents of a data overlay, you
13898 must copy its contents back out to its load address before you can copy a
13899 different data overlay into the same mapped area.
13904 @node Overlay Commands
13905 @section Overlay Commands
13907 To use @value{GDBN}'s overlay support, each overlay in your program must
13908 correspond to a separate section of the executable file. The section's
13909 virtual memory address and load memory address must be the overlay's
13910 mapped and load addresses. Identifying overlays with sections allows
13911 @value{GDBN} to determine the appropriate address of a function or
13912 variable, depending on whether the overlay is mapped or not.
13914 @value{GDBN}'s overlay commands all start with the word @code{overlay};
13915 you can abbreviate this as @code{ov} or @code{ovly}. The commands are:
13920 Disable @value{GDBN}'s overlay support. When overlay support is
13921 disabled, @value{GDBN} assumes that all functions and variables are
13922 always present at their mapped addresses. By default, @value{GDBN}'s
13923 overlay support is disabled.
13925 @item overlay manual
13926 @cindex manual overlay debugging
13927 Enable @dfn{manual} overlay debugging. In this mode, @value{GDBN}
13928 relies on you to tell it which overlays are mapped, and which are not,
13929 using the @code{overlay map-overlay} and @code{overlay unmap-overlay}
13930 commands described below.
13932 @item overlay map-overlay @var{overlay}
13933 @itemx overlay map @var{overlay}
13934 @cindex map an overlay
13935 Tell @value{GDBN} that @var{overlay} is now mapped; @var{overlay} must
13936 be the name of the object file section containing the overlay. When an
13937 overlay is mapped, @value{GDBN} assumes it can find the overlay's
13938 functions and variables at their mapped addresses. @value{GDBN} assumes
13939 that any other overlays whose mapped ranges overlap that of
13940 @var{overlay} are now unmapped.
13942 @item overlay unmap-overlay @var{overlay}
13943 @itemx overlay unmap @var{overlay}
13944 @cindex unmap an overlay
13945 Tell @value{GDBN} that @var{overlay} is no longer mapped; @var{overlay}
13946 must be the name of the object file section containing the overlay.
13947 When an overlay is unmapped, @value{GDBN} assumes it can find the
13948 overlay's functions and variables at their load addresses.
13951 Enable @dfn{automatic} overlay debugging. In this mode, @value{GDBN}
13952 consults a data structure the overlay manager maintains in the inferior
13953 to see which overlays are mapped. For details, see @ref{Automatic
13954 Overlay Debugging}.
13956 @item overlay load-target
13957 @itemx overlay load
13958 @cindex reloading the overlay table
13959 Re-read the overlay table from the inferior. Normally, @value{GDBN}
13960 re-reads the table @value{GDBN} automatically each time the inferior
13961 stops, so this command should only be necessary if you have changed the
13962 overlay mapping yourself using @value{GDBN}. This command is only
13963 useful when using automatic overlay debugging.
13965 @item overlay list-overlays
13966 @itemx overlay list
13967 @cindex listing mapped overlays
13968 Display a list of the overlays currently mapped, along with their mapped
13969 addresses, load addresses, and sizes.
13973 Normally, when @value{GDBN} prints a code address, it includes the name
13974 of the function the address falls in:
13977 (@value{GDBP}) print main
13978 $3 = @{int ()@} 0x11a0 <main>
13981 When overlay debugging is enabled, @value{GDBN} recognizes code in
13982 unmapped overlays, and prints the names of unmapped functions with
13983 asterisks around them. For example, if @code{foo} is a function in an
13984 unmapped overlay, @value{GDBN} prints it this way:
13987 (@value{GDBP}) overlay list
13988 No sections are mapped.
13989 (@value{GDBP}) print foo
13990 $5 = @{int (int)@} 0x100000 <*foo*>
13993 When @code{foo}'s overlay is mapped, @value{GDBN} prints the function's
13997 (@value{GDBP}) overlay list
13998 Section .ov.foo.text, loaded at 0x100000 - 0x100034,
13999 mapped at 0x1016 - 0x104a
14000 (@value{GDBP}) print foo
14001 $6 = @{int (int)@} 0x1016 <foo>
14004 When overlay debugging is enabled, @value{GDBN} can find the correct
14005 address for functions and variables in an overlay, whether or not the
14006 overlay is mapped. This allows most @value{GDBN} commands, like
14007 @code{break} and @code{disassemble}, to work normally, even on unmapped
14008 code. However, @value{GDBN}'s breakpoint support has some limitations:
14012 @cindex breakpoints in overlays
14013 @cindex overlays, setting breakpoints in
14014 You can set breakpoints in functions in unmapped overlays, as long as
14015 @value{GDBN} can write to the overlay at its load address.
14017 @value{GDBN} can not set hardware or simulator-based breakpoints in
14018 unmapped overlays. However, if you set a breakpoint at the end of your
14019 overlay manager (and tell @value{GDBN} which overlays are now mapped, if
14020 you are using manual overlay management), @value{GDBN} will re-set its
14021 breakpoints properly.
14025 @node Automatic Overlay Debugging
14026 @section Automatic Overlay Debugging
14027 @cindex automatic overlay debugging
14029 @value{GDBN} can automatically track which overlays are mapped and which
14030 are not, given some simple co-operation from the overlay manager in the
14031 inferior. If you enable automatic overlay debugging with the
14032 @code{overlay auto} command (@pxref{Overlay Commands}), @value{GDBN}
14033 looks in the inferior's memory for certain variables describing the
14034 current state of the overlays.
14036 Here are the variables your overlay manager must define to support
14037 @value{GDBN}'s automatic overlay debugging:
14041 @item @code{_ovly_table}:
14042 This variable must be an array of the following structures:
14047 /* The overlay's mapped address. */
14050 /* The size of the overlay, in bytes. */
14051 unsigned long size;
14053 /* The overlay's load address. */
14056 /* Non-zero if the overlay is currently mapped;
14058 unsigned long mapped;
14062 @item @code{_novlys}:
14063 This variable must be a four-byte signed integer, holding the total
14064 number of elements in @code{_ovly_table}.
14068 To decide whether a particular overlay is mapped or not, @value{GDBN}
14069 looks for an entry in @w{@code{_ovly_table}} whose @code{vma} and
14070 @code{lma} members equal the VMA and LMA of the overlay's section in the
14071 executable file. When @value{GDBN} finds a matching entry, it consults
14072 the entry's @code{mapped} member to determine whether the overlay is
14075 In addition, your overlay manager may define a function called
14076 @code{_ovly_debug_event}. If this function is defined, @value{GDBN}
14077 will silently set a breakpoint there. If the overlay manager then
14078 calls this function whenever it has changed the overlay table, this
14079 will enable @value{GDBN} to accurately keep track of which overlays
14080 are in program memory, and update any breakpoints that may be set
14081 in overlays. This will allow breakpoints to work even if the
14082 overlays are kept in ROM or other non-writable memory while they
14083 are not being executed.
14085 @node Overlay Sample Program
14086 @section Overlay Sample Program
14087 @cindex overlay example program
14089 When linking a program which uses overlays, you must place the overlays
14090 at their load addresses, while relocating them to run at their mapped
14091 addresses. To do this, you must write a linker script (@pxref{Overlay
14092 Description,,, ld.info, Using ld: the GNU linker}). Unfortunately,
14093 since linker scripts are specific to a particular host system, target
14094 architecture, and target memory layout, this manual cannot provide
14095 portable sample code demonstrating @value{GDBN}'s overlay support.
14097 However, the @value{GDBN} source distribution does contain an overlaid
14098 program, with linker scripts for a few systems, as part of its test
14099 suite. The program consists of the following files from
14100 @file{gdb/testsuite/gdb.base}:
14104 The main program file.
14106 A simple overlay manager, used by @file{overlays.c}.
14111 Overlay modules, loaded and used by @file{overlays.c}.
14114 Linker scripts for linking the test program on the @code{d10v-elf}
14115 and @code{m32r-elf} targets.
14118 You can build the test program using the @code{d10v-elf} GCC
14119 cross-compiler like this:
14122 $ d10v-elf-gcc -g -c overlays.c
14123 $ d10v-elf-gcc -g -c ovlymgr.c
14124 $ d10v-elf-gcc -g -c foo.c
14125 $ d10v-elf-gcc -g -c bar.c
14126 $ d10v-elf-gcc -g -c baz.c
14127 $ d10v-elf-gcc -g -c grbx.c
14128 $ d10v-elf-gcc -g overlays.o ovlymgr.o foo.o bar.o \
14129 baz.o grbx.o -Wl,-Td10v.ld -o overlays
14132 The build process is identical for any other architecture, except that
14133 you must substitute the appropriate compiler and linker script for the
14134 target system for @code{d10v-elf-gcc} and @code{d10v.ld}.
14138 @chapter Using @value{GDBN} with Different Languages
14141 Although programming languages generally have common aspects, they are
14142 rarely expressed in the same manner. For instance, in ANSI C,
14143 dereferencing a pointer @code{p} is accomplished by @code{*p}, but in
14144 Modula-2, it is accomplished by @code{p^}. Values can also be
14145 represented (and displayed) differently. Hex numbers in C appear as
14146 @samp{0x1ae}, while in Modula-2 they appear as @samp{1AEH}.
14148 @cindex working language
14149 Language-specific information is built into @value{GDBN} for some languages,
14150 allowing you to express operations like the above in your program's
14151 native language, and allowing @value{GDBN} to output values in a manner
14152 consistent with the syntax of your program's native language. The
14153 language you use to build expressions is called the @dfn{working
14157 * Setting:: Switching between source languages
14158 * Show:: Displaying the language
14159 * Checks:: Type and range checks
14160 * Supported Languages:: Supported languages
14161 * Unsupported Languages:: Unsupported languages
14165 @section Switching Between Source Languages
14167 There are two ways to control the working language---either have @value{GDBN}
14168 set it automatically, or select it manually yourself. You can use the
14169 @code{set language} command for either purpose. On startup, @value{GDBN}
14170 defaults to setting the language automatically. The working language is
14171 used to determine how expressions you type are interpreted, how values
14174 In addition to the working language, every source file that
14175 @value{GDBN} knows about has its own working language. For some object
14176 file formats, the compiler might indicate which language a particular
14177 source file is in. However, most of the time @value{GDBN} infers the
14178 language from the name of the file. The language of a source file
14179 controls whether C@t{++} names are demangled---this way @code{backtrace} can
14180 show each frame appropriately for its own language. There is no way to
14181 set the language of a source file from within @value{GDBN}, but you can
14182 set the language associated with a filename extension. @xref{Show, ,
14183 Displaying the Language}.
14185 This is most commonly a problem when you use a program, such
14186 as @code{cfront} or @code{f2c}, that generates C but is written in
14187 another language. In that case, make the
14188 program use @code{#line} directives in its C output; that way
14189 @value{GDBN} will know the correct language of the source code of the original
14190 program, and will display that source code, not the generated C code.
14193 * Filenames:: Filename extensions and languages.
14194 * Manually:: Setting the working language manually
14195 * Automatically:: Having @value{GDBN} infer the source language
14199 @subsection List of Filename Extensions and Languages
14201 If a source file name ends in one of the following extensions, then
14202 @value{GDBN} infers that its language is the one indicated.
14220 C@t{++} source file
14226 Objective-C source file
14230 Fortran source file
14233 Modula-2 source file
14237 Assembler source file. This actually behaves almost like C, but
14238 @value{GDBN} does not skip over function prologues when stepping.
14241 In addition, you may set the language associated with a filename
14242 extension. @xref{Show, , Displaying the Language}.
14245 @subsection Setting the Working Language
14247 If you allow @value{GDBN} to set the language automatically,
14248 expressions are interpreted the same way in your debugging session and
14251 @kindex set language
14252 If you wish, you may set the language manually. To do this, issue the
14253 command @samp{set language @var{lang}}, where @var{lang} is the name of
14254 a language, such as
14255 @code{c} or @code{modula-2}.
14256 For a list of the supported languages, type @samp{set language}.
14258 Setting the language manually prevents @value{GDBN} from updating the working
14259 language automatically. This can lead to confusion if you try
14260 to debug a program when the working language is not the same as the
14261 source language, when an expression is acceptable to both
14262 languages---but means different things. For instance, if the current
14263 source file were written in C, and @value{GDBN} was parsing Modula-2, a
14271 might not have the effect you intended. In C, this means to add
14272 @code{b} and @code{c} and place the result in @code{a}. The result
14273 printed would be the value of @code{a}. In Modula-2, this means to compare
14274 @code{a} to the result of @code{b+c}, yielding a @code{BOOLEAN} value.
14276 @node Automatically
14277 @subsection Having @value{GDBN} Infer the Source Language
14279 To have @value{GDBN} set the working language automatically, use
14280 @samp{set language local} or @samp{set language auto}. @value{GDBN}
14281 then infers the working language. That is, when your program stops in a
14282 frame (usually by encountering a breakpoint), @value{GDBN} sets the
14283 working language to the language recorded for the function in that
14284 frame. If the language for a frame is unknown (that is, if the function
14285 or block corresponding to the frame was defined in a source file that
14286 does not have a recognized extension), the current working language is
14287 not changed, and @value{GDBN} issues a warning.
14289 This may not seem necessary for most programs, which are written
14290 entirely in one source language. However, program modules and libraries
14291 written in one source language can be used by a main program written in
14292 a different source language. Using @samp{set language auto} in this
14293 case frees you from having to set the working language manually.
14296 @section Displaying the Language
14298 The following commands help you find out which language is the
14299 working language, and also what language source files were written in.
14302 @item show language
14303 @anchor{show language}
14304 @kindex show language
14305 Display the current working language. This is the
14306 language you can use with commands such as @code{print} to
14307 build and compute expressions that may involve variables in your program.
14310 @kindex info frame@r{, show the source language}
14311 Display the source language for this frame. This language becomes the
14312 working language if you use an identifier from this frame.
14313 @xref{Frame Info, ,Information about a Frame}, to identify the other
14314 information listed here.
14317 @kindex info source@r{, show the source language}
14318 Display the source language of this source file.
14319 @xref{Symbols, ,Examining the Symbol Table}, to identify the other
14320 information listed here.
14323 In unusual circumstances, you may have source files with extensions
14324 not in the standard list. You can then set the extension associated
14325 with a language explicitly:
14328 @item set extension-language @var{ext} @var{language}
14329 @kindex set extension-language
14330 Tell @value{GDBN} that source files with extension @var{ext} are to be
14331 assumed as written in the source language @var{language}.
14333 @item info extensions
14334 @kindex info extensions
14335 List all the filename extensions and the associated languages.
14339 @section Type and Range Checking
14341 Some languages are designed to guard you against making seemingly common
14342 errors through a series of compile- and run-time checks. These include
14343 checking the type of arguments to functions and operators and making
14344 sure mathematical overflows are caught at run time. Checks such as
14345 these help to ensure a program's correctness once it has been compiled
14346 by eliminating type mismatches and providing active checks for range
14347 errors when your program is running.
14349 By default @value{GDBN} checks for these errors according to the
14350 rules of the current source language. Although @value{GDBN} does not check
14351 the statements in your program, it can check expressions entered directly
14352 into @value{GDBN} for evaluation via the @code{print} command, for example.
14355 * Type Checking:: An overview of type checking
14356 * Range Checking:: An overview of range checking
14359 @cindex type checking
14360 @cindex checks, type
14361 @node Type Checking
14362 @subsection An Overview of Type Checking
14364 Some languages, such as C and C@t{++}, are strongly typed, meaning that the
14365 arguments to operators and functions have to be of the correct type,
14366 otherwise an error occurs. These checks prevent type mismatch
14367 errors from ever causing any run-time problems. For example,
14370 int klass::my_method(char *b) @{ return b ? 1 : 2; @}
14372 (@value{GDBP}) print obj.my_method (0)
14375 (@value{GDBP}) print obj.my_method (0x1234)
14376 Cannot resolve method klass::my_method to any overloaded instance
14379 The second example fails because in C@t{++} the integer constant
14380 @samp{0x1234} is not type-compatible with the pointer parameter type.
14382 For the expressions you use in @value{GDBN} commands, you can tell
14383 @value{GDBN} to not enforce strict type checking or
14384 to treat any mismatches as errors and abandon the expression;
14385 When type checking is disabled, @value{GDBN} successfully evaluates
14386 expressions like the second example above.
14388 Even if type checking is off, there may be other reasons
14389 related to type that prevent @value{GDBN} from evaluating an expression.
14390 For instance, @value{GDBN} does not know how to add an @code{int} and
14391 a @code{struct foo}. These particular type errors have nothing to do
14392 with the language in use and usually arise from expressions which make
14393 little sense to evaluate anyway.
14395 @value{GDBN} provides some additional commands for controlling type checking:
14397 @kindex set check type
14398 @kindex show check type
14400 @item set check type on
14401 @itemx set check type off
14402 Set strict type checking on or off. If any type mismatches occur in
14403 evaluating an expression while type checking is on, @value{GDBN} prints a
14404 message and aborts evaluation of the expression.
14406 @item show check type
14407 Show the current setting of type checking and whether @value{GDBN}
14408 is enforcing strict type checking rules.
14411 @cindex range checking
14412 @cindex checks, range
14413 @node Range Checking
14414 @subsection An Overview of Range Checking
14416 In some languages (such as Modula-2), it is an error to exceed the
14417 bounds of a type; this is enforced with run-time checks. Such range
14418 checking is meant to ensure program correctness by making sure
14419 computations do not overflow, or indices on an array element access do
14420 not exceed the bounds of the array.
14422 For expressions you use in @value{GDBN} commands, you can tell
14423 @value{GDBN} to treat range errors in one of three ways: ignore them,
14424 always treat them as errors and abandon the expression, or issue
14425 warnings but evaluate the expression anyway.
14427 A range error can result from numerical overflow, from exceeding an
14428 array index bound, or when you type a constant that is not a member
14429 of any type. Some languages, however, do not treat overflows as an
14430 error. In many implementations of C, mathematical overflow causes the
14431 result to ``wrap around'' to lower values---for example, if @var{m} is
14432 the largest integer value, and @var{s} is the smallest, then
14435 @var{m} + 1 @result{} @var{s}
14438 This, too, is specific to individual languages, and in some cases
14439 specific to individual compilers or machines. @xref{Supported Languages, ,
14440 Supported Languages}, for further details on specific languages.
14442 @value{GDBN} provides some additional commands for controlling the range checker:
14444 @kindex set check range
14445 @kindex show check range
14447 @item set check range auto
14448 Set range checking on or off based on the current working language.
14449 @xref{Supported Languages, ,Supported Languages}, for the default settings for
14452 @item set check range on
14453 @itemx set check range off
14454 Set range checking on or off, overriding the default setting for the
14455 current working language. A warning is issued if the setting does not
14456 match the language default. If a range error occurs and range checking is on,
14457 then a message is printed and evaluation of the expression is aborted.
14459 @item set check range warn
14460 Output messages when the @value{GDBN} range checker detects a range error,
14461 but attempt to evaluate the expression anyway. Evaluating the
14462 expression may still be impossible for other reasons, such as accessing
14463 memory that the process does not own (a typical example from many Unix
14467 Show the current setting of the range checker, and whether or not it is
14468 being set automatically by @value{GDBN}.
14471 @node Supported Languages
14472 @section Supported Languages
14474 @value{GDBN} supports C, C@t{++}, D, Go, Objective-C, Fortran,
14475 OpenCL C, Pascal, Rust, assembly, Modula-2, and Ada.
14476 @c This is false ...
14477 Some @value{GDBN} features may be used in expressions regardless of the
14478 language you use: the @value{GDBN} @code{@@} and @code{::} operators,
14479 and the @samp{@{type@}addr} construct (@pxref{Expressions,
14480 ,Expressions}) can be used with the constructs of any supported
14483 The following sections detail to what degree each source language is
14484 supported by @value{GDBN}. These sections are not meant to be language
14485 tutorials or references, but serve only as a reference guide to what the
14486 @value{GDBN} expression parser accepts, and what input and output
14487 formats should look like for different languages. There are many good
14488 books written on each of these languages; please look to these for a
14489 language reference or tutorial.
14492 * C:: C and C@t{++}
14495 * Objective-C:: Objective-C
14496 * OpenCL C:: OpenCL C
14497 * Fortran:: Fortran
14500 * Modula-2:: Modula-2
14505 @subsection C and C@t{++}
14507 @cindex C and C@t{++}
14508 @cindex expressions in C or C@t{++}
14510 Since C and C@t{++} are so closely related, many features of @value{GDBN} apply
14511 to both languages. Whenever this is the case, we discuss those languages
14515 @cindex @code{g++}, @sc{gnu} C@t{++} compiler
14516 @cindex @sc{gnu} C@t{++}
14517 The C@t{++} debugging facilities are jointly implemented by the C@t{++}
14518 compiler and @value{GDBN}. Therefore, to debug your C@t{++} code
14519 effectively, you must compile your C@t{++} programs with a supported
14520 C@t{++} compiler, such as @sc{gnu} @code{g++}, or the HP ANSI C@t{++}
14521 compiler (@code{aCC}).
14524 * C Operators:: C and C@t{++} operators
14525 * C Constants:: C and C@t{++} constants
14526 * C Plus Plus Expressions:: C@t{++} expressions
14527 * C Defaults:: Default settings for C and C@t{++}
14528 * C Checks:: C and C@t{++} type and range checks
14529 * Debugging C:: @value{GDBN} and C
14530 * Debugging C Plus Plus:: @value{GDBN} features for C@t{++}
14531 * Decimal Floating Point:: Numbers in Decimal Floating Point format
14535 @subsubsection C and C@t{++} Operators
14537 @cindex C and C@t{++} operators
14539 Operators must be defined on values of specific types. For instance,
14540 @code{+} is defined on numbers, but not on structures. Operators are
14541 often defined on groups of types.
14543 For the purposes of C and C@t{++}, the following definitions hold:
14548 @emph{Integral types} include @code{int} with any of its storage-class
14549 specifiers; @code{char}; @code{enum}; and, for C@t{++}, @code{bool}.
14552 @emph{Floating-point types} include @code{float}, @code{double}, and
14553 @code{long double} (if supported by the target platform).
14556 @emph{Pointer types} include all types defined as @code{(@var{type} *)}.
14559 @emph{Scalar types} include all of the above.
14564 The following operators are supported. They are listed here
14565 in order of increasing precedence:
14569 The comma or sequencing operator. Expressions in a comma-separated list
14570 are evaluated from left to right, with the result of the entire
14571 expression being the last expression evaluated.
14574 Assignment. The value of an assignment expression is the value
14575 assigned. Defined on scalar types.
14578 Used in an expression of the form @w{@code{@var{a} @var{op}= @var{b}}},
14579 and translated to @w{@code{@var{a} = @var{a op b}}}.
14580 @w{@code{@var{op}=}} and @code{=} have the same precedence. The operator
14581 @var{op} is any one of the operators @code{|}, @code{^}, @code{&},
14582 @code{<<}, @code{>>}, @code{+}, @code{-}, @code{*}, @code{/}, @code{%}.
14585 The ternary operator. @code{@var{a} ? @var{b} : @var{c}} can be thought
14586 of as: if @var{a} then @var{b} else @var{c}. The argument @var{a}
14587 should be of an integral type.
14590 Logical @sc{or}. Defined on integral types.
14593 Logical @sc{and}. Defined on integral types.
14596 Bitwise @sc{or}. Defined on integral types.
14599 Bitwise exclusive-@sc{or}. Defined on integral types.
14602 Bitwise @sc{and}. Defined on integral types.
14605 Equality and inequality. Defined on scalar types. The value of these
14606 expressions is 0 for false and non-zero for true.
14608 @item <@r{, }>@r{, }<=@r{, }>=
14609 Less than, greater than, less than or equal, greater than or equal.
14610 Defined on scalar types. The value of these expressions is 0 for false
14611 and non-zero for true.
14614 left shift, and right shift. Defined on integral types.
14617 The @value{GDBN} ``artificial array'' operator (@pxref{Expressions, ,Expressions}).
14620 Addition and subtraction. Defined on integral types, floating-point types and
14623 @item *@r{, }/@r{, }%
14624 Multiplication, division, and modulus. Multiplication and division are
14625 defined on integral and floating-point types. Modulus is defined on
14629 Increment and decrement. When appearing before a variable, the
14630 operation is performed before the variable is used in an expression;
14631 when appearing after it, the variable's value is used before the
14632 operation takes place.
14635 Pointer dereferencing. Defined on pointer types. Same precedence as
14639 Address operator. Defined on variables. Same precedence as @code{++}.
14641 For debugging C@t{++}, @value{GDBN} implements a use of @samp{&} beyond what is
14642 allowed in the C@t{++} language itself: you can use @samp{&(&@var{ref})}
14643 to examine the address
14644 where a C@t{++} reference variable (declared with @samp{&@var{ref}}) is
14648 Negative. Defined on integral and floating-point types. Same
14649 precedence as @code{++}.
14652 Logical negation. Defined on integral types. Same precedence as
14656 Bitwise complement operator. Defined on integral types. Same precedence as
14661 Structure member, and pointer-to-structure member. For convenience,
14662 @value{GDBN} regards the two as equivalent, choosing whether to dereference a
14663 pointer based on the stored type information.
14664 Defined on @code{struct} and @code{union} data.
14667 Dereferences of pointers to members.
14670 Array indexing. @code{@var{a}[@var{i}]} is defined as
14671 @code{*(@var{a}+@var{i})}. Same precedence as @code{->}.
14674 Function parameter list. Same precedence as @code{->}.
14677 C@t{++} scope resolution operator. Defined on @code{struct}, @code{union},
14678 and @code{class} types.
14681 Doubled colons also represent the @value{GDBN} scope operator
14682 (@pxref{Expressions, ,Expressions}). Same precedence as @code{::},
14686 If an operator is redefined in the user code, @value{GDBN} usually
14687 attempts to invoke the redefined version instead of using the operator's
14688 predefined meaning.
14691 @subsubsection C and C@t{++} Constants
14693 @cindex C and C@t{++} constants
14695 @value{GDBN} allows you to express the constants of C and C@t{++} in the
14700 Integer constants are a sequence of digits. Octal constants are
14701 specified by a leading @samp{0} (i.e.@: zero), and hexadecimal constants
14702 by a leading @samp{0x} or @samp{0X}. Constants may also end with a letter
14703 @samp{l}, specifying that the constant should be treated as a
14707 Floating point constants are a sequence of digits, followed by a decimal
14708 point, followed by a sequence of digits, and optionally followed by an
14709 exponent. An exponent is of the form:
14710 @samp{@w{e@r{[[}+@r{]|}-@r{]}@var{nnn}}}, where @var{nnn} is another
14711 sequence of digits. The @samp{+} is optional for positive exponents.
14712 A floating-point constant may also end with a letter @samp{f} or
14713 @samp{F}, specifying that the constant should be treated as being of
14714 the @code{float} (as opposed to the default @code{double}) type; or with
14715 a letter @samp{l} or @samp{L}, which specifies a @code{long double}
14719 Enumerated constants consist of enumerated identifiers, or their
14720 integral equivalents.
14723 Character constants are a single character surrounded by single quotes
14724 (@code{'}), or a number---the ordinal value of the corresponding character
14725 (usually its @sc{ascii} value). Within quotes, the single character may
14726 be represented by a letter or by @dfn{escape sequences}, which are of
14727 the form @samp{\@var{nnn}}, where @var{nnn} is the octal representation
14728 of the character's ordinal value; or of the form @samp{\@var{x}}, where
14729 @samp{@var{x}} is a predefined special character---for example,
14730 @samp{\n} for newline.
14732 Wide character constants can be written by prefixing a character
14733 constant with @samp{L}, as in C. For example, @samp{L'x'} is the wide
14734 form of @samp{x}. The target wide character set is used when
14735 computing the value of this constant (@pxref{Character Sets}).
14738 String constants are a sequence of character constants surrounded by
14739 double quotes (@code{"}). Any valid character constant (as described
14740 above) may appear. Double quotes within the string must be preceded by
14741 a backslash, so for instance @samp{"a\"b'c"} is a string of five
14744 Wide string constants can be written by prefixing a string constant
14745 with @samp{L}, as in C. The target wide character set is used when
14746 computing the value of this constant (@pxref{Character Sets}).
14749 Pointer constants are an integral value. You can also write pointers
14750 to constants using the C operator @samp{&}.
14753 Array constants are comma-separated lists surrounded by braces @samp{@{}
14754 and @samp{@}}; for example, @samp{@{1,2,3@}} is a three-element array of
14755 integers, @samp{@{@{1,2@}, @{3,4@}, @{5,6@}@}} is a three-by-two array,
14756 and @samp{@{&"hi", &"there", &"fred"@}} is a three-element array of pointers.
14759 @node C Plus Plus Expressions
14760 @subsubsection C@t{++} Expressions
14762 @cindex expressions in C@t{++}
14763 @value{GDBN} expression handling can interpret most C@t{++} expressions.
14765 @cindex debugging C@t{++} programs
14766 @cindex C@t{++} compilers
14767 @cindex debug formats and C@t{++}
14768 @cindex @value{NGCC} and C@t{++}
14770 @emph{Warning:} @value{GDBN} can only debug C@t{++} code if you use
14771 the proper compiler and the proper debug format. Currently,
14772 @value{GDBN} works best when debugging C@t{++} code that is compiled
14773 with the most recent version of @value{NGCC} possible. The DWARF
14774 debugging format is preferred; @value{NGCC} defaults to this on most
14775 popular platforms. Other compilers and/or debug formats are likely to
14776 work badly or not at all when using @value{GDBN} to debug C@t{++}
14777 code. @xref{Compilation}.
14782 @cindex member functions
14784 Member function calls are allowed; you can use expressions like
14787 count = aml->GetOriginal(x, y)
14790 @vindex this@r{, inside C@t{++} member functions}
14791 @cindex namespace in C@t{++}
14793 While a member function is active (in the selected stack frame), your
14794 expressions have the same namespace available as the member function;
14795 that is, @value{GDBN} allows implicit references to the class instance
14796 pointer @code{this} following the same rules as C@t{++}. @code{using}
14797 declarations in the current scope are also respected by @value{GDBN}.
14799 @cindex call overloaded functions
14800 @cindex overloaded functions, calling
14801 @cindex type conversions in C@t{++}
14803 You can call overloaded functions; @value{GDBN} resolves the function
14804 call to the right definition, with some restrictions. @value{GDBN} does not
14805 perform overload resolution involving user-defined type conversions,
14806 calls to constructors, or instantiations of templates that do not exist
14807 in the program. It also cannot handle ellipsis argument lists or
14810 It does perform integral conversions and promotions, floating-point
14811 promotions, arithmetic conversions, pointer conversions, conversions of
14812 class objects to base classes, and standard conversions such as those of
14813 functions or arrays to pointers; it requires an exact match on the
14814 number of function arguments.
14816 Overload resolution is always performed, unless you have specified
14817 @code{set overload-resolution off}. @xref{Debugging C Plus Plus,
14818 ,@value{GDBN} Features for C@t{++}}.
14820 You must specify @code{set overload-resolution off} in order to use an
14821 explicit function signature to call an overloaded function, as in
14823 p 'foo(char,int)'('x', 13)
14826 The @value{GDBN} command-completion facility can simplify this;
14827 see @ref{Completion, ,Command Completion}.
14829 @cindex reference declarations
14831 @value{GDBN} understands variables declared as C@t{++} lvalue or rvalue
14832 references; you can use them in expressions just as you do in C@t{++}
14833 source---they are automatically dereferenced.
14835 In the parameter list shown when @value{GDBN} displays a frame, the values of
14836 reference variables are not displayed (unlike other variables); this
14837 avoids clutter, since references are often used for large structures.
14838 The @emph{address} of a reference variable is always shown, unless
14839 you have specified @samp{set print address off}.
14842 @value{GDBN} supports the C@t{++} name resolution operator @code{::}---your
14843 expressions can use it just as expressions in your program do. Since
14844 one scope may be defined in another, you can use @code{::} repeatedly if
14845 necessary, for example in an expression like
14846 @samp{@var{scope1}::@var{scope2}::@var{name}}. @value{GDBN} also allows
14847 resolving name scope by reference to source files, in both C and C@t{++}
14848 debugging (@pxref{Variables, ,Program Variables}).
14851 @value{GDBN} performs argument-dependent lookup, following the C@t{++}
14856 @subsubsection C and C@t{++} Defaults
14858 @cindex C and C@t{++} defaults
14860 If you allow @value{GDBN} to set range checking automatically, it
14861 defaults to @code{off} whenever the working language changes to
14862 C or C@t{++}. This happens regardless of whether you or @value{GDBN}
14863 selects the working language.
14865 If you allow @value{GDBN} to set the language automatically, it
14866 recognizes source files whose names end with @file{.c}, @file{.C}, or
14867 @file{.cc}, etc, and when @value{GDBN} enters code compiled from one of
14868 these files, it sets the working language to C or C@t{++}.
14869 @xref{Automatically, ,Having @value{GDBN} Infer the Source Language},
14870 for further details.
14873 @subsubsection C and C@t{++} Type and Range Checks
14875 @cindex C and C@t{++} checks
14877 By default, when @value{GDBN} parses C or C@t{++} expressions, strict type
14878 checking is used. However, if you turn type checking off, @value{GDBN}
14879 will allow certain non-standard conversions, such as promoting integer
14880 constants to pointers.
14882 Range checking, if turned on, is done on mathematical operations. Array
14883 indices are not checked, since they are often used to index a pointer
14884 that is not itself an array.
14887 @subsubsection @value{GDBN} and C
14889 The @code{set print union} and @code{show print union} commands apply to
14890 the @code{union} type. When set to @samp{on}, any @code{union} that is
14891 inside a @code{struct} or @code{class} is also printed. Otherwise, it
14892 appears as @samp{@{...@}}.
14894 The @code{@@} operator aids in the debugging of dynamic arrays, formed
14895 with pointers and a memory allocation function. @xref{Expressions,
14898 @node Debugging C Plus Plus
14899 @subsubsection @value{GDBN} Features for C@t{++}
14901 @cindex commands for C@t{++}
14903 Some @value{GDBN} commands are particularly useful with C@t{++}, and some are
14904 designed specifically for use with C@t{++}. Here is a summary:
14907 @cindex break in overloaded functions
14908 @item @r{breakpoint menus}
14909 When you want a breakpoint in a function whose name is overloaded,
14910 @value{GDBN} has the capability to display a menu of possible breakpoint
14911 locations to help you specify which function definition you want.
14912 @xref{Ambiguous Expressions,,Ambiguous Expressions}.
14914 @cindex overloading in C@t{++}
14915 @item rbreak @var{regex}
14916 Setting breakpoints using regular expressions is helpful for setting
14917 breakpoints on overloaded functions that are not members of any special
14919 @xref{Set Breaks, ,Setting Breakpoints}.
14921 @cindex C@t{++} exception handling
14923 @itemx catch rethrow
14925 Debug C@t{++} exception handling using these commands. @xref{Set
14926 Catchpoints, , Setting Catchpoints}.
14928 @cindex inheritance
14929 @item ptype @var{typename}
14930 Print inheritance relationships as well as other information for type
14932 @xref{Symbols, ,Examining the Symbol Table}.
14934 @item info vtbl @var{expression}.
14935 The @code{info vtbl} command can be used to display the virtual
14936 method tables of the object computed by @var{expression}. This shows
14937 one entry per virtual table; there may be multiple virtual tables when
14938 multiple inheritance is in use.
14940 @cindex C@t{++} demangling
14941 @item demangle @var{name}
14942 Demangle @var{name}.
14943 @xref{Symbols}, for a more complete description of the @code{demangle} command.
14945 @cindex C@t{++} symbol display
14946 @item set print demangle
14947 @itemx show print demangle
14948 @itemx set print asm-demangle
14949 @itemx show print asm-demangle
14950 Control whether C@t{++} symbols display in their source form, both when
14951 displaying code as C@t{++} source and when displaying disassemblies.
14952 @xref{Print Settings, ,Print Settings}.
14954 @item set print object
14955 @itemx show print object
14956 Choose whether to print derived (actual) or declared types of objects.
14957 @xref{Print Settings, ,Print Settings}.
14959 @item set print vtbl
14960 @itemx show print vtbl
14961 Control the format for printing virtual function tables.
14962 @xref{Print Settings, ,Print Settings}.
14963 (The @code{vtbl} commands do not work on programs compiled with the HP
14964 ANSI C@t{++} compiler (@code{aCC}).)
14966 @kindex set overload-resolution
14967 @cindex overloaded functions, overload resolution
14968 @item set overload-resolution on
14969 Enable overload resolution for C@t{++} expression evaluation. The default
14970 is on. For overloaded functions, @value{GDBN} evaluates the arguments
14971 and searches for a function whose signature matches the argument types,
14972 using the standard C@t{++} conversion rules (see @ref{C Plus Plus
14973 Expressions, ,C@t{++} Expressions}, for details).
14974 If it cannot find a match, it emits a message.
14976 @item set overload-resolution off
14977 Disable overload resolution for C@t{++} expression evaluation. For
14978 overloaded functions that are not class member functions, @value{GDBN}
14979 chooses the first function of the specified name that it finds in the
14980 symbol table, whether or not its arguments are of the correct type. For
14981 overloaded functions that are class member functions, @value{GDBN}
14982 searches for a function whose signature @emph{exactly} matches the
14985 @kindex show overload-resolution
14986 @item show overload-resolution
14987 Show the current setting of overload resolution.
14989 @item @r{Overloaded symbol names}
14990 You can specify a particular definition of an overloaded symbol, using
14991 the same notation that is used to declare such symbols in C@t{++}: type
14992 @code{@var{symbol}(@var{types})} rather than just @var{symbol}. You can
14993 also use the @value{GDBN} command-line word completion facilities to list the
14994 available choices, or to finish the type list for you.
14995 @xref{Completion,, Command Completion}, for details on how to do this.
14998 @node Decimal Floating Point
14999 @subsubsection Decimal Floating Point format
15000 @cindex decimal floating point format
15002 @value{GDBN} can examine, set and perform computations with numbers in
15003 decimal floating point format, which in the C language correspond to the
15004 @code{_Decimal32}, @code{_Decimal64} and @code{_Decimal128} types as
15005 specified by the extension to support decimal floating-point arithmetic.
15007 There are two encodings in use, depending on the architecture: BID (Binary
15008 Integer Decimal) for x86 and x86-64, and DPD (Densely Packed Decimal) for
15009 PowerPC and S/390. @value{GDBN} will use the appropriate encoding for the
15012 Because of a limitation in @file{libdecnumber}, the library used by @value{GDBN}
15013 to manipulate decimal floating point numbers, it is not possible to convert
15014 (using a cast, for example) integers wider than 32-bit to decimal float.
15016 In addition, in order to imitate @value{GDBN}'s behaviour with binary floating
15017 point computations, error checking in decimal float operations ignores
15018 underflow, overflow and divide by zero exceptions.
15020 In the PowerPC architecture, @value{GDBN} provides a set of pseudo-registers
15021 to inspect @code{_Decimal128} values stored in floating point registers.
15022 See @ref{PowerPC,,PowerPC} for more details.
15028 @value{GDBN} can be used to debug programs written in D and compiled with
15029 GDC, LDC or DMD compilers. Currently @value{GDBN} supports only one D
15030 specific feature --- dynamic arrays.
15035 @cindex Go (programming language)
15036 @value{GDBN} can be used to debug programs written in Go and compiled with
15037 @file{gccgo} or @file{6g} compilers.
15039 Here is a summary of the Go-specific features and restrictions:
15042 @cindex current Go package
15043 @item The current Go package
15044 The name of the current package does not need to be specified when
15045 specifying global variables and functions.
15047 For example, given the program:
15051 var myglob = "Shall we?"
15057 When stopped inside @code{main} either of these work:
15061 (gdb) p main.myglob
15064 @cindex builtin Go types
15065 @item Builtin Go types
15066 The @code{string} type is recognized by @value{GDBN} and is printed
15069 @cindex builtin Go functions
15070 @item Builtin Go functions
15071 The @value{GDBN} expression parser recognizes the @code{unsafe.Sizeof}
15072 function and handles it internally.
15074 @cindex restrictions on Go expressions
15075 @item Restrictions on Go expressions
15076 All Go operators are supported except @code{&^}.
15077 The Go @code{_} ``blank identifier'' is not supported.
15078 Automatic dereferencing of pointers is not supported.
15082 @subsection Objective-C
15084 @cindex Objective-C
15085 This section provides information about some commands and command
15086 options that are useful for debugging Objective-C code. See also
15087 @ref{Symbols, info classes}, and @ref{Symbols, info selectors}, for a
15088 few more commands specific to Objective-C support.
15091 * Method Names in Commands::
15092 * The Print Command with Objective-C::
15095 @node Method Names in Commands
15096 @subsubsection Method Names in Commands
15098 The following commands have been extended to accept Objective-C method
15099 names as line specifications:
15101 @kindex clear@r{, and Objective-C}
15102 @kindex break@r{, and Objective-C}
15103 @kindex info line@r{, and Objective-C}
15104 @kindex jump@r{, and Objective-C}
15105 @kindex list@r{, and Objective-C}
15109 @item @code{info line}
15114 A fully qualified Objective-C method name is specified as
15117 -[@var{Class} @var{methodName}]
15120 where the minus sign is used to indicate an instance method and a
15121 plus sign (not shown) is used to indicate a class method. The class
15122 name @var{Class} and method name @var{methodName} are enclosed in
15123 brackets, similar to the way messages are specified in Objective-C
15124 source code. For example, to set a breakpoint at the @code{create}
15125 instance method of class @code{Fruit} in the program currently being
15129 break -[Fruit create]
15132 To list ten program lines around the @code{initialize} class method,
15136 list +[NSText initialize]
15139 In the current version of @value{GDBN}, the plus or minus sign is
15140 required. In future versions of @value{GDBN}, the plus or minus
15141 sign will be optional, but you can use it to narrow the search. It
15142 is also possible to specify just a method name:
15148 You must specify the complete method name, including any colons. If
15149 your program's source files contain more than one @code{create} method,
15150 you'll be presented with a numbered list of classes that implement that
15151 method. Indicate your choice by number, or type @samp{0} to exit if
15154 As another example, to clear a breakpoint established at the
15155 @code{makeKeyAndOrderFront:} method of the @code{NSWindow} class, enter:
15158 clear -[NSWindow makeKeyAndOrderFront:]
15161 @node The Print Command with Objective-C
15162 @subsubsection The Print Command With Objective-C
15163 @cindex Objective-C, print objects
15164 @kindex print-object
15165 @kindex po @r{(@code{print-object})}
15167 The print command has also been extended to accept methods. For example:
15170 print -[@var{object} hash]
15173 @cindex print an Objective-C object description
15174 @cindex @code{_NSPrintForDebugger}, and printing Objective-C objects
15176 will tell @value{GDBN} to send the @code{hash} message to @var{object}
15177 and print the result. Also, an additional command has been added,
15178 @code{print-object} or @code{po} for short, which is meant to print
15179 the description of an object. However, this command may only work
15180 with certain Objective-C libraries that have a particular hook
15181 function, @code{_NSPrintForDebugger}, defined.
15184 @subsection OpenCL C
15187 This section provides information about @value{GDBN}s OpenCL C support.
15190 * OpenCL C Datatypes::
15191 * OpenCL C Expressions::
15192 * OpenCL C Operators::
15195 @node OpenCL C Datatypes
15196 @subsubsection OpenCL C Datatypes
15198 @cindex OpenCL C Datatypes
15199 @value{GDBN} supports the builtin scalar and vector datatypes specified
15200 by OpenCL 1.1. In addition the half- and double-precision floating point
15201 data types of the @code{cl_khr_fp16} and @code{cl_khr_fp64} OpenCL
15202 extensions are also known to @value{GDBN}.
15204 @node OpenCL C Expressions
15205 @subsubsection OpenCL C Expressions
15207 @cindex OpenCL C Expressions
15208 @value{GDBN} supports accesses to vector components including the access as
15209 lvalue where possible. Since OpenCL C is based on C99 most C expressions
15210 supported by @value{GDBN} can be used as well.
15212 @node OpenCL C Operators
15213 @subsubsection OpenCL C Operators
15215 @cindex OpenCL C Operators
15216 @value{GDBN} supports the operators specified by OpenCL 1.1 for scalar and
15220 @subsection Fortran
15221 @cindex Fortran-specific support in @value{GDBN}
15223 @value{GDBN} can be used to debug programs written in Fortran, but it
15224 currently supports only the features of Fortran 77 language.
15226 @cindex trailing underscore, in Fortran symbols
15227 Some Fortran compilers (@sc{gnu} Fortran 77 and Fortran 95 compilers
15228 among them) append an underscore to the names of variables and
15229 functions. When you debug programs compiled by those compilers, you
15230 will need to refer to variables and functions with a trailing
15234 * Fortran Operators:: Fortran operators and expressions
15235 * Fortran Defaults:: Default settings for Fortran
15236 * Special Fortran Commands:: Special @value{GDBN} commands for Fortran
15239 @node Fortran Operators
15240 @subsubsection Fortran Operators and Expressions
15242 @cindex Fortran operators and expressions
15244 Operators must be defined on values of specific types. For instance,
15245 @code{+} is defined on numbers, but not on characters or other non-
15246 arithmetic types. Operators are often defined on groups of types.
15250 The exponentiation operator. It raises the first operand to the power
15254 The range operator. Normally used in the form of array(low:high) to
15255 represent a section of array.
15258 The access component operator. Normally used to access elements in derived
15259 types. Also suitable for unions. As unions aren't part of regular Fortran,
15260 this can only happen when accessing a register that uses a gdbarch-defined
15264 @node Fortran Defaults
15265 @subsubsection Fortran Defaults
15267 @cindex Fortran Defaults
15269 Fortran symbols are usually case-insensitive, so @value{GDBN} by
15270 default uses case-insensitive matches for Fortran symbols. You can
15271 change that with the @samp{set case-insensitive} command, see
15272 @ref{Symbols}, for the details.
15274 @node Special Fortran Commands
15275 @subsubsection Special Fortran Commands
15277 @cindex Special Fortran commands
15279 @value{GDBN} has some commands to support Fortran-specific features,
15280 such as displaying common blocks.
15283 @cindex @code{COMMON} blocks, Fortran
15284 @kindex info common
15285 @item info common @r{[}@var{common-name}@r{]}
15286 This command prints the values contained in the Fortran @code{COMMON}
15287 block whose name is @var{common-name}. With no argument, the names of
15288 all @code{COMMON} blocks visible at the current program location are
15295 @cindex Pascal support in @value{GDBN}, limitations
15296 Debugging Pascal programs which use sets, subranges, file variables, or
15297 nested functions does not currently work. @value{GDBN} does not support
15298 entering expressions, printing values, or similar features using Pascal
15301 The Pascal-specific command @code{set print pascal_static-members}
15302 controls whether static members of Pascal objects are displayed.
15303 @xref{Print Settings, pascal_static-members}.
15308 @value{GDBN} supports the @url{https://www.rust-lang.org/, Rust
15309 Programming Language}. Type- and value-printing, and expression
15310 parsing, are reasonably complete. However, there are a few
15311 peculiarities and holes to be aware of.
15315 Linespecs (@pxref{Specify Location}) are never relative to the current
15316 crate. Instead, they act as if there were a global namespace of
15317 crates, somewhat similar to the way @code{extern crate} behaves.
15319 That is, if @value{GDBN} is stopped at a breakpoint in a function in
15320 crate @samp{A}, module @samp{B}, then @code{break B::f} will attempt
15321 to set a breakpoint in a function named @samp{f} in a crate named
15324 As a consequence of this approach, linespecs also cannot refer to
15325 items using @samp{self::} or @samp{super::}.
15328 Because @value{GDBN} implements Rust name-lookup semantics in
15329 expressions, it will sometimes prepend the current crate to a name.
15330 For example, if @value{GDBN} is stopped at a breakpoint in the crate
15331 @samp{K}, then @code{print ::x::y} will try to find the symbol
15334 However, since it is useful to be able to refer to other crates when
15335 debugging, @value{GDBN} provides the @code{extern} extension to
15336 circumvent this. To use the extension, just put @code{extern} before
15337 a path expression to refer to the otherwise unavailable ``global''
15340 In the above example, if you wanted to refer to the symbol @samp{y} in
15341 the crate @samp{x}, you would use @code{print extern x::y}.
15344 The Rust expression evaluator does not support ``statement-like''
15345 expressions such as @code{if} or @code{match}, or lambda expressions.
15348 Tuple expressions are not implemented.
15351 The Rust expression evaluator does not currently implement the
15352 @code{Drop} trait. Objects that may be created by the evaluator will
15353 never be destroyed.
15356 @value{GDBN} does not implement type inference for generics. In order
15357 to call generic functions or otherwise refer to generic items, you
15358 will have to specify the type parameters manually.
15361 @value{GDBN} currently uses the C@t{++} demangler for Rust. In most
15362 cases this does not cause any problems. However, in an expression
15363 context, completing a generic function name will give syntactically
15364 invalid results. This happens because Rust requires the @samp{::}
15365 operator between the function name and its generic arguments. For
15366 example, @value{GDBN} might provide a completion like
15367 @code{crate::f<u32>}, where the parser would require
15368 @code{crate::f::<u32>}.
15371 As of this writing, the Rust compiler (version 1.8) has a few holes in
15372 the debugging information it generates. These holes prevent certain
15373 features from being implemented by @value{GDBN}:
15377 Method calls cannot be made via traits.
15380 Trait objects cannot be created or inspected.
15383 Operator overloading is not implemented.
15386 When debugging in a monomorphized function, you cannot use the generic
15390 The type @code{Self} is not available.
15393 @code{use} statements are not available, so some names may not be
15394 available in the crate.
15399 @subsection Modula-2
15401 @cindex Modula-2, @value{GDBN} support
15403 The extensions made to @value{GDBN} to support Modula-2 only support
15404 output from the @sc{gnu} Modula-2 compiler (which is currently being
15405 developed). Other Modula-2 compilers are not currently supported, and
15406 attempting to debug executables produced by them is most likely
15407 to give an error as @value{GDBN} reads in the executable's symbol
15410 @cindex expressions in Modula-2
15412 * M2 Operators:: Built-in operators
15413 * Built-In Func/Proc:: Built-in functions and procedures
15414 * M2 Constants:: Modula-2 constants
15415 * M2 Types:: Modula-2 types
15416 * M2 Defaults:: Default settings for Modula-2
15417 * Deviations:: Deviations from standard Modula-2
15418 * M2 Checks:: Modula-2 type and range checks
15419 * M2 Scope:: The scope operators @code{::} and @code{.}
15420 * GDB/M2:: @value{GDBN} and Modula-2
15424 @subsubsection Operators
15425 @cindex Modula-2 operators
15427 Operators must be defined on values of specific types. For instance,
15428 @code{+} is defined on numbers, but not on structures. Operators are
15429 often defined on groups of types. For the purposes of Modula-2, the
15430 following definitions hold:
15435 @emph{Integral types} consist of @code{INTEGER}, @code{CARDINAL}, and
15439 @emph{Character types} consist of @code{CHAR} and its subranges.
15442 @emph{Floating-point types} consist of @code{REAL}.
15445 @emph{Pointer types} consist of anything declared as @code{POINTER TO
15449 @emph{Scalar types} consist of all of the above.
15452 @emph{Set types} consist of @code{SET} and @code{BITSET} types.
15455 @emph{Boolean types} consist of @code{BOOLEAN}.
15459 The following operators are supported, and appear in order of
15460 increasing precedence:
15464 Function argument or array index separator.
15467 Assignment. The value of @var{var} @code{:=} @var{value} is
15471 Less than, greater than on integral, floating-point, or enumerated
15475 Less than or equal to, greater than or equal to
15476 on integral, floating-point and enumerated types, or set inclusion on
15477 set types. Same precedence as @code{<}.
15479 @item =@r{, }<>@r{, }#
15480 Equality and two ways of expressing inequality, valid on scalar types.
15481 Same precedence as @code{<}. In @value{GDBN} scripts, only @code{<>} is
15482 available for inequality, since @code{#} conflicts with the script
15486 Set membership. Defined on set types and the types of their members.
15487 Same precedence as @code{<}.
15490 Boolean disjunction. Defined on boolean types.
15493 Boolean conjunction. Defined on boolean types.
15496 The @value{GDBN} ``artificial array'' operator (@pxref{Expressions, ,Expressions}).
15499 Addition and subtraction on integral and floating-point types, or union
15500 and difference on set types.
15503 Multiplication on integral and floating-point types, or set intersection
15507 Division on floating-point types, or symmetric set difference on set
15508 types. Same precedence as @code{*}.
15511 Integer division and remainder. Defined on integral types. Same
15512 precedence as @code{*}.
15515 Negative. Defined on @code{INTEGER} and @code{REAL} data.
15518 Pointer dereferencing. Defined on pointer types.
15521 Boolean negation. Defined on boolean types. Same precedence as
15525 @code{RECORD} field selector. Defined on @code{RECORD} data. Same
15526 precedence as @code{^}.
15529 Array indexing. Defined on @code{ARRAY} data. Same precedence as @code{^}.
15532 Procedure argument list. Defined on @code{PROCEDURE} objects. Same precedence
15536 @value{GDBN} and Modula-2 scope operators.
15540 @emph{Warning:} Set expressions and their operations are not yet supported, so @value{GDBN}
15541 treats the use of the operator @code{IN}, or the use of operators
15542 @code{+}, @code{-}, @code{*}, @code{/}, @code{=}, , @code{<>}, @code{#},
15543 @code{<=}, and @code{>=} on sets as an error.
15547 @node Built-In Func/Proc
15548 @subsubsection Built-in Functions and Procedures
15549 @cindex Modula-2 built-ins
15551 Modula-2 also makes available several built-in procedures and functions.
15552 In describing these, the following metavariables are used:
15557 represents an @code{ARRAY} variable.
15560 represents a @code{CHAR} constant or variable.
15563 represents a variable or constant of integral type.
15566 represents an identifier that belongs to a set. Generally used in the
15567 same function with the metavariable @var{s}. The type of @var{s} should
15568 be @code{SET OF @var{mtype}} (where @var{mtype} is the type of @var{m}).
15571 represents a variable or constant of integral or floating-point type.
15574 represents a variable or constant of floating-point type.
15580 represents a variable.
15583 represents a variable or constant of one of many types. See the
15584 explanation of the function for details.
15587 All Modula-2 built-in procedures also return a result, described below.
15591 Returns the absolute value of @var{n}.
15594 If @var{c} is a lower case letter, it returns its upper case
15595 equivalent, otherwise it returns its argument.
15598 Returns the character whose ordinal value is @var{i}.
15601 Decrements the value in the variable @var{v} by one. Returns the new value.
15603 @item DEC(@var{v},@var{i})
15604 Decrements the value in the variable @var{v} by @var{i}. Returns the
15607 @item EXCL(@var{m},@var{s})
15608 Removes the element @var{m} from the set @var{s}. Returns the new
15611 @item FLOAT(@var{i})
15612 Returns the floating point equivalent of the integer @var{i}.
15614 @item HIGH(@var{a})
15615 Returns the index of the last member of @var{a}.
15618 Increments the value in the variable @var{v} by one. Returns the new value.
15620 @item INC(@var{v},@var{i})
15621 Increments the value in the variable @var{v} by @var{i}. Returns the
15624 @item INCL(@var{m},@var{s})
15625 Adds the element @var{m} to the set @var{s} if it is not already
15626 there. Returns the new set.
15629 Returns the maximum value of the type @var{t}.
15632 Returns the minimum value of the type @var{t}.
15635 Returns boolean TRUE if @var{i} is an odd number.
15638 Returns the ordinal value of its argument. For example, the ordinal
15639 value of a character is its @sc{ascii} value (on machines supporting
15640 the @sc{ascii} character set). The argument @var{x} must be of an
15641 ordered type, which include integral, character and enumerated types.
15643 @item SIZE(@var{x})
15644 Returns the size of its argument. The argument @var{x} can be a
15645 variable or a type.
15647 @item TRUNC(@var{r})
15648 Returns the integral part of @var{r}.
15650 @item TSIZE(@var{x})
15651 Returns the size of its argument. The argument @var{x} can be a
15652 variable or a type.
15654 @item VAL(@var{t},@var{i})
15655 Returns the member of the type @var{t} whose ordinal value is @var{i}.
15659 @emph{Warning:} Sets and their operations are not yet supported, so
15660 @value{GDBN} treats the use of procedures @code{INCL} and @code{EXCL} as
15664 @cindex Modula-2 constants
15666 @subsubsection Constants
15668 @value{GDBN} allows you to express the constants of Modula-2 in the following
15674 Integer constants are simply a sequence of digits. When used in an
15675 expression, a constant is interpreted to be type-compatible with the
15676 rest of the expression. Hexadecimal integers are specified by a
15677 trailing @samp{H}, and octal integers by a trailing @samp{B}.
15680 Floating point constants appear as a sequence of digits, followed by a
15681 decimal point and another sequence of digits. An optional exponent can
15682 then be specified, in the form @samp{E@r{[}+@r{|}-@r{]}@var{nnn}}, where
15683 @samp{@r{[}+@r{|}-@r{]}@var{nnn}} is the desired exponent. All of the
15684 digits of the floating point constant must be valid decimal (base 10)
15688 Character constants consist of a single character enclosed by a pair of
15689 like quotes, either single (@code{'}) or double (@code{"}). They may
15690 also be expressed by their ordinal value (their @sc{ascii} value, usually)
15691 followed by a @samp{C}.
15694 String constants consist of a sequence of characters enclosed by a
15695 pair of like quotes, either single (@code{'}) or double (@code{"}).
15696 Escape sequences in the style of C are also allowed. @xref{C
15697 Constants, ,C and C@t{++} Constants}, for a brief explanation of escape
15701 Enumerated constants consist of an enumerated identifier.
15704 Boolean constants consist of the identifiers @code{TRUE} and
15708 Pointer constants consist of integral values only.
15711 Set constants are not yet supported.
15715 @subsubsection Modula-2 Types
15716 @cindex Modula-2 types
15718 Currently @value{GDBN} can print the following data types in Modula-2
15719 syntax: array types, record types, set types, pointer types, procedure
15720 types, enumerated types, subrange types and base types. You can also
15721 print the contents of variables declared using these type.
15722 This section gives a number of simple source code examples together with
15723 sample @value{GDBN} sessions.
15725 The first example contains the following section of code:
15734 and you can request @value{GDBN} to interrogate the type and value of
15735 @code{r} and @code{s}.
15738 (@value{GDBP}) print s
15740 (@value{GDBP}) ptype s
15742 (@value{GDBP}) print r
15744 (@value{GDBP}) ptype r
15749 Likewise if your source code declares @code{s} as:
15753 s: SET ['A'..'Z'] ;
15757 then you may query the type of @code{s} by:
15760 (@value{GDBP}) ptype s
15761 type = SET ['A'..'Z']
15765 Note that at present you cannot interactively manipulate set
15766 expressions using the debugger.
15768 The following example shows how you might declare an array in Modula-2
15769 and how you can interact with @value{GDBN} to print its type and contents:
15773 s: ARRAY [-10..10] OF CHAR ;
15777 (@value{GDBP}) ptype s
15778 ARRAY [-10..10] OF CHAR
15781 Note that the array handling is not yet complete and although the type
15782 is printed correctly, expression handling still assumes that all
15783 arrays have a lower bound of zero and not @code{-10} as in the example
15786 Here are some more type related Modula-2 examples:
15790 colour = (blue, red, yellow, green) ;
15791 t = [blue..yellow] ;
15799 The @value{GDBN} interaction shows how you can query the data type
15800 and value of a variable.
15803 (@value{GDBP}) print s
15805 (@value{GDBP}) ptype t
15806 type = [blue..yellow]
15810 In this example a Modula-2 array is declared and its contents
15811 displayed. Observe that the contents are written in the same way as
15812 their @code{C} counterparts.
15816 s: ARRAY [1..5] OF CARDINAL ;
15822 (@value{GDBP}) print s
15823 $1 = @{1, 0, 0, 0, 0@}
15824 (@value{GDBP}) ptype s
15825 type = ARRAY [1..5] OF CARDINAL
15828 The Modula-2 language interface to @value{GDBN} also understands
15829 pointer types as shown in this example:
15833 s: POINTER TO ARRAY [1..5] OF CARDINAL ;
15840 and you can request that @value{GDBN} describes the type of @code{s}.
15843 (@value{GDBP}) ptype s
15844 type = POINTER TO ARRAY [1..5] OF CARDINAL
15847 @value{GDBN} handles compound types as we can see in this example.
15848 Here we combine array types, record types, pointer types and subrange
15859 myarray = ARRAY myrange OF CARDINAL ;
15860 myrange = [-2..2] ;
15862 s: POINTER TO ARRAY myrange OF foo ;
15866 and you can ask @value{GDBN} to describe the type of @code{s} as shown
15870 (@value{GDBP}) ptype s
15871 type = POINTER TO ARRAY [-2..2] OF foo = RECORD
15874 f3 : ARRAY [-2..2] OF CARDINAL;
15879 @subsubsection Modula-2 Defaults
15880 @cindex Modula-2 defaults
15882 If type and range checking are set automatically by @value{GDBN}, they
15883 both default to @code{on} whenever the working language changes to
15884 Modula-2. This happens regardless of whether you or @value{GDBN}
15885 selected the working language.
15887 If you allow @value{GDBN} to set the language automatically, then entering
15888 code compiled from a file whose name ends with @file{.mod} sets the
15889 working language to Modula-2. @xref{Automatically, ,Having @value{GDBN}
15890 Infer the Source Language}, for further details.
15893 @subsubsection Deviations from Standard Modula-2
15894 @cindex Modula-2, deviations from
15896 A few changes have been made to make Modula-2 programs easier to debug.
15897 This is done primarily via loosening its type strictness:
15901 Unlike in standard Modula-2, pointer constants can be formed by
15902 integers. This allows you to modify pointer variables during
15903 debugging. (In standard Modula-2, the actual address contained in a
15904 pointer variable is hidden from you; it can only be modified
15905 through direct assignment to another pointer variable or expression that
15906 returned a pointer.)
15909 C escape sequences can be used in strings and characters to represent
15910 non-printable characters. @value{GDBN} prints out strings with these
15911 escape sequences embedded. Single non-printable characters are
15912 printed using the @samp{CHR(@var{nnn})} format.
15915 The assignment operator (@code{:=}) returns the value of its right-hand
15919 All built-in procedures both modify @emph{and} return their argument.
15923 @subsubsection Modula-2 Type and Range Checks
15924 @cindex Modula-2 checks
15927 @emph{Warning:} in this release, @value{GDBN} does not yet perform type or
15930 @c FIXME remove warning when type/range checks added
15932 @value{GDBN} considers two Modula-2 variables type equivalent if:
15936 They are of types that have been declared equivalent via a @code{TYPE
15937 @var{t1} = @var{t2}} statement
15940 They have been declared on the same line. (Note: This is true of the
15941 @sc{gnu} Modula-2 compiler, but it may not be true of other compilers.)
15944 As long as type checking is enabled, any attempt to combine variables
15945 whose types are not equivalent is an error.
15947 Range checking is done on all mathematical operations, assignment, array
15948 index bounds, and all built-in functions and procedures.
15951 @subsubsection The Scope Operators @code{::} and @code{.}
15953 @cindex @code{.}, Modula-2 scope operator
15954 @cindex colon, doubled as scope operator
15956 @vindex colon-colon@r{, in Modula-2}
15957 @c Info cannot handle :: but TeX can.
15960 @vindex ::@r{, in Modula-2}
15963 There are a few subtle differences between the Modula-2 scope operator
15964 (@code{.}) and the @value{GDBN} scope operator (@code{::}). The two have
15969 @var{module} . @var{id}
15970 @var{scope} :: @var{id}
15974 where @var{scope} is the name of a module or a procedure,
15975 @var{module} the name of a module, and @var{id} is any declared
15976 identifier within your program, except another module.
15978 Using the @code{::} operator makes @value{GDBN} search the scope
15979 specified by @var{scope} for the identifier @var{id}. If it is not
15980 found in the specified scope, then @value{GDBN} searches all scopes
15981 enclosing the one specified by @var{scope}.
15983 Using the @code{.} operator makes @value{GDBN} search the current scope for
15984 the identifier specified by @var{id} that was imported from the
15985 definition module specified by @var{module}. With this operator, it is
15986 an error if the identifier @var{id} was not imported from definition
15987 module @var{module}, or if @var{id} is not an identifier in
15991 @subsubsection @value{GDBN} and Modula-2
15993 Some @value{GDBN} commands have little use when debugging Modula-2 programs.
15994 Five subcommands of @code{set print} and @code{show print} apply
15995 specifically to C and C@t{++}: @samp{vtbl}, @samp{demangle},
15996 @samp{asm-demangle}, @samp{object}, and @samp{union}. The first four
15997 apply to C@t{++}, and the last to the C @code{union} type, which has no direct
15998 analogue in Modula-2.
16000 The @code{@@} operator (@pxref{Expressions, ,Expressions}), while available
16001 with any language, is not useful with Modula-2. Its
16002 intent is to aid the debugging of @dfn{dynamic arrays}, which cannot be
16003 created in Modula-2 as they can in C or C@t{++}. However, because an
16004 address can be specified by an integral constant, the construct
16005 @samp{@{@var{type}@}@var{adrexp}} is still useful.
16007 @cindex @code{#} in Modula-2
16008 In @value{GDBN} scripts, the Modula-2 inequality operator @code{#} is
16009 interpreted as the beginning of a comment. Use @code{<>} instead.
16015 The extensions made to @value{GDBN} for Ada only support
16016 output from the @sc{gnu} Ada (GNAT) compiler.
16017 Other Ada compilers are not currently supported, and
16018 attempting to debug executables produced by them is most likely
16022 @cindex expressions in Ada
16024 * Ada Mode Intro:: General remarks on the Ada syntax
16025 and semantics supported by Ada mode
16027 * Omissions from Ada:: Restrictions on the Ada expression syntax.
16028 * Additions to Ada:: Extensions of the Ada expression syntax.
16029 * Overloading support for Ada:: Support for expressions involving overloaded
16031 * Stopping Before Main Program:: Debugging the program during elaboration.
16032 * Ada Exceptions:: Ada Exceptions
16033 * Ada Tasks:: Listing and setting breakpoints in tasks.
16034 * Ada Tasks and Core Files:: Tasking Support when Debugging Core Files
16035 * Ravenscar Profile:: Tasking Support when using the Ravenscar
16037 * Ada Glitches:: Known peculiarities of Ada mode.
16040 @node Ada Mode Intro
16041 @subsubsection Introduction
16042 @cindex Ada mode, general
16044 The Ada mode of @value{GDBN} supports a fairly large subset of Ada expression
16045 syntax, with some extensions.
16046 The philosophy behind the design of this subset is
16050 That @value{GDBN} should provide basic literals and access to operations for
16051 arithmetic, dereferencing, field selection, indexing, and subprogram calls,
16052 leaving more sophisticated computations to subprograms written into the
16053 program (which therefore may be called from @value{GDBN}).
16056 That type safety and strict adherence to Ada language restrictions
16057 are not particularly important to the @value{GDBN} user.
16060 That brevity is important to the @value{GDBN} user.
16063 Thus, for brevity, the debugger acts as if all names declared in
16064 user-written packages are directly visible, even if they are not visible
16065 according to Ada rules, thus making it unnecessary to fully qualify most
16066 names with their packages, regardless of context. Where this causes
16067 ambiguity, @value{GDBN} asks the user's intent.
16069 The debugger will start in Ada mode if it detects an Ada main program.
16070 As for other languages, it will enter Ada mode when stopped in a program that
16071 was translated from an Ada source file.
16073 While in Ada mode, you may use `@t{--}' for comments. This is useful
16074 mostly for documenting command files. The standard @value{GDBN} comment
16075 (@samp{#}) still works at the beginning of a line in Ada mode, but not in the
16076 middle (to allow based literals).
16078 @node Omissions from Ada
16079 @subsubsection Omissions from Ada
16080 @cindex Ada, omissions from
16082 Here are the notable omissions from the subset:
16086 Only a subset of the attributes are supported:
16090 @t{'First}, @t{'Last}, and @t{'Length}
16091 on array objects (not on types and subtypes).
16094 @t{'Min} and @t{'Max}.
16097 @t{'Pos} and @t{'Val}.
16103 @t{'Range} on array objects (not subtypes), but only as the right
16104 operand of the membership (@code{in}) operator.
16107 @t{'Access}, @t{'Unchecked_Access}, and
16108 @t{'Unrestricted_Access} (a GNAT extension).
16116 @code{Characters.Latin_1} are not available and
16117 concatenation is not implemented. Thus, escape characters in strings are
16118 not currently available.
16121 Equality tests (@samp{=} and @samp{/=}) on arrays test for bitwise
16122 equality of representations. They will generally work correctly
16123 for strings and arrays whose elements have integer or enumeration types.
16124 They may not work correctly for arrays whose element
16125 types have user-defined equality, for arrays of real values
16126 (in particular, IEEE-conformant floating point, because of negative
16127 zeroes and NaNs), and for arrays whose elements contain unused bits with
16128 indeterminate values.
16131 The other component-by-component array operations (@code{and}, @code{or},
16132 @code{xor}, @code{not}, and relational tests other than equality)
16133 are not implemented.
16136 @cindex array aggregates (Ada)
16137 @cindex record aggregates (Ada)
16138 @cindex aggregates (Ada)
16139 There is limited support for array and record aggregates. They are
16140 permitted only on the right sides of assignments, as in these examples:
16143 (@value{GDBP}) set An_Array := (1, 2, 3, 4, 5, 6)
16144 (@value{GDBP}) set An_Array := (1, others => 0)
16145 (@value{GDBP}) set An_Array := (0|4 => 1, 1..3 => 2, 5 => 6)
16146 (@value{GDBP}) set A_2D_Array := ((1, 2, 3), (4, 5, 6), (7, 8, 9))
16147 (@value{GDBP}) set A_Record := (1, "Peter", True);
16148 (@value{GDBP}) set A_Record := (Name => "Peter", Id => 1, Alive => True)
16152 discriminant's value by assigning an aggregate has an
16153 undefined effect if that discriminant is used within the record.
16154 However, you can first modify discriminants by directly assigning to
16155 them (which normally would not be allowed in Ada), and then performing an
16156 aggregate assignment. For example, given a variable @code{A_Rec}
16157 declared to have a type such as:
16160 type Rec (Len : Small_Integer := 0) is record
16162 Vals : IntArray (1 .. Len);
16166 you can assign a value with a different size of @code{Vals} with two
16170 (@value{GDBP}) set A_Rec.Len := 4
16171 (@value{GDBP}) set A_Rec := (Id => 42, Vals => (1, 2, 3, 4))
16174 As this example also illustrates, @value{GDBN} is very loose about the usual
16175 rules concerning aggregates. You may leave out some of the
16176 components of an array or record aggregate (such as the @code{Len}
16177 component in the assignment to @code{A_Rec} above); they will retain their
16178 original values upon assignment. You may freely use dynamic values as
16179 indices in component associations. You may even use overlapping or
16180 redundant component associations, although which component values are
16181 assigned in such cases is not defined.
16184 Calls to dispatching subprograms are not implemented.
16187 The overloading algorithm is much more limited (i.e., less selective)
16188 than that of real Ada. It makes only limited use of the context in
16189 which a subexpression appears to resolve its meaning, and it is much
16190 looser in its rules for allowing type matches. As a result, some
16191 function calls will be ambiguous, and the user will be asked to choose
16192 the proper resolution.
16195 The @code{new} operator is not implemented.
16198 Entry calls are not implemented.
16201 Aside from printing, arithmetic operations on the native VAX floating-point
16202 formats are not supported.
16205 It is not possible to slice a packed array.
16208 The names @code{True} and @code{False}, when not part of a qualified name,
16209 are interpreted as if implicitly prefixed by @code{Standard}, regardless of
16211 Should your program
16212 redefine these names in a package or procedure (at best a dubious practice),
16213 you will have to use fully qualified names to access their new definitions.
16216 @node Additions to Ada
16217 @subsubsection Additions to Ada
16218 @cindex Ada, deviations from
16220 As it does for other languages, @value{GDBN} makes certain generic
16221 extensions to Ada (@pxref{Expressions}):
16225 If the expression @var{E} is a variable residing in memory (typically
16226 a local variable or array element) and @var{N} is a positive integer,
16227 then @code{@var{E}@@@var{N}} displays the values of @var{E} and the
16228 @var{N}-1 adjacent variables following it in memory as an array. In
16229 Ada, this operator is generally not necessary, since its prime use is
16230 in displaying parts of an array, and slicing will usually do this in
16231 Ada. However, there are occasional uses when debugging programs in
16232 which certain debugging information has been optimized away.
16235 @code{@var{B}::@var{var}} means ``the variable named @var{var} that
16236 appears in function or file @var{B}.'' When @var{B} is a file name,
16237 you must typically surround it in single quotes.
16240 The expression @code{@{@var{type}@} @var{addr}} means ``the variable of type
16241 @var{type} that appears at address @var{addr}.''
16244 A name starting with @samp{$} is a convenience variable
16245 (@pxref{Convenience Vars}) or a machine register (@pxref{Registers}).
16248 In addition, @value{GDBN} provides a few other shortcuts and outright
16249 additions specific to Ada:
16253 The assignment statement is allowed as an expression, returning
16254 its right-hand operand as its value. Thus, you may enter
16257 (@value{GDBP}) set x := y + 3
16258 (@value{GDBP}) print A(tmp := y + 1)
16262 The semicolon is allowed as an ``operator,'' returning as its value
16263 the value of its right-hand operand.
16264 This allows, for example,
16265 complex conditional breaks:
16268 (@value{GDBP}) break f
16269 (@value{GDBP}) condition 1 (report(i); k += 1; A(k) > 100)
16273 Rather than use catenation and symbolic character names to introduce special
16274 characters into strings, one may instead use a special bracket notation,
16275 which is also used to print strings. A sequence of characters of the form
16276 @samp{["@var{XX}"]} within a string or character literal denotes the
16277 (single) character whose numeric encoding is @var{XX} in hexadecimal. The
16278 sequence of characters @samp{["""]} also denotes a single quotation mark
16279 in strings. For example,
16281 "One line.["0a"]Next line.["0a"]"
16284 contains an ASCII newline character (@code{Ada.Characters.Latin_1.LF})
16288 The subtype used as a prefix for the attributes @t{'Pos}, @t{'Min}, and
16289 @t{'Max} is optional (and is ignored in any case). For example, it is valid
16293 (@value{GDBP}) print 'max(x, y)
16297 When printing arrays, @value{GDBN} uses positional notation when the
16298 array has a lower bound of 1, and uses a modified named notation otherwise.
16299 For example, a one-dimensional array of three integers with a lower bound
16300 of 3 might print as
16307 That is, in contrast to valid Ada, only the first component has a @code{=>}
16311 You may abbreviate attributes in expressions with any unique,
16312 multi-character subsequence of
16313 their names (an exact match gets preference).
16314 For example, you may use @t{a'len}, @t{a'gth}, or @t{a'lh}
16315 in place of @t{a'length}.
16318 @cindex quoting Ada internal identifiers
16319 Since Ada is case-insensitive, the debugger normally maps identifiers you type
16320 to lower case. The GNAT compiler uses upper-case characters for
16321 some of its internal identifiers, which are normally of no interest to users.
16322 For the rare occasions when you actually have to look at them,
16323 enclose them in angle brackets to avoid the lower-case mapping.
16326 (@value{GDBP}) print <JMPBUF_SAVE>[0]
16330 Printing an object of class-wide type or dereferencing an
16331 access-to-class-wide value will display all the components of the object's
16332 specific type (as indicated by its run-time tag). Likewise, component
16333 selection on such a value will operate on the specific type of the
16338 @node Overloading support for Ada
16339 @subsubsection Overloading support for Ada
16340 @cindex overloading, Ada
16342 The debugger supports limited overloading. Given a subprogram call in which
16343 the function symbol has multiple definitions, it will use the number of
16344 actual parameters and some information about their types to attempt to narrow
16345 the set of definitions. It also makes very limited use of context, preferring
16346 procedures to functions in the context of the @code{call} command, and
16347 functions to procedures elsewhere.
16349 If, after narrowing, the set of matching definitions still contains more than
16350 one definition, @value{GDBN} will display a menu to query which one it should
16354 (@value{GDBP}) print f(1)
16355 Multiple matches for f
16357 [1] foo.f (integer) return boolean at foo.adb:23
16358 [2] foo.f (foo.new_integer) return boolean at foo.adb:28
16362 In this case, just select one menu entry either to cancel expression evaluation
16363 (type @kbd{0} and press @key{RET}) or to continue evaluation with a specific
16364 instance (type the corresponding number and press @key{RET}).
16366 Here are a couple of commands to customize @value{GDBN}'s behavior in this
16371 @kindex set ada print-signatures
16372 @item set ada print-signatures
16373 Control whether parameter types and return types are displayed in overloads
16374 selection menus. It is @code{on} by default.
16375 @xref{Overloading support for Ada}.
16377 @kindex show ada print-signatures
16378 @item show ada print-signatures
16379 Show the current setting for displaying parameter types and return types in
16380 overloads selection menu.
16381 @xref{Overloading support for Ada}.
16385 @node Stopping Before Main Program
16386 @subsubsection Stopping at the Very Beginning
16388 @cindex breakpointing Ada elaboration code
16389 It is sometimes necessary to debug the program during elaboration, and
16390 before reaching the main procedure.
16391 As defined in the Ada Reference
16392 Manual, the elaboration code is invoked from a procedure called
16393 @code{adainit}. To run your program up to the beginning of
16394 elaboration, simply use the following two commands:
16395 @code{tbreak adainit} and @code{run}.
16397 @node Ada Exceptions
16398 @subsubsection Ada Exceptions
16400 A command is provided to list all Ada exceptions:
16403 @kindex info exceptions
16404 @item info exceptions
16405 @itemx info exceptions @var{regexp}
16406 The @code{info exceptions} command allows you to list all Ada exceptions
16407 defined within the program being debugged, as well as their addresses.
16408 With a regular expression, @var{regexp}, as argument, only those exceptions
16409 whose names match @var{regexp} are listed.
16412 Below is a small example, showing how the command can be used, first
16413 without argument, and next with a regular expression passed as an
16417 (@value{GDBP}) info exceptions
16418 All defined Ada exceptions:
16419 constraint_error: 0x613da0
16420 program_error: 0x613d20
16421 storage_error: 0x613ce0
16422 tasking_error: 0x613ca0
16423 const.aint_global_e: 0x613b00
16424 (@value{GDBP}) info exceptions const.aint
16425 All Ada exceptions matching regular expression "const.aint":
16426 constraint_error: 0x613da0
16427 const.aint_global_e: 0x613b00
16430 It is also possible to ask @value{GDBN} to stop your program's execution
16431 when an exception is raised. For more details, see @ref{Set Catchpoints}.
16434 @subsubsection Extensions for Ada Tasks
16435 @cindex Ada, tasking
16437 Support for Ada tasks is analogous to that for threads (@pxref{Threads}).
16438 @value{GDBN} provides the following task-related commands:
16443 This command shows a list of current Ada tasks, as in the following example:
16450 (@value{GDBP}) info tasks
16451 ID TID P-ID Pri State Name
16452 1 8088000 0 15 Child Activation Wait main_task
16453 2 80a4000 1 15 Accept Statement b
16454 3 809a800 1 15 Child Activation Wait a
16455 * 4 80ae800 3 15 Runnable c
16460 In this listing, the asterisk before the last task indicates it to be the
16461 task currently being inspected.
16465 Represents @value{GDBN}'s internal task number.
16471 The parent's task ID (@value{GDBN}'s internal task number).
16474 The base priority of the task.
16477 Current state of the task.
16481 The task has been created but has not been activated. It cannot be
16485 The task is not blocked for any reason known to Ada. (It may be waiting
16486 for a mutex, though.) It is conceptually "executing" in normal mode.
16489 The task is terminated, in the sense of ARM 9.3 (5). Any dependents
16490 that were waiting on terminate alternatives have been awakened and have
16491 terminated themselves.
16493 @item Child Activation Wait
16494 The task is waiting for created tasks to complete activation.
16496 @item Accept Statement
16497 The task is waiting on an accept or selective wait statement.
16499 @item Waiting on entry call
16500 The task is waiting on an entry call.
16502 @item Async Select Wait
16503 The task is waiting to start the abortable part of an asynchronous
16507 The task is waiting on a select statement with only a delay
16510 @item Child Termination Wait
16511 The task is sleeping having completed a master within itself, and is
16512 waiting for the tasks dependent on that master to become terminated or
16513 waiting on a terminate Phase.
16515 @item Wait Child in Term Alt
16516 The task is sleeping waiting for tasks on terminate alternatives to
16517 finish terminating.
16519 @item Accepting RV with @var{taskno}
16520 The task is accepting a rendez-vous with the task @var{taskno}.
16524 Name of the task in the program.
16528 @kindex info task @var{taskno}
16529 @item info task @var{taskno}
16530 This command shows detailled informations on the specified task, as in
16531 the following example:
16536 (@value{GDBP}) info tasks
16537 ID TID P-ID Pri State Name
16538 1 8077880 0 15 Child Activation Wait main_task
16539 * 2 807c468 1 15 Runnable task_1
16540 (@value{GDBP}) info task 2
16541 Ada Task: 0x807c468
16544 Parent: 1 (main_task)
16550 @kindex task@r{ (Ada)}
16551 @cindex current Ada task ID
16552 This command prints the ID of the current task.
16558 (@value{GDBP}) info tasks
16559 ID TID P-ID Pri State Name
16560 1 8077870 0 15 Child Activation Wait main_task
16561 * 2 807c458 1 15 Runnable t
16562 (@value{GDBP}) task
16563 [Current task is 2]
16566 @item task @var{taskno}
16567 @cindex Ada task switching
16568 This command is like the @code{thread @var{thread-id}}
16569 command (@pxref{Threads}). It switches the context of debugging
16570 from the current task to the given task.
16576 (@value{GDBP}) info tasks
16577 ID TID P-ID Pri State Name
16578 1 8077870 0 15 Child Activation Wait main_task
16579 * 2 807c458 1 15 Runnable t
16580 (@value{GDBP}) task 1
16581 [Switching to task 1]
16582 #0 0x8067726 in pthread_cond_wait ()
16584 #0 0x8067726 in pthread_cond_wait ()
16585 #1 0x8056714 in system.os_interface.pthread_cond_wait ()
16586 #2 0x805cb63 in system.task_primitives.operations.sleep ()
16587 #3 0x806153e in system.tasking.stages.activate_tasks ()
16588 #4 0x804aacc in un () at un.adb:5
16591 @item break @var{location} task @var{taskno}
16592 @itemx break @var{location} task @var{taskno} if @dots{}
16593 @cindex breakpoints and tasks, in Ada
16594 @cindex task breakpoints, in Ada
16595 @kindex break @dots{} task @var{taskno}@r{ (Ada)}
16596 These commands are like the @code{break @dots{} thread @dots{}}
16597 command (@pxref{Thread Stops}). The
16598 @var{location} argument specifies source lines, as described
16599 in @ref{Specify Location}.
16601 Use the qualifier @samp{task @var{taskno}} with a breakpoint command
16602 to specify that you only want @value{GDBN} to stop the program when a
16603 particular Ada task reaches this breakpoint. The @var{taskno} is one of the
16604 numeric task identifiers assigned by @value{GDBN}, shown in the first
16605 column of the @samp{info tasks} display.
16607 If you do not specify @samp{task @var{taskno}} when you set a
16608 breakpoint, the breakpoint applies to @emph{all} tasks of your
16611 You can use the @code{task} qualifier on conditional breakpoints as
16612 well; in this case, place @samp{task @var{taskno}} before the
16613 breakpoint condition (before the @code{if}).
16621 (@value{GDBP}) info tasks
16622 ID TID P-ID Pri State Name
16623 1 140022020 0 15 Child Activation Wait main_task
16624 2 140045060 1 15 Accept/Select Wait t2
16625 3 140044840 1 15 Runnable t1
16626 * 4 140056040 1 15 Runnable t3
16627 (@value{GDBP}) b 15 task 2
16628 Breakpoint 5 at 0x120044cb0: file test_task_debug.adb, line 15.
16629 (@value{GDBP}) cont
16634 Breakpoint 5, test_task_debug () at test_task_debug.adb:15
16636 (@value{GDBP}) info tasks
16637 ID TID P-ID Pri State Name
16638 1 140022020 0 15 Child Activation Wait main_task
16639 * 2 140045060 1 15 Runnable t2
16640 3 140044840 1 15 Runnable t1
16641 4 140056040 1 15 Delay Sleep t3
16645 @node Ada Tasks and Core Files
16646 @subsubsection Tasking Support when Debugging Core Files
16647 @cindex Ada tasking and core file debugging
16649 When inspecting a core file, as opposed to debugging a live program,
16650 tasking support may be limited or even unavailable, depending on
16651 the platform being used.
16652 For instance, on x86-linux, the list of tasks is available, but task
16653 switching is not supported.
16655 On certain platforms, the debugger needs to perform some
16656 memory writes in order to provide Ada tasking support. When inspecting
16657 a core file, this means that the core file must be opened with read-write
16658 privileges, using the command @samp{"set write on"} (@pxref{Patching}).
16659 Under these circumstances, you should make a backup copy of the core
16660 file before inspecting it with @value{GDBN}.
16662 @node Ravenscar Profile
16663 @subsubsection Tasking Support when using the Ravenscar Profile
16664 @cindex Ravenscar Profile
16666 The @dfn{Ravenscar Profile} is a subset of the Ada tasking features,
16667 specifically designed for systems with safety-critical real-time
16671 @kindex set ravenscar task-switching on
16672 @cindex task switching with program using Ravenscar Profile
16673 @item set ravenscar task-switching on
16674 Allows task switching when debugging a program that uses the Ravenscar
16675 Profile. This is the default.
16677 @kindex set ravenscar task-switching off
16678 @item set ravenscar task-switching off
16679 Turn off task switching when debugging a program that uses the Ravenscar
16680 Profile. This is mostly intended to disable the code that adds support
16681 for the Ravenscar Profile, in case a bug in either @value{GDBN} or in
16682 the Ravenscar runtime is preventing @value{GDBN} from working properly.
16683 To be effective, this command should be run before the program is started.
16685 @kindex show ravenscar task-switching
16686 @item show ravenscar task-switching
16687 Show whether it is possible to switch from task to task in a program
16688 using the Ravenscar Profile.
16693 @subsubsection Known Peculiarities of Ada Mode
16694 @cindex Ada, problems
16696 Besides the omissions listed previously (@pxref{Omissions from Ada}),
16697 we know of several problems with and limitations of Ada mode in
16699 some of which will be fixed with planned future releases of the debugger
16700 and the GNU Ada compiler.
16704 Static constants that the compiler chooses not to materialize as objects in
16705 storage are invisible to the debugger.
16708 Named parameter associations in function argument lists are ignored (the
16709 argument lists are treated as positional).
16712 Many useful library packages are currently invisible to the debugger.
16715 Fixed-point arithmetic, conversions, input, and output is carried out using
16716 floating-point arithmetic, and may give results that only approximate those on
16720 The GNAT compiler never generates the prefix @code{Standard} for any of
16721 the standard symbols defined by the Ada language. @value{GDBN} knows about
16722 this: it will strip the prefix from names when you use it, and will never
16723 look for a name you have so qualified among local symbols, nor match against
16724 symbols in other packages or subprograms. If you have
16725 defined entities anywhere in your program other than parameters and
16726 local variables whose simple names match names in @code{Standard},
16727 GNAT's lack of qualification here can cause confusion. When this happens,
16728 you can usually resolve the confusion
16729 by qualifying the problematic names with package
16730 @code{Standard} explicitly.
16733 Older versions of the compiler sometimes generate erroneous debugging
16734 information, resulting in the debugger incorrectly printing the value
16735 of affected entities. In some cases, the debugger is able to work
16736 around an issue automatically. In other cases, the debugger is able
16737 to work around the issue, but the work-around has to be specifically
16740 @kindex set ada trust-PAD-over-XVS
16741 @kindex show ada trust-PAD-over-XVS
16744 @item set ada trust-PAD-over-XVS on
16745 Configure GDB to strictly follow the GNAT encoding when computing the
16746 value of Ada entities, particularly when @code{PAD} and @code{PAD___XVS}
16747 types are involved (see @code{ada/exp_dbug.ads} in the GCC sources for
16748 a complete description of the encoding used by the GNAT compiler).
16749 This is the default.
16751 @item set ada trust-PAD-over-XVS off
16752 This is related to the encoding using by the GNAT compiler. If @value{GDBN}
16753 sometimes prints the wrong value for certain entities, changing @code{ada
16754 trust-PAD-over-XVS} to @code{off} activates a work-around which may fix
16755 the issue. It is always safe to set @code{ada trust-PAD-over-XVS} to
16756 @code{off}, but this incurs a slight performance penalty, so it is
16757 recommended to leave this setting to @code{on} unless necessary.
16761 @cindex GNAT descriptive types
16762 @cindex GNAT encoding
16763 Internally, the debugger also relies on the compiler following a number
16764 of conventions known as the @samp{GNAT Encoding}, all documented in
16765 @file{gcc/ada/exp_dbug.ads} in the GCC sources. This encoding describes
16766 how the debugging information should be generated for certain types.
16767 In particular, this convention makes use of @dfn{descriptive types},
16768 which are artificial types generated purely to help the debugger.
16770 These encodings were defined at a time when the debugging information
16771 format used was not powerful enough to describe some of the more complex
16772 types available in Ada. Since DWARF allows us to express nearly all
16773 Ada features, the long-term goal is to slowly replace these descriptive
16774 types by their pure DWARF equivalent. To facilitate that transition,
16775 a new maintenance option is available to force the debugger to ignore
16776 those descriptive types. It allows the user to quickly evaluate how
16777 well @value{GDBN} works without them.
16781 @kindex maint ada set ignore-descriptive-types
16782 @item maintenance ada set ignore-descriptive-types [on|off]
16783 Control whether the debugger should ignore descriptive types.
16784 The default is not to ignore descriptives types (@code{off}).
16786 @kindex maint ada show ignore-descriptive-types
16787 @item maintenance ada show ignore-descriptive-types
16788 Show if descriptive types are ignored by @value{GDBN}.
16792 @node Unsupported Languages
16793 @section Unsupported Languages
16795 @cindex unsupported languages
16796 @cindex minimal language
16797 In addition to the other fully-supported programming languages,
16798 @value{GDBN} also provides a pseudo-language, called @code{minimal}.
16799 It does not represent a real programming language, but provides a set
16800 of capabilities close to what the C or assembly languages provide.
16801 This should allow most simple operations to be performed while debugging
16802 an application that uses a language currently not supported by @value{GDBN}.
16804 If the language is set to @code{auto}, @value{GDBN} will automatically
16805 select this language if the current frame corresponds to an unsupported
16809 @chapter Examining the Symbol Table
16811 The commands described in this chapter allow you to inquire about the
16812 symbols (names of variables, functions and types) defined in your
16813 program. This information is inherent in the text of your program and
16814 does not change as your program executes. @value{GDBN} finds it in your
16815 program's symbol table, in the file indicated when you started @value{GDBN}
16816 (@pxref{File Options, ,Choosing Files}), or by one of the
16817 file-management commands (@pxref{Files, ,Commands to Specify Files}).
16819 @cindex symbol names
16820 @cindex names of symbols
16821 @cindex quoting names
16822 Occasionally, you may need to refer to symbols that contain unusual
16823 characters, which @value{GDBN} ordinarily treats as word delimiters. The
16824 most frequent case is in referring to static variables in other
16825 source files (@pxref{Variables,,Program Variables}). File names
16826 are recorded in object files as debugging symbols, but @value{GDBN} would
16827 ordinarily parse a typical file name, like @file{foo.c}, as the three words
16828 @samp{foo} @samp{.} @samp{c}. To allow @value{GDBN} to recognize
16829 @samp{foo.c} as a single symbol, enclose it in single quotes; for example,
16836 looks up the value of @code{x} in the scope of the file @file{foo.c}.
16839 @cindex case-insensitive symbol names
16840 @cindex case sensitivity in symbol names
16841 @kindex set case-sensitive
16842 @item set case-sensitive on
16843 @itemx set case-sensitive off
16844 @itemx set case-sensitive auto
16845 Normally, when @value{GDBN} looks up symbols, it matches their names
16846 with case sensitivity determined by the current source language.
16847 Occasionally, you may wish to control that. The command @code{set
16848 case-sensitive} lets you do that by specifying @code{on} for
16849 case-sensitive matches or @code{off} for case-insensitive ones. If
16850 you specify @code{auto}, case sensitivity is reset to the default
16851 suitable for the source language. The default is case-sensitive
16852 matches for all languages except for Fortran, for which the default is
16853 case-insensitive matches.
16855 @kindex show case-sensitive
16856 @item show case-sensitive
16857 This command shows the current setting of case sensitivity for symbols
16860 @kindex set print type methods
16861 @item set print type methods
16862 @itemx set print type methods on
16863 @itemx set print type methods off
16864 Normally, when @value{GDBN} prints a class, it displays any methods
16865 declared in that class. You can control this behavior either by
16866 passing the appropriate flag to @code{ptype}, or using @command{set
16867 print type methods}. Specifying @code{on} will cause @value{GDBN} to
16868 display the methods; this is the default. Specifying @code{off} will
16869 cause @value{GDBN} to omit the methods.
16871 @kindex show print type methods
16872 @item show print type methods
16873 This command shows the current setting of method display when printing
16876 @kindex set print type typedefs
16877 @item set print type typedefs
16878 @itemx set print type typedefs on
16879 @itemx set print type typedefs off
16881 Normally, when @value{GDBN} prints a class, it displays any typedefs
16882 defined in that class. You can control this behavior either by
16883 passing the appropriate flag to @code{ptype}, or using @command{set
16884 print type typedefs}. Specifying @code{on} will cause @value{GDBN} to
16885 display the typedef definitions; this is the default. Specifying
16886 @code{off} will cause @value{GDBN} to omit the typedef definitions.
16887 Note that this controls whether the typedef definition itself is
16888 printed, not whether typedef names are substituted when printing other
16891 @kindex show print type typedefs
16892 @item show print type typedefs
16893 This command shows the current setting of typedef display when
16896 @kindex info address
16897 @cindex address of a symbol
16898 @item info address @var{symbol}
16899 Describe where the data for @var{symbol} is stored. For a register
16900 variable, this says which register it is kept in. For a non-register
16901 local variable, this prints the stack-frame offset at which the variable
16904 Note the contrast with @samp{print &@var{symbol}}, which does not work
16905 at all for a register variable, and for a stack local variable prints
16906 the exact address of the current instantiation of the variable.
16908 @kindex info symbol
16909 @cindex symbol from address
16910 @cindex closest symbol and offset for an address
16911 @item info symbol @var{addr}
16912 Print the name of a symbol which is stored at the address @var{addr}.
16913 If no symbol is stored exactly at @var{addr}, @value{GDBN} prints the
16914 nearest symbol and an offset from it:
16917 (@value{GDBP}) info symbol 0x54320
16918 _initialize_vx + 396 in section .text
16922 This is the opposite of the @code{info address} command. You can use
16923 it to find out the name of a variable or a function given its address.
16925 For dynamically linked executables, the name of executable or shared
16926 library containing the symbol is also printed:
16929 (@value{GDBP}) info symbol 0x400225
16930 _start + 5 in section .text of /tmp/a.out
16931 (@value{GDBP}) info symbol 0x2aaaac2811cf
16932 __read_nocancel + 6 in section .text of /usr/lib64/libc.so.6
16937 @item demangle @r{[}-l @var{language}@r{]} @r{[}@var{--}@r{]} @var{name}
16938 Demangle @var{name}.
16939 If @var{language} is provided it is the name of the language to demangle
16940 @var{name} in. Otherwise @var{name} is demangled in the current language.
16942 The @samp{--} option specifies the end of options,
16943 and is useful when @var{name} begins with a dash.
16945 The parameter @code{demangle-style} specifies how to interpret the kind
16946 of mangling used. @xref{Print Settings}.
16949 @item whatis[/@var{flags}] [@var{arg}]
16950 Print the data type of @var{arg}, which can be either an expression
16951 or a name of a data type. With no argument, print the data type of
16952 @code{$}, the last value in the value history.
16954 If @var{arg} is an expression (@pxref{Expressions, ,Expressions}), it
16955 is not actually evaluated, and any side-effecting operations (such as
16956 assignments or function calls) inside it do not take place.
16958 If @var{arg} is a variable or an expression, @code{whatis} prints its
16959 literal type as it is used in the source code. If the type was
16960 defined using a @code{typedef}, @code{whatis} will @emph{not} print
16961 the data type underlying the @code{typedef}. If the type of the
16962 variable or the expression is a compound data type, such as
16963 @code{struct} or @code{class}, @code{whatis} never prints their
16964 fields or methods. It just prints the @code{struct}/@code{class}
16965 name (a.k.a.@: its @dfn{tag}). If you want to see the members of
16966 such a compound data type, use @code{ptype}.
16968 If @var{arg} is a type name that was defined using @code{typedef},
16969 @code{whatis} @dfn{unrolls} only one level of that @code{typedef}.
16970 Unrolling means that @code{whatis} will show the underlying type used
16971 in the @code{typedef} declaration of @var{arg}. However, if that
16972 underlying type is also a @code{typedef}, @code{whatis} will not
16975 For C code, the type names may also have the form @samp{class
16976 @var{class-name}}, @samp{struct @var{struct-tag}}, @samp{union
16977 @var{union-tag}} or @samp{enum @var{enum-tag}}.
16979 @var{flags} can be used to modify how the type is displayed.
16980 Available flags are:
16984 Display in ``raw'' form. Normally, @value{GDBN} substitutes template
16985 parameters and typedefs defined in a class when printing the class'
16986 members. The @code{/r} flag disables this.
16989 Do not print methods defined in the class.
16992 Print methods defined in the class. This is the default, but the flag
16993 exists in case you change the default with @command{set print type methods}.
16996 Do not print typedefs defined in the class. Note that this controls
16997 whether the typedef definition itself is printed, not whether typedef
16998 names are substituted when printing other types.
17001 Print typedefs defined in the class. This is the default, but the flag
17002 exists in case you change the default with @command{set print type typedefs}.
17006 @item ptype[/@var{flags}] [@var{arg}]
17007 @code{ptype} accepts the same arguments as @code{whatis}, but prints a
17008 detailed description of the type, instead of just the name of the type.
17009 @xref{Expressions, ,Expressions}.
17011 Contrary to @code{whatis}, @code{ptype} always unrolls any
17012 @code{typedef}s in its argument declaration, whether the argument is
17013 a variable, expression, or a data type. This means that @code{ptype}
17014 of a variable or an expression will not print literally its type as
17015 present in the source code---use @code{whatis} for that. @code{typedef}s at
17016 the pointer or reference targets are also unrolled. Only @code{typedef}s of
17017 fields, methods and inner @code{class typedef}s of @code{struct}s,
17018 @code{class}es and @code{union}s are not unrolled even with @code{ptype}.
17020 For example, for this variable declaration:
17023 typedef double real_t;
17024 struct complex @{ real_t real; double imag; @};
17025 typedef struct complex complex_t;
17027 real_t *real_pointer_var;
17031 the two commands give this output:
17035 (@value{GDBP}) whatis var
17037 (@value{GDBP}) ptype var
17038 type = struct complex @{
17042 (@value{GDBP}) whatis complex_t
17043 type = struct complex
17044 (@value{GDBP}) whatis struct complex
17045 type = struct complex
17046 (@value{GDBP}) ptype struct complex
17047 type = struct complex @{
17051 (@value{GDBP}) whatis real_pointer_var
17053 (@value{GDBP}) ptype real_pointer_var
17059 As with @code{whatis}, using @code{ptype} without an argument refers to
17060 the type of @code{$}, the last value in the value history.
17062 @cindex incomplete type
17063 Sometimes, programs use opaque data types or incomplete specifications
17064 of complex data structure. If the debug information included in the
17065 program does not allow @value{GDBN} to display a full declaration of
17066 the data type, it will say @samp{<incomplete type>}. For example,
17067 given these declarations:
17071 struct foo *fooptr;
17075 but no definition for @code{struct foo} itself, @value{GDBN} will say:
17078 (@value{GDBP}) ptype foo
17079 $1 = <incomplete type>
17083 ``Incomplete type'' is C terminology for data types that are not
17084 completely specified.
17087 @item info types @var{regexp}
17089 Print a brief description of all types whose names match the regular
17090 expression @var{regexp} (or all types in your program, if you supply
17091 no argument). Each complete typename is matched as though it were a
17092 complete line; thus, @samp{i type value} gives information on all
17093 types in your program whose names include the string @code{value}, but
17094 @samp{i type ^value$} gives information only on types whose complete
17095 name is @code{value}.
17097 This command differs from @code{ptype} in two ways: first, like
17098 @code{whatis}, it does not print a detailed description; second, it
17099 lists all source files where a type is defined.
17101 @kindex info type-printers
17102 @item info type-printers
17103 Versions of @value{GDBN} that ship with Python scripting enabled may
17104 have ``type printers'' available. When using @command{ptype} or
17105 @command{whatis}, these printers are consulted when the name of a type
17106 is needed. @xref{Type Printing API}, for more information on writing
17109 @code{info type-printers} displays all the available type printers.
17111 @kindex enable type-printer
17112 @kindex disable type-printer
17113 @item enable type-printer @var{name}@dots{}
17114 @item disable type-printer @var{name}@dots{}
17115 These commands can be used to enable or disable type printers.
17118 @cindex local variables
17119 @item info scope @var{location}
17120 List all the variables local to a particular scope. This command
17121 accepts a @var{location} argument---a function name, a source line, or
17122 an address preceded by a @samp{*}, and prints all the variables local
17123 to the scope defined by that location. (@xref{Specify Location}, for
17124 details about supported forms of @var{location}.) For example:
17127 (@value{GDBP}) @b{info scope command_line_handler}
17128 Scope for command_line_handler:
17129 Symbol rl is an argument at stack/frame offset 8, length 4.
17130 Symbol linebuffer is in static storage at address 0x150a18, length 4.
17131 Symbol linelength is in static storage at address 0x150a1c, length 4.
17132 Symbol p is a local variable in register $esi, length 4.
17133 Symbol p1 is a local variable in register $ebx, length 4.
17134 Symbol nline is a local variable in register $edx, length 4.
17135 Symbol repeat is a local variable at frame offset -8, length 4.
17139 This command is especially useful for determining what data to collect
17140 during a @dfn{trace experiment}, see @ref{Tracepoint Actions,
17143 @kindex info source
17145 Show information about the current source file---that is, the source file for
17146 the function containing the current point of execution:
17149 the name of the source file, and the directory containing it,
17151 the directory it was compiled in,
17153 its length, in lines,
17155 which programming language it is written in,
17157 if the debug information provides it, the program that compiled the file
17158 (which may include, e.g., the compiler version and command line arguments),
17160 whether the executable includes debugging information for that file, and
17161 if so, what format the information is in (e.g., STABS, Dwarf 2, etc.), and
17163 whether the debugging information includes information about
17164 preprocessor macros.
17168 @kindex info sources
17170 Print the names of all source files in your program for which there is
17171 debugging information, organized into two lists: files whose symbols
17172 have already been read, and files whose symbols will be read when needed.
17174 @kindex info functions
17175 @item info functions
17176 Print the names and data types of all defined functions.
17178 @item info functions @var{regexp}
17179 Print the names and data types of all defined functions
17180 whose names contain a match for regular expression @var{regexp}.
17181 Thus, @samp{info fun step} finds all functions whose names
17182 include @code{step}; @samp{info fun ^step} finds those whose names
17183 start with @code{step}. If a function name contains characters
17184 that conflict with the regular expression language (e.g.@:
17185 @samp{operator*()}), they may be quoted with a backslash.
17187 @kindex info variables
17188 @item info variables
17189 Print the names and data types of all variables that are defined
17190 outside of functions (i.e.@: excluding local variables).
17192 @item info variables @var{regexp}
17193 Print the names and data types of all variables (except for local
17194 variables) whose names contain a match for regular expression
17197 @kindex info classes
17198 @cindex Objective-C, classes and selectors
17200 @itemx info classes @var{regexp}
17201 Display all Objective-C classes in your program, or
17202 (with the @var{regexp} argument) all those matching a particular regular
17205 @kindex info selectors
17206 @item info selectors
17207 @itemx info selectors @var{regexp}
17208 Display all Objective-C selectors in your program, or
17209 (with the @var{regexp} argument) all those matching a particular regular
17213 This was never implemented.
17214 @kindex info methods
17216 @itemx info methods @var{regexp}
17217 The @code{info methods} command permits the user to examine all defined
17218 methods within C@t{++} program, or (with the @var{regexp} argument) a
17219 specific set of methods found in the various C@t{++} classes. Many
17220 C@t{++} classes provide a large number of methods. Thus, the output
17221 from the @code{ptype} command can be overwhelming and hard to use. The
17222 @code{info-methods} command filters the methods, printing only those
17223 which match the regular-expression @var{regexp}.
17226 @cindex opaque data types
17227 @kindex set opaque-type-resolution
17228 @item set opaque-type-resolution on
17229 Tell @value{GDBN} to resolve opaque types. An opaque type is a type
17230 declared as a pointer to a @code{struct}, @code{class}, or
17231 @code{union}---for example, @code{struct MyType *}---that is used in one
17232 source file although the full declaration of @code{struct MyType} is in
17233 another source file. The default is on.
17235 A change in the setting of this subcommand will not take effect until
17236 the next time symbols for a file are loaded.
17238 @item set opaque-type-resolution off
17239 Tell @value{GDBN} not to resolve opaque types. In this case, the type
17240 is printed as follows:
17242 @{<no data fields>@}
17245 @kindex show opaque-type-resolution
17246 @item show opaque-type-resolution
17247 Show whether opaque types are resolved or not.
17249 @kindex set print symbol-loading
17250 @cindex print messages when symbols are loaded
17251 @item set print symbol-loading
17252 @itemx set print symbol-loading full
17253 @itemx set print symbol-loading brief
17254 @itemx set print symbol-loading off
17255 The @code{set print symbol-loading} command allows you to control the
17256 printing of messages when @value{GDBN} loads symbol information.
17257 By default a message is printed for the executable and one for each
17258 shared library, and normally this is what you want. However, when
17259 debugging apps with large numbers of shared libraries these messages
17261 When set to @code{brief} a message is printed for each executable,
17262 and when @value{GDBN} loads a collection of shared libraries at once
17263 it will only print one message regardless of the number of shared
17264 libraries. When set to @code{off} no messages are printed.
17266 @kindex show print symbol-loading
17267 @item show print symbol-loading
17268 Show whether messages will be printed when a @value{GDBN} command
17269 entered from the keyboard causes symbol information to be loaded.
17271 @kindex maint print symbols
17272 @cindex symbol dump
17273 @kindex maint print psymbols
17274 @cindex partial symbol dump
17275 @kindex maint print msymbols
17276 @cindex minimal symbol dump
17277 @item maint print symbols @r{[}-pc @var{address}@r{]} @r{[}@var{filename}@r{]}
17278 @itemx maint print symbols @r{[}-objfile @var{objfile}@r{]} @r{[}-source @var{source}@r{]} @r{[}--@r{]} @r{[}@var{filename}@r{]}
17279 @itemx maint print psymbols @r{[}-objfile @var{objfile}@r{]} @r{[}-pc @var{address}@r{]} @r{[}--@r{]} @r{[}@var{filename}@r{]}
17280 @itemx maint print psymbols @r{[}-objfile @var{objfile}@r{]} @r{[}-source @var{source}@r{]} @r{[}--@r{]} @r{[}@var{filename}@r{]}
17281 @itemx maint print msymbols @r{[}-objfile @var{objfile}@r{]} @r{[}--@r{]} @r{[}@var{filename}@r{]}
17282 Write a dump of debugging symbol data into the file @var{filename} or
17283 the terminal if @var{filename} is unspecified.
17284 If @code{-objfile @var{objfile}} is specified, only dump symbols for
17286 If @code{-pc @var{address}} is specified, only dump symbols for the file
17287 with code at that address. Note that @var{address} may be a symbol like
17289 If @code{-source @var{source}} is specified, only dump symbols for that
17292 These commands are used to debug the @value{GDBN} symbol-reading code.
17293 These commands do not modify internal @value{GDBN} state, therefore
17294 @samp{maint print symbols} will only print symbols for already expanded symbol
17296 You can use the command @code{info sources} to find out which files these are.
17297 If you use @samp{maint print psymbols} instead, the dump shows information
17298 about symbols that @value{GDBN} only knows partially---that is, symbols
17299 defined in files that @value{GDBN} has skimmed, but not yet read completely.
17300 Finally, @samp{maint print msymbols} just dumps ``minimal symbols'', e.g.,
17303 @xref{Files, ,Commands to Specify Files}, for a discussion of how
17304 @value{GDBN} reads symbols (in the description of @code{symbol-file}).
17306 @kindex maint info symtabs
17307 @kindex maint info psymtabs
17308 @cindex listing @value{GDBN}'s internal symbol tables
17309 @cindex symbol tables, listing @value{GDBN}'s internal
17310 @cindex full symbol tables, listing @value{GDBN}'s internal
17311 @cindex partial symbol tables, listing @value{GDBN}'s internal
17312 @item maint info symtabs @r{[} @var{regexp} @r{]}
17313 @itemx maint info psymtabs @r{[} @var{regexp} @r{]}
17315 List the @code{struct symtab} or @code{struct partial_symtab}
17316 structures whose names match @var{regexp}. If @var{regexp} is not
17317 given, list them all. The output includes expressions which you can
17318 copy into a @value{GDBN} debugging this one to examine a particular
17319 structure in more detail. For example:
17322 (@value{GDBP}) maint info psymtabs dwarf2read
17323 @{ objfile /home/gnu/build/gdb/gdb
17324 ((struct objfile *) 0x82e69d0)
17325 @{ psymtab /home/gnu/src/gdb/dwarf2read.c
17326 ((struct partial_symtab *) 0x8474b10)
17329 text addresses 0x814d3c8 -- 0x8158074
17330 globals (* (struct partial_symbol **) 0x8507a08 @@ 9)
17331 statics (* (struct partial_symbol **) 0x40e95b78 @@ 2882)
17332 dependencies (none)
17335 (@value{GDBP}) maint info symtabs
17339 We see that there is one partial symbol table whose filename contains
17340 the string @samp{dwarf2read}, belonging to the @samp{gdb} executable;
17341 and we see that @value{GDBN} has not read in any symtabs yet at all.
17342 If we set a breakpoint on a function, that will cause @value{GDBN} to
17343 read the symtab for the compilation unit containing that function:
17346 (@value{GDBP}) break dwarf2_psymtab_to_symtab
17347 Breakpoint 1 at 0x814e5da: file /home/gnu/src/gdb/dwarf2read.c,
17349 (@value{GDBP}) maint info symtabs
17350 @{ objfile /home/gnu/build/gdb/gdb
17351 ((struct objfile *) 0x82e69d0)
17352 @{ symtab /home/gnu/src/gdb/dwarf2read.c
17353 ((struct symtab *) 0x86c1f38)
17356 blockvector ((struct blockvector *) 0x86c1bd0) (primary)
17357 linetable ((struct linetable *) 0x8370fa0)
17358 debugformat DWARF 2
17364 @kindex maint info line-table
17365 @cindex listing @value{GDBN}'s internal line tables
17366 @cindex line tables, listing @value{GDBN}'s internal
17367 @item maint info line-table @r{[} @var{regexp} @r{]}
17369 List the @code{struct linetable} from all @code{struct symtab}
17370 instances whose name matches @var{regexp}. If @var{regexp} is not
17371 given, list the @code{struct linetable} from all @code{struct symtab}.
17373 @kindex maint set symbol-cache-size
17374 @cindex symbol cache size
17375 @item maint set symbol-cache-size @var{size}
17376 Set the size of the symbol cache to @var{size}.
17377 The default size is intended to be good enough for debugging
17378 most applications. This option exists to allow for experimenting
17379 with different sizes.
17381 @kindex maint show symbol-cache-size
17382 @item maint show symbol-cache-size
17383 Show the size of the symbol cache.
17385 @kindex maint print symbol-cache
17386 @cindex symbol cache, printing its contents
17387 @item maint print symbol-cache
17388 Print the contents of the symbol cache.
17389 This is useful when debugging symbol cache issues.
17391 @kindex maint print symbol-cache-statistics
17392 @cindex symbol cache, printing usage statistics
17393 @item maint print symbol-cache-statistics
17394 Print symbol cache usage statistics.
17395 This helps determine how well the cache is being utilized.
17397 @kindex maint flush-symbol-cache
17398 @cindex symbol cache, flushing
17399 @item maint flush-symbol-cache
17400 Flush the contents of the symbol cache, all entries are removed.
17401 This command is useful when debugging the symbol cache.
17402 It is also useful when collecting performance data.
17407 @chapter Altering Execution
17409 Once you think you have found an error in your program, you might want to
17410 find out for certain whether correcting the apparent error would lead to
17411 correct results in the rest of the run. You can find the answer by
17412 experiment, using the @value{GDBN} features for altering execution of the
17415 For example, you can store new values into variables or memory
17416 locations, give your program a signal, restart it at a different
17417 address, or even return prematurely from a function.
17420 * Assignment:: Assignment to variables
17421 * Jumping:: Continuing at a different address
17422 * Signaling:: Giving your program a signal
17423 * Returning:: Returning from a function
17424 * Calling:: Calling your program's functions
17425 * Patching:: Patching your program
17426 * Compiling and Injecting Code:: Compiling and injecting code in @value{GDBN}
17430 @section Assignment to Variables
17433 @cindex setting variables
17434 To alter the value of a variable, evaluate an assignment expression.
17435 @xref{Expressions, ,Expressions}. For example,
17442 stores the value 4 into the variable @code{x}, and then prints the
17443 value of the assignment expression (which is 4).
17444 @xref{Languages, ,Using @value{GDBN} with Different Languages}, for more
17445 information on operators in supported languages.
17447 @kindex set variable
17448 @cindex variables, setting
17449 If you are not interested in seeing the value of the assignment, use the
17450 @code{set} command instead of the @code{print} command. @code{set} is
17451 really the same as @code{print} except that the expression's value is
17452 not printed and is not put in the value history (@pxref{Value History,
17453 ,Value History}). The expression is evaluated only for its effects.
17455 If the beginning of the argument string of the @code{set} command
17456 appears identical to a @code{set} subcommand, use the @code{set
17457 variable} command instead of just @code{set}. This command is identical
17458 to @code{set} except for its lack of subcommands. For example, if your
17459 program has a variable @code{width}, you get an error if you try to set
17460 a new value with just @samp{set width=13}, because @value{GDBN} has the
17461 command @code{set width}:
17464 (@value{GDBP}) whatis width
17466 (@value{GDBP}) p width
17468 (@value{GDBP}) set width=47
17469 Invalid syntax in expression.
17473 The invalid expression, of course, is @samp{=47}. In
17474 order to actually set the program's variable @code{width}, use
17477 (@value{GDBP}) set var width=47
17480 Because the @code{set} command has many subcommands that can conflict
17481 with the names of program variables, it is a good idea to use the
17482 @code{set variable} command instead of just @code{set}. For example, if
17483 your program has a variable @code{g}, you run into problems if you try
17484 to set a new value with just @samp{set g=4}, because @value{GDBN} has
17485 the command @code{set gnutarget}, abbreviated @code{set g}:
17489 (@value{GDBP}) whatis g
17493 (@value{GDBP}) set g=4
17497 The program being debugged has been started already.
17498 Start it from the beginning? (y or n) y
17499 Starting program: /home/smith/cc_progs/a.out
17500 "/home/smith/cc_progs/a.out": can't open to read symbols:
17501 Invalid bfd target.
17502 (@value{GDBP}) show g
17503 The current BFD target is "=4".
17508 The program variable @code{g} did not change, and you silently set the
17509 @code{gnutarget} to an invalid value. In order to set the variable
17513 (@value{GDBP}) set var g=4
17516 @value{GDBN} allows more implicit conversions in assignments than C; you can
17517 freely store an integer value into a pointer variable or vice versa,
17518 and you can convert any structure to any other structure that is the
17519 same length or shorter.
17520 @comment FIXME: how do structs align/pad in these conversions?
17521 @comment /doc@cygnus.com 18dec1990
17523 To store values into arbitrary places in memory, use the @samp{@{@dots{}@}}
17524 construct to generate a value of specified type at a specified address
17525 (@pxref{Expressions, ,Expressions}). For example, @code{@{int@}0x83040} refers
17526 to memory location @code{0x83040} as an integer (which implies a certain size
17527 and representation in memory), and
17530 set @{int@}0x83040 = 4
17534 stores the value 4 into that memory location.
17537 @section Continuing at a Different Address
17539 Ordinarily, when you continue your program, you do so at the place where
17540 it stopped, with the @code{continue} command. You can instead continue at
17541 an address of your own choosing, with the following commands:
17545 @kindex j @r{(@code{jump})}
17546 @item jump @var{location}
17547 @itemx j @var{location}
17548 Resume execution at @var{location}. Execution stops again immediately
17549 if there is a breakpoint there. @xref{Specify Location}, for a description
17550 of the different forms of @var{location}. It is common
17551 practice to use the @code{tbreak} command in conjunction with
17552 @code{jump}. @xref{Set Breaks, ,Setting Breakpoints}.
17554 The @code{jump} command does not change the current stack frame, or
17555 the stack pointer, or the contents of any memory location or any
17556 register other than the program counter. If @var{location} is in
17557 a different function from the one currently executing, the results may
17558 be bizarre if the two functions expect different patterns of arguments or
17559 of local variables. For this reason, the @code{jump} command requests
17560 confirmation if the specified line is not in the function currently
17561 executing. However, even bizarre results are predictable if you are
17562 well acquainted with the machine-language code of your program.
17565 On many systems, you can get much the same effect as the @code{jump}
17566 command by storing a new value into the register @code{$pc}. The
17567 difference is that this does not start your program running; it only
17568 changes the address of where it @emph{will} run when you continue. For
17576 makes the next @code{continue} command or stepping command execute at
17577 address @code{0x485}, rather than at the address where your program stopped.
17578 @xref{Continuing and Stepping, ,Continuing and Stepping}.
17580 The most common occasion to use the @code{jump} command is to back
17581 up---perhaps with more breakpoints set---over a portion of a program
17582 that has already executed, in order to examine its execution in more
17587 @section Giving your Program a Signal
17588 @cindex deliver a signal to a program
17592 @item signal @var{signal}
17593 Resume execution where your program is stopped, but immediately give it the
17594 signal @var{signal}. The @var{signal} can be the name or the number of a
17595 signal. For example, on many systems @code{signal 2} and @code{signal
17596 SIGINT} are both ways of sending an interrupt signal.
17598 Alternatively, if @var{signal} is zero, continue execution without
17599 giving a signal. This is useful when your program stopped on account of
17600 a signal and would ordinarily see the signal when resumed with the
17601 @code{continue} command; @samp{signal 0} causes it to resume without a
17604 @emph{Note:} When resuming a multi-threaded program, @var{signal} is
17605 delivered to the currently selected thread, not the thread that last
17606 reported a stop. This includes the situation where a thread was
17607 stopped due to a signal. So if you want to continue execution
17608 suppressing the signal that stopped a thread, you should select that
17609 same thread before issuing the @samp{signal 0} command. If you issue
17610 the @samp{signal 0} command with another thread as the selected one,
17611 @value{GDBN} detects that and asks for confirmation.
17613 Invoking the @code{signal} command is not the same as invoking the
17614 @code{kill} utility from the shell. Sending a signal with @code{kill}
17615 causes @value{GDBN} to decide what to do with the signal depending on
17616 the signal handling tables (@pxref{Signals}). The @code{signal} command
17617 passes the signal directly to your program.
17619 @code{signal} does not repeat when you press @key{RET} a second time
17620 after executing the command.
17622 @kindex queue-signal
17623 @item queue-signal @var{signal}
17624 Queue @var{signal} to be delivered immediately to the current thread
17625 when execution of the thread resumes. The @var{signal} can be the name or
17626 the number of a signal. For example, on many systems @code{signal 2} and
17627 @code{signal SIGINT} are both ways of sending an interrupt signal.
17628 The handling of the signal must be set to pass the signal to the program,
17629 otherwise @value{GDBN} will report an error.
17630 You can control the handling of signals from @value{GDBN} with the
17631 @code{handle} command (@pxref{Signals}).
17633 Alternatively, if @var{signal} is zero, any currently queued signal
17634 for the current thread is discarded and when execution resumes no signal
17635 will be delivered. This is useful when your program stopped on account
17636 of a signal and would ordinarily see the signal when resumed with the
17637 @code{continue} command.
17639 This command differs from the @code{signal} command in that the signal
17640 is just queued, execution is not resumed. And @code{queue-signal} cannot
17641 be used to pass a signal whose handling state has been set to @code{nopass}
17646 @xref{stepping into signal handlers}, for information on how stepping
17647 commands behave when the thread has a signal queued.
17650 @section Returning from a Function
17653 @cindex returning from a function
17656 @itemx return @var{expression}
17657 You can cancel execution of a function call with the @code{return}
17658 command. If you give an
17659 @var{expression} argument, its value is used as the function's return
17663 When you use @code{return}, @value{GDBN} discards the selected stack frame
17664 (and all frames within it). You can think of this as making the
17665 discarded frame return prematurely. If you wish to specify a value to
17666 be returned, give that value as the argument to @code{return}.
17668 This pops the selected stack frame (@pxref{Selection, ,Selecting a
17669 Frame}), and any other frames inside of it, leaving its caller as the
17670 innermost remaining frame. That frame becomes selected. The
17671 specified value is stored in the registers used for returning values
17674 The @code{return} command does not resume execution; it leaves the
17675 program stopped in the state that would exist if the function had just
17676 returned. In contrast, the @code{finish} command (@pxref{Continuing
17677 and Stepping, ,Continuing and Stepping}) resumes execution until the
17678 selected stack frame returns naturally.
17680 @value{GDBN} needs to know how the @var{expression} argument should be set for
17681 the inferior. The concrete registers assignment depends on the OS ABI and the
17682 type being returned by the selected stack frame. For example it is common for
17683 OS ABI to return floating point values in FPU registers while integer values in
17684 CPU registers. Still some ABIs return even floating point values in CPU
17685 registers. Larger integer widths (such as @code{long long int}) also have
17686 specific placement rules. @value{GDBN} already knows the OS ABI from its
17687 current target so it needs to find out also the type being returned to make the
17688 assignment into the right register(s).
17690 Normally, the selected stack frame has debug info. @value{GDBN} will always
17691 use the debug info instead of the implicit type of @var{expression} when the
17692 debug info is available. For example, if you type @kbd{return -1}, and the
17693 function in the current stack frame is declared to return a @code{long long
17694 int}, @value{GDBN} transparently converts the implicit @code{int} value of -1
17695 into a @code{long long int}:
17698 Breakpoint 1, func () at gdb.base/return-nodebug.c:29
17700 (@value{GDBP}) return -1
17701 Make func return now? (y or n) y
17702 #0 0x004004f6 in main () at gdb.base/return-nodebug.c:43
17703 43 printf ("result=%lld\n", func ());
17707 However, if the selected stack frame does not have a debug info, e.g., if the
17708 function was compiled without debug info, @value{GDBN} has to find out the type
17709 to return from user. Specifying a different type by mistake may set the value
17710 in different inferior registers than the caller code expects. For example,
17711 typing @kbd{return -1} with its implicit type @code{int} would set only a part
17712 of a @code{long long int} result for a debug info less function (on 32-bit
17713 architectures). Therefore the user is required to specify the return type by
17714 an appropriate cast explicitly:
17717 Breakpoint 2, 0x0040050b in func ()
17718 (@value{GDBP}) return -1
17719 Return value type not available for selected stack frame.
17720 Please use an explicit cast of the value to return.
17721 (@value{GDBP}) return (long long int) -1
17722 Make selected stack frame return now? (y or n) y
17723 #0 0x00400526 in main ()
17728 @section Calling Program Functions
17731 @cindex calling functions
17732 @cindex inferior functions, calling
17733 @item print @var{expr}
17734 Evaluate the expression @var{expr} and display the resulting value.
17735 The expression may include calls to functions in the program being
17739 @item call @var{expr}
17740 Evaluate the expression @var{expr} without displaying @code{void}
17743 You can use this variant of the @code{print} command if you want to
17744 execute a function from your program that does not return anything
17745 (a.k.a.@: @dfn{a void function}), but without cluttering the output
17746 with @code{void} returned values that @value{GDBN} will otherwise
17747 print. If the result is not void, it is printed and saved in the
17751 It is possible for the function you call via the @code{print} or
17752 @code{call} command to generate a signal (e.g., if there's a bug in
17753 the function, or if you passed it incorrect arguments). What happens
17754 in that case is controlled by the @code{set unwindonsignal} command.
17756 Similarly, with a C@t{++} program it is possible for the function you
17757 call via the @code{print} or @code{call} command to generate an
17758 exception that is not handled due to the constraints of the dummy
17759 frame. In this case, any exception that is raised in the frame, but has
17760 an out-of-frame exception handler will not be found. GDB builds a
17761 dummy-frame for the inferior function call, and the unwinder cannot
17762 seek for exception handlers outside of this dummy-frame. What happens
17763 in that case is controlled by the
17764 @code{set unwind-on-terminating-exception} command.
17767 @item set unwindonsignal
17768 @kindex set unwindonsignal
17769 @cindex unwind stack in called functions
17770 @cindex call dummy stack unwinding
17771 Set unwinding of the stack if a signal is received while in a function
17772 that @value{GDBN} called in the program being debugged. If set to on,
17773 @value{GDBN} unwinds the stack it created for the call and restores
17774 the context to what it was before the call. If set to off (the
17775 default), @value{GDBN} stops in the frame where the signal was
17778 @item show unwindonsignal
17779 @kindex show unwindonsignal
17780 Show the current setting of stack unwinding in the functions called by
17783 @item set unwind-on-terminating-exception
17784 @kindex set unwind-on-terminating-exception
17785 @cindex unwind stack in called functions with unhandled exceptions
17786 @cindex call dummy stack unwinding on unhandled exception.
17787 Set unwinding of the stack if a C@t{++} exception is raised, but left
17788 unhandled while in a function that @value{GDBN} called in the program being
17789 debugged. If set to on (the default), @value{GDBN} unwinds the stack
17790 it created for the call and restores the context to what it was before
17791 the call. If set to off, @value{GDBN} the exception is delivered to
17792 the default C@t{++} exception handler and the inferior terminated.
17794 @item show unwind-on-terminating-exception
17795 @kindex show unwind-on-terminating-exception
17796 Show the current setting of stack unwinding in the functions called by
17801 @cindex weak alias functions
17802 Sometimes, a function you wish to call is actually a @dfn{weak alias}
17803 for another function. In such case, @value{GDBN} might not pick up
17804 the type information, including the types of the function arguments,
17805 which causes @value{GDBN} to call the inferior function incorrectly.
17806 As a result, the called function will function erroneously and may
17807 even crash. A solution to that is to use the name of the aliased
17811 @section Patching Programs
17813 @cindex patching binaries
17814 @cindex writing into executables
17815 @cindex writing into corefiles
17817 By default, @value{GDBN} opens the file containing your program's
17818 executable code (or the corefile) read-only. This prevents accidental
17819 alterations to machine code; but it also prevents you from intentionally
17820 patching your program's binary.
17822 If you'd like to be able to patch the binary, you can specify that
17823 explicitly with the @code{set write} command. For example, you might
17824 want to turn on internal debugging flags, or even to make emergency
17830 @itemx set write off
17831 If you specify @samp{set write on}, @value{GDBN} opens executable and
17832 core files for both reading and writing; if you specify @kbd{set write
17833 off} (the default), @value{GDBN} opens them read-only.
17835 If you have already loaded a file, you must load it again (using the
17836 @code{exec-file} or @code{core-file} command) after changing @code{set
17837 write}, for your new setting to take effect.
17841 Display whether executable files and core files are opened for writing
17842 as well as reading.
17845 @node Compiling and Injecting Code
17846 @section Compiling and injecting code in @value{GDBN}
17847 @cindex injecting code
17848 @cindex writing into executables
17849 @cindex compiling code
17851 @value{GDBN} supports on-demand compilation and code injection into
17852 programs running under @value{GDBN}. GCC 5.0 or higher built with
17853 @file{libcc1.so} must be installed for this functionality to be enabled.
17854 This functionality is implemented with the following commands.
17857 @kindex compile code
17858 @item compile code @var{source-code}
17859 @itemx compile code -raw @var{--} @var{source-code}
17860 Compile @var{source-code} with the compiler language found as the current
17861 language in @value{GDBN} (@pxref{Languages}). If compilation and
17862 injection is not supported with the current language specified in
17863 @value{GDBN}, or the compiler does not support this feature, an error
17864 message will be printed. If @var{source-code} compiles and links
17865 successfully, @value{GDBN} will load the object-code emitted,
17866 and execute it within the context of the currently selected inferior.
17867 It is important to note that the compiled code is executed immediately.
17868 After execution, the compiled code is removed from @value{GDBN} and any
17869 new types or variables you have defined will be deleted.
17871 The command allows you to specify @var{source-code} in two ways.
17872 The simplest method is to provide a single line of code to the command.
17876 compile code printf ("hello world\n");
17879 If you specify options on the command line as well as source code, they
17880 may conflict. The @samp{--} delimiter can be used to separate options
17881 from actual source code. E.g.:
17884 compile code -r -- printf ("hello world\n");
17887 Alternatively you can enter source code as multiple lines of text. To
17888 enter this mode, invoke the @samp{compile code} command without any text
17889 following the command. This will start the multiple-line editor and
17890 allow you to type as many lines of source code as required. When you
17891 have completed typing, enter @samp{end} on its own line to exit the
17896 >printf ("hello\n");
17897 >printf ("world\n");
17901 Specifying @samp{-raw}, prohibits @value{GDBN} from wrapping the
17902 provided @var{source-code} in a callable scope. In this case, you must
17903 specify the entry point of the code by defining a function named
17904 @code{_gdb_expr_}. The @samp{-raw} code cannot access variables of the
17905 inferior. Using @samp{-raw} option may be needed for example when
17906 @var{source-code} requires @samp{#include} lines which may conflict with
17907 inferior symbols otherwise.
17909 @kindex compile file
17910 @item compile file @var{filename}
17911 @itemx compile file -raw @var{filename}
17912 Like @code{compile code}, but take the source code from @var{filename}.
17915 compile file /home/user/example.c
17920 @item compile print @var{expr}
17921 @itemx compile print /@var{f} @var{expr}
17922 Compile and execute @var{expr} with the compiler language found as the
17923 current language in @value{GDBN} (@pxref{Languages}). By default the
17924 value of @var{expr} is printed in a format appropriate to its data type;
17925 you can choose a different format by specifying @samp{/@var{f}}, where
17926 @var{f} is a letter specifying the format; see @ref{Output Formats,,Output
17929 @item compile print
17930 @itemx compile print /@var{f}
17931 @cindex reprint the last value
17932 Alternatively you can enter the expression (source code producing it) as
17933 multiple lines of text. To enter this mode, invoke the @samp{compile print}
17934 command without any text following the command. This will start the
17935 multiple-line editor.
17939 The process of compiling and injecting the code can be inspected using:
17942 @anchor{set debug compile}
17943 @item set debug compile
17944 @cindex compile command debugging info
17945 Turns on or off display of @value{GDBN} process of compiling and
17946 injecting the code. The default is off.
17948 @item show debug compile
17949 Displays the current state of displaying @value{GDBN} process of
17950 compiling and injecting the code.
17953 @subsection Compilation options for the @code{compile} command
17955 @value{GDBN} needs to specify the right compilation options for the code
17956 to be injected, in part to make its ABI compatible with the inferior
17957 and in part to make the injected code compatible with @value{GDBN}'s
17961 The options used, in increasing precedence:
17964 @item target architecture and OS options (@code{gdbarch})
17965 These options depend on target processor type and target operating
17966 system, usually they specify at least 32-bit (@code{-m32}) or 64-bit
17967 (@code{-m64}) compilation option.
17969 @item compilation options recorded in the target
17970 @value{NGCC} (since version 4.7) stores the options used for compilation
17971 into @code{DW_AT_producer} part of DWARF debugging information according
17972 to the @value{NGCC} option @code{-grecord-gcc-switches}. One has to
17973 explicitly specify @code{-g} during inferior compilation otherwise
17974 @value{NGCC} produces no DWARF. This feature is only relevant for
17975 platforms where @code{-g} produces DWARF by default, otherwise one may
17976 try to enforce DWARF by using @code{-gdwarf-4}.
17978 @item compilation options set by @code{set compile-args}
17982 You can override compilation options using the following command:
17985 @item set compile-args
17986 @cindex compile command options override
17987 Set compilation options used for compiling and injecting code with the
17988 @code{compile} commands. These options override any conflicting ones
17989 from the target architecture and/or options stored during inferior
17992 @item show compile-args
17993 Displays the current state of compilation options override.
17994 This does not show all the options actually used during compilation,
17995 use @ref{set debug compile} for that.
17998 @subsection Caveats when using the @code{compile} command
18000 There are a few caveats to keep in mind when using the @code{compile}
18001 command. As the caveats are different per language, the table below
18002 highlights specific issues on a per language basis.
18005 @item C code examples and caveats
18006 When the language in @value{GDBN} is set to @samp{C}, the compiler will
18007 attempt to compile the source code with a @samp{C} compiler. The source
18008 code provided to the @code{compile} command will have much the same
18009 access to variables and types as it normally would if it were part of
18010 the program currently being debugged in @value{GDBN}.
18012 Below is a sample program that forms the basis of the examples that
18013 follow. This program has been compiled and loaded into @value{GDBN},
18014 much like any other normal debugging session.
18017 void function1 (void)
18020 printf ("function 1\n");
18023 void function2 (void)
18038 For the purposes of the examples in this section, the program above has
18039 been compiled, loaded into @value{GDBN}, stopped at the function
18040 @code{main}, and @value{GDBN} is awaiting input from the user.
18042 To access variables and types for any program in @value{GDBN}, the
18043 program must be compiled and packaged with debug information. The
18044 @code{compile} command is not an exception to this rule. Without debug
18045 information, you can still use the @code{compile} command, but you will
18046 be very limited in what variables and types you can access.
18048 So with that in mind, the example above has been compiled with debug
18049 information enabled. The @code{compile} command will have access to
18050 all variables and types (except those that may have been optimized
18051 out). Currently, as @value{GDBN} has stopped the program in the
18052 @code{main} function, the @code{compile} command would have access to
18053 the variable @code{k}. You could invoke the @code{compile} command
18054 and type some source code to set the value of @code{k}. You can also
18055 read it, or do anything with that variable you would normally do in
18056 @code{C}. Be aware that changes to inferior variables in the
18057 @code{compile} command are persistent. In the following example:
18060 compile code k = 3;
18064 the variable @code{k} is now 3. It will retain that value until
18065 something else in the example program changes it, or another
18066 @code{compile} command changes it.
18068 Normal scope and access rules apply to source code compiled and
18069 injected by the @code{compile} command. In the example, the variables
18070 @code{j} and @code{k} are not accessible yet, because the program is
18071 currently stopped in the @code{main} function, where these variables
18072 are not in scope. Therefore, the following command
18075 compile code j = 3;
18079 will result in a compilation error message.
18081 Once the program is continued, execution will bring these variables in
18082 scope, and they will become accessible; then the code you specify via
18083 the @code{compile} command will be able to access them.
18085 You can create variables and types with the @code{compile} command as
18086 part of your source code. Variables and types that are created as part
18087 of the @code{compile} command are not visible to the rest of the program for
18088 the duration of its run. This example is valid:
18091 compile code int ff = 5; printf ("ff is %d\n", ff);
18094 However, if you were to type the following into @value{GDBN} after that
18095 command has completed:
18098 compile code printf ("ff is %d\n'', ff);
18102 a compiler error would be raised as the variable @code{ff} no longer
18103 exists. Object code generated and injected by the @code{compile}
18104 command is removed when its execution ends. Caution is advised
18105 when assigning to program variables values of variables created by the
18106 code submitted to the @code{compile} command. This example is valid:
18109 compile code int ff = 5; k = ff;
18112 The value of the variable @code{ff} is assigned to @code{k}. The variable
18113 @code{k} does not require the existence of @code{ff} to maintain the value
18114 it has been assigned. However, pointers require particular care in
18115 assignment. If the source code compiled with the @code{compile} command
18116 changed the address of a pointer in the example program, perhaps to a
18117 variable created in the @code{compile} command, that pointer would point
18118 to an invalid location when the command exits. The following example
18119 would likely cause issues with your debugged program:
18122 compile code int ff = 5; p = &ff;
18125 In this example, @code{p} would point to @code{ff} when the
18126 @code{compile} command is executing the source code provided to it.
18127 However, as variables in the (example) program persist with their
18128 assigned values, the variable @code{p} would point to an invalid
18129 location when the command exists. A general rule should be followed
18130 in that you should either assign @code{NULL} to any assigned pointers,
18131 or restore a valid location to the pointer before the command exits.
18133 Similar caution must be exercised with any structs, unions, and typedefs
18134 defined in @code{compile} command. Types defined in the @code{compile}
18135 command will no longer be available in the next @code{compile} command.
18136 Therefore, if you cast a variable to a type defined in the
18137 @code{compile} command, care must be taken to ensure that any future
18138 need to resolve the type can be achieved.
18141 (gdb) compile code static struct a @{ int a; @} v = @{ 42 @}; argv = &v;
18142 (gdb) compile code printf ("%d\n", ((struct a *) argv)->a);
18143 gdb command line:1:36: error: dereferencing pointer to incomplete type ‘struct a’
18144 Compilation failed.
18145 (gdb) compile code struct a @{ int a; @}; printf ("%d\n", ((struct a *) argv)->a);
18149 Variables that have been optimized away by the compiler are not
18150 accessible to the code submitted to the @code{compile} command.
18151 Access to those variables will generate a compiler error which @value{GDBN}
18152 will print to the console.
18155 @subsection Compiler search for the @code{compile} command
18157 @value{GDBN} needs to find @value{NGCC} for the inferior being debugged which
18158 may not be obvious for remote targets of different architecture than where
18159 @value{GDBN} is running. Environment variable @code{PATH} (@code{PATH} from
18160 shell that executed @value{GDBN}, not the one set by @value{GDBN}
18161 command @code{set environment}). @xref{Environment}. @code{PATH} on
18162 @value{GDBN} host is searched for @value{NGCC} binary matching the
18163 target architecture and operating system.
18165 Specifically @code{PATH} is searched for binaries matching regular expression
18166 @code{@var{arch}(-[^-]*)?-@var{os}-gcc} according to the inferior target being
18167 debugged. @var{arch} is processor name --- multiarch is supported, so for
18168 example both @code{i386} and @code{x86_64} targets look for pattern
18169 @code{(x86_64|i.86)} and both @code{s390} and @code{s390x} targets look
18170 for pattern @code{s390x?}. @var{os} is currently supported only for
18171 pattern @code{linux(-gnu)?}.
18174 @chapter @value{GDBN} Files
18176 @value{GDBN} needs to know the file name of the program to be debugged,
18177 both in order to read its symbol table and in order to start your
18178 program. To debug a core dump of a previous run, you must also tell
18179 @value{GDBN} the name of the core dump file.
18182 * Files:: Commands to specify files
18183 * File Caching:: Information about @value{GDBN}'s file caching
18184 * Separate Debug Files:: Debugging information in separate files
18185 * MiniDebugInfo:: Debugging information in a special section
18186 * Index Files:: Index files speed up GDB
18187 * Symbol Errors:: Errors reading symbol files
18188 * Data Files:: GDB data files
18192 @section Commands to Specify Files
18194 @cindex symbol table
18195 @cindex core dump file
18197 You may want to specify executable and core dump file names. The usual
18198 way to do this is at start-up time, using the arguments to
18199 @value{GDBN}'s start-up commands (@pxref{Invocation, , Getting In and
18200 Out of @value{GDBN}}).
18202 Occasionally it is necessary to change to a different file during a
18203 @value{GDBN} session. Or you may run @value{GDBN} and forget to
18204 specify a file you want to use. Or you are debugging a remote target
18205 via @code{gdbserver} (@pxref{Server, file, Using the @code{gdbserver}
18206 Program}). In these situations the @value{GDBN} commands to specify
18207 new files are useful.
18210 @cindex executable file
18212 @item file @var{filename}
18213 Use @var{filename} as the program to be debugged. It is read for its
18214 symbols and for the contents of pure memory. It is also the program
18215 executed when you use the @code{run} command. If you do not specify a
18216 directory and the file is not found in the @value{GDBN} working directory,
18217 @value{GDBN} uses the environment variable @code{PATH} as a list of
18218 directories to search, just as the shell does when looking for a program
18219 to run. You can change the value of this variable, for both @value{GDBN}
18220 and your program, using the @code{path} command.
18222 @cindex unlinked object files
18223 @cindex patching object files
18224 You can load unlinked object @file{.o} files into @value{GDBN} using
18225 the @code{file} command. You will not be able to ``run'' an object
18226 file, but you can disassemble functions and inspect variables. Also,
18227 if the underlying BFD functionality supports it, you could use
18228 @kbd{gdb -write} to patch object files using this technique. Note
18229 that @value{GDBN} can neither interpret nor modify relocations in this
18230 case, so branches and some initialized variables will appear to go to
18231 the wrong place. But this feature is still handy from time to time.
18234 @code{file} with no argument makes @value{GDBN} discard any information it
18235 has on both executable file and the symbol table.
18238 @item exec-file @r{[} @var{filename} @r{]}
18239 Specify that the program to be run (but not the symbol table) is found
18240 in @var{filename}. @value{GDBN} searches the environment variable @code{PATH}
18241 if necessary to locate your program. Omitting @var{filename} means to
18242 discard information on the executable file.
18244 @kindex symbol-file
18245 @item symbol-file @r{[} @var{filename} @r{]}
18246 Read symbol table information from file @var{filename}. @code{PATH} is
18247 searched when necessary. Use the @code{file} command to get both symbol
18248 table and program to run from the same file.
18250 @code{symbol-file} with no argument clears out @value{GDBN} information on your
18251 program's symbol table.
18253 The @code{symbol-file} command causes @value{GDBN} to forget the contents of
18254 some breakpoints and auto-display expressions. This is because they may
18255 contain pointers to the internal data recording symbols and data types,
18256 which are part of the old symbol table data being discarded inside
18259 @code{symbol-file} does not repeat if you press @key{RET} again after
18262 When @value{GDBN} is configured for a particular environment, it
18263 understands debugging information in whatever format is the standard
18264 generated for that environment; you may use either a @sc{gnu} compiler, or
18265 other compilers that adhere to the local conventions.
18266 Best results are usually obtained from @sc{gnu} compilers; for example,
18267 using @code{@value{NGCC}} you can generate debugging information for
18270 For most kinds of object files, with the exception of old SVR3 systems
18271 using COFF, the @code{symbol-file} command does not normally read the
18272 symbol table in full right away. Instead, it scans the symbol table
18273 quickly to find which source files and which symbols are present. The
18274 details are read later, one source file at a time, as they are needed.
18276 The purpose of this two-stage reading strategy is to make @value{GDBN}
18277 start up faster. For the most part, it is invisible except for
18278 occasional pauses while the symbol table details for a particular source
18279 file are being read. (The @code{set verbose} command can turn these
18280 pauses into messages if desired. @xref{Messages/Warnings, ,Optional
18281 Warnings and Messages}.)
18283 We have not implemented the two-stage strategy for COFF yet. When the
18284 symbol table is stored in COFF format, @code{symbol-file} reads the
18285 symbol table data in full right away. Note that ``stabs-in-COFF''
18286 still does the two-stage strategy, since the debug info is actually
18290 @cindex reading symbols immediately
18291 @cindex symbols, reading immediately
18292 @item symbol-file @r{[} -readnow @r{]} @var{filename}
18293 @itemx file @r{[} -readnow @r{]} @var{filename}
18294 You can override the @value{GDBN} two-stage strategy for reading symbol
18295 tables by using the @samp{-readnow} option with any of the commands that
18296 load symbol table information, if you want to be sure @value{GDBN} has the
18297 entire symbol table available.
18299 @c FIXME: for now no mention of directories, since this seems to be in
18300 @c flux. 13mar1992 status is that in theory GDB would look either in
18301 @c current dir or in same dir as myprog; but issues like competing
18302 @c GDB's, or clutter in system dirs, mean that in practice right now
18303 @c only current dir is used. FFish says maybe a special GDB hierarchy
18304 @c (eg rooted in val of env var GDBSYMS) could exist for mappable symbol
18308 @item core-file @r{[}@var{filename}@r{]}
18310 Specify the whereabouts of a core dump file to be used as the ``contents
18311 of memory''. Traditionally, core files contain only some parts of the
18312 address space of the process that generated them; @value{GDBN} can access the
18313 executable file itself for other parts.
18315 @code{core-file} with no argument specifies that no core file is
18318 Note that the core file is ignored when your program is actually running
18319 under @value{GDBN}. So, if you have been running your program and you
18320 wish to debug a core file instead, you must kill the subprocess in which
18321 the program is running. To do this, use the @code{kill} command
18322 (@pxref{Kill Process, ,Killing the Child Process}).
18324 @kindex add-symbol-file
18325 @cindex dynamic linking
18326 @item add-symbol-file @var{filename} @var{address}
18327 @itemx add-symbol-file @var{filename} @var{address} @r{[} -readnow @r{]}
18328 @itemx add-symbol-file @var{filename} @var{address} -s @var{section} @var{address} @dots{}
18329 The @code{add-symbol-file} command reads additional symbol table
18330 information from the file @var{filename}. You would use this command
18331 when @var{filename} has been dynamically loaded (by some other means)
18332 into the program that is running. The @var{address} should give the memory
18333 address at which the file has been loaded; @value{GDBN} cannot figure
18334 this out for itself. You can additionally specify an arbitrary number
18335 of @samp{-s @var{section} @var{address}} pairs, to give an explicit
18336 section name and base address for that section. You can specify any
18337 @var{address} as an expression.
18339 The symbol table of the file @var{filename} is added to the symbol table
18340 originally read with the @code{symbol-file} command. You can use the
18341 @code{add-symbol-file} command any number of times; the new symbol data
18342 thus read is kept in addition to the old.
18344 Changes can be reverted using the command @code{remove-symbol-file}.
18346 @cindex relocatable object files, reading symbols from
18347 @cindex object files, relocatable, reading symbols from
18348 @cindex reading symbols from relocatable object files
18349 @cindex symbols, reading from relocatable object files
18350 @cindex @file{.o} files, reading symbols from
18351 Although @var{filename} is typically a shared library file, an
18352 executable file, or some other object file which has been fully
18353 relocated for loading into a process, you can also load symbolic
18354 information from relocatable @file{.o} files, as long as:
18358 the file's symbolic information refers only to linker symbols defined in
18359 that file, not to symbols defined by other object files,
18361 every section the file's symbolic information refers to has actually
18362 been loaded into the inferior, as it appears in the file, and
18364 you can determine the address at which every section was loaded, and
18365 provide these to the @code{add-symbol-file} command.
18369 Some embedded operating systems, like Sun Chorus and VxWorks, can load
18370 relocatable files into an already running program; such systems
18371 typically make the requirements above easy to meet. However, it's
18372 important to recognize that many native systems use complex link
18373 procedures (@code{.linkonce} section factoring and C@t{++} constructor table
18374 assembly, for example) that make the requirements difficult to meet. In
18375 general, one cannot assume that using @code{add-symbol-file} to read a
18376 relocatable object file's symbolic information will have the same effect
18377 as linking the relocatable object file into the program in the normal
18380 @code{add-symbol-file} does not repeat if you press @key{RET} after using it.
18382 @kindex remove-symbol-file
18383 @item remove-symbol-file @var{filename}
18384 @item remove-symbol-file -a @var{address}
18385 Remove a symbol file added via the @code{add-symbol-file} command. The
18386 file to remove can be identified by its @var{filename} or by an @var{address}
18387 that lies within the boundaries of this symbol file in memory. Example:
18390 (gdb) add-symbol-file /home/user/gdb/mylib.so 0x7ffff7ff9480
18391 add symbol table from file "/home/user/gdb/mylib.so" at
18392 .text_addr = 0x7ffff7ff9480
18394 Reading symbols from /home/user/gdb/mylib.so...done.
18395 (gdb) remove-symbol-file -a 0x7ffff7ff9480
18396 Remove symbol table from file "/home/user/gdb/mylib.so"? (y or n) y
18401 @code{remove-symbol-file} does not repeat if you press @key{RET} after using it.
18403 @kindex add-symbol-file-from-memory
18404 @cindex @code{syscall DSO}
18405 @cindex load symbols from memory
18406 @item add-symbol-file-from-memory @var{address}
18407 Load symbols from the given @var{address} in a dynamically loaded
18408 object file whose image is mapped directly into the inferior's memory.
18409 For example, the Linux kernel maps a @code{syscall DSO} into each
18410 process's address space; this DSO provides kernel-specific code for
18411 some system calls. The argument can be any expression whose
18412 evaluation yields the address of the file's shared object file header.
18413 For this command to work, you must have used @code{symbol-file} or
18414 @code{exec-file} commands in advance.
18417 @item section @var{section} @var{addr}
18418 The @code{section} command changes the base address of the named
18419 @var{section} of the exec file to @var{addr}. This can be used if the
18420 exec file does not contain section addresses, (such as in the
18421 @code{a.out} format), or when the addresses specified in the file
18422 itself are wrong. Each section must be changed separately. The
18423 @code{info files} command, described below, lists all the sections and
18427 @kindex info target
18430 @code{info files} and @code{info target} are synonymous; both print the
18431 current target (@pxref{Targets, ,Specifying a Debugging Target}),
18432 including the names of the executable and core dump files currently in
18433 use by @value{GDBN}, and the files from which symbols were loaded. The
18434 command @code{help target} lists all possible targets rather than
18437 @kindex maint info sections
18438 @item maint info sections
18439 Another command that can give you extra information about program sections
18440 is @code{maint info sections}. In addition to the section information
18441 displayed by @code{info files}, this command displays the flags and file
18442 offset of each section in the executable and core dump files. In addition,
18443 @code{maint info sections} provides the following command options (which
18444 may be arbitrarily combined):
18448 Display sections for all loaded object files, including shared libraries.
18449 @item @var{sections}
18450 Display info only for named @var{sections}.
18451 @item @var{section-flags}
18452 Display info only for sections for which @var{section-flags} are true.
18453 The section flags that @value{GDBN} currently knows about are:
18456 Section will have space allocated in the process when loaded.
18457 Set for all sections except those containing debug information.
18459 Section will be loaded from the file into the child process memory.
18460 Set for pre-initialized code and data, clear for @code{.bss} sections.
18462 Section needs to be relocated before loading.
18464 Section cannot be modified by the child process.
18466 Section contains executable code only.
18468 Section contains data only (no executable code).
18470 Section will reside in ROM.
18472 Section contains data for constructor/destructor lists.
18474 Section is not empty.
18476 An instruction to the linker to not output the section.
18477 @item COFF_SHARED_LIBRARY
18478 A notification to the linker that the section contains
18479 COFF shared library information.
18481 Section contains common symbols.
18484 @kindex set trust-readonly-sections
18485 @cindex read-only sections
18486 @item set trust-readonly-sections on
18487 Tell @value{GDBN} that readonly sections in your object file
18488 really are read-only (i.e.@: that their contents will not change).
18489 In that case, @value{GDBN} can fetch values from these sections
18490 out of the object file, rather than from the target program.
18491 For some targets (notably embedded ones), this can be a significant
18492 enhancement to debugging performance.
18494 The default is off.
18496 @item set trust-readonly-sections off
18497 Tell @value{GDBN} not to trust readonly sections. This means that
18498 the contents of the section might change while the program is running,
18499 and must therefore be fetched from the target when needed.
18501 @item show trust-readonly-sections
18502 Show the current setting of trusting readonly sections.
18505 All file-specifying commands allow both absolute and relative file names
18506 as arguments. @value{GDBN} always converts the file name to an absolute file
18507 name and remembers it that way.
18509 @cindex shared libraries
18510 @anchor{Shared Libraries}
18511 @value{GDBN} supports @sc{gnu}/Linux, MS-Windows, SunOS,
18512 Darwin/Mach-O, SVr4, IBM RS/6000 AIX, QNX Neutrino, FDPIC (FR-V), and
18513 DSBT (TIC6X) shared libraries.
18515 On MS-Windows @value{GDBN} must be linked with the Expat library to support
18516 shared libraries. @xref{Expat}.
18518 @value{GDBN} automatically loads symbol definitions from shared libraries
18519 when you use the @code{run} command, or when you examine a core file.
18520 (Before you issue the @code{run} command, @value{GDBN} does not understand
18521 references to a function in a shared library, however---unless you are
18522 debugging a core file).
18524 @c FIXME: some @value{GDBN} release may permit some refs to undef
18525 @c FIXME...symbols---eg in a break cmd---assuming they are from a shared
18526 @c FIXME...lib; check this from time to time when updating manual
18528 There are times, however, when you may wish to not automatically load
18529 symbol definitions from shared libraries, such as when they are
18530 particularly large or there are many of them.
18532 To control the automatic loading of shared library symbols, use the
18536 @kindex set auto-solib-add
18537 @item set auto-solib-add @var{mode}
18538 If @var{mode} is @code{on}, symbols from all shared object libraries
18539 will be loaded automatically when the inferior begins execution, you
18540 attach to an independently started inferior, or when the dynamic linker
18541 informs @value{GDBN} that a new library has been loaded. If @var{mode}
18542 is @code{off}, symbols must be loaded manually, using the
18543 @code{sharedlibrary} command. The default value is @code{on}.
18545 @cindex memory used for symbol tables
18546 If your program uses lots of shared libraries with debug info that
18547 takes large amounts of memory, you can decrease the @value{GDBN}
18548 memory footprint by preventing it from automatically loading the
18549 symbols from shared libraries. To that end, type @kbd{set
18550 auto-solib-add off} before running the inferior, then load each
18551 library whose debug symbols you do need with @kbd{sharedlibrary
18552 @var{regexp}}, where @var{regexp} is a regular expression that matches
18553 the libraries whose symbols you want to be loaded.
18555 @kindex show auto-solib-add
18556 @item show auto-solib-add
18557 Display the current autoloading mode.
18560 @cindex load shared library
18561 To explicitly load shared library symbols, use the @code{sharedlibrary}
18565 @kindex info sharedlibrary
18567 @item info share @var{regex}
18568 @itemx info sharedlibrary @var{regex}
18569 Print the names of the shared libraries which are currently loaded
18570 that match @var{regex}. If @var{regex} is omitted then print
18571 all shared libraries that are loaded.
18574 @item info dll @var{regex}
18575 This is an alias of @code{info sharedlibrary}.
18577 @kindex sharedlibrary
18579 @item sharedlibrary @var{regex}
18580 @itemx share @var{regex}
18581 Load shared object library symbols for files matching a
18582 Unix regular expression.
18583 As with files loaded automatically, it only loads shared libraries
18584 required by your program for a core file or after typing @code{run}. If
18585 @var{regex} is omitted all shared libraries required by your program are
18588 @item nosharedlibrary
18589 @kindex nosharedlibrary
18590 @cindex unload symbols from shared libraries
18591 Unload all shared object library symbols. This discards all symbols
18592 that have been loaded from all shared libraries. Symbols from shared
18593 libraries that were loaded by explicit user requests are not
18597 Sometimes you may wish that @value{GDBN} stops and gives you control
18598 when any of shared library events happen. The best way to do this is
18599 to use @code{catch load} and @code{catch unload} (@pxref{Set
18602 @value{GDBN} also supports the the @code{set stop-on-solib-events}
18603 command for this. This command exists for historical reasons. It is
18604 less useful than setting a catchpoint, because it does not allow for
18605 conditions or commands as a catchpoint does.
18608 @item set stop-on-solib-events
18609 @kindex set stop-on-solib-events
18610 This command controls whether @value{GDBN} should give you control
18611 when the dynamic linker notifies it about some shared library event.
18612 The most common event of interest is loading or unloading of a new
18615 @item show stop-on-solib-events
18616 @kindex show stop-on-solib-events
18617 Show whether @value{GDBN} stops and gives you control when shared
18618 library events happen.
18621 Shared libraries are also supported in many cross or remote debugging
18622 configurations. @value{GDBN} needs to have access to the target's libraries;
18623 this can be accomplished either by providing copies of the libraries
18624 on the host system, or by asking @value{GDBN} to automatically retrieve the
18625 libraries from the target. If copies of the target libraries are
18626 provided, they need to be the same as the target libraries, although the
18627 copies on the target can be stripped as long as the copies on the host are
18630 @cindex where to look for shared libraries
18631 For remote debugging, you need to tell @value{GDBN} where the target
18632 libraries are, so that it can load the correct copies---otherwise, it
18633 may try to load the host's libraries. @value{GDBN} has two variables
18634 to specify the search directories for target libraries.
18637 @cindex prefix for executable and shared library file names
18638 @cindex system root, alternate
18639 @kindex set solib-absolute-prefix
18640 @kindex set sysroot
18641 @item set sysroot @var{path}
18642 Use @var{path} as the system root for the program being debugged. Any
18643 absolute shared library paths will be prefixed with @var{path}; many
18644 runtime loaders store the absolute paths to the shared library in the
18645 target program's memory. When starting processes remotely, and when
18646 attaching to already-running processes (local or remote), their
18647 executable filenames will be prefixed with @var{path} if reported to
18648 @value{GDBN} as absolute by the operating system. If you use
18649 @code{set sysroot} to find executables and shared libraries, they need
18650 to be laid out in the same way that they are on the target, with
18651 e.g.@: a @file{/bin}, @file{/lib} and @file{/usr/lib} hierarchy under
18654 If @var{path} starts with the sequence @file{target:} and the target
18655 system is remote then @value{GDBN} will retrieve the target binaries
18656 from the remote system. This is only supported when using a remote
18657 target that supports the @code{remote get} command (@pxref{File
18658 Transfer,,Sending files to a remote system}). The part of @var{path}
18659 following the initial @file{target:} (if present) is used as system
18660 root prefix on the remote file system. If @var{path} starts with the
18661 sequence @file{remote:} this is converted to the sequence
18662 @file{target:} by @code{set sysroot}@footnote{Historically the
18663 functionality to retrieve binaries from the remote system was
18664 provided by prefixing @var{path} with @file{remote:}}. If you want
18665 to specify a local system root using a directory that happens to be
18666 named @file{target:} or @file{remote:}, you need to use some
18667 equivalent variant of the name like @file{./target:}.
18669 For targets with an MS-DOS based filesystem, such as MS-Windows and
18670 SymbianOS, @value{GDBN} tries prefixing a few variants of the target
18671 absolute file name with @var{path}. But first, on Unix hosts,
18672 @value{GDBN} converts all backslash directory separators into forward
18673 slashes, because the backslash is not a directory separator on Unix:
18676 c:\foo\bar.dll @result{} c:/foo/bar.dll
18679 Then, @value{GDBN} attempts prefixing the target file name with
18680 @var{path}, and looks for the resulting file name in the host file
18684 c:/foo/bar.dll @result{} /path/to/sysroot/c:/foo/bar.dll
18687 If that does not find the binary, @value{GDBN} tries removing
18688 the @samp{:} character from the drive spec, both for convenience, and,
18689 for the case of the host file system not supporting file names with
18693 c:/foo/bar.dll @result{} /path/to/sysroot/c/foo/bar.dll
18696 This makes it possible to have a system root that mirrors a target
18697 with more than one drive. E.g., you may want to setup your local
18698 copies of the target system shared libraries like so (note @samp{c} vs
18702 @file{/path/to/sysroot/c/sys/bin/foo.dll}
18703 @file{/path/to/sysroot/c/sys/bin/bar.dll}
18704 @file{/path/to/sysroot/z/sys/bin/bar.dll}
18708 and point the system root at @file{/path/to/sysroot}, so that
18709 @value{GDBN} can find the correct copies of both
18710 @file{c:\sys\bin\foo.dll}, and @file{z:\sys\bin\bar.dll}.
18712 If that still does not find the binary, @value{GDBN} tries
18713 removing the whole drive spec from the target file name:
18716 c:/foo/bar.dll @result{} /path/to/sysroot/foo/bar.dll
18719 This last lookup makes it possible to not care about the drive name,
18720 if you don't want or need to.
18722 The @code{set solib-absolute-prefix} command is an alias for @code{set
18725 @cindex default system root
18726 @cindex @samp{--with-sysroot}
18727 You can set the default system root by using the configure-time
18728 @samp{--with-sysroot} option. If the system root is inside
18729 @value{GDBN}'s configured binary prefix (set with @samp{--prefix} or
18730 @samp{--exec-prefix}), then the default system root will be updated
18731 automatically if the installed @value{GDBN} is moved to a new
18734 @kindex show sysroot
18736 Display the current executable and shared library prefix.
18738 @kindex set solib-search-path
18739 @item set solib-search-path @var{path}
18740 If this variable is set, @var{path} is a colon-separated list of
18741 directories to search for shared libraries. @samp{solib-search-path}
18742 is used after @samp{sysroot} fails to locate the library, or if the
18743 path to the library is relative instead of absolute. If you want to
18744 use @samp{solib-search-path} instead of @samp{sysroot}, be sure to set
18745 @samp{sysroot} to a nonexistent directory to prevent @value{GDBN} from
18746 finding your host's libraries. @samp{sysroot} is preferred; setting
18747 it to a nonexistent directory may interfere with automatic loading
18748 of shared library symbols.
18750 @kindex show solib-search-path
18751 @item show solib-search-path
18752 Display the current shared library search path.
18754 @cindex DOS file-name semantics of file names.
18755 @kindex set target-file-system-kind (unix|dos-based|auto)
18756 @kindex show target-file-system-kind
18757 @item set target-file-system-kind @var{kind}
18758 Set assumed file system kind for target reported file names.
18760 Shared library file names as reported by the target system may not
18761 make sense as is on the system @value{GDBN} is running on. For
18762 example, when remote debugging a target that has MS-DOS based file
18763 system semantics, from a Unix host, the target may be reporting to
18764 @value{GDBN} a list of loaded shared libraries with file names such as
18765 @file{c:\Windows\kernel32.dll}. On Unix hosts, there's no concept of
18766 drive letters, so the @samp{c:\} prefix is not normally understood as
18767 indicating an absolute file name, and neither is the backslash
18768 normally considered a directory separator character. In that case,
18769 the native file system would interpret this whole absolute file name
18770 as a relative file name with no directory components. This would make
18771 it impossible to point @value{GDBN} at a copy of the remote target's
18772 shared libraries on the host using @code{set sysroot}, and impractical
18773 with @code{set solib-search-path}. Setting
18774 @code{target-file-system-kind} to @code{dos-based} tells @value{GDBN}
18775 to interpret such file names similarly to how the target would, and to
18776 map them to file names valid on @value{GDBN}'s native file system
18777 semantics. The value of @var{kind} can be @code{"auto"}, in addition
18778 to one of the supported file system kinds. In that case, @value{GDBN}
18779 tries to determine the appropriate file system variant based on the
18780 current target's operating system (@pxref{ABI, ,Configuring the
18781 Current ABI}). The supported file system settings are:
18785 Instruct @value{GDBN} to assume the target file system is of Unix
18786 kind. Only file names starting the forward slash (@samp{/}) character
18787 are considered absolute, and the directory separator character is also
18791 Instruct @value{GDBN} to assume the target file system is DOS based.
18792 File names starting with either a forward slash, or a drive letter
18793 followed by a colon (e.g., @samp{c:}), are considered absolute, and
18794 both the slash (@samp{/}) and the backslash (@samp{\\}) characters are
18795 considered directory separators.
18798 Instruct @value{GDBN} to use the file system kind associated with the
18799 target operating system (@pxref{ABI, ,Configuring the Current ABI}).
18800 This is the default.
18804 @cindex file name canonicalization
18805 @cindex base name differences
18806 When processing file names provided by the user, @value{GDBN}
18807 frequently needs to compare them to the file names recorded in the
18808 program's debug info. Normally, @value{GDBN} compares just the
18809 @dfn{base names} of the files as strings, which is reasonably fast
18810 even for very large programs. (The base name of a file is the last
18811 portion of its name, after stripping all the leading directories.)
18812 This shortcut in comparison is based upon the assumption that files
18813 cannot have more than one base name. This is usually true, but
18814 references to files that use symlinks or similar filesystem
18815 facilities violate that assumption. If your program records files
18816 using such facilities, or if you provide file names to @value{GDBN}
18817 using symlinks etc., you can set @code{basenames-may-differ} to
18818 @code{true} to instruct @value{GDBN} to completely canonicalize each
18819 pair of file names it needs to compare. This will make file-name
18820 comparisons accurate, but at a price of a significant slowdown.
18823 @item set basenames-may-differ
18824 @kindex set basenames-may-differ
18825 Set whether a source file may have multiple base names.
18827 @item show basenames-may-differ
18828 @kindex show basenames-may-differ
18829 Show whether a source file may have multiple base names.
18833 @section File Caching
18834 @cindex caching of opened files
18835 @cindex caching of bfd objects
18837 To speed up file loading, and reduce memory usage, @value{GDBN} will
18838 reuse the @code{bfd} objects used to track open files. @xref{Top, ,
18839 BFD, bfd, The Binary File Descriptor Library}. The following commands
18840 allow visibility and control of the caching behavior.
18843 @kindex maint info bfds
18844 @item maint info bfds
18845 This prints information about each @code{bfd} object that is known to
18848 @kindex maint set bfd-sharing
18849 @kindex maint show bfd-sharing
18850 @kindex bfd caching
18851 @item maint set bfd-sharing
18852 @item maint show bfd-sharing
18853 Control whether @code{bfd} objects can be shared. When sharing is
18854 enabled @value{GDBN} reuses already open @code{bfd} objects rather
18855 than reopening the same file. Turning sharing off does not cause
18856 already shared @code{bfd} objects to be unshared, but all future files
18857 that are opened will create a new @code{bfd} object. Similarly,
18858 re-enabling sharing does not cause multiple existing @code{bfd}
18859 objects to be collapsed into a single shared @code{bfd} object.
18861 @kindex set debug bfd-cache @var{level}
18862 @kindex bfd caching
18863 @item set debug bfd-cache @var{level}
18864 Turns on debugging of the bfd cache, setting the level to @var{level}.
18866 @kindex show debug bfd-cache
18867 @kindex bfd caching
18868 @item show debug bfd-cache
18869 Show the current debugging level of the bfd cache.
18872 @node Separate Debug Files
18873 @section Debugging Information in Separate Files
18874 @cindex separate debugging information files
18875 @cindex debugging information in separate files
18876 @cindex @file{.debug} subdirectories
18877 @cindex debugging information directory, global
18878 @cindex global debugging information directories
18879 @cindex build ID, and separate debugging files
18880 @cindex @file{.build-id} directory
18882 @value{GDBN} allows you to put a program's debugging information in a
18883 file separate from the executable itself, in a way that allows
18884 @value{GDBN} to find and load the debugging information automatically.
18885 Since debugging information can be very large---sometimes larger
18886 than the executable code itself---some systems distribute debugging
18887 information for their executables in separate files, which users can
18888 install only when they need to debug a problem.
18890 @value{GDBN} supports two ways of specifying the separate debug info
18895 The executable contains a @dfn{debug link} that specifies the name of
18896 the separate debug info file. The separate debug file's name is
18897 usually @file{@var{executable}.debug}, where @var{executable} is the
18898 name of the corresponding executable file without leading directories
18899 (e.g., @file{ls.debug} for @file{/usr/bin/ls}). In addition, the
18900 debug link specifies a 32-bit @dfn{Cyclic Redundancy Check} (CRC)
18901 checksum for the debug file, which @value{GDBN} uses to validate that
18902 the executable and the debug file came from the same build.
18905 The executable contains a @dfn{build ID}, a unique bit string that is
18906 also present in the corresponding debug info file. (This is supported
18907 only on some operating systems, when using the ELF or PE file formats
18908 for binary files and the @sc{gnu} Binutils.) For more details about
18909 this feature, see the description of the @option{--build-id}
18910 command-line option in @ref{Options, , Command Line Options, ld.info,
18911 The GNU Linker}. The debug info file's name is not specified
18912 explicitly by the build ID, but can be computed from the build ID, see
18916 Depending on the way the debug info file is specified, @value{GDBN}
18917 uses two different methods of looking for the debug file:
18921 For the ``debug link'' method, @value{GDBN} looks up the named file in
18922 the directory of the executable file, then in a subdirectory of that
18923 directory named @file{.debug}, and finally under each one of the global debug
18924 directories, in a subdirectory whose name is identical to the leading
18925 directories of the executable's absolute file name.
18928 For the ``build ID'' method, @value{GDBN} looks in the
18929 @file{.build-id} subdirectory of each one of the global debug directories for
18930 a file named @file{@var{nn}/@var{nnnnnnnn}.debug}, where @var{nn} are the
18931 first 2 hex characters of the build ID bit string, and @var{nnnnnnnn}
18932 are the rest of the bit string. (Real build ID strings are 32 or more
18933 hex characters, not 10.)
18936 So, for example, suppose you ask @value{GDBN} to debug
18937 @file{/usr/bin/ls}, which has a debug link that specifies the
18938 file @file{ls.debug}, and a build ID whose value in hex is
18939 @code{abcdef1234}. If the list of the global debug directories includes
18940 @file{/usr/lib/debug}, then @value{GDBN} will look for the following
18941 debug information files, in the indicated order:
18945 @file{/usr/lib/debug/.build-id/ab/cdef1234.debug}
18947 @file{/usr/bin/ls.debug}
18949 @file{/usr/bin/.debug/ls.debug}
18951 @file{/usr/lib/debug/usr/bin/ls.debug}.
18954 @anchor{debug-file-directory}
18955 Global debugging info directories default to what is set by @value{GDBN}
18956 configure option @option{--with-separate-debug-dir}. During @value{GDBN} run
18957 you can also set the global debugging info directories, and view the list
18958 @value{GDBN} is currently using.
18962 @kindex set debug-file-directory
18963 @item set debug-file-directory @var{directories}
18964 Set the directories which @value{GDBN} searches for separate debugging
18965 information files to @var{directory}. Multiple path components can be set
18966 concatenating them by a path separator.
18968 @kindex show debug-file-directory
18969 @item show debug-file-directory
18970 Show the directories @value{GDBN} searches for separate debugging
18975 @cindex @code{.gnu_debuglink} sections
18976 @cindex debug link sections
18977 A debug link is a special section of the executable file named
18978 @code{.gnu_debuglink}. The section must contain:
18982 A filename, with any leading directory components removed, followed by
18985 zero to three bytes of padding, as needed to reach the next four-byte
18986 boundary within the section, and
18988 a four-byte CRC checksum, stored in the same endianness used for the
18989 executable file itself. The checksum is computed on the debugging
18990 information file's full contents by the function given below, passing
18991 zero as the @var{crc} argument.
18994 Any executable file format can carry a debug link, as long as it can
18995 contain a section named @code{.gnu_debuglink} with the contents
18998 @cindex @code{.note.gnu.build-id} sections
18999 @cindex build ID sections
19000 The build ID is a special section in the executable file (and in other
19001 ELF binary files that @value{GDBN} may consider). This section is
19002 often named @code{.note.gnu.build-id}, but that name is not mandatory.
19003 It contains unique identification for the built files---the ID remains
19004 the same across multiple builds of the same build tree. The default
19005 algorithm SHA1 produces 160 bits (40 hexadecimal characters) of the
19006 content for the build ID string. The same section with an identical
19007 value is present in the original built binary with symbols, in its
19008 stripped variant, and in the separate debugging information file.
19010 The debugging information file itself should be an ordinary
19011 executable, containing a full set of linker symbols, sections, and
19012 debugging information. The sections of the debugging information file
19013 should have the same names, addresses, and sizes as the original file,
19014 but they need not contain any data---much like a @code{.bss} section
19015 in an ordinary executable.
19017 The @sc{gnu} binary utilities (Binutils) package includes the
19018 @samp{objcopy} utility that can produce
19019 the separated executable / debugging information file pairs using the
19020 following commands:
19023 @kbd{objcopy --only-keep-debug foo foo.debug}
19028 These commands remove the debugging
19029 information from the executable file @file{foo} and place it in the file
19030 @file{foo.debug}. You can use the first, second or both methods to link the
19035 The debug link method needs the following additional command to also leave
19036 behind a debug link in @file{foo}:
19039 @kbd{objcopy --add-gnu-debuglink=foo.debug foo}
19042 Ulrich Drepper's @file{elfutils} package, starting with version 0.53, contains
19043 a version of the @code{strip} command such that the command @kbd{strip foo -f
19044 foo.debug} has the same functionality as the two @code{objcopy} commands and
19045 the @code{ln -s} command above, together.
19048 Build ID gets embedded into the main executable using @code{ld --build-id} or
19049 the @value{NGCC} counterpart @code{gcc -Wl,--build-id}. Build ID support plus
19050 compatibility fixes for debug files separation are present in @sc{gnu} binary
19051 utilities (Binutils) package since version 2.18.
19056 @cindex CRC algorithm definition
19057 The CRC used in @code{.gnu_debuglink} is the CRC-32 defined in
19058 IEEE 802.3 using the polynomial:
19060 @c TexInfo requires naked braces for multi-digit exponents for Tex
19061 @c output, but this causes HTML output to barf. HTML has to be set using
19062 @c raw commands. So we end up having to specify this equation in 2
19067 <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>
19068 + <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
19074 @math{x^{32} + x^{26} + x^{23} + x^{22} + x^{16} + x^{12} + x^{11}}
19075 @math{+ x^{10} + x^8 + x^7 + x^5 + x^4 + x^2 + x + 1}
19079 The function is computed byte at a time, taking the least
19080 significant bit of each byte first. The initial pattern
19081 @code{0xffffffff} is used, to ensure leading zeros affect the CRC and
19082 the final result is inverted to ensure trailing zeros also affect the
19085 @emph{Note:} This is the same CRC polynomial as used in handling the
19086 @dfn{Remote Serial Protocol} @code{qCRC} packet (@pxref{qCRC packet}).
19087 However in the case of the Remote Serial Protocol, the CRC is computed
19088 @emph{most} significant bit first, and the result is not inverted, so
19089 trailing zeros have no effect on the CRC value.
19091 To complete the description, we show below the code of the function
19092 which produces the CRC used in @code{.gnu_debuglink}. Inverting the
19093 initially supplied @code{crc} argument means that an initial call to
19094 this function passing in zero will start computing the CRC using
19097 @kindex gnu_debuglink_crc32
19100 gnu_debuglink_crc32 (unsigned long crc,
19101 unsigned char *buf, size_t len)
19103 static const unsigned long crc32_table[256] =
19105 0x00000000, 0x77073096, 0xee0e612c, 0x990951ba, 0x076dc419,
19106 0x706af48f, 0xe963a535, 0x9e6495a3, 0x0edb8832, 0x79dcb8a4,
19107 0xe0d5e91e, 0x97d2d988, 0x09b64c2b, 0x7eb17cbd, 0xe7b82d07,
19108 0x90bf1d91, 0x1db71064, 0x6ab020f2, 0xf3b97148, 0x84be41de,
19109 0x1adad47d, 0x6ddde4eb, 0xf4d4b551, 0x83d385c7, 0x136c9856,
19110 0x646ba8c0, 0xfd62f97a, 0x8a65c9ec, 0x14015c4f, 0x63066cd9,
19111 0xfa0f3d63, 0x8d080df5, 0x3b6e20c8, 0x4c69105e, 0xd56041e4,
19112 0xa2677172, 0x3c03e4d1, 0x4b04d447, 0xd20d85fd, 0xa50ab56b,
19113 0x35b5a8fa, 0x42b2986c, 0xdbbbc9d6, 0xacbcf940, 0x32d86ce3,
19114 0x45df5c75, 0xdcd60dcf, 0xabd13d59, 0x26d930ac, 0x51de003a,
19115 0xc8d75180, 0xbfd06116, 0x21b4f4b5, 0x56b3c423, 0xcfba9599,
19116 0xb8bda50f, 0x2802b89e, 0x5f058808, 0xc60cd9b2, 0xb10be924,
19117 0x2f6f7c87, 0x58684c11, 0xc1611dab, 0xb6662d3d, 0x76dc4190,
19118 0x01db7106, 0x98d220bc, 0xefd5102a, 0x71b18589, 0x06b6b51f,
19119 0x9fbfe4a5, 0xe8b8d433, 0x7807c9a2, 0x0f00f934, 0x9609a88e,
19120 0xe10e9818, 0x7f6a0dbb, 0x086d3d2d, 0x91646c97, 0xe6635c01,
19121 0x6b6b51f4, 0x1c6c6162, 0x856530d8, 0xf262004e, 0x6c0695ed,
19122 0x1b01a57b, 0x8208f4c1, 0xf50fc457, 0x65b0d9c6, 0x12b7e950,
19123 0x8bbeb8ea, 0xfcb9887c, 0x62dd1ddf, 0x15da2d49, 0x8cd37cf3,
19124 0xfbd44c65, 0x4db26158, 0x3ab551ce, 0xa3bc0074, 0xd4bb30e2,
19125 0x4adfa541, 0x3dd895d7, 0xa4d1c46d, 0xd3d6f4fb, 0x4369e96a,
19126 0x346ed9fc, 0xad678846, 0xda60b8d0, 0x44042d73, 0x33031de5,
19127 0xaa0a4c5f, 0xdd0d7cc9, 0x5005713c, 0x270241aa, 0xbe0b1010,
19128 0xc90c2086, 0x5768b525, 0x206f85b3, 0xb966d409, 0xce61e49f,
19129 0x5edef90e, 0x29d9c998, 0xb0d09822, 0xc7d7a8b4, 0x59b33d17,
19130 0x2eb40d81, 0xb7bd5c3b, 0xc0ba6cad, 0xedb88320, 0x9abfb3b6,
19131 0x03b6e20c, 0x74b1d29a, 0xead54739, 0x9dd277af, 0x04db2615,
19132 0x73dc1683, 0xe3630b12, 0x94643b84, 0x0d6d6a3e, 0x7a6a5aa8,
19133 0xe40ecf0b, 0x9309ff9d, 0x0a00ae27, 0x7d079eb1, 0xf00f9344,
19134 0x8708a3d2, 0x1e01f268, 0x6906c2fe, 0xf762575d, 0x806567cb,
19135 0x196c3671, 0x6e6b06e7, 0xfed41b76, 0x89d32be0, 0x10da7a5a,
19136 0x67dd4acc, 0xf9b9df6f, 0x8ebeeff9, 0x17b7be43, 0x60b08ed5,
19137 0xd6d6a3e8, 0xa1d1937e, 0x38d8c2c4, 0x4fdff252, 0xd1bb67f1,
19138 0xa6bc5767, 0x3fb506dd, 0x48b2364b, 0xd80d2bda, 0xaf0a1b4c,
19139 0x36034af6, 0x41047a60, 0xdf60efc3, 0xa867df55, 0x316e8eef,
19140 0x4669be79, 0xcb61b38c, 0xbc66831a, 0x256fd2a0, 0x5268e236,
19141 0xcc0c7795, 0xbb0b4703, 0x220216b9, 0x5505262f, 0xc5ba3bbe,
19142 0xb2bd0b28, 0x2bb45a92, 0x5cb36a04, 0xc2d7ffa7, 0xb5d0cf31,
19143 0x2cd99e8b, 0x5bdeae1d, 0x9b64c2b0, 0xec63f226, 0x756aa39c,
19144 0x026d930a, 0x9c0906a9, 0xeb0e363f, 0x72076785, 0x05005713,
19145 0x95bf4a82, 0xe2b87a14, 0x7bb12bae, 0x0cb61b38, 0x92d28e9b,
19146 0xe5d5be0d, 0x7cdcefb7, 0x0bdbdf21, 0x86d3d2d4, 0xf1d4e242,
19147 0x68ddb3f8, 0x1fda836e, 0x81be16cd, 0xf6b9265b, 0x6fb077e1,
19148 0x18b74777, 0x88085ae6, 0xff0f6a70, 0x66063bca, 0x11010b5c,
19149 0x8f659eff, 0xf862ae69, 0x616bffd3, 0x166ccf45, 0xa00ae278,
19150 0xd70dd2ee, 0x4e048354, 0x3903b3c2, 0xa7672661, 0xd06016f7,
19151 0x4969474d, 0x3e6e77db, 0xaed16a4a, 0xd9d65adc, 0x40df0b66,
19152 0x37d83bf0, 0xa9bcae53, 0xdebb9ec5, 0x47b2cf7f, 0x30b5ffe9,
19153 0xbdbdf21c, 0xcabac28a, 0x53b39330, 0x24b4a3a6, 0xbad03605,
19154 0xcdd70693, 0x54de5729, 0x23d967bf, 0xb3667a2e, 0xc4614ab8,
19155 0x5d681b02, 0x2a6f2b94, 0xb40bbe37, 0xc30c8ea1, 0x5a05df1b,
19158 unsigned char *end;
19160 crc = ~crc & 0xffffffff;
19161 for (end = buf + len; buf < end; ++buf)
19162 crc = crc32_table[(crc ^ *buf) & 0xff] ^ (crc >> 8);
19163 return ~crc & 0xffffffff;
19168 This computation does not apply to the ``build ID'' method.
19170 @node MiniDebugInfo
19171 @section Debugging information in a special section
19172 @cindex separate debug sections
19173 @cindex @samp{.gnu_debugdata} section
19175 Some systems ship pre-built executables and libraries that have a
19176 special @samp{.gnu_debugdata} section. This feature is called
19177 @dfn{MiniDebugInfo}. This section holds an LZMA-compressed object and
19178 is used to supply extra symbols for backtraces.
19180 The intent of this section is to provide extra minimal debugging
19181 information for use in simple backtraces. It is not intended to be a
19182 replacement for full separate debugging information (@pxref{Separate
19183 Debug Files}). The example below shows the intended use; however,
19184 @value{GDBN} does not currently put restrictions on what sort of
19185 debugging information might be included in the section.
19187 @value{GDBN} has support for this extension. If the section exists,
19188 then it is used provided that no other source of debugging information
19189 can be found, and that @value{GDBN} was configured with LZMA support.
19191 This section can be easily created using @command{objcopy} and other
19192 standard utilities:
19195 # Extract the dynamic symbols from the main binary, there is no need
19196 # to also have these in the normal symbol table.
19197 nm -D @var{binary} --format=posix --defined-only \
19198 | awk '@{ print $1 @}' | sort > dynsyms
19200 # Extract all the text (i.e. function) symbols from the debuginfo.
19201 # (Note that we actually also accept "D" symbols, for the benefit
19202 # of platforms like PowerPC64 that use function descriptors.)
19203 nm @var{binary} --format=posix --defined-only \
19204 | awk '@{ if ($2 == "T" || $2 == "t" || $2 == "D") print $1 @}' \
19207 # Keep all the function symbols not already in the dynamic symbol
19209 comm -13 dynsyms funcsyms > keep_symbols
19211 # Separate full debug info into debug binary.
19212 objcopy --only-keep-debug @var{binary} debug
19214 # Copy the full debuginfo, keeping only a minimal set of symbols and
19215 # removing some unnecessary sections.
19216 objcopy -S --remove-section .gdb_index --remove-section .comment \
19217 --keep-symbols=keep_symbols debug mini_debuginfo
19219 # Drop the full debug info from the original binary.
19220 strip --strip-all -R .comment @var{binary}
19222 # Inject the compressed data into the .gnu_debugdata section of the
19225 objcopy --add-section .gnu_debugdata=mini_debuginfo.xz @var{binary}
19229 @section Index Files Speed Up @value{GDBN}
19230 @cindex index files
19231 @cindex @samp{.gdb_index} section
19233 When @value{GDBN} finds a symbol file, it scans the symbols in the
19234 file in order to construct an internal symbol table. This lets most
19235 @value{GDBN} operations work quickly---at the cost of a delay early
19236 on. For large programs, this delay can be quite lengthy, so
19237 @value{GDBN} provides a way to build an index, which speeds up
19240 The index is stored as a section in the symbol file. @value{GDBN} can
19241 write the index to a file, then you can put it into the symbol file
19242 using @command{objcopy}.
19244 To create an index file, use the @code{save gdb-index} command:
19247 @item save gdb-index @var{directory}
19248 @kindex save gdb-index
19249 Create an index file for each symbol file currently known by
19250 @value{GDBN}. Each file is named after its corresponding symbol file,
19251 with @samp{.gdb-index} appended, and is written into the given
19255 Once you have created an index file you can merge it into your symbol
19256 file, here named @file{symfile}, using @command{objcopy}:
19259 $ objcopy --add-section .gdb_index=symfile.gdb-index \
19260 --set-section-flags .gdb_index=readonly symfile symfile
19263 @value{GDBN} will normally ignore older versions of @file{.gdb_index}
19264 sections that have been deprecated. Usually they are deprecated because
19265 they are missing a new feature or have performance issues.
19266 To tell @value{GDBN} to use a deprecated index section anyway
19267 specify @code{set use-deprecated-index-sections on}.
19268 The default is @code{off}.
19269 This can speed up startup, but may result in some functionality being lost.
19270 @xref{Index Section Format}.
19272 @emph{Warning:} Setting @code{use-deprecated-index-sections} to @code{on}
19273 must be done before gdb reads the file. The following will not work:
19276 $ gdb -ex "set use-deprecated-index-sections on" <program>
19279 Instead you must do, for example,
19282 $ gdb -iex "set use-deprecated-index-sections on" <program>
19285 There are currently some limitation on indices. They only work when
19286 for DWARF debugging information, not stabs. And, they do not
19287 currently work for programs using Ada.
19289 @node Symbol Errors
19290 @section Errors Reading Symbol Files
19292 While reading a symbol file, @value{GDBN} occasionally encounters problems,
19293 such as symbol types it does not recognize, or known bugs in compiler
19294 output. By default, @value{GDBN} does not notify you of such problems, since
19295 they are relatively common and primarily of interest to people
19296 debugging compilers. If you are interested in seeing information
19297 about ill-constructed symbol tables, you can either ask @value{GDBN} to print
19298 only one message about each such type of problem, no matter how many
19299 times the problem occurs; or you can ask @value{GDBN} to print more messages,
19300 to see how many times the problems occur, with the @code{set
19301 complaints} command (@pxref{Messages/Warnings, ,Optional Warnings and
19304 The messages currently printed, and their meanings, include:
19307 @item inner block not inside outer block in @var{symbol}
19309 The symbol information shows where symbol scopes begin and end
19310 (such as at the start of a function or a block of statements). This
19311 error indicates that an inner scope block is not fully contained
19312 in its outer scope blocks.
19314 @value{GDBN} circumvents the problem by treating the inner block as if it had
19315 the same scope as the outer block. In the error message, @var{symbol}
19316 may be shown as ``@code{(don't know)}'' if the outer block is not a
19319 @item block at @var{address} out of order
19321 The symbol information for symbol scope blocks should occur in
19322 order of increasing addresses. This error indicates that it does not
19325 @value{GDBN} does not circumvent this problem, and has trouble
19326 locating symbols in the source file whose symbols it is reading. (You
19327 can often determine what source file is affected by specifying
19328 @code{set verbose on}. @xref{Messages/Warnings, ,Optional Warnings and
19331 @item bad block start address patched
19333 The symbol information for a symbol scope block has a start address
19334 smaller than the address of the preceding source line. This is known
19335 to occur in the SunOS 4.1.1 (and earlier) C compiler.
19337 @value{GDBN} circumvents the problem by treating the symbol scope block as
19338 starting on the previous source line.
19340 @item bad string table offset in symbol @var{n}
19343 Symbol number @var{n} contains a pointer into the string table which is
19344 larger than the size of the string table.
19346 @value{GDBN} circumvents the problem by considering the symbol to have the
19347 name @code{foo}, which may cause other problems if many symbols end up
19350 @item unknown symbol type @code{0x@var{nn}}
19352 The symbol information contains new data types that @value{GDBN} does
19353 not yet know how to read. @code{0x@var{nn}} is the symbol type of the
19354 uncomprehended information, in hexadecimal.
19356 @value{GDBN} circumvents the error by ignoring this symbol information.
19357 This usually allows you to debug your program, though certain symbols
19358 are not accessible. If you encounter such a problem and feel like
19359 debugging it, you can debug @code{@value{GDBP}} with itself, breakpoint
19360 on @code{complain}, then go up to the function @code{read_dbx_symtab}
19361 and examine @code{*bufp} to see the symbol.
19363 @item stub type has NULL name
19365 @value{GDBN} could not find the full definition for a struct or class.
19367 @item const/volatile indicator missing (ok if using g++ v1.x), got@dots{}
19368 The symbol information for a C@t{++} member function is missing some
19369 information that recent versions of the compiler should have output for
19372 @item info mismatch between compiler and debugger
19374 @value{GDBN} could not parse a type specification output by the compiler.
19379 @section GDB Data Files
19381 @cindex prefix for data files
19382 @value{GDBN} will sometimes read an auxiliary data file. These files
19383 are kept in a directory known as the @dfn{data directory}.
19385 You can set the data directory's name, and view the name @value{GDBN}
19386 is currently using.
19389 @kindex set data-directory
19390 @item set data-directory @var{directory}
19391 Set the directory which @value{GDBN} searches for auxiliary data files
19392 to @var{directory}.
19394 @kindex show data-directory
19395 @item show data-directory
19396 Show the directory @value{GDBN} searches for auxiliary data files.
19399 @cindex default data directory
19400 @cindex @samp{--with-gdb-datadir}
19401 You can set the default data directory by using the configure-time
19402 @samp{--with-gdb-datadir} option. If the data directory is inside
19403 @value{GDBN}'s configured binary prefix (set with @samp{--prefix} or
19404 @samp{--exec-prefix}), then the default data directory will be updated
19405 automatically if the installed @value{GDBN} is moved to a new
19408 The data directory may also be specified with the
19409 @code{--data-directory} command line option.
19410 @xref{Mode Options}.
19413 @chapter Specifying a Debugging Target
19415 @cindex debugging target
19416 A @dfn{target} is the execution environment occupied by your program.
19418 Often, @value{GDBN} runs in the same host environment as your program;
19419 in that case, the debugging target is specified as a side effect when
19420 you use the @code{file} or @code{core} commands. When you need more
19421 flexibility---for example, running @value{GDBN} on a physically separate
19422 host, or controlling a standalone system over a serial port or a
19423 realtime system over a TCP/IP connection---you can use the @code{target}
19424 command to specify one of the target types configured for @value{GDBN}
19425 (@pxref{Target Commands, ,Commands for Managing Targets}).
19427 @cindex target architecture
19428 It is possible to build @value{GDBN} for several different @dfn{target
19429 architectures}. When @value{GDBN} is built like that, you can choose
19430 one of the available architectures with the @kbd{set architecture}
19434 @kindex set architecture
19435 @kindex show architecture
19436 @item set architecture @var{arch}
19437 This command sets the current target architecture to @var{arch}. The
19438 value of @var{arch} can be @code{"auto"}, in addition to one of the
19439 supported architectures.
19441 @item show architecture
19442 Show the current target architecture.
19444 @item set processor
19446 @kindex set processor
19447 @kindex show processor
19448 These are alias commands for, respectively, @code{set architecture}
19449 and @code{show architecture}.
19453 * Active Targets:: Active targets
19454 * Target Commands:: Commands for managing targets
19455 * Byte Order:: Choosing target byte order
19458 @node Active Targets
19459 @section Active Targets
19461 @cindex stacking targets
19462 @cindex active targets
19463 @cindex multiple targets
19465 There are multiple classes of targets such as: processes, executable files or
19466 recording sessions. Core files belong to the process class, making core file
19467 and process mutually exclusive. Otherwise, @value{GDBN} can work concurrently
19468 on multiple active targets, one in each class. This allows you to (for
19469 example) start a process and inspect its activity, while still having access to
19470 the executable file after the process finishes. Or if you start process
19471 recording (@pxref{Reverse Execution}) and @code{reverse-step} there, you are
19472 presented a virtual layer of the recording target, while the process target
19473 remains stopped at the chronologically last point of the process execution.
19475 Use the @code{core-file} and @code{exec-file} commands to select a new core
19476 file or executable target (@pxref{Files, ,Commands to Specify Files}). To
19477 specify as a target a process that is already running, use the @code{attach}
19478 command (@pxref{Attach, ,Debugging an Already-running Process}).
19480 @node Target Commands
19481 @section Commands for Managing Targets
19484 @item target @var{type} @var{parameters}
19485 Connects the @value{GDBN} host environment to a target machine or
19486 process. A target is typically a protocol for talking to debugging
19487 facilities. You use the argument @var{type} to specify the type or
19488 protocol of the target machine.
19490 Further @var{parameters} are interpreted by the target protocol, but
19491 typically include things like device names or host names to connect
19492 with, process numbers, and baud rates.
19494 The @code{target} command does not repeat if you press @key{RET} again
19495 after executing the command.
19497 @kindex help target
19499 Displays the names of all targets available. To display targets
19500 currently selected, use either @code{info target} or @code{info files}
19501 (@pxref{Files, ,Commands to Specify Files}).
19503 @item help target @var{name}
19504 Describe a particular target, including any parameters necessary to
19507 @kindex set gnutarget
19508 @item set gnutarget @var{args}
19509 @value{GDBN} uses its own library BFD to read your files. @value{GDBN}
19510 knows whether it is reading an @dfn{executable},
19511 a @dfn{core}, or a @dfn{.o} file; however, you can specify the file format
19512 with the @code{set gnutarget} command. Unlike most @code{target} commands,
19513 with @code{gnutarget} the @code{target} refers to a program, not a machine.
19516 @emph{Warning:} To specify a file format with @code{set gnutarget},
19517 you must know the actual BFD name.
19521 @xref{Files, , Commands to Specify Files}.
19523 @kindex show gnutarget
19524 @item show gnutarget
19525 Use the @code{show gnutarget} command to display what file format
19526 @code{gnutarget} is set to read. If you have not set @code{gnutarget},
19527 @value{GDBN} will determine the file format for each file automatically,
19528 and @code{show gnutarget} displays @samp{The current BFD target is "auto"}.
19531 @cindex common targets
19532 Here are some common targets (available, or not, depending on the GDB
19537 @item target exec @var{program}
19538 @cindex executable file target
19539 An executable file. @samp{target exec @var{program}} is the same as
19540 @samp{exec-file @var{program}}.
19542 @item target core @var{filename}
19543 @cindex core dump file target
19544 A core dump file. @samp{target core @var{filename}} is the same as
19545 @samp{core-file @var{filename}}.
19547 @item target remote @var{medium}
19548 @cindex remote target
19549 A remote system connected to @value{GDBN} via a serial line or network
19550 connection. This command tells @value{GDBN} to use its own remote
19551 protocol over @var{medium} for debugging. @xref{Remote Debugging}.
19553 For example, if you have a board connected to @file{/dev/ttya} on the
19554 machine running @value{GDBN}, you could say:
19557 target remote /dev/ttya
19560 @code{target remote} supports the @code{load} command. This is only
19561 useful if you have some other way of getting the stub to the target
19562 system, and you can put it somewhere in memory where it won't get
19563 clobbered by the download.
19565 @item target sim @r{[}@var{simargs}@r{]} @dots{}
19566 @cindex built-in simulator target
19567 Builtin CPU simulator. @value{GDBN} includes simulators for most architectures.
19575 works; however, you cannot assume that a specific memory map, device
19576 drivers, or even basic I/O is available, although some simulators do
19577 provide these. For info about any processor-specific simulator details,
19578 see the appropriate section in @ref{Embedded Processors, ,Embedded
19581 @item target native
19582 @cindex native target
19583 Setup for local/native process debugging. Useful to make the
19584 @code{run} command spawn native processes (likewise @code{attach},
19585 etc.@:) even when @code{set auto-connect-native-target} is @code{off}
19586 (@pxref{set auto-connect-native-target}).
19590 Different targets are available on different configurations of @value{GDBN};
19591 your configuration may have more or fewer targets.
19593 Many remote targets require you to download the executable's code once
19594 you've successfully established a connection. You may wish to control
19595 various aspects of this process.
19600 @kindex set hash@r{, for remote monitors}
19601 @cindex hash mark while downloading
19602 This command controls whether a hash mark @samp{#} is displayed while
19603 downloading a file to the remote monitor. If on, a hash mark is
19604 displayed after each S-record is successfully downloaded to the
19608 @kindex show hash@r{, for remote monitors}
19609 Show the current status of displaying the hash mark.
19611 @item set debug monitor
19612 @kindex set debug monitor
19613 @cindex display remote monitor communications
19614 Enable or disable display of communications messages between
19615 @value{GDBN} and the remote monitor.
19617 @item show debug monitor
19618 @kindex show debug monitor
19619 Show the current status of displaying communications between
19620 @value{GDBN} and the remote monitor.
19625 @kindex load @var{filename} @var{offset}
19626 @item load @var{filename} @var{offset}
19628 Depending on what remote debugging facilities are configured into
19629 @value{GDBN}, the @code{load} command may be available. Where it exists, it
19630 is meant to make @var{filename} (an executable) available for debugging
19631 on the remote system---by downloading, or dynamic linking, for example.
19632 @code{load} also records the @var{filename} symbol table in @value{GDBN}, like
19633 the @code{add-symbol-file} command.
19635 If your @value{GDBN} does not have a @code{load} command, attempting to
19636 execute it gets the error message ``@code{You can't do that when your
19637 target is @dots{}}''
19639 The file is loaded at whatever address is specified in the executable.
19640 For some object file formats, you can specify the load address when you
19641 link the program; for other formats, like a.out, the object file format
19642 specifies a fixed address.
19643 @c FIXME! This would be a good place for an xref to the GNU linker doc.
19645 It is also possible to tell @value{GDBN} to load the executable file at a
19646 specific offset described by the optional argument @var{offset}. When
19647 @var{offset} is provided, @var{filename} must also be provided.
19649 Depending on the remote side capabilities, @value{GDBN} may be able to
19650 load programs into flash memory.
19652 @code{load} does not repeat if you press @key{RET} again after using it.
19657 @kindex flash-erase
19659 @anchor{flash-erase}
19661 Erases all known flash memory regions on the target.
19666 @section Choosing Target Byte Order
19668 @cindex choosing target byte order
19669 @cindex target byte order
19671 Some types of processors, such as the @acronym{MIPS}, PowerPC, and Renesas SH,
19672 offer the ability to run either big-endian or little-endian byte
19673 orders. Usually the executable or symbol will include a bit to
19674 designate the endian-ness, and you will not need to worry about
19675 which to use. However, you may still find it useful to adjust
19676 @value{GDBN}'s idea of processor endian-ness manually.
19680 @item set endian big
19681 Instruct @value{GDBN} to assume the target is big-endian.
19683 @item set endian little
19684 Instruct @value{GDBN} to assume the target is little-endian.
19686 @item set endian auto
19687 Instruct @value{GDBN} to use the byte order associated with the
19691 Display @value{GDBN}'s current idea of the target byte order.
19695 Note that these commands merely adjust interpretation of symbolic
19696 data on the host, and that they have absolutely no effect on the
19700 @node Remote Debugging
19701 @chapter Debugging Remote Programs
19702 @cindex remote debugging
19704 If you are trying to debug a program running on a machine that cannot run
19705 @value{GDBN} in the usual way, it is often useful to use remote debugging.
19706 For example, you might use remote debugging on an operating system kernel,
19707 or on a small system which does not have a general purpose operating system
19708 powerful enough to run a full-featured debugger.
19710 Some configurations of @value{GDBN} have special serial or TCP/IP interfaces
19711 to make this work with particular debugging targets. In addition,
19712 @value{GDBN} comes with a generic serial protocol (specific to @value{GDBN},
19713 but not specific to any particular target system) which you can use if you
19714 write the remote stubs---the code that runs on the remote system to
19715 communicate with @value{GDBN}.
19717 Other remote targets may be available in your
19718 configuration of @value{GDBN}; use @code{help target} to list them.
19721 * Connecting:: Connecting to a remote target
19722 * File Transfer:: Sending files to a remote system
19723 * Server:: Using the gdbserver program
19724 * Remote Configuration:: Remote configuration
19725 * Remote Stub:: Implementing a remote stub
19729 @section Connecting to a Remote Target
19730 @cindex remote debugging, connecting
19731 @cindex @code{gdbserver}, connecting
19732 @cindex remote debugging, types of connections
19733 @cindex @code{gdbserver}, types of connections
19734 @cindex @code{gdbserver}, @code{target remote} mode
19735 @cindex @code{gdbserver}, @code{target extended-remote} mode
19737 This section describes how to connect to a remote target, including the
19738 types of connections and their differences, how to set up executable and
19739 symbol files on the host and target, and the commands used for
19740 connecting to and disconnecting from the remote target.
19742 @subsection Types of Remote Connections
19744 @value{GDBN} supports two types of remote connections, @code{target remote}
19745 mode and @code{target extended-remote} mode. Note that many remote targets
19746 support only @code{target remote} mode. There are several major
19747 differences between the two types of connections, enumerated here:
19751 @cindex remote debugging, detach and program exit
19752 @item Result of detach or program exit
19753 @strong{With target remote mode:} When the debugged program exits or you
19754 detach from it, @value{GDBN} disconnects from the target. When using
19755 @code{gdbserver}, @code{gdbserver} will exit.
19757 @strong{With target extended-remote mode:} When the debugged program exits or
19758 you detach from it, @value{GDBN} remains connected to the target, even
19759 though no program is running. You can rerun the program, attach to a
19760 running program, or use @code{monitor} commands specific to the target.
19762 When using @code{gdbserver} in this case, it does not exit unless it was
19763 invoked using the @option{--once} option. If the @option{--once} option
19764 was not used, you can ask @code{gdbserver} to exit using the
19765 @code{monitor exit} command (@pxref{Monitor Commands for gdbserver}).
19767 @item Specifying the program to debug
19768 For both connection types you use the @code{file} command to specify the
19769 program on the host system. If you are using @code{gdbserver} there are
19770 some differences in how to specify the location of the program on the
19773 @strong{With target remote mode:} You must either specify the program to debug
19774 on the @code{gdbserver} command line or use the @option{--attach} option
19775 (@pxref{Attaching to a program,,Attaching to a Running Program}).
19777 @cindex @option{--multi}, @code{gdbserver} option
19778 @strong{With target extended-remote mode:} You may specify the program to debug
19779 on the @code{gdbserver} command line, or you can load the program or attach
19780 to it using @value{GDBN} commands after connecting to @code{gdbserver}.
19782 @anchor{--multi Option in Types of Remote Connnections}
19783 You can start @code{gdbserver} without supplying an initial command to run
19784 or process ID to attach. To do this, use the @option{--multi} command line
19785 option. Then you can connect using @code{target extended-remote} and start
19786 the program you want to debug (see below for details on using the
19787 @code{run} command in this scenario). Note that the conditions under which
19788 @code{gdbserver} terminates depend on how @value{GDBN} connects to it
19789 (@code{target remote} or @code{target extended-remote}). The
19790 @option{--multi} option to @code{gdbserver} has no influence on that.
19792 @item The @code{run} command
19793 @strong{With target remote mode:} The @code{run} command is not
19794 supported. Once a connection has been established, you can use all
19795 the usual @value{GDBN} commands to examine and change data. The
19796 remote program is already running, so you can use commands like
19797 @kbd{step} and @kbd{continue}.
19799 @strong{With target extended-remote mode:} The @code{run} command is
19800 supported. The @code{run} command uses the value set by
19801 @code{set remote exec-file} (@pxref{set remote exec-file}) to select
19802 the program to run. Command line arguments are supported, except for
19803 wildcard expansion and I/O redirection (@pxref{Arguments}).
19805 If you specify the program to debug on the command line, then the
19806 @code{run} command is not required to start execution, and you can
19807 resume using commands like @kbd{step} and @kbd{continue} as with
19808 @code{target remote} mode.
19810 @anchor{Attaching in Types of Remote Connections}
19812 @strong{With target remote mode:} The @value{GDBN} command @code{attach} is
19813 not supported. To attach to a running program using @code{gdbserver}, you
19814 must use the @option{--attach} option (@pxref{Running gdbserver}).
19816 @strong{With target extended-remote mode:} To attach to a running program,
19817 you may use the @code{attach} command after the connection has been
19818 established. If you are using @code{gdbserver}, you may also invoke
19819 @code{gdbserver} using the @option{--attach} option
19820 (@pxref{Running gdbserver}).
19824 @anchor{Host and target files}
19825 @subsection Host and Target Files
19826 @cindex remote debugging, symbol files
19827 @cindex symbol files, remote debugging
19829 @value{GDBN}, running on the host, needs access to symbol and debugging
19830 information for your program running on the target. This requires
19831 access to an unstripped copy of your program, and possibly any associated
19832 symbol files. Note that this section applies equally to both @code{target
19833 remote} mode and @code{target extended-remote} mode.
19835 Some remote targets (@pxref{qXfer executable filename read}, and
19836 @pxref{Host I/O Packets}) allow @value{GDBN} to access program files over
19837 the same connection used to communicate with @value{GDBN}. With such a
19838 target, if the remote program is unstripped, the only command you need is
19839 @code{target remote} (or @code{target extended-remote}).
19841 If the remote program is stripped, or the target does not support remote
19842 program file access, start up @value{GDBN} using the name of the local
19843 unstripped copy of your program as the first argument, or use the
19844 @code{file} command. Use @code{set sysroot} to specify the location (on
19845 the host) of target libraries (unless your @value{GDBN} was compiled with
19846 the correct sysroot using @code{--with-sysroot}). Alternatively, you
19847 may use @code{set solib-search-path} to specify how @value{GDBN} locates
19850 The symbol file and target libraries must exactly match the executable
19851 and libraries on the target, with one exception: the files on the host
19852 system should not be stripped, even if the files on the target system
19853 are. Mismatched or missing files will lead to confusing results
19854 during debugging. On @sc{gnu}/Linux targets, mismatched or missing
19855 files may also prevent @code{gdbserver} from debugging multi-threaded
19858 @subsection Remote Connection Commands
19859 @cindex remote connection commands
19860 @value{GDBN} can communicate with the target over a serial line, or
19861 over an @acronym{IP} network using @acronym{TCP} or @acronym{UDP}. In
19862 each case, @value{GDBN} uses the same protocol for debugging your
19863 program; only the medium carrying the debugging packets varies. The
19864 @code{target remote} and @code{target extended-remote} commands
19865 establish a connection to the target. Both commands accept the same
19866 arguments, which indicate the medium to use:
19870 @item target remote @var{serial-device}
19871 @itemx target extended-remote @var{serial-device}
19872 @cindex serial line, @code{target remote}
19873 Use @var{serial-device} to communicate with the target. For example,
19874 to use a serial line connected to the device named @file{/dev/ttyb}:
19877 target remote /dev/ttyb
19880 If you're using a serial line, you may want to give @value{GDBN} the
19881 @samp{--baud} option, or use the @code{set serial baud} command
19882 (@pxref{Remote Configuration, set serial baud}) before the
19883 @code{target} command.
19885 @item target remote @code{@var{host}:@var{port}}
19886 @itemx target remote @code{tcp:@var{host}:@var{port}}
19887 @itemx target extended-remote @code{@var{host}:@var{port}}
19888 @itemx target extended-remote @code{tcp:@var{host}:@var{port}}
19889 @cindex @acronym{TCP} port, @code{target remote}
19890 Debug using a @acronym{TCP} connection to @var{port} on @var{host}.
19891 The @var{host} may be either a host name or a numeric @acronym{IP}
19892 address; @var{port} must be a decimal number. The @var{host} could be
19893 the target machine itself, if it is directly connected to the net, or
19894 it might be a terminal server which in turn has a serial line to the
19897 For example, to connect to port 2828 on a terminal server named
19901 target remote manyfarms:2828
19904 If your remote target is actually running on the same machine as your
19905 debugger session (e.g.@: a simulator for your target running on the
19906 same host), you can omit the hostname. For example, to connect to
19907 port 1234 on your local machine:
19910 target remote :1234
19914 Note that the colon is still required here.
19916 @item target remote @code{udp:@var{host}:@var{port}}
19917 @itemx target extended-remote @code{udp:@var{host}:@var{port}}
19918 @cindex @acronym{UDP} port, @code{target remote}
19919 Debug using @acronym{UDP} packets to @var{port} on @var{host}. For example, to
19920 connect to @acronym{UDP} port 2828 on a terminal server named @code{manyfarms}:
19923 target remote udp:manyfarms:2828
19926 When using a @acronym{UDP} connection for remote debugging, you should
19927 keep in mind that the `U' stands for ``Unreliable''. @acronym{UDP}
19928 can silently drop packets on busy or unreliable networks, which will
19929 cause havoc with your debugging session.
19931 @item target remote | @var{command}
19932 @itemx target extended-remote | @var{command}
19933 @cindex pipe, @code{target remote} to
19934 Run @var{command} in the background and communicate with it using a
19935 pipe. The @var{command} is a shell command, to be parsed and expanded
19936 by the system's command shell, @code{/bin/sh}; it should expect remote
19937 protocol packets on its standard input, and send replies on its
19938 standard output. You could use this to run a stand-alone simulator
19939 that speaks the remote debugging protocol, to make net connections
19940 using programs like @code{ssh}, or for other similar tricks.
19942 If @var{command} closes its standard output (perhaps by exiting),
19943 @value{GDBN} will try to send it a @code{SIGTERM} signal. (If the
19944 program has already exited, this will have no effect.)
19948 @cindex interrupting remote programs
19949 @cindex remote programs, interrupting
19950 Whenever @value{GDBN} is waiting for the remote program, if you type the
19951 interrupt character (often @kbd{Ctrl-c}), @value{GDBN} attempts to stop the
19952 program. This may or may not succeed, depending in part on the hardware
19953 and the serial drivers the remote system uses. If you type the
19954 interrupt character once again, @value{GDBN} displays this prompt:
19957 Interrupted while waiting for the program.
19958 Give up (and stop debugging it)? (y or n)
19961 In @code{target remote} mode, if you type @kbd{y}, @value{GDBN} abandons
19962 the remote debugging session. (If you decide you want to try again later,
19963 you can use @kbd{target remote} again to connect once more.) If you type
19964 @kbd{n}, @value{GDBN} goes back to waiting.
19966 In @code{target extended-remote} mode, typing @kbd{n} will leave
19967 @value{GDBN} connected to the target.
19970 @kindex detach (remote)
19972 When you have finished debugging the remote program, you can use the
19973 @code{detach} command to release it from @value{GDBN} control.
19974 Detaching from the target normally resumes its execution, but the results
19975 will depend on your particular remote stub. After the @code{detach}
19976 command in @code{target remote} mode, @value{GDBN} is free to connect to
19977 another target. In @code{target extended-remote} mode, @value{GDBN} is
19978 still connected to the target.
19982 The @code{disconnect} command closes the connection to the target, and
19983 the target is generally not resumed. It will wait for @value{GDBN}
19984 (this instance or another one) to connect and continue debugging. After
19985 the @code{disconnect} command, @value{GDBN} is again free to connect to
19988 @cindex send command to remote monitor
19989 @cindex extend @value{GDBN} for remote targets
19990 @cindex add new commands for external monitor
19992 @item monitor @var{cmd}
19993 This command allows you to send arbitrary commands directly to the
19994 remote monitor. Since @value{GDBN} doesn't care about the commands it
19995 sends like this, this command is the way to extend @value{GDBN}---you
19996 can add new commands that only the external monitor will understand
20000 @node File Transfer
20001 @section Sending files to a remote system
20002 @cindex remote target, file transfer
20003 @cindex file transfer
20004 @cindex sending files to remote systems
20006 Some remote targets offer the ability to transfer files over the same
20007 connection used to communicate with @value{GDBN}. This is convenient
20008 for targets accessible through other means, e.g.@: @sc{gnu}/Linux systems
20009 running @code{gdbserver} over a network interface. For other targets,
20010 e.g.@: embedded devices with only a single serial port, this may be
20011 the only way to upload or download files.
20013 Not all remote targets support these commands.
20017 @item remote put @var{hostfile} @var{targetfile}
20018 Copy file @var{hostfile} from the host system (the machine running
20019 @value{GDBN}) to @var{targetfile} on the target system.
20022 @item remote get @var{targetfile} @var{hostfile}
20023 Copy file @var{targetfile} from the target system to @var{hostfile}
20024 on the host system.
20026 @kindex remote delete
20027 @item remote delete @var{targetfile}
20028 Delete @var{targetfile} from the target system.
20033 @section Using the @code{gdbserver} Program
20036 @cindex remote connection without stubs
20037 @code{gdbserver} is a control program for Unix-like systems, which
20038 allows you to connect your program with a remote @value{GDBN} via
20039 @code{target remote} or @code{target extended-remote}---but without
20040 linking in the usual debugging stub.
20042 @code{gdbserver} is not a complete replacement for the debugging stubs,
20043 because it requires essentially the same operating-system facilities
20044 that @value{GDBN} itself does. In fact, a system that can run
20045 @code{gdbserver} to connect to a remote @value{GDBN} could also run
20046 @value{GDBN} locally! @code{gdbserver} is sometimes useful nevertheless,
20047 because it is a much smaller program than @value{GDBN} itself. It is
20048 also easier to port than all of @value{GDBN}, so you may be able to get
20049 started more quickly on a new system by using @code{gdbserver}.
20050 Finally, if you develop code for real-time systems, you may find that
20051 the tradeoffs involved in real-time operation make it more convenient to
20052 do as much development work as possible on another system, for example
20053 by cross-compiling. You can use @code{gdbserver} to make a similar
20054 choice for debugging.
20056 @value{GDBN} and @code{gdbserver} communicate via either a serial line
20057 or a TCP connection, using the standard @value{GDBN} remote serial
20061 @emph{Warning:} @code{gdbserver} does not have any built-in security.
20062 Do not run @code{gdbserver} connected to any public network; a
20063 @value{GDBN} connection to @code{gdbserver} provides access to the
20064 target system with the same privileges as the user running
20068 @anchor{Running gdbserver}
20069 @subsection Running @code{gdbserver}
20070 @cindex arguments, to @code{gdbserver}
20071 @cindex @code{gdbserver}, command-line arguments
20073 Run @code{gdbserver} on the target system. You need a copy of the
20074 program you want to debug, including any libraries it requires.
20075 @code{gdbserver} does not need your program's symbol table, so you can
20076 strip the program if necessary to save space. @value{GDBN} on the host
20077 system does all the symbol handling.
20079 To use the server, you must tell it how to communicate with @value{GDBN};
20080 the name of your program; and the arguments for your program. The usual
20084 target> gdbserver @var{comm} @var{program} [ @var{args} @dots{} ]
20087 @var{comm} is either a device name (to use a serial line), or a TCP
20088 hostname and portnumber, or @code{-} or @code{stdio} to use
20089 stdin/stdout of @code{gdbserver}.
20090 For example, to debug Emacs with the argument
20091 @samp{foo.txt} and communicate with @value{GDBN} over the serial port
20095 target> gdbserver /dev/com1 emacs foo.txt
20098 @code{gdbserver} waits passively for the host @value{GDBN} to communicate
20101 To use a TCP connection instead of a serial line:
20104 target> gdbserver host:2345 emacs foo.txt
20107 The only difference from the previous example is the first argument,
20108 specifying that you are communicating with the host @value{GDBN} via
20109 TCP. The @samp{host:2345} argument means that @code{gdbserver} is to
20110 expect a TCP connection from machine @samp{host} to local TCP port 2345.
20111 (Currently, the @samp{host} part is ignored.) You can choose any number
20112 you want for the port number as long as it does not conflict with any
20113 TCP ports already in use on the target system (for example, @code{23} is
20114 reserved for @code{telnet}).@footnote{If you choose a port number that
20115 conflicts with another service, @code{gdbserver} prints an error message
20116 and exits.} You must use the same port number with the host @value{GDBN}
20117 @code{target remote} command.
20119 The @code{stdio} connection is useful when starting @code{gdbserver}
20123 (gdb) target remote | ssh -T hostname gdbserver - hello
20126 The @samp{-T} option to ssh is provided because we don't need a remote pty,
20127 and we don't want escape-character handling. Ssh does this by default when
20128 a command is provided, the flag is provided to make it explicit.
20129 You could elide it if you want to.
20131 Programs started with stdio-connected gdbserver have @file{/dev/null} for
20132 @code{stdin}, and @code{stdout},@code{stderr} are sent back to gdb for
20133 display through a pipe connected to gdbserver.
20134 Both @code{stdout} and @code{stderr} use the same pipe.
20136 @anchor{Attaching to a program}
20137 @subsubsection Attaching to a Running Program
20138 @cindex attach to a program, @code{gdbserver}
20139 @cindex @option{--attach}, @code{gdbserver} option
20141 On some targets, @code{gdbserver} can also attach to running programs.
20142 This is accomplished via the @code{--attach} argument. The syntax is:
20145 target> gdbserver --attach @var{comm} @var{pid}
20148 @var{pid} is the process ID of a currently running process. It isn't
20149 necessary to point @code{gdbserver} at a binary for the running process.
20151 In @code{target extended-remote} mode, you can also attach using the
20152 @value{GDBN} attach command
20153 (@pxref{Attaching in Types of Remote Connections}).
20156 You can debug processes by name instead of process ID if your target has the
20157 @code{pidof} utility:
20160 target> gdbserver --attach @var{comm} `pidof @var{program}`
20163 In case more than one copy of @var{program} is running, or @var{program}
20164 has multiple threads, most versions of @code{pidof} support the
20165 @code{-s} option to only return the first process ID.
20167 @subsubsection TCP port allocation lifecycle of @code{gdbserver}
20169 This section applies only when @code{gdbserver} is run to listen on a TCP
20172 @code{gdbserver} normally terminates after all of its debugged processes have
20173 terminated in @kbd{target remote} mode. On the other hand, for @kbd{target
20174 extended-remote}, @code{gdbserver} stays running even with no processes left.
20175 @value{GDBN} normally terminates the spawned debugged process on its exit,
20176 which normally also terminates @code{gdbserver} in the @kbd{target remote}
20177 mode. Therefore, when the connection drops unexpectedly, and @value{GDBN}
20178 cannot ask @code{gdbserver} to kill its debugged processes, @code{gdbserver}
20179 stays running even in the @kbd{target remote} mode.
20181 When @code{gdbserver} stays running, @value{GDBN} can connect to it again later.
20182 Such reconnecting is useful for features like @ref{disconnected tracing}. For
20183 completeness, at most one @value{GDBN} can be connected at a time.
20185 @cindex @option{--once}, @code{gdbserver} option
20186 By default, @code{gdbserver} keeps the listening TCP port open, so that
20187 subsequent connections are possible. However, if you start @code{gdbserver}
20188 with the @option{--once} option, it will stop listening for any further
20189 connection attempts after connecting to the first @value{GDBN} session. This
20190 means no further connections to @code{gdbserver} will be possible after the
20191 first one. It also means @code{gdbserver} will terminate after the first
20192 connection with remote @value{GDBN} has closed, even for unexpectedly closed
20193 connections and even in the @kbd{target extended-remote} mode. The
20194 @option{--once} option allows reusing the same port number for connecting to
20195 multiple instances of @code{gdbserver} running on the same host, since each
20196 instance closes its port after the first connection.
20198 @anchor{Other Command-Line Arguments for gdbserver}
20199 @subsubsection Other Command-Line Arguments for @code{gdbserver}
20201 You can use the @option{--multi} option to start @code{gdbserver} without
20202 specifying a program to debug or a process to attach to. Then you can
20203 attach in @code{target extended-remote} mode and run or attach to a
20204 program. For more information,
20205 @pxref{--multi Option in Types of Remote Connnections}.
20207 @cindex @option{--debug}, @code{gdbserver} option
20208 The @option{--debug} option tells @code{gdbserver} to display extra
20209 status information about the debugging process.
20210 @cindex @option{--remote-debug}, @code{gdbserver} option
20211 The @option{--remote-debug} option tells @code{gdbserver} to display
20212 remote protocol debug output. These options are intended for
20213 @code{gdbserver} development and for bug reports to the developers.
20215 @cindex @option{--debug-format}, @code{gdbserver} option
20216 The @option{--debug-format=option1[,option2,...]} option tells
20217 @code{gdbserver} to include additional information in each output.
20218 Possible options are:
20222 Turn off all extra information in debugging output.
20224 Turn on all extra information in debugging output.
20226 Include a timestamp in each line of debugging output.
20229 Options are processed in order. Thus, for example, if @option{none}
20230 appears last then no additional information is added to debugging output.
20232 @cindex @option{--wrapper}, @code{gdbserver} option
20233 The @option{--wrapper} option specifies a wrapper to launch programs
20234 for debugging. The option should be followed by the name of the
20235 wrapper, then any command-line arguments to pass to the wrapper, then
20236 @kbd{--} indicating the end of the wrapper arguments.
20238 @code{gdbserver} runs the specified wrapper program with a combined
20239 command line including the wrapper arguments, then the name of the
20240 program to debug, then any arguments to the program. The wrapper
20241 runs until it executes your program, and then @value{GDBN} gains control.
20243 You can use any program that eventually calls @code{execve} with
20244 its arguments as a wrapper. Several standard Unix utilities do
20245 this, e.g.@: @code{env} and @code{nohup}. Any Unix shell script ending
20246 with @code{exec "$@@"} will also work.
20248 For example, you can use @code{env} to pass an environment variable to
20249 the debugged program, without setting the variable in @code{gdbserver}'s
20253 $ gdbserver --wrapper env LD_PRELOAD=libtest.so -- :2222 ./testprog
20256 @subsection Connecting to @code{gdbserver}
20258 The basic procedure for connecting to the remote target is:
20262 Run @value{GDBN} on the host system.
20265 Make sure you have the necessary symbol files
20266 (@pxref{Host and target files}).
20267 Load symbols for your application using the @code{file} command before you
20268 connect. Use @code{set sysroot} to locate target libraries (unless your
20269 @value{GDBN} was compiled with the correct sysroot using
20270 @code{--with-sysroot}).
20273 Connect to your target (@pxref{Connecting,,Connecting to a Remote Target}).
20274 For TCP connections, you must start up @code{gdbserver} prior to using
20275 the @code{target} command. Otherwise you may get an error whose
20276 text depends on the host system, but which usually looks something like
20277 @samp{Connection refused}. Don't use the @code{load}
20278 command in @value{GDBN} when using @code{target remote} mode, since the
20279 program is already on the target.
20283 @anchor{Monitor Commands for gdbserver}
20284 @subsection Monitor Commands for @code{gdbserver}
20285 @cindex monitor commands, for @code{gdbserver}
20287 During a @value{GDBN} session using @code{gdbserver}, you can use the
20288 @code{monitor} command to send special requests to @code{gdbserver}.
20289 Here are the available commands.
20293 List the available monitor commands.
20295 @item monitor set debug 0
20296 @itemx monitor set debug 1
20297 Disable or enable general debugging messages.
20299 @item monitor set remote-debug 0
20300 @itemx monitor set remote-debug 1
20301 Disable or enable specific debugging messages associated with the remote
20302 protocol (@pxref{Remote Protocol}).
20304 @item monitor set debug-format option1@r{[},option2,...@r{]}
20305 Specify additional text to add to debugging messages.
20306 Possible options are:
20310 Turn off all extra information in debugging output.
20312 Turn on all extra information in debugging output.
20314 Include a timestamp in each line of debugging output.
20317 Options are processed in order. Thus, for example, if @option{none}
20318 appears last then no additional information is added to debugging output.
20320 @item monitor set libthread-db-search-path [PATH]
20321 @cindex gdbserver, search path for @code{libthread_db}
20322 When this command is issued, @var{path} is a colon-separated list of
20323 directories to search for @code{libthread_db} (@pxref{Threads,,set
20324 libthread-db-search-path}). If you omit @var{path},
20325 @samp{libthread-db-search-path} will be reset to its default value.
20327 The special entry @samp{$pdir} for @samp{libthread-db-search-path} is
20328 not supported in @code{gdbserver}.
20331 Tell gdbserver to exit immediately. This command should be followed by
20332 @code{disconnect} to close the debugging session. @code{gdbserver} will
20333 detach from any attached processes and kill any processes it created.
20334 Use @code{monitor exit} to terminate @code{gdbserver} at the end
20335 of a multi-process mode debug session.
20339 @subsection Tracepoints support in @code{gdbserver}
20340 @cindex tracepoints support in @code{gdbserver}
20342 On some targets, @code{gdbserver} supports tracepoints, fast
20343 tracepoints and static tracepoints.
20345 For fast or static tracepoints to work, a special library called the
20346 @dfn{in-process agent} (IPA), must be loaded in the inferior process.
20347 This library is built and distributed as an integral part of
20348 @code{gdbserver}. In addition, support for static tracepoints
20349 requires building the in-process agent library with static tracepoints
20350 support. At present, the UST (LTTng Userspace Tracer,
20351 @url{http://lttng.org/ust}) tracing engine is supported. This support
20352 is automatically available if UST development headers are found in the
20353 standard include path when @code{gdbserver} is built, or if
20354 @code{gdbserver} was explicitly configured using @option{--with-ust}
20355 to point at such headers. You can explicitly disable the support
20356 using @option{--with-ust=no}.
20358 There are several ways to load the in-process agent in your program:
20361 @item Specifying it as dependency at link time
20363 You can link your program dynamically with the in-process agent
20364 library. On most systems, this is accomplished by adding
20365 @code{-linproctrace} to the link command.
20367 @item Using the system's preloading mechanisms
20369 You can force loading the in-process agent at startup time by using
20370 your system's support for preloading shared libraries. Many Unixes
20371 support the concept of preloading user defined libraries. In most
20372 cases, you do that by specifying @code{LD_PRELOAD=libinproctrace.so}
20373 in the environment. See also the description of @code{gdbserver}'s
20374 @option{--wrapper} command line option.
20376 @item Using @value{GDBN} to force loading the agent at run time
20378 On some systems, you can force the inferior to load a shared library,
20379 by calling a dynamic loader function in the inferior that takes care
20380 of dynamically looking up and loading a shared library. On most Unix
20381 systems, the function is @code{dlopen}. You'll use the @code{call}
20382 command for that. For example:
20385 (@value{GDBP}) call dlopen ("libinproctrace.so", ...)
20388 Note that on most Unix systems, for the @code{dlopen} function to be
20389 available, the program needs to be linked with @code{-ldl}.
20392 On systems that have a userspace dynamic loader, like most Unix
20393 systems, when you connect to @code{gdbserver} using @code{target
20394 remote}, you'll find that the program is stopped at the dynamic
20395 loader's entry point, and no shared library has been loaded in the
20396 program's address space yet, including the in-process agent. In that
20397 case, before being able to use any of the fast or static tracepoints
20398 features, you need to let the loader run and load the shared
20399 libraries. The simplest way to do that is to run the program to the
20400 main procedure. E.g., if debugging a C or C@t{++} program, start
20401 @code{gdbserver} like so:
20404 $ gdbserver :9999 myprogram
20407 Start GDB and connect to @code{gdbserver} like so, and run to main:
20411 (@value{GDBP}) target remote myhost:9999
20412 0x00007f215893ba60 in ?? () from /lib64/ld-linux-x86-64.so.2
20413 (@value{GDBP}) b main
20414 (@value{GDBP}) continue
20417 The in-process tracing agent library should now be loaded into the
20418 process; you can confirm it with the @code{info sharedlibrary}
20419 command, which will list @file{libinproctrace.so} as loaded in the
20420 process. You are now ready to install fast tracepoints, list static
20421 tracepoint markers, probe static tracepoints markers, and start
20424 @node Remote Configuration
20425 @section Remote Configuration
20428 @kindex show remote
20429 This section documents the configuration options available when
20430 debugging remote programs. For the options related to the File I/O
20431 extensions of the remote protocol, see @ref{system,
20432 system-call-allowed}.
20435 @item set remoteaddresssize @var{bits}
20436 @cindex address size for remote targets
20437 @cindex bits in remote address
20438 Set the maximum size of address in a memory packet to the specified
20439 number of bits. @value{GDBN} will mask off the address bits above
20440 that number, when it passes addresses to the remote target. The
20441 default value is the number of bits in the target's address.
20443 @item show remoteaddresssize
20444 Show the current value of remote address size in bits.
20446 @item set serial baud @var{n}
20447 @cindex baud rate for remote targets
20448 Set the baud rate for the remote serial I/O to @var{n} baud. The
20449 value is used to set the speed of the serial port used for debugging
20452 @item show serial baud
20453 Show the current speed of the remote connection.
20455 @item set serial parity @var{parity}
20456 Set the parity for the remote serial I/O. Supported values of @var{parity} are:
20457 @code{even}, @code{none}, and @code{odd}. The default is @code{none}.
20459 @item show serial parity
20460 Show the current parity of the serial port.
20462 @item set remotebreak
20463 @cindex interrupt remote programs
20464 @cindex BREAK signal instead of Ctrl-C
20465 @anchor{set remotebreak}
20466 If set to on, @value{GDBN} sends a @code{BREAK} signal to the remote
20467 when you type @kbd{Ctrl-c} to interrupt the program running
20468 on the remote. If set to off, @value{GDBN} sends the @samp{Ctrl-C}
20469 character instead. The default is off, since most remote systems
20470 expect to see @samp{Ctrl-C} as the interrupt signal.
20472 @item show remotebreak
20473 Show whether @value{GDBN} sends @code{BREAK} or @samp{Ctrl-C} to
20474 interrupt the remote program.
20476 @item set remoteflow on
20477 @itemx set remoteflow off
20478 @kindex set remoteflow
20479 Enable or disable hardware flow control (@code{RTS}/@code{CTS})
20480 on the serial port used to communicate to the remote target.
20482 @item show remoteflow
20483 @kindex show remoteflow
20484 Show the current setting of hardware flow control.
20486 @item set remotelogbase @var{base}
20487 Set the base (a.k.a.@: radix) of logging serial protocol
20488 communications to @var{base}. Supported values of @var{base} are:
20489 @code{ascii}, @code{octal}, and @code{hex}. The default is
20492 @item show remotelogbase
20493 Show the current setting of the radix for logging remote serial
20496 @item set remotelogfile @var{file}
20497 @cindex record serial communications on file
20498 Record remote serial communications on the named @var{file}. The
20499 default is not to record at all.
20501 @item show remotelogfile.
20502 Show the current setting of the file name on which to record the
20503 serial communications.
20505 @item set remotetimeout @var{num}
20506 @cindex timeout for serial communications
20507 @cindex remote timeout
20508 Set the timeout limit to wait for the remote target to respond to
20509 @var{num} seconds. The default is 2 seconds.
20511 @item show remotetimeout
20512 Show the current number of seconds to wait for the remote target
20515 @cindex limit hardware breakpoints and watchpoints
20516 @cindex remote target, limit break- and watchpoints
20517 @anchor{set remote hardware-watchpoint-limit}
20518 @anchor{set remote hardware-breakpoint-limit}
20519 @item set remote hardware-watchpoint-limit @var{limit}
20520 @itemx set remote hardware-breakpoint-limit @var{limit}
20521 Restrict @value{GDBN} to using @var{limit} remote hardware breakpoint or
20522 watchpoints. A limit of -1, the default, is treated as unlimited.
20524 @cindex limit hardware watchpoints length
20525 @cindex remote target, limit watchpoints length
20526 @anchor{set remote hardware-watchpoint-length-limit}
20527 @item set remote hardware-watchpoint-length-limit @var{limit}
20528 Restrict @value{GDBN} to using @var{limit} bytes for the maximum length of
20529 a remote hardware watchpoint. A limit of -1, the default, is treated
20532 @item show remote hardware-watchpoint-length-limit
20533 Show the current limit (in bytes) of the maximum length of
20534 a remote hardware watchpoint.
20536 @item set remote exec-file @var{filename}
20537 @itemx show remote exec-file
20538 @anchor{set remote exec-file}
20539 @cindex executable file, for remote target
20540 Select the file used for @code{run} with @code{target
20541 extended-remote}. This should be set to a filename valid on the
20542 target system. If it is not set, the target will use a default
20543 filename (e.g.@: the last program run).
20545 @item set remote interrupt-sequence
20546 @cindex interrupt remote programs
20547 @cindex select Ctrl-C, BREAK or BREAK-g
20548 Allow the user to select one of @samp{Ctrl-C}, a @code{BREAK} or
20549 @samp{BREAK-g} as the
20550 sequence to the remote target in order to interrupt the execution.
20551 @samp{Ctrl-C} is a default. Some system prefers @code{BREAK} which
20552 is high level of serial line for some certain time.
20553 Linux kernel prefers @samp{BREAK-g}, a.k.a Magic SysRq g.
20554 It is @code{BREAK} signal followed by character @code{g}.
20556 @item show interrupt-sequence
20557 Show which of @samp{Ctrl-C}, @code{BREAK} or @code{BREAK-g}
20558 is sent by @value{GDBN} to interrupt the remote program.
20559 @code{BREAK-g} is BREAK signal followed by @code{g} and
20560 also known as Magic SysRq g.
20562 @item set remote interrupt-on-connect
20563 @cindex send interrupt-sequence on start
20564 Specify whether interrupt-sequence is sent to remote target when
20565 @value{GDBN} connects to it. This is mostly needed when you debug
20566 Linux kernel. Linux kernel expects @code{BREAK} followed by @code{g}
20567 which is known as Magic SysRq g in order to connect @value{GDBN}.
20569 @item show interrupt-on-connect
20570 Show whether interrupt-sequence is sent
20571 to remote target when @value{GDBN} connects to it.
20575 @item set tcp auto-retry on
20576 @cindex auto-retry, for remote TCP target
20577 Enable auto-retry for remote TCP connections. This is useful if the remote
20578 debugging agent is launched in parallel with @value{GDBN}; there is a race
20579 condition because the agent may not become ready to accept the connection
20580 before @value{GDBN} attempts to connect. When auto-retry is
20581 enabled, if the initial attempt to connect fails, @value{GDBN} reattempts
20582 to establish the connection using the timeout specified by
20583 @code{set tcp connect-timeout}.
20585 @item set tcp auto-retry off
20586 Do not auto-retry failed TCP connections.
20588 @item show tcp auto-retry
20589 Show the current auto-retry setting.
20591 @item set tcp connect-timeout @var{seconds}
20592 @itemx set tcp connect-timeout unlimited
20593 @cindex connection timeout, for remote TCP target
20594 @cindex timeout, for remote target connection
20595 Set the timeout for establishing a TCP connection to the remote target to
20596 @var{seconds}. The timeout affects both polling to retry failed connections
20597 (enabled by @code{set tcp auto-retry on}) and waiting for connections
20598 that are merely slow to complete, and represents an approximate cumulative
20599 value. If @var{seconds} is @code{unlimited}, there is no timeout and
20600 @value{GDBN} will keep attempting to establish a connection forever,
20601 unless interrupted with @kbd{Ctrl-c}. The default is 15 seconds.
20603 @item show tcp connect-timeout
20604 Show the current connection timeout setting.
20607 @cindex remote packets, enabling and disabling
20608 The @value{GDBN} remote protocol autodetects the packets supported by
20609 your debugging stub. If you need to override the autodetection, you
20610 can use these commands to enable or disable individual packets. Each
20611 packet can be set to @samp{on} (the remote target supports this
20612 packet), @samp{off} (the remote target does not support this packet),
20613 or @samp{auto} (detect remote target support for this packet). They
20614 all default to @samp{auto}. For more information about each packet,
20615 see @ref{Remote Protocol}.
20617 During normal use, you should not have to use any of these commands.
20618 If you do, that may be a bug in your remote debugging stub, or a bug
20619 in @value{GDBN}. You may want to report the problem to the
20620 @value{GDBN} developers.
20622 For each packet @var{name}, the command to enable or disable the
20623 packet is @code{set remote @var{name}-packet}. The available settings
20626 @multitable @columnfractions 0.28 0.32 0.25
20629 @tab Related Features
20631 @item @code{fetch-register}
20633 @tab @code{info registers}
20635 @item @code{set-register}
20639 @item @code{binary-download}
20641 @tab @code{load}, @code{set}
20643 @item @code{read-aux-vector}
20644 @tab @code{qXfer:auxv:read}
20645 @tab @code{info auxv}
20647 @item @code{symbol-lookup}
20648 @tab @code{qSymbol}
20649 @tab Detecting multiple threads
20651 @item @code{attach}
20652 @tab @code{vAttach}
20655 @item @code{verbose-resume}
20657 @tab Stepping or resuming multiple threads
20663 @item @code{software-breakpoint}
20667 @item @code{hardware-breakpoint}
20671 @item @code{write-watchpoint}
20675 @item @code{read-watchpoint}
20679 @item @code{access-watchpoint}
20683 @item @code{pid-to-exec-file}
20684 @tab @code{qXfer:exec-file:read}
20685 @tab @code{attach}, @code{run}
20687 @item @code{target-features}
20688 @tab @code{qXfer:features:read}
20689 @tab @code{set architecture}
20691 @item @code{library-info}
20692 @tab @code{qXfer:libraries:read}
20693 @tab @code{info sharedlibrary}
20695 @item @code{memory-map}
20696 @tab @code{qXfer:memory-map:read}
20697 @tab @code{info mem}
20699 @item @code{read-sdata-object}
20700 @tab @code{qXfer:sdata:read}
20701 @tab @code{print $_sdata}
20703 @item @code{read-spu-object}
20704 @tab @code{qXfer:spu:read}
20705 @tab @code{info spu}
20707 @item @code{write-spu-object}
20708 @tab @code{qXfer:spu:write}
20709 @tab @code{info spu}
20711 @item @code{read-siginfo-object}
20712 @tab @code{qXfer:siginfo:read}
20713 @tab @code{print $_siginfo}
20715 @item @code{write-siginfo-object}
20716 @tab @code{qXfer:siginfo:write}
20717 @tab @code{set $_siginfo}
20719 @item @code{threads}
20720 @tab @code{qXfer:threads:read}
20721 @tab @code{info threads}
20723 @item @code{get-thread-local-@*storage-address}
20724 @tab @code{qGetTLSAddr}
20725 @tab Displaying @code{__thread} variables
20727 @item @code{get-thread-information-block-address}
20728 @tab @code{qGetTIBAddr}
20729 @tab Display MS-Windows Thread Information Block.
20731 @item @code{search-memory}
20732 @tab @code{qSearch:memory}
20735 @item @code{supported-packets}
20736 @tab @code{qSupported}
20737 @tab Remote communications parameters
20739 @item @code{catch-syscalls}
20740 @tab @code{QCatchSyscalls}
20741 @tab @code{catch syscall}
20743 @item @code{pass-signals}
20744 @tab @code{QPassSignals}
20745 @tab @code{handle @var{signal}}
20747 @item @code{program-signals}
20748 @tab @code{QProgramSignals}
20749 @tab @code{handle @var{signal}}
20751 @item @code{hostio-close-packet}
20752 @tab @code{vFile:close}
20753 @tab @code{remote get}, @code{remote put}
20755 @item @code{hostio-open-packet}
20756 @tab @code{vFile:open}
20757 @tab @code{remote get}, @code{remote put}
20759 @item @code{hostio-pread-packet}
20760 @tab @code{vFile:pread}
20761 @tab @code{remote get}, @code{remote put}
20763 @item @code{hostio-pwrite-packet}
20764 @tab @code{vFile:pwrite}
20765 @tab @code{remote get}, @code{remote put}
20767 @item @code{hostio-unlink-packet}
20768 @tab @code{vFile:unlink}
20769 @tab @code{remote delete}
20771 @item @code{hostio-readlink-packet}
20772 @tab @code{vFile:readlink}
20775 @item @code{hostio-fstat-packet}
20776 @tab @code{vFile:fstat}
20779 @item @code{hostio-setfs-packet}
20780 @tab @code{vFile:setfs}
20783 @item @code{noack-packet}
20784 @tab @code{QStartNoAckMode}
20785 @tab Packet acknowledgment
20787 @item @code{osdata}
20788 @tab @code{qXfer:osdata:read}
20789 @tab @code{info os}
20791 @item @code{query-attached}
20792 @tab @code{qAttached}
20793 @tab Querying remote process attach state.
20795 @item @code{trace-buffer-size}
20796 @tab @code{QTBuffer:size}
20797 @tab @code{set trace-buffer-size}
20799 @item @code{trace-status}
20800 @tab @code{qTStatus}
20801 @tab @code{tstatus}
20803 @item @code{traceframe-info}
20804 @tab @code{qXfer:traceframe-info:read}
20805 @tab Traceframe info
20807 @item @code{install-in-trace}
20808 @tab @code{InstallInTrace}
20809 @tab Install tracepoint in tracing
20811 @item @code{disable-randomization}
20812 @tab @code{QDisableRandomization}
20813 @tab @code{set disable-randomization}
20815 @item @code{startup-with-shell}
20816 @tab @code{QStartupWithShell}
20817 @tab @code{set startup-with-shell}
20819 @item @code{conditional-breakpoints-packet}
20820 @tab @code{Z0 and Z1}
20821 @tab @code{Support for target-side breakpoint condition evaluation}
20823 @item @code{multiprocess-extensions}
20824 @tab @code{multiprocess extensions}
20825 @tab Debug multiple processes and remote process PID awareness
20827 @item @code{swbreak-feature}
20828 @tab @code{swbreak stop reason}
20831 @item @code{hwbreak-feature}
20832 @tab @code{hwbreak stop reason}
20835 @item @code{fork-event-feature}
20836 @tab @code{fork stop reason}
20839 @item @code{vfork-event-feature}
20840 @tab @code{vfork stop reason}
20843 @item @code{exec-event-feature}
20844 @tab @code{exec stop reason}
20847 @item @code{thread-events}
20848 @tab @code{QThreadEvents}
20849 @tab Tracking thread lifetime.
20851 @item @code{no-resumed-stop-reply}
20852 @tab @code{no resumed thread left stop reply}
20853 @tab Tracking thread lifetime.
20858 @section Implementing a Remote Stub
20860 @cindex debugging stub, example
20861 @cindex remote stub, example
20862 @cindex stub example, remote debugging
20863 The stub files provided with @value{GDBN} implement the target side of the
20864 communication protocol, and the @value{GDBN} side is implemented in the
20865 @value{GDBN} source file @file{remote.c}. Normally, you can simply allow
20866 these subroutines to communicate, and ignore the details. (If you're
20867 implementing your own stub file, you can still ignore the details: start
20868 with one of the existing stub files. @file{sparc-stub.c} is the best
20869 organized, and therefore the easiest to read.)
20871 @cindex remote serial debugging, overview
20872 To debug a program running on another machine (the debugging
20873 @dfn{target} machine), you must first arrange for all the usual
20874 prerequisites for the program to run by itself. For example, for a C
20879 A startup routine to set up the C runtime environment; these usually
20880 have a name like @file{crt0}. The startup routine may be supplied by
20881 your hardware supplier, or you may have to write your own.
20884 A C subroutine library to support your program's
20885 subroutine calls, notably managing input and output.
20888 A way of getting your program to the other machine---for example, a
20889 download program. These are often supplied by the hardware
20890 manufacturer, but you may have to write your own from hardware
20894 The next step is to arrange for your program to use a serial port to
20895 communicate with the machine where @value{GDBN} is running (the @dfn{host}
20896 machine). In general terms, the scheme looks like this:
20900 @value{GDBN} already understands how to use this protocol; when everything
20901 else is set up, you can simply use the @samp{target remote} command
20902 (@pxref{Targets,,Specifying a Debugging Target}).
20904 @item On the target,
20905 you must link with your program a few special-purpose subroutines that
20906 implement the @value{GDBN} remote serial protocol. The file containing these
20907 subroutines is called a @dfn{debugging stub}.
20909 On certain remote targets, you can use an auxiliary program
20910 @code{gdbserver} instead of linking a stub into your program.
20911 @xref{Server,,Using the @code{gdbserver} Program}, for details.
20914 The debugging stub is specific to the architecture of the remote
20915 machine; for example, use @file{sparc-stub.c} to debug programs on
20918 @cindex remote serial stub list
20919 These working remote stubs are distributed with @value{GDBN}:
20924 @cindex @file{i386-stub.c}
20927 For Intel 386 and compatible architectures.
20930 @cindex @file{m68k-stub.c}
20931 @cindex Motorola 680x0
20933 For Motorola 680x0 architectures.
20936 @cindex @file{sh-stub.c}
20939 For Renesas SH architectures.
20942 @cindex @file{sparc-stub.c}
20944 For @sc{sparc} architectures.
20946 @item sparcl-stub.c
20947 @cindex @file{sparcl-stub.c}
20950 For Fujitsu @sc{sparclite} architectures.
20954 The @file{README} file in the @value{GDBN} distribution may list other
20955 recently added stubs.
20958 * Stub Contents:: What the stub can do for you
20959 * Bootstrapping:: What you must do for the stub
20960 * Debug Session:: Putting it all together
20963 @node Stub Contents
20964 @subsection What the Stub Can Do for You
20966 @cindex remote serial stub
20967 The debugging stub for your architecture supplies these three
20971 @item set_debug_traps
20972 @findex set_debug_traps
20973 @cindex remote serial stub, initialization
20974 This routine arranges for @code{handle_exception} to run when your
20975 program stops. You must call this subroutine explicitly in your
20976 program's startup code.
20978 @item handle_exception
20979 @findex handle_exception
20980 @cindex remote serial stub, main routine
20981 This is the central workhorse, but your program never calls it
20982 explicitly---the setup code arranges for @code{handle_exception} to
20983 run when a trap is triggered.
20985 @code{handle_exception} takes control when your program stops during
20986 execution (for example, on a breakpoint), and mediates communications
20987 with @value{GDBN} on the host machine. This is where the communications
20988 protocol is implemented; @code{handle_exception} acts as the @value{GDBN}
20989 representative on the target machine. It begins by sending summary
20990 information on the state of your program, then continues to execute,
20991 retrieving and transmitting any information @value{GDBN} needs, until you
20992 execute a @value{GDBN} command that makes your program resume; at that point,
20993 @code{handle_exception} returns control to your own code on the target
20997 @cindex @code{breakpoint} subroutine, remote
20998 Use this auxiliary subroutine to make your program contain a
20999 breakpoint. Depending on the particular situation, this may be the only
21000 way for @value{GDBN} to get control. For instance, if your target
21001 machine has some sort of interrupt button, you won't need to call this;
21002 pressing the interrupt button transfers control to
21003 @code{handle_exception}---in effect, to @value{GDBN}. On some machines,
21004 simply receiving characters on the serial port may also trigger a trap;
21005 again, in that situation, you don't need to call @code{breakpoint} from
21006 your own program---simply running @samp{target remote} from the host
21007 @value{GDBN} session gets control.
21009 Call @code{breakpoint} if none of these is true, or if you simply want
21010 to make certain your program stops at a predetermined point for the
21011 start of your debugging session.
21014 @node Bootstrapping
21015 @subsection What You Must Do for the Stub
21017 @cindex remote stub, support routines
21018 The debugging stubs that come with @value{GDBN} are set up for a particular
21019 chip architecture, but they have no information about the rest of your
21020 debugging target machine.
21022 First of all you need to tell the stub how to communicate with the
21026 @item int getDebugChar()
21027 @findex getDebugChar
21028 Write this subroutine to read a single character from the serial port.
21029 It may be identical to @code{getchar} for your target system; a
21030 different name is used to allow you to distinguish the two if you wish.
21032 @item void putDebugChar(int)
21033 @findex putDebugChar
21034 Write this subroutine to write a single character to the serial port.
21035 It may be identical to @code{putchar} for your target system; a
21036 different name is used to allow you to distinguish the two if you wish.
21039 @cindex control C, and remote debugging
21040 @cindex interrupting remote targets
21041 If you want @value{GDBN} to be able to stop your program while it is
21042 running, you need to use an interrupt-driven serial driver, and arrange
21043 for it to stop when it receives a @code{^C} (@samp{\003}, the control-C
21044 character). That is the character which @value{GDBN} uses to tell the
21045 remote system to stop.
21047 Getting the debugging target to return the proper status to @value{GDBN}
21048 probably requires changes to the standard stub; one quick and dirty way
21049 is to just execute a breakpoint instruction (the ``dirty'' part is that
21050 @value{GDBN} reports a @code{SIGTRAP} instead of a @code{SIGINT}).
21052 Other routines you need to supply are:
21055 @item void exceptionHandler (int @var{exception_number}, void *@var{exception_address})
21056 @findex exceptionHandler
21057 Write this function to install @var{exception_address} in the exception
21058 handling tables. You need to do this because the stub does not have any
21059 way of knowing what the exception handling tables on your target system
21060 are like (for example, the processor's table might be in @sc{rom},
21061 containing entries which point to a table in @sc{ram}).
21062 The @var{exception_number} specifies the exception which should be changed;
21063 its meaning is architecture-dependent (for example, different numbers
21064 might represent divide by zero, misaligned access, etc). When this
21065 exception occurs, control should be transferred directly to
21066 @var{exception_address}, and the processor state (stack, registers,
21067 and so on) should be just as it is when a processor exception occurs. So if
21068 you want to use a jump instruction to reach @var{exception_address}, it
21069 should be a simple jump, not a jump to subroutine.
21071 For the 386, @var{exception_address} should be installed as an interrupt
21072 gate so that interrupts are masked while the handler runs. The gate
21073 should be at privilege level 0 (the most privileged level). The
21074 @sc{sparc} and 68k stubs are able to mask interrupts themselves without
21075 help from @code{exceptionHandler}.
21077 @item void flush_i_cache()
21078 @findex flush_i_cache
21079 On @sc{sparc} and @sc{sparclite} only, write this subroutine to flush the
21080 instruction cache, if any, on your target machine. If there is no
21081 instruction cache, this subroutine may be a no-op.
21083 On target machines that have instruction caches, @value{GDBN} requires this
21084 function to make certain that the state of your program is stable.
21088 You must also make sure this library routine is available:
21091 @item void *memset(void *, int, int)
21093 This is the standard library function @code{memset} that sets an area of
21094 memory to a known value. If you have one of the free versions of
21095 @code{libc.a}, @code{memset} can be found there; otherwise, you must
21096 either obtain it from your hardware manufacturer, or write your own.
21099 If you do not use the GNU C compiler, you may need other standard
21100 library subroutines as well; this varies from one stub to another,
21101 but in general the stubs are likely to use any of the common library
21102 subroutines which @code{@value{NGCC}} generates as inline code.
21105 @node Debug Session
21106 @subsection Putting it All Together
21108 @cindex remote serial debugging summary
21109 In summary, when your program is ready to debug, you must follow these
21114 Make sure you have defined the supporting low-level routines
21115 (@pxref{Bootstrapping,,What You Must Do for the Stub}):
21117 @code{getDebugChar}, @code{putDebugChar},
21118 @code{flush_i_cache}, @code{memset}, @code{exceptionHandler}.
21122 Insert these lines in your program's startup code, before the main
21123 procedure is called:
21130 On some machines, when a breakpoint trap is raised, the hardware
21131 automatically makes the PC point to the instruction after the
21132 breakpoint. If your machine doesn't do that, you may need to adjust
21133 @code{handle_exception} to arrange for it to return to the instruction
21134 after the breakpoint on this first invocation, so that your program
21135 doesn't keep hitting the initial breakpoint instead of making
21139 For the 680x0 stub only, you need to provide a variable called
21140 @code{exceptionHook}. Normally you just use:
21143 void (*exceptionHook)() = 0;
21147 but if before calling @code{set_debug_traps}, you set it to point to a
21148 function in your program, that function is called when
21149 @code{@value{GDBN}} continues after stopping on a trap (for example, bus
21150 error). The function indicated by @code{exceptionHook} is called with
21151 one parameter: an @code{int} which is the exception number.
21154 Compile and link together: your program, the @value{GDBN} debugging stub for
21155 your target architecture, and the supporting subroutines.
21158 Make sure you have a serial connection between your target machine and
21159 the @value{GDBN} host, and identify the serial port on the host.
21162 @c The "remote" target now provides a `load' command, so we should
21163 @c document that. FIXME.
21164 Download your program to your target machine (or get it there by
21165 whatever means the manufacturer provides), and start it.
21168 Start @value{GDBN} on the host, and connect to the target
21169 (@pxref{Connecting,,Connecting to a Remote Target}).
21173 @node Configurations
21174 @chapter Configuration-Specific Information
21176 While nearly all @value{GDBN} commands are available for all native and
21177 cross versions of the debugger, there are some exceptions. This chapter
21178 describes things that are only available in certain configurations.
21180 There are three major categories of configurations: native
21181 configurations, where the host and target are the same, embedded
21182 operating system configurations, which are usually the same for several
21183 different processor architectures, and bare embedded processors, which
21184 are quite different from each other.
21189 * Embedded Processors::
21196 This section describes details specific to particular native
21200 * BSD libkvm Interface:: Debugging BSD kernel memory images
21201 * SVR4 Process Information:: SVR4 process information
21202 * DJGPP Native:: Features specific to the DJGPP port
21203 * Cygwin Native:: Features specific to the Cygwin port
21204 * Hurd Native:: Features specific to @sc{gnu} Hurd
21205 * Darwin:: Features specific to Darwin
21208 @node BSD libkvm Interface
21209 @subsection BSD libkvm Interface
21212 @cindex kernel memory image
21213 @cindex kernel crash dump
21215 BSD-derived systems (FreeBSD/NetBSD/OpenBSD) have a kernel memory
21216 interface that provides a uniform interface for accessing kernel virtual
21217 memory images, including live systems and crash dumps. @value{GDBN}
21218 uses this interface to allow you to debug live kernels and kernel crash
21219 dumps on many native BSD configurations. This is implemented as a
21220 special @code{kvm} debugging target. For debugging a live system, load
21221 the currently running kernel into @value{GDBN} and connect to the
21225 (@value{GDBP}) @b{target kvm}
21228 For debugging crash dumps, provide the file name of the crash dump as an
21232 (@value{GDBP}) @b{target kvm /var/crash/bsd.0}
21235 Once connected to the @code{kvm} target, the following commands are
21241 Set current context from the @dfn{Process Control Block} (PCB) address.
21244 Set current context from proc address. This command isn't available on
21245 modern FreeBSD systems.
21248 @node SVR4 Process Information
21249 @subsection SVR4 Process Information
21251 @cindex examine process image
21252 @cindex process info via @file{/proc}
21254 Many versions of SVR4 and compatible systems provide a facility called
21255 @samp{/proc} that can be used to examine the image of a running
21256 process using file-system subroutines.
21258 If @value{GDBN} is configured for an operating system with this
21259 facility, the command @code{info proc} is available to report
21260 information about the process running your program, or about any
21261 process running on your system. This includes, as of this writing,
21262 @sc{gnu}/Linux and Solaris, for example.
21264 This command may also work on core files that were created on a system
21265 that has the @samp{/proc} facility.
21271 @itemx info proc @var{process-id}
21272 Summarize available information about any running process. If a
21273 process ID is specified by @var{process-id}, display information about
21274 that process; otherwise display information about the program being
21275 debugged. The summary includes the debugged process ID, the command
21276 line used to invoke it, its current working directory, and its
21277 executable file's absolute file name.
21279 On some systems, @var{process-id} can be of the form
21280 @samp{[@var{pid}]/@var{tid}} which specifies a certain thread ID
21281 within a process. If the optional @var{pid} part is missing, it means
21282 a thread from the process being debugged (the leading @samp{/} still
21283 needs to be present, or else @value{GDBN} will interpret the number as
21284 a process ID rather than a thread ID).
21286 @item info proc cmdline
21287 @cindex info proc cmdline
21288 Show the original command line of the process. This command is
21289 specific to @sc{gnu}/Linux.
21291 @item info proc cwd
21292 @cindex info proc cwd
21293 Show the current working directory of the process. This command is
21294 specific to @sc{gnu}/Linux.
21296 @item info proc exe
21297 @cindex info proc exe
21298 Show the name of executable of the process. This command is specific
21301 @item info proc mappings
21302 @cindex memory address space mappings
21303 Report the memory address space ranges accessible in the program, with
21304 information on whether the process has read, write, or execute access
21305 rights to each range. On @sc{gnu}/Linux systems, each memory range
21306 includes the object file which is mapped to that range, instead of the
21307 memory access rights to that range.
21309 @item info proc stat
21310 @itemx info proc status
21311 @cindex process detailed status information
21312 These subcommands are specific to @sc{gnu}/Linux systems. They show
21313 the process-related information, including the user ID and group ID;
21314 how many threads are there in the process; its virtual memory usage;
21315 the signals that are pending, blocked, and ignored; its TTY; its
21316 consumption of system and user time; its stack size; its @samp{nice}
21317 value; etc. For more information, see the @samp{proc} man page
21318 (type @kbd{man 5 proc} from your shell prompt).
21320 @item info proc all
21321 Show all the information about the process described under all of the
21322 above @code{info proc} subcommands.
21325 @comment These sub-options of 'info proc' were not included when
21326 @comment procfs.c was re-written. Keep their descriptions around
21327 @comment against the day when someone finds the time to put them back in.
21328 @kindex info proc times
21329 @item info proc times
21330 Starting time, user CPU time, and system CPU time for your program and
21333 @kindex info proc id
21335 Report on the process IDs related to your program: its own process ID,
21336 the ID of its parent, the process group ID, and the session ID.
21339 @item set procfs-trace
21340 @kindex set procfs-trace
21341 @cindex @code{procfs} API calls
21342 This command enables and disables tracing of @code{procfs} API calls.
21344 @item show procfs-trace
21345 @kindex show procfs-trace
21346 Show the current state of @code{procfs} API call tracing.
21348 @item set procfs-file @var{file}
21349 @kindex set procfs-file
21350 Tell @value{GDBN} to write @code{procfs} API trace to the named
21351 @var{file}. @value{GDBN} appends the trace info to the previous
21352 contents of the file. The default is to display the trace on the
21355 @item show procfs-file
21356 @kindex show procfs-file
21357 Show the file to which @code{procfs} API trace is written.
21359 @item proc-trace-entry
21360 @itemx proc-trace-exit
21361 @itemx proc-untrace-entry
21362 @itemx proc-untrace-exit
21363 @kindex proc-trace-entry
21364 @kindex proc-trace-exit
21365 @kindex proc-untrace-entry
21366 @kindex proc-untrace-exit
21367 These commands enable and disable tracing of entries into and exits
21368 from the @code{syscall} interface.
21371 @kindex info pidlist
21372 @cindex process list, QNX Neutrino
21373 For QNX Neutrino only, this command displays the list of all the
21374 processes and all the threads within each process.
21377 @kindex info meminfo
21378 @cindex mapinfo list, QNX Neutrino
21379 For QNX Neutrino only, this command displays the list of all mapinfos.
21383 @subsection Features for Debugging @sc{djgpp} Programs
21384 @cindex @sc{djgpp} debugging
21385 @cindex native @sc{djgpp} debugging
21386 @cindex MS-DOS-specific commands
21389 @sc{djgpp} is a port of the @sc{gnu} development tools to MS-DOS and
21390 MS-Windows. @sc{djgpp} programs are 32-bit protected-mode programs
21391 that use the @dfn{DPMI} (DOS Protected-Mode Interface) API to run on
21392 top of real-mode DOS systems and their emulations.
21394 @value{GDBN} supports native debugging of @sc{djgpp} programs, and
21395 defines a few commands specific to the @sc{djgpp} port. This
21396 subsection describes those commands.
21401 This is a prefix of @sc{djgpp}-specific commands which print
21402 information about the target system and important OS structures.
21405 @cindex MS-DOS system info
21406 @cindex free memory information (MS-DOS)
21407 @item info dos sysinfo
21408 This command displays assorted information about the underlying
21409 platform: the CPU type and features, the OS version and flavor, the
21410 DPMI version, and the available conventional and DPMI memory.
21415 @cindex segment descriptor tables
21416 @cindex descriptor tables display
21418 @itemx info dos ldt
21419 @itemx info dos idt
21420 These 3 commands display entries from, respectively, Global, Local,
21421 and Interrupt Descriptor Tables (GDT, LDT, and IDT). The descriptor
21422 tables are data structures which store a descriptor for each segment
21423 that is currently in use. The segment's selector is an index into a
21424 descriptor table; the table entry for that index holds the
21425 descriptor's base address and limit, and its attributes and access
21428 A typical @sc{djgpp} program uses 3 segments: a code segment, a data
21429 segment (used for both data and the stack), and a DOS segment (which
21430 allows access to DOS/BIOS data structures and absolute addresses in
21431 conventional memory). However, the DPMI host will usually define
21432 additional segments in order to support the DPMI environment.
21434 @cindex garbled pointers
21435 These commands allow to display entries from the descriptor tables.
21436 Without an argument, all entries from the specified table are
21437 displayed. An argument, which should be an integer expression, means
21438 display a single entry whose index is given by the argument. For
21439 example, here's a convenient way to display information about the
21440 debugged program's data segment:
21443 @exdent @code{(@value{GDBP}) info dos ldt $ds}
21444 @exdent @code{0x13f: base=0x11970000 limit=0x0009ffff 32-Bit Data (Read/Write, Exp-up)}
21448 This comes in handy when you want to see whether a pointer is outside
21449 the data segment's limit (i.e.@: @dfn{garbled}).
21451 @cindex page tables display (MS-DOS)
21453 @itemx info dos pte
21454 These two commands display entries from, respectively, the Page
21455 Directory and the Page Tables. Page Directories and Page Tables are
21456 data structures which control how virtual memory addresses are mapped
21457 into physical addresses. A Page Table includes an entry for every
21458 page of memory that is mapped into the program's address space; there
21459 may be several Page Tables, each one holding up to 4096 entries. A
21460 Page Directory has up to 4096 entries, one each for every Page Table
21461 that is currently in use.
21463 Without an argument, @kbd{info dos pde} displays the entire Page
21464 Directory, and @kbd{info dos pte} displays all the entries in all of
21465 the Page Tables. An argument, an integer expression, given to the
21466 @kbd{info dos pde} command means display only that entry from the Page
21467 Directory table. An argument given to the @kbd{info dos pte} command
21468 means display entries from a single Page Table, the one pointed to by
21469 the specified entry in the Page Directory.
21471 @cindex direct memory access (DMA) on MS-DOS
21472 These commands are useful when your program uses @dfn{DMA} (Direct
21473 Memory Access), which needs physical addresses to program the DMA
21476 These commands are supported only with some DPMI servers.
21478 @cindex physical address from linear address
21479 @item info dos address-pte @var{addr}
21480 This command displays the Page Table entry for a specified linear
21481 address. The argument @var{addr} is a linear address which should
21482 already have the appropriate segment's base address added to it,
21483 because this command accepts addresses which may belong to @emph{any}
21484 segment. For example, here's how to display the Page Table entry for
21485 the page where a variable @code{i} is stored:
21488 @exdent @code{(@value{GDBP}) info dos address-pte __djgpp_base_address + (char *)&i}
21489 @exdent @code{Page Table entry for address 0x11a00d30:}
21490 @exdent @code{Base=0x02698000 Dirty Acc. Not-Cached Write-Back Usr Read-Write +0xd30}
21494 This says that @code{i} is stored at offset @code{0xd30} from the page
21495 whose physical base address is @code{0x02698000}, and shows all the
21496 attributes of that page.
21498 Note that you must cast the addresses of variables to a @code{char *},
21499 since otherwise the value of @code{__djgpp_base_address}, the base
21500 address of all variables and functions in a @sc{djgpp} program, will
21501 be added using the rules of C pointer arithmetics: if @code{i} is
21502 declared an @code{int}, @value{GDBN} will add 4 times the value of
21503 @code{__djgpp_base_address} to the address of @code{i}.
21505 Here's another example, it displays the Page Table entry for the
21509 @exdent @code{(@value{GDBP}) info dos address-pte *((unsigned *)&_go32_info_block + 3)}
21510 @exdent @code{Page Table entry for address 0x29110:}
21511 @exdent @code{Base=0x00029000 Dirty Acc. Not-Cached Write-Back Usr Read-Write +0x110}
21515 (The @code{+ 3} offset is because the transfer buffer's address is the
21516 3rd member of the @code{_go32_info_block} structure.) The output
21517 clearly shows that this DPMI server maps the addresses in conventional
21518 memory 1:1, i.e.@: the physical (@code{0x00029000} + @code{0x110}) and
21519 linear (@code{0x29110}) addresses are identical.
21521 This command is supported only with some DPMI servers.
21524 @cindex DOS serial data link, remote debugging
21525 In addition to native debugging, the DJGPP port supports remote
21526 debugging via a serial data link. The following commands are specific
21527 to remote serial debugging in the DJGPP port of @value{GDBN}.
21530 @kindex set com1base
21531 @kindex set com1irq
21532 @kindex set com2base
21533 @kindex set com2irq
21534 @kindex set com3base
21535 @kindex set com3irq
21536 @kindex set com4base
21537 @kindex set com4irq
21538 @item set com1base @var{addr}
21539 This command sets the base I/O port address of the @file{COM1} serial
21542 @item set com1irq @var{irq}
21543 This command sets the @dfn{Interrupt Request} (@code{IRQ}) line to use
21544 for the @file{COM1} serial port.
21546 There are similar commands @samp{set com2base}, @samp{set com3irq},
21547 etc.@: for setting the port address and the @code{IRQ} lines for the
21550 @kindex show com1base
21551 @kindex show com1irq
21552 @kindex show com2base
21553 @kindex show com2irq
21554 @kindex show com3base
21555 @kindex show com3irq
21556 @kindex show com4base
21557 @kindex show com4irq
21558 The related commands @samp{show com1base}, @samp{show com1irq} etc.@:
21559 display the current settings of the base address and the @code{IRQ}
21560 lines used by the COM ports.
21563 @kindex info serial
21564 @cindex DOS serial port status
21565 This command prints the status of the 4 DOS serial ports. For each
21566 port, it prints whether it's active or not, its I/O base address and
21567 IRQ number, whether it uses a 16550-style FIFO, its baudrate, and the
21568 counts of various errors encountered so far.
21572 @node Cygwin Native
21573 @subsection Features for Debugging MS Windows PE Executables
21574 @cindex MS Windows debugging
21575 @cindex native Cygwin debugging
21576 @cindex Cygwin-specific commands
21578 @value{GDBN} supports native debugging of MS Windows programs, including
21579 DLLs with and without symbolic debugging information.
21581 @cindex Ctrl-BREAK, MS-Windows
21582 @cindex interrupt debuggee on MS-Windows
21583 MS-Windows programs that call @code{SetConsoleMode} to switch off the
21584 special meaning of the @samp{Ctrl-C} keystroke cannot be interrupted
21585 by typing @kbd{C-c}. For this reason, @value{GDBN} on MS-Windows
21586 supports @kbd{C-@key{BREAK}} as an alternative interrupt key
21587 sequence, which can be used to interrupt the debuggee even if it
21590 There are various additional Cygwin-specific commands, described in
21591 this section. Working with DLLs that have no debugging symbols is
21592 described in @ref{Non-debug DLL Symbols}.
21597 This is a prefix of MS Windows-specific commands which print
21598 information about the target system and important OS structures.
21600 @item info w32 selector
21601 This command displays information returned by
21602 the Win32 API @code{GetThreadSelectorEntry} function.
21603 It takes an optional argument that is evaluated to
21604 a long value to give the information about this given selector.
21605 Without argument, this command displays information
21606 about the six segment registers.
21608 @item info w32 thread-information-block
21609 This command displays thread specific information stored in the
21610 Thread Information Block (readable on the X86 CPU family using @code{$fs}
21611 selector for 32-bit programs and @code{$gs} for 64-bit programs).
21613 @kindex signal-event
21614 @item signal-event @var{id}
21615 This command signals an event with user-provided @var{id}. Used to resume
21616 crashing process when attached to it using MS-Windows JIT debugging (AeDebug).
21618 To use it, create or edit the following keys in
21619 @code{HKLM\SOFTWARE\Microsoft\Windows NT\CurrentVersion\AeDebug} and/or
21620 @code{HKLM\SOFTWARE\Wow6432Node\Microsoft\Windows NT\CurrentVersion\AeDebug}
21621 (for x86_64 versions):
21625 @code{Debugger} (REG_SZ) --- a command to launch the debugger.
21626 Suggested command is: @code{@var{fully-qualified-path-to-gdb.exe} -ex
21627 "attach %ld" -ex "signal-event %ld" -ex "continue"}.
21629 The first @code{%ld} will be replaced by the process ID of the
21630 crashing process, the second @code{%ld} will be replaced by the ID of
21631 the event that blocks the crashing process, waiting for @value{GDBN}
21635 @code{Auto} (REG_SZ) --- either @code{1} or @code{0}. @code{1} will
21636 make the system run debugger specified by the Debugger key
21637 automatically, @code{0} will cause a dialog box with ``OK'' and
21638 ``Cancel'' buttons to appear, which allows the user to either
21639 terminate the crashing process (OK) or debug it (Cancel).
21642 @kindex set cygwin-exceptions
21643 @cindex debugging the Cygwin DLL
21644 @cindex Cygwin DLL, debugging
21645 @item set cygwin-exceptions @var{mode}
21646 If @var{mode} is @code{on}, @value{GDBN} will break on exceptions that
21647 happen inside the Cygwin DLL. If @var{mode} is @code{off},
21648 @value{GDBN} will delay recognition of exceptions, and may ignore some
21649 exceptions which seem to be caused by internal Cygwin DLL
21650 ``bookkeeping''. This option is meant primarily for debugging the
21651 Cygwin DLL itself; the default value is @code{off} to avoid annoying
21652 @value{GDBN} users with false @code{SIGSEGV} signals.
21654 @kindex show cygwin-exceptions
21655 @item show cygwin-exceptions
21656 Displays whether @value{GDBN} will break on exceptions that happen
21657 inside the Cygwin DLL itself.
21659 @kindex set new-console
21660 @item set new-console @var{mode}
21661 If @var{mode} is @code{on} the debuggee will
21662 be started in a new console on next start.
21663 If @var{mode} is @code{off}, the debuggee will
21664 be started in the same console as the debugger.
21666 @kindex show new-console
21667 @item show new-console
21668 Displays whether a new console is used
21669 when the debuggee is started.
21671 @kindex set new-group
21672 @item set new-group @var{mode}
21673 This boolean value controls whether the debuggee should
21674 start a new group or stay in the same group as the debugger.
21675 This affects the way the Windows OS handles
21678 @kindex show new-group
21679 @item show new-group
21680 Displays current value of new-group boolean.
21682 @kindex set debugevents
21683 @item set debugevents
21684 This boolean value adds debug output concerning kernel events related
21685 to the debuggee seen by the debugger. This includes events that
21686 signal thread and process creation and exit, DLL loading and
21687 unloading, console interrupts, and debugging messages produced by the
21688 Windows @code{OutputDebugString} API call.
21690 @kindex set debugexec
21691 @item set debugexec
21692 This boolean value adds debug output concerning execute events
21693 (such as resume thread) seen by the debugger.
21695 @kindex set debugexceptions
21696 @item set debugexceptions
21697 This boolean value adds debug output concerning exceptions in the
21698 debuggee seen by the debugger.
21700 @kindex set debugmemory
21701 @item set debugmemory
21702 This boolean value adds debug output concerning debuggee memory reads
21703 and writes by the debugger.
21707 This boolean values specifies whether the debuggee is called
21708 via a shell or directly (default value is on).
21712 Displays if the debuggee will be started with a shell.
21717 * Non-debug DLL Symbols:: Support for DLLs without debugging symbols
21720 @node Non-debug DLL Symbols
21721 @subsubsection Support for DLLs without Debugging Symbols
21722 @cindex DLLs with no debugging symbols
21723 @cindex Minimal symbols and DLLs
21725 Very often on windows, some of the DLLs that your program relies on do
21726 not include symbolic debugging information (for example,
21727 @file{kernel32.dll}). When @value{GDBN} doesn't recognize any debugging
21728 symbols in a DLL, it relies on the minimal amount of symbolic
21729 information contained in the DLL's export table. This section
21730 describes working with such symbols, known internally to @value{GDBN} as
21731 ``minimal symbols''.
21733 Note that before the debugged program has started execution, no DLLs
21734 will have been loaded. The easiest way around this problem is simply to
21735 start the program --- either by setting a breakpoint or letting the
21736 program run once to completion.
21738 @subsubsection DLL Name Prefixes
21740 In keeping with the naming conventions used by the Microsoft debugging
21741 tools, DLL export symbols are made available with a prefix based on the
21742 DLL name, for instance @code{KERNEL32!CreateFileA}. The plain name is
21743 also entered into the symbol table, so @code{CreateFileA} is often
21744 sufficient. In some cases there will be name clashes within a program
21745 (particularly if the executable itself includes full debugging symbols)
21746 necessitating the use of the fully qualified name when referring to the
21747 contents of the DLL. Use single-quotes around the name to avoid the
21748 exclamation mark (``!'') being interpreted as a language operator.
21750 Note that the internal name of the DLL may be all upper-case, even
21751 though the file name of the DLL is lower-case, or vice-versa. Since
21752 symbols within @value{GDBN} are @emph{case-sensitive} this may cause
21753 some confusion. If in doubt, try the @code{info functions} and
21754 @code{info variables} commands or even @code{maint print msymbols}
21755 (@pxref{Symbols}). Here's an example:
21758 (@value{GDBP}) info function CreateFileA
21759 All functions matching regular expression "CreateFileA":
21761 Non-debugging symbols:
21762 0x77e885f4 CreateFileA
21763 0x77e885f4 KERNEL32!CreateFileA
21767 (@value{GDBP}) info function !
21768 All functions matching regular expression "!":
21770 Non-debugging symbols:
21771 0x6100114c cygwin1!__assert
21772 0x61004034 cygwin1!_dll_crt0@@0
21773 0x61004240 cygwin1!dll_crt0(per_process *)
21777 @subsubsection Working with Minimal Symbols
21779 Symbols extracted from a DLL's export table do not contain very much
21780 type information. All that @value{GDBN} can do is guess whether a symbol
21781 refers to a function or variable depending on the linker section that
21782 contains the symbol. Also note that the actual contents of the memory
21783 contained in a DLL are not available unless the program is running. This
21784 means that you cannot examine the contents of a variable or disassemble
21785 a function within a DLL without a running program.
21787 Variables are generally treated as pointers and dereferenced
21788 automatically. For this reason, it is often necessary to prefix a
21789 variable name with the address-of operator (``&'') and provide explicit
21790 type information in the command. Here's an example of the type of
21794 (@value{GDBP}) print 'cygwin1!__argv'
21799 (@value{GDBP}) x 'cygwin1!__argv'
21800 0x10021610: "\230y\""
21803 And two possible solutions:
21806 (@value{GDBP}) print ((char **)'cygwin1!__argv')[0]
21807 $2 = 0x22fd98 "/cygdrive/c/mydirectory/myprogram"
21811 (@value{GDBP}) x/2x &'cygwin1!__argv'
21812 0x610c0aa8 <cygwin1!__argv>: 0x10021608 0x00000000
21813 (@value{GDBP}) x/x 0x10021608
21814 0x10021608: 0x0022fd98
21815 (@value{GDBP}) x/s 0x0022fd98
21816 0x22fd98: "/cygdrive/c/mydirectory/myprogram"
21819 Setting a break point within a DLL is possible even before the program
21820 starts execution. However, under these circumstances, @value{GDBN} can't
21821 examine the initial instructions of the function in order to skip the
21822 function's frame set-up code. You can work around this by using ``*&''
21823 to set the breakpoint at a raw memory address:
21826 (@value{GDBP}) break *&'python22!PyOS_Readline'
21827 Breakpoint 1 at 0x1e04eff0
21830 The author of these extensions is not entirely convinced that setting a
21831 break point within a shared DLL like @file{kernel32.dll} is completely
21835 @subsection Commands Specific to @sc{gnu} Hurd Systems
21836 @cindex @sc{gnu} Hurd debugging
21838 This subsection describes @value{GDBN} commands specific to the
21839 @sc{gnu} Hurd native debugging.
21844 @kindex set signals@r{, Hurd command}
21845 @kindex set sigs@r{, Hurd command}
21846 This command toggles the state of inferior signal interception by
21847 @value{GDBN}. Mach exceptions, such as breakpoint traps, are not
21848 affected by this command. @code{sigs} is a shorthand alias for
21853 @kindex show signals@r{, Hurd command}
21854 @kindex show sigs@r{, Hurd command}
21855 Show the current state of intercepting inferior's signals.
21857 @item set signal-thread
21858 @itemx set sigthread
21859 @kindex set signal-thread
21860 @kindex set sigthread
21861 This command tells @value{GDBN} which thread is the @code{libc} signal
21862 thread. That thread is run when a signal is delivered to a running
21863 process. @code{set sigthread} is the shorthand alias of @code{set
21866 @item show signal-thread
21867 @itemx show sigthread
21868 @kindex show signal-thread
21869 @kindex show sigthread
21870 These two commands show which thread will run when the inferior is
21871 delivered a signal.
21874 @kindex set stopped@r{, Hurd command}
21875 This commands tells @value{GDBN} that the inferior process is stopped,
21876 as with the @code{SIGSTOP} signal. The stopped process can be
21877 continued by delivering a signal to it.
21880 @kindex show stopped@r{, Hurd command}
21881 This command shows whether @value{GDBN} thinks the debuggee is
21884 @item set exceptions
21885 @kindex set exceptions@r{, Hurd command}
21886 Use this command to turn off trapping of exceptions in the inferior.
21887 When exception trapping is off, neither breakpoints nor
21888 single-stepping will work. To restore the default, set exception
21891 @item show exceptions
21892 @kindex show exceptions@r{, Hurd command}
21893 Show the current state of trapping exceptions in the inferior.
21895 @item set task pause
21896 @kindex set task@r{, Hurd commands}
21897 @cindex task attributes (@sc{gnu} Hurd)
21898 @cindex pause current task (@sc{gnu} Hurd)
21899 This command toggles task suspension when @value{GDBN} has control.
21900 Setting it to on takes effect immediately, and the task is suspended
21901 whenever @value{GDBN} gets control. Setting it to off will take
21902 effect the next time the inferior is continued. If this option is set
21903 to off, you can use @code{set thread default pause on} or @code{set
21904 thread pause on} (see below) to pause individual threads.
21906 @item show task pause
21907 @kindex show task@r{, Hurd commands}
21908 Show the current state of task suspension.
21910 @item set task detach-suspend-count
21911 @cindex task suspend count
21912 @cindex detach from task, @sc{gnu} Hurd
21913 This command sets the suspend count the task will be left with when
21914 @value{GDBN} detaches from it.
21916 @item show task detach-suspend-count
21917 Show the suspend count the task will be left with when detaching.
21919 @item set task exception-port
21920 @itemx set task excp
21921 @cindex task exception port, @sc{gnu} Hurd
21922 This command sets the task exception port to which @value{GDBN} will
21923 forward exceptions. The argument should be the value of the @dfn{send
21924 rights} of the task. @code{set task excp} is a shorthand alias.
21926 @item set noninvasive
21927 @cindex noninvasive task options
21928 This command switches @value{GDBN} to a mode that is the least
21929 invasive as far as interfering with the inferior is concerned. This
21930 is the same as using @code{set task pause}, @code{set exceptions}, and
21931 @code{set signals} to values opposite to the defaults.
21933 @item info send-rights
21934 @itemx info receive-rights
21935 @itemx info port-rights
21936 @itemx info port-sets
21937 @itemx info dead-names
21940 @cindex send rights, @sc{gnu} Hurd
21941 @cindex receive rights, @sc{gnu} Hurd
21942 @cindex port rights, @sc{gnu} Hurd
21943 @cindex port sets, @sc{gnu} Hurd
21944 @cindex dead names, @sc{gnu} Hurd
21945 These commands display information about, respectively, send rights,
21946 receive rights, port rights, port sets, and dead names of a task.
21947 There are also shorthand aliases: @code{info ports} for @code{info
21948 port-rights} and @code{info psets} for @code{info port-sets}.
21950 @item set thread pause
21951 @kindex set thread@r{, Hurd command}
21952 @cindex thread properties, @sc{gnu} Hurd
21953 @cindex pause current thread (@sc{gnu} Hurd)
21954 This command toggles current thread suspension when @value{GDBN} has
21955 control. Setting it to on takes effect immediately, and the current
21956 thread is suspended whenever @value{GDBN} gets control. Setting it to
21957 off will take effect the next time the inferior is continued.
21958 Normally, this command has no effect, since when @value{GDBN} has
21959 control, the whole task is suspended. However, if you used @code{set
21960 task pause off} (see above), this command comes in handy to suspend
21961 only the current thread.
21963 @item show thread pause
21964 @kindex show thread@r{, Hurd command}
21965 This command shows the state of current thread suspension.
21967 @item set thread run
21968 This command sets whether the current thread is allowed to run.
21970 @item show thread run
21971 Show whether the current thread is allowed to run.
21973 @item set thread detach-suspend-count
21974 @cindex thread suspend count, @sc{gnu} Hurd
21975 @cindex detach from thread, @sc{gnu} Hurd
21976 This command sets the suspend count @value{GDBN} will leave on a
21977 thread when detaching. This number is relative to the suspend count
21978 found by @value{GDBN} when it notices the thread; use @code{set thread
21979 takeover-suspend-count} to force it to an absolute value.
21981 @item show thread detach-suspend-count
21982 Show the suspend count @value{GDBN} will leave on the thread when
21985 @item set thread exception-port
21986 @itemx set thread excp
21987 Set the thread exception port to which to forward exceptions. This
21988 overrides the port set by @code{set task exception-port} (see above).
21989 @code{set thread excp} is the shorthand alias.
21991 @item set thread takeover-suspend-count
21992 Normally, @value{GDBN}'s thread suspend counts are relative to the
21993 value @value{GDBN} finds when it notices each thread. This command
21994 changes the suspend counts to be absolute instead.
21996 @item set thread default
21997 @itemx show thread default
21998 @cindex thread default settings, @sc{gnu} Hurd
21999 Each of the above @code{set thread} commands has a @code{set thread
22000 default} counterpart (e.g., @code{set thread default pause}, @code{set
22001 thread default exception-port}, etc.). The @code{thread default}
22002 variety of commands sets the default thread properties for all
22003 threads; you can then change the properties of individual threads with
22004 the non-default commands.
22011 @value{GDBN} provides the following commands specific to the Darwin target:
22014 @item set debug darwin @var{num}
22015 @kindex set debug darwin
22016 When set to a non zero value, enables debugging messages specific to
22017 the Darwin support. Higher values produce more verbose output.
22019 @item show debug darwin
22020 @kindex show debug darwin
22021 Show the current state of Darwin messages.
22023 @item set debug mach-o @var{num}
22024 @kindex set debug mach-o
22025 When set to a non zero value, enables debugging messages while
22026 @value{GDBN} is reading Darwin object files. (@dfn{Mach-O} is the
22027 file format used on Darwin for object and executable files.) Higher
22028 values produce more verbose output. This is a command to diagnose
22029 problems internal to @value{GDBN} and should not be needed in normal
22032 @item show debug mach-o
22033 @kindex show debug mach-o
22034 Show the current state of Mach-O file messages.
22036 @item set mach-exceptions on
22037 @itemx set mach-exceptions off
22038 @kindex set mach-exceptions
22039 On Darwin, faults are first reported as a Mach exception and are then
22040 mapped to a Posix signal. Use this command to turn on trapping of
22041 Mach exceptions in the inferior. This might be sometimes useful to
22042 better understand the cause of a fault. The default is off.
22044 @item show mach-exceptions
22045 @kindex show mach-exceptions
22046 Show the current state of exceptions trapping.
22051 @section Embedded Operating Systems
22053 This section describes configurations involving the debugging of
22054 embedded operating systems that are available for several different
22057 @value{GDBN} includes the ability to debug programs running on
22058 various real-time operating systems.
22060 @node Embedded Processors
22061 @section Embedded Processors
22063 This section goes into details specific to particular embedded
22066 @cindex send command to simulator
22067 Whenever a specific embedded processor has a simulator, @value{GDBN}
22068 allows to send an arbitrary command to the simulator.
22071 @item sim @var{command}
22072 @kindex sim@r{, a command}
22073 Send an arbitrary @var{command} string to the simulator. Consult the
22074 documentation for the specific simulator in use for information about
22075 acceptable commands.
22080 * ARC:: Synopsys ARC
22082 * M68K:: Motorola M68K
22083 * MicroBlaze:: Xilinx MicroBlaze
22084 * MIPS Embedded:: MIPS Embedded
22085 * PowerPC Embedded:: PowerPC Embedded
22088 * Super-H:: Renesas Super-H
22092 @subsection Synopsys ARC
22093 @cindex Synopsys ARC
22094 @cindex ARC specific commands
22100 @value{GDBN} provides the following ARC-specific commands:
22103 @item set debug arc
22104 @kindex set debug arc
22105 Control the level of ARC specific debug messages. Use 0 for no messages (the
22106 default), 1 for debug messages, and 2 for even more debug messages.
22108 @item show debug arc
22109 @kindex show debug arc
22110 Show the level of ARC specific debugging in operation.
22112 @item maint print arc arc-instruction @var{address}
22113 @kindex maint print arc arc-instruction
22114 Print internal disassembler information about instruction at a given address.
22121 @value{GDBN} provides the following ARM-specific commands:
22124 @item set arm disassembler
22126 This commands selects from a list of disassembly styles. The
22127 @code{"std"} style is the standard style.
22129 @item show arm disassembler
22131 Show the current disassembly style.
22133 @item set arm apcs32
22134 @cindex ARM 32-bit mode
22135 This command toggles ARM operation mode between 32-bit and 26-bit.
22137 @item show arm apcs32
22138 Display the current usage of the ARM 32-bit mode.
22140 @item set arm fpu @var{fputype}
22141 This command sets the ARM floating-point unit (FPU) type. The
22142 argument @var{fputype} can be one of these:
22146 Determine the FPU type by querying the OS ABI.
22148 Software FPU, with mixed-endian doubles on little-endian ARM
22151 GCC-compiled FPA co-processor.
22153 Software FPU with pure-endian doubles.
22159 Show the current type of the FPU.
22162 This command forces @value{GDBN} to use the specified ABI.
22165 Show the currently used ABI.
22167 @item set arm fallback-mode (arm|thumb|auto)
22168 @value{GDBN} uses the symbol table, when available, to determine
22169 whether instructions are ARM or Thumb. This command controls
22170 @value{GDBN}'s default behavior when the symbol table is not
22171 available. The default is @samp{auto}, which causes @value{GDBN} to
22172 use the current execution mode (from the @code{T} bit in the @code{CPSR}
22175 @item show arm fallback-mode
22176 Show the current fallback instruction mode.
22178 @item set arm force-mode (arm|thumb|auto)
22179 This command overrides use of the symbol table to determine whether
22180 instructions are ARM or Thumb. The default is @samp{auto}, which
22181 causes @value{GDBN} to use the symbol table and then the setting
22182 of @samp{set arm fallback-mode}.
22184 @item show arm force-mode
22185 Show the current forced instruction mode.
22187 @item set debug arm
22188 Toggle whether to display ARM-specific debugging messages from the ARM
22189 target support subsystem.
22191 @item show debug arm
22192 Show whether ARM-specific debugging messages are enabled.
22196 @item target sim @r{[}@var{simargs}@r{]} @dots{}
22197 The @value{GDBN} ARM simulator accepts the following optional arguments.
22200 @item --swi-support=@var{type}
22201 Tell the simulator which SWI interfaces to support. The argument
22202 @var{type} may be a comma separated list of the following values.
22203 The default value is @code{all}.
22218 The Motorola m68k configuration includes ColdFire support.
22221 @subsection MicroBlaze
22222 @cindex Xilinx MicroBlaze
22223 @cindex XMD, Xilinx Microprocessor Debugger
22225 The MicroBlaze is a soft-core processor supported on various Xilinx
22226 FPGAs, such as Spartan or Virtex series. Boards with these processors
22227 usually have JTAG ports which connect to a host system running the Xilinx
22228 Embedded Development Kit (EDK) or Software Development Kit (SDK).
22229 This host system is used to download the configuration bitstream to
22230 the target FPGA. The Xilinx Microprocessor Debugger (XMD) program
22231 communicates with the target board using the JTAG interface and
22232 presents a @code{gdbserver} interface to the board. By default
22233 @code{xmd} uses port @code{1234}. (While it is possible to change
22234 this default port, it requires the use of undocumented @code{xmd}
22235 commands. Contact Xilinx support if you need to do this.)
22237 Use these GDB commands to connect to the MicroBlaze target processor.
22240 @item target remote :1234
22241 Use this command to connect to the target if you are running @value{GDBN}
22242 on the same system as @code{xmd}.
22244 @item target remote @var{xmd-host}:1234
22245 Use this command to connect to the target if it is connected to @code{xmd}
22246 running on a different system named @var{xmd-host}.
22249 Use this command to download a program to the MicroBlaze target.
22251 @item set debug microblaze @var{n}
22252 Enable MicroBlaze-specific debugging messages if non-zero.
22254 @item show debug microblaze @var{n}
22255 Show MicroBlaze-specific debugging level.
22258 @node MIPS Embedded
22259 @subsection @acronym{MIPS} Embedded
22262 @value{GDBN} supports these special commands for @acronym{MIPS} targets:
22265 @item set mipsfpu double
22266 @itemx set mipsfpu single
22267 @itemx set mipsfpu none
22268 @itemx set mipsfpu auto
22269 @itemx show mipsfpu
22270 @kindex set mipsfpu
22271 @kindex show mipsfpu
22272 @cindex @acronym{MIPS} remote floating point
22273 @cindex floating point, @acronym{MIPS} remote
22274 If your target board does not support the @acronym{MIPS} floating point
22275 coprocessor, you should use the command @samp{set mipsfpu none} (if you
22276 need this, you may wish to put the command in your @value{GDBN} init
22277 file). This tells @value{GDBN} how to find the return value of
22278 functions which return floating point values. It also allows
22279 @value{GDBN} to avoid saving the floating point registers when calling
22280 functions on the board. If you are using a floating point coprocessor
22281 with only single precision floating point support, as on the @sc{r4650}
22282 processor, use the command @samp{set mipsfpu single}. The default
22283 double precision floating point coprocessor may be selected using
22284 @samp{set mipsfpu double}.
22286 In previous versions the only choices were double precision or no
22287 floating point, so @samp{set mipsfpu on} will select double precision
22288 and @samp{set mipsfpu off} will select no floating point.
22290 As usual, you can inquire about the @code{mipsfpu} variable with
22291 @samp{show mipsfpu}.
22294 @node PowerPC Embedded
22295 @subsection PowerPC Embedded
22297 @cindex DVC register
22298 @value{GDBN} supports using the DVC (Data Value Compare) register to
22299 implement in hardware simple hardware watchpoint conditions of the form:
22302 (@value{GDBP}) watch @var{ADDRESS|VARIABLE} \
22303 if @var{ADDRESS|VARIABLE} == @var{CONSTANT EXPRESSION}
22306 The DVC register will be automatically used when @value{GDBN} detects
22307 such pattern in a condition expression, and the created watchpoint uses one
22308 debug register (either the @code{exact-watchpoints} option is on and the
22309 variable is scalar, or the variable has a length of one byte). This feature
22310 is available in native @value{GDBN} running on a Linux kernel version 2.6.34
22313 When running on PowerPC embedded processors, @value{GDBN} automatically uses
22314 ranged hardware watchpoints, unless the @code{exact-watchpoints} option is on,
22315 in which case watchpoints using only one debug register are created when
22316 watching variables of scalar types.
22318 You can create an artificial array to watch an arbitrary memory
22319 region using one of the following commands (@pxref{Expressions}):
22322 (@value{GDBP}) watch *((char *) @var{address})@@@var{length}
22323 (@value{GDBP}) watch @{char[@var{length}]@} @var{address}
22326 PowerPC embedded processors support masked watchpoints. See the discussion
22327 about the @code{mask} argument in @ref{Set Watchpoints}.
22329 @cindex ranged breakpoint
22330 PowerPC embedded processors support hardware accelerated
22331 @dfn{ranged breakpoints}. A ranged breakpoint stops execution of
22332 the inferior whenever it executes an instruction at any address within
22333 the range it specifies. To set a ranged breakpoint in @value{GDBN},
22334 use the @code{break-range} command.
22336 @value{GDBN} provides the following PowerPC-specific commands:
22339 @kindex break-range
22340 @item break-range @var{start-location}, @var{end-location}
22341 Set a breakpoint for an address range given by
22342 @var{start-location} and @var{end-location}, which can specify a function name,
22343 a line number, an offset of lines from the current line or from the start
22344 location, or an address of an instruction (see @ref{Specify Location},
22345 for a list of all the possible ways to specify a @var{location}.)
22346 The breakpoint will stop execution of the inferior whenever it
22347 executes an instruction at any address within the specified range,
22348 (including @var{start-location} and @var{end-location}.)
22350 @kindex set powerpc
22351 @item set powerpc soft-float
22352 @itemx show powerpc soft-float
22353 Force @value{GDBN} to use (or not use) a software floating point calling
22354 convention. By default, @value{GDBN} selects the calling convention based
22355 on the selected architecture and the provided executable file.
22357 @item set powerpc vector-abi
22358 @itemx show powerpc vector-abi
22359 Force @value{GDBN} to use the specified calling convention for vector
22360 arguments and return values. The valid options are @samp{auto};
22361 @samp{generic}, to avoid vector registers even if they are present;
22362 @samp{altivec}, to use AltiVec registers; and @samp{spe} to use SPE
22363 registers. By default, @value{GDBN} selects the calling convention
22364 based on the selected architecture and the provided executable file.
22366 @item set powerpc exact-watchpoints
22367 @itemx show powerpc exact-watchpoints
22368 Allow @value{GDBN} to use only one debug register when watching a variable
22369 of scalar type, thus assuming that the variable is accessed through the
22370 address of its first byte.
22375 @subsection Atmel AVR
22378 When configured for debugging the Atmel AVR, @value{GDBN} supports the
22379 following AVR-specific commands:
22382 @item info io_registers
22383 @kindex info io_registers@r{, AVR}
22384 @cindex I/O registers (Atmel AVR)
22385 This command displays information about the AVR I/O registers. For
22386 each register, @value{GDBN} prints its number and value.
22393 When configured for debugging CRIS, @value{GDBN} provides the
22394 following CRIS-specific commands:
22397 @item set cris-version @var{ver}
22398 @cindex CRIS version
22399 Set the current CRIS version to @var{ver}, either @samp{10} or @samp{32}.
22400 The CRIS version affects register names and sizes. This command is useful in
22401 case autodetection of the CRIS version fails.
22403 @item show cris-version
22404 Show the current CRIS version.
22406 @item set cris-dwarf2-cfi
22407 @cindex DWARF-2 CFI and CRIS
22408 Set the usage of DWARF-2 CFI for CRIS debugging. The default is @samp{on}.
22409 Change to @samp{off} when using @code{gcc-cris} whose version is below
22412 @item show cris-dwarf2-cfi
22413 Show the current state of using DWARF-2 CFI.
22415 @item set cris-mode @var{mode}
22417 Set the current CRIS mode to @var{mode}. It should only be changed when
22418 debugging in guru mode, in which case it should be set to
22419 @samp{guru} (the default is @samp{normal}).
22421 @item show cris-mode
22422 Show the current CRIS mode.
22426 @subsection Renesas Super-H
22429 For the Renesas Super-H processor, @value{GDBN} provides these
22433 @item set sh calling-convention @var{convention}
22434 @kindex set sh calling-convention
22435 Set the calling-convention used when calling functions from @value{GDBN}.
22436 Allowed values are @samp{gcc}, which is the default setting, and @samp{renesas}.
22437 With the @samp{gcc} setting, functions are called using the @value{NGCC} calling
22438 convention. If the DWARF-2 information of the called function specifies
22439 that the function follows the Renesas calling convention, the function
22440 is called using the Renesas calling convention. If the calling convention
22441 is set to @samp{renesas}, the Renesas calling convention is always used,
22442 regardless of the DWARF-2 information. This can be used to override the
22443 default of @samp{gcc} if debug information is missing, or the compiler
22444 does not emit the DWARF-2 calling convention entry for a function.
22446 @item show sh calling-convention
22447 @kindex show sh calling-convention
22448 Show the current calling convention setting.
22453 @node Architectures
22454 @section Architectures
22456 This section describes characteristics of architectures that affect
22457 all uses of @value{GDBN} with the architecture, both native and cross.
22464 * HPPA:: HP PA architecture
22465 * SPU:: Cell Broadband Engine SPU architecture
22471 @subsection AArch64
22472 @cindex AArch64 support
22474 When @value{GDBN} is debugging the AArch64 architecture, it provides the
22475 following special commands:
22478 @item set debug aarch64
22479 @kindex set debug aarch64
22480 This command determines whether AArch64 architecture-specific debugging
22481 messages are to be displayed.
22483 @item show debug aarch64
22484 Show whether AArch64 debugging messages are displayed.
22489 @subsection x86 Architecture-specific Issues
22492 @item set struct-convention @var{mode}
22493 @kindex set struct-convention
22494 @cindex struct return convention
22495 @cindex struct/union returned in registers
22496 Set the convention used by the inferior to return @code{struct}s and
22497 @code{union}s from functions to @var{mode}. Possible values of
22498 @var{mode} are @code{"pcc"}, @code{"reg"}, and @code{"default"} (the
22499 default). @code{"default"} or @code{"pcc"} means that @code{struct}s
22500 are returned on the stack, while @code{"reg"} means that a
22501 @code{struct} or a @code{union} whose size is 1, 2, 4, or 8 bytes will
22502 be returned in a register.
22504 @item show struct-convention
22505 @kindex show struct-convention
22506 Show the current setting of the convention to return @code{struct}s
22511 @subsubsection Intel @dfn{Memory Protection Extensions} (MPX).
22512 @cindex Intel Memory Protection Extensions (MPX).
22514 Memory Protection Extension (MPX) adds the bound registers @samp{BND0}
22515 @footnote{The register named with capital letters represent the architecture
22516 registers.} through @samp{BND3}. Bound registers store a pair of 64-bit values
22517 which are the lower bound and upper bound. Bounds are effective addresses or
22518 memory locations. The upper bounds are architecturally represented in 1's
22519 complement form. A bound having lower bound = 0, and upper bound = 0
22520 (1's complement of all bits set) will allow access to the entire address space.
22522 @samp{BND0} through @samp{BND3} are represented in @value{GDBN} as @samp{bnd0raw}
22523 through @samp{bnd3raw}. Pseudo registers @samp{bnd0} through @samp{bnd3}
22524 display the upper bound performing the complement of one operation on the
22525 upper bound value, i.e.@ when upper bound in @samp{bnd0raw} is 0 in the
22526 @value{GDBN} @samp{bnd0} it will be @code{0xfff@dots{}}. In this sense it
22527 can also be noted that the upper bounds are inclusive.
22529 As an example, assume that the register BND0 holds bounds for a pointer having
22530 access allowed for the range between 0x32 and 0x71. The values present on
22531 bnd0raw and bnd registers are presented as follows:
22534 bnd0raw = @{0x32, 0xffffffff8e@}
22535 bnd0 = @{lbound = 0x32, ubound = 0x71@} : size 64
22538 This way the raw value can be accessed via bnd0raw@dots{}bnd3raw. Any
22539 change on bnd0@dots{}bnd3 or bnd0raw@dots{}bnd3raw is reflect on its
22540 counterpart. When the bnd0@dots{}bnd3 registers are displayed via
22541 Python, the display includes the memory size, in bits, accessible to
22544 Bounds can also be stored in bounds tables, which are stored in
22545 application memory. These tables store bounds for pointers by specifying
22546 the bounds pointer's value along with its bounds. Evaluating and changing
22547 bounds located in bound tables is therefore interesting while investigating
22548 bugs on MPX context. @value{GDBN} provides commands for this purpose:
22551 @item show mpx bound @var{pointer}
22552 @kindex show mpx bound
22553 Display bounds of the given @var{pointer}.
22555 @item set mpx bound @var{pointer}, @var{lbound}, @var{ubound}
22556 @kindex set mpx bound
22557 Set the bounds of a pointer in the bound table.
22558 This command takes three parameters: @var{pointer} is the pointers
22559 whose bounds are to be changed, @var{lbound} and @var{ubound} are new values
22560 for lower and upper bounds respectively.
22563 When you call an inferior function on an Intel MPX enabled program,
22564 GDB sets the inferior's bound registers to the init (disabled) state
22565 before calling the function. As a consequence, bounds checks for the
22566 pointer arguments passed to the function will always pass.
22568 This is necessary because when you call an inferior function, the
22569 program is usually in the middle of the execution of other function.
22570 Since at that point bound registers are in an arbitrary state, not
22571 clearing them would lead to random bound violations in the called
22574 You can still examine the influence of the bound registers on the
22575 execution of the called function by stopping the execution of the
22576 called function at its prologue, setting bound registers, and
22577 continuing the execution. For example:
22581 Breakpoint 2 at 0x4009de: file i386-mpx-call.c, line 47.
22582 $ print upper (a, b, c, d, 1)
22583 Breakpoint 2, upper (a=0x0, b=0x6e0000005b, c=0x0, d=0x0, len=48)....
22585 @{lbound = 0x0, ubound = ffffffff@} : size -1
22588 At this last step the value of bnd0 can be changed for investigation of bound
22589 violations caused along the execution of the call. In order to know how to
22590 set the bound registers or bound table for the call consult the ABI.
22595 See the following section.
22598 @subsection @acronym{MIPS}
22600 @cindex stack on Alpha
22601 @cindex stack on @acronym{MIPS}
22602 @cindex Alpha stack
22603 @cindex @acronym{MIPS} stack
22604 Alpha- and @acronym{MIPS}-based computers use an unusual stack frame, which
22605 sometimes requires @value{GDBN} to search backward in the object code to
22606 find the beginning of a function.
22608 @cindex response time, @acronym{MIPS} debugging
22609 To improve response time (especially for embedded applications, where
22610 @value{GDBN} may be restricted to a slow serial line for this search)
22611 you may want to limit the size of this search, using one of these
22615 @cindex @code{heuristic-fence-post} (Alpha, @acronym{MIPS})
22616 @item set heuristic-fence-post @var{limit}
22617 Restrict @value{GDBN} to examining at most @var{limit} bytes in its
22618 search for the beginning of a function. A value of @var{0} (the
22619 default) means there is no limit. However, except for @var{0}, the
22620 larger the limit the more bytes @code{heuristic-fence-post} must search
22621 and therefore the longer it takes to run. You should only need to use
22622 this command when debugging a stripped executable.
22624 @item show heuristic-fence-post
22625 Display the current limit.
22629 These commands are available @emph{only} when @value{GDBN} is configured
22630 for debugging programs on Alpha or @acronym{MIPS} processors.
22632 Several @acronym{MIPS}-specific commands are available when debugging @acronym{MIPS}
22636 @item set mips abi @var{arg}
22637 @kindex set mips abi
22638 @cindex set ABI for @acronym{MIPS}
22639 Tell @value{GDBN} which @acronym{MIPS} ABI is used by the inferior. Possible
22640 values of @var{arg} are:
22644 The default ABI associated with the current binary (this is the
22654 @item show mips abi
22655 @kindex show mips abi
22656 Show the @acronym{MIPS} ABI used by @value{GDBN} to debug the inferior.
22658 @item set mips compression @var{arg}
22659 @kindex set mips compression
22660 @cindex code compression, @acronym{MIPS}
22661 Tell @value{GDBN} which @acronym{MIPS} compressed
22662 @acronym{ISA, Instruction Set Architecture} encoding is used by the
22663 inferior. @value{GDBN} uses this for code disassembly and other
22664 internal interpretation purposes. This setting is only referred to
22665 when no executable has been associated with the debugging session or
22666 the executable does not provide information about the encoding it uses.
22667 Otherwise this setting is automatically updated from information
22668 provided by the executable.
22670 Possible values of @var{arg} are @samp{mips16} and @samp{micromips}.
22671 The default compressed @acronym{ISA} encoding is @samp{mips16}, as
22672 executables containing @acronym{MIPS16} code frequently are not
22673 identified as such.
22675 This setting is ``sticky''; that is, it retains its value across
22676 debugging sessions until reset either explicitly with this command or
22677 implicitly from an executable.
22679 The compiler and/or assembler typically add symbol table annotations to
22680 identify functions compiled for the @acronym{MIPS16} or
22681 @acronym{microMIPS} @acronym{ISA}s. If these function-scope annotations
22682 are present, @value{GDBN} uses them in preference to the global
22683 compressed @acronym{ISA} encoding setting.
22685 @item show mips compression
22686 @kindex show mips compression
22687 Show the @acronym{MIPS} compressed @acronym{ISA} encoding used by
22688 @value{GDBN} to debug the inferior.
22691 @itemx show mipsfpu
22692 @xref{MIPS Embedded, set mipsfpu}.
22694 @item set mips mask-address @var{arg}
22695 @kindex set mips mask-address
22696 @cindex @acronym{MIPS} addresses, masking
22697 This command determines whether the most-significant 32 bits of 64-bit
22698 @acronym{MIPS} addresses are masked off. The argument @var{arg} can be
22699 @samp{on}, @samp{off}, or @samp{auto}. The latter is the default
22700 setting, which lets @value{GDBN} determine the correct value.
22702 @item show mips mask-address
22703 @kindex show mips mask-address
22704 Show whether the upper 32 bits of @acronym{MIPS} addresses are masked off or
22707 @item set remote-mips64-transfers-32bit-regs
22708 @kindex set remote-mips64-transfers-32bit-regs
22709 This command controls compatibility with 64-bit @acronym{MIPS} targets that
22710 transfer data in 32-bit quantities. If you have an old @acronym{MIPS} 64 target
22711 that transfers 32 bits for some registers, like @sc{sr} and @sc{fsr},
22712 and 64 bits for other registers, set this option to @samp{on}.
22714 @item show remote-mips64-transfers-32bit-regs
22715 @kindex show remote-mips64-transfers-32bit-regs
22716 Show the current setting of compatibility with older @acronym{MIPS} 64 targets.
22718 @item set debug mips
22719 @kindex set debug mips
22720 This command turns on and off debugging messages for the @acronym{MIPS}-specific
22721 target code in @value{GDBN}.
22723 @item show debug mips
22724 @kindex show debug mips
22725 Show the current setting of @acronym{MIPS} debugging messages.
22731 @cindex HPPA support
22733 When @value{GDBN} is debugging the HP PA architecture, it provides the
22734 following special commands:
22737 @item set debug hppa
22738 @kindex set debug hppa
22739 This command determines whether HPPA architecture-specific debugging
22740 messages are to be displayed.
22742 @item show debug hppa
22743 Show whether HPPA debugging messages are displayed.
22745 @item maint print unwind @var{address}
22746 @kindex maint print unwind@r{, HPPA}
22747 This command displays the contents of the unwind table entry at the
22748 given @var{address}.
22754 @subsection Cell Broadband Engine SPU architecture
22755 @cindex Cell Broadband Engine
22758 When @value{GDBN} is debugging the Cell Broadband Engine SPU architecture,
22759 it provides the following special commands:
22762 @item info spu event
22764 Display SPU event facility status. Shows current event mask
22765 and pending event status.
22767 @item info spu signal
22768 Display SPU signal notification facility status. Shows pending
22769 signal-control word and signal notification mode of both signal
22770 notification channels.
22772 @item info spu mailbox
22773 Display SPU mailbox facility status. Shows all pending entries,
22774 in order of processing, in each of the SPU Write Outbound,
22775 SPU Write Outbound Interrupt, and SPU Read Inbound mailboxes.
22778 Display MFC DMA status. Shows all pending commands in the MFC
22779 DMA queue. For each entry, opcode, tag, class IDs, effective
22780 and local store addresses and transfer size are shown.
22782 @item info spu proxydma
22783 Display MFC Proxy-DMA status. Shows all pending commands in the MFC
22784 Proxy-DMA queue. For each entry, opcode, tag, class IDs, effective
22785 and local store addresses and transfer size are shown.
22789 When @value{GDBN} is debugging a combined PowerPC/SPU application
22790 on the Cell Broadband Engine, it provides in addition the following
22794 @item set spu stop-on-load @var{arg}
22796 Set whether to stop for new SPE threads. When set to @code{on}, @value{GDBN}
22797 will give control to the user when a new SPE thread enters its @code{main}
22798 function. The default is @code{off}.
22800 @item show spu stop-on-load
22802 Show whether to stop for new SPE threads.
22804 @item set spu auto-flush-cache @var{arg}
22805 Set whether to automatically flush the software-managed cache. When set to
22806 @code{on}, @value{GDBN} will automatically cause the SPE software-managed
22807 cache to be flushed whenever SPE execution stops. This provides a consistent
22808 view of PowerPC memory that is accessed via the cache. If an application
22809 does not use the software-managed cache, this option has no effect.
22811 @item show spu auto-flush-cache
22812 Show whether to automatically flush the software-managed cache.
22817 @subsection PowerPC
22818 @cindex PowerPC architecture
22820 When @value{GDBN} is debugging the PowerPC architecture, it provides a set of
22821 pseudo-registers to enable inspection of 128-bit wide Decimal Floating Point
22822 numbers stored in the floating point registers. These values must be stored
22823 in two consecutive registers, always starting at an even register like
22824 @code{f0} or @code{f2}.
22826 The pseudo-registers go from @code{$dl0} through @code{$dl15}, and are formed
22827 by joining the even/odd register pairs @code{f0} and @code{f1} for @code{$dl0},
22828 @code{f2} and @code{f3} for @code{$dl1} and so on.
22830 For POWER7 processors, @value{GDBN} provides a set of pseudo-registers, the 64-bit
22831 wide Extended Floating Point Registers (@samp{f32} through @samp{f63}).
22834 @subsection Nios II
22835 @cindex Nios II architecture
22837 When @value{GDBN} is debugging the Nios II architecture,
22838 it provides the following special commands:
22842 @item set debug nios2
22843 @kindex set debug nios2
22844 This command turns on and off debugging messages for the Nios II
22845 target code in @value{GDBN}.
22847 @item show debug nios2
22848 @kindex show debug nios2
22849 Show the current setting of Nios II debugging messages.
22852 @node Controlling GDB
22853 @chapter Controlling @value{GDBN}
22855 You can alter the way @value{GDBN} interacts with you by using the
22856 @code{set} command. For commands controlling how @value{GDBN} displays
22857 data, see @ref{Print Settings, ,Print Settings}. Other settings are
22862 * Editing:: Command editing
22863 * Command History:: Command history
22864 * Screen Size:: Screen size
22865 * Numbers:: Numbers
22866 * ABI:: Configuring the current ABI
22867 * Auto-loading:: Automatically loading associated files
22868 * Messages/Warnings:: Optional warnings and messages
22869 * Debugging Output:: Optional messages about internal happenings
22870 * Other Misc Settings:: Other Miscellaneous Settings
22878 @value{GDBN} indicates its readiness to read a command by printing a string
22879 called the @dfn{prompt}. This string is normally @samp{(@value{GDBP})}. You
22880 can change the prompt string with the @code{set prompt} command. For
22881 instance, when debugging @value{GDBN} with @value{GDBN}, it is useful to change
22882 the prompt in one of the @value{GDBN} sessions so that you can always tell
22883 which one you are talking to.
22885 @emph{Note:} @code{set prompt} does not add a space for you after the
22886 prompt you set. This allows you to set a prompt which ends in a space
22887 or a prompt that does not.
22891 @item set prompt @var{newprompt}
22892 Directs @value{GDBN} to use @var{newprompt} as its prompt string henceforth.
22894 @kindex show prompt
22896 Prints a line of the form: @samp{Gdb's prompt is: @var{your-prompt}}
22899 Versions of @value{GDBN} that ship with Python scripting enabled have
22900 prompt extensions. The commands for interacting with these extensions
22904 @kindex set extended-prompt
22905 @item set extended-prompt @var{prompt}
22906 Set an extended prompt that allows for substitutions.
22907 @xref{gdb.prompt}, for a list of escape sequences that can be used for
22908 substitution. Any escape sequences specified as part of the prompt
22909 string are replaced with the corresponding strings each time the prompt
22915 set extended-prompt Current working directory: \w (gdb)
22918 Note that when an extended-prompt is set, it takes control of the
22919 @var{prompt_hook} hook. @xref{prompt_hook}, for further information.
22921 @kindex show extended-prompt
22922 @item show extended-prompt
22923 Prints the extended prompt. Any escape sequences specified as part of
22924 the prompt string with @code{set extended-prompt}, are replaced with the
22925 corresponding strings each time the prompt is displayed.
22929 @section Command Editing
22931 @cindex command line editing
22933 @value{GDBN} reads its input commands via the @dfn{Readline} interface. This
22934 @sc{gnu} library provides consistent behavior for programs which provide a
22935 command line interface to the user. Advantages are @sc{gnu} Emacs-style
22936 or @dfn{vi}-style inline editing of commands, @code{csh}-like history
22937 substitution, and a storage and recall of command history across
22938 debugging sessions.
22940 You may control the behavior of command line editing in @value{GDBN} with the
22941 command @code{set}.
22944 @kindex set editing
22947 @itemx set editing on
22948 Enable command line editing (enabled by default).
22950 @item set editing off
22951 Disable command line editing.
22953 @kindex show editing
22955 Show whether command line editing is enabled.
22958 @ifset SYSTEM_READLINE
22959 @xref{Command Line Editing, , , rluserman, GNU Readline Library},
22961 @ifclear SYSTEM_READLINE
22962 @xref{Command Line Editing},
22964 for more details about the Readline
22965 interface. Users unfamiliar with @sc{gnu} Emacs or @code{vi} are
22966 encouraged to read that chapter.
22968 @node Command History
22969 @section Command History
22970 @cindex command history
22972 @value{GDBN} can keep track of the commands you type during your
22973 debugging sessions, so that you can be certain of precisely what
22974 happened. Use these commands to manage the @value{GDBN} command
22977 @value{GDBN} uses the @sc{gnu} History library, a part of the Readline
22978 package, to provide the history facility.
22979 @ifset SYSTEM_READLINE
22980 @xref{Using History Interactively, , , history, GNU History Library},
22982 @ifclear SYSTEM_READLINE
22983 @xref{Using History Interactively},
22985 for the detailed description of the History library.
22987 To issue a command to @value{GDBN} without affecting certain aspects of
22988 the state which is seen by users, prefix it with @samp{server }
22989 (@pxref{Server Prefix}). This
22990 means that this command will not affect the command history, nor will it
22991 affect @value{GDBN}'s notion of which command to repeat if @key{RET} is
22992 pressed on a line by itself.
22994 @cindex @code{server}, command prefix
22995 The server prefix does not affect the recording of values into the value
22996 history; to print a value without recording it into the value history,
22997 use the @code{output} command instead of the @code{print} command.
22999 Here is the description of @value{GDBN} commands related to command
23003 @cindex history substitution
23004 @cindex history file
23005 @kindex set history filename
23006 @cindex @env{GDBHISTFILE}, environment variable
23007 @item set history filename @var{fname}
23008 Set the name of the @value{GDBN} command history file to @var{fname}.
23009 This is the file where @value{GDBN} reads an initial command history
23010 list, and where it writes the command history from this session when it
23011 exits. You can access this list through history expansion or through
23012 the history command editing characters listed below. This file defaults
23013 to the value of the environment variable @code{GDBHISTFILE}, or to
23014 @file{./.gdb_history} (@file{./_gdb_history} on MS-DOS) if this variable
23017 @cindex save command history
23018 @kindex set history save
23019 @item set history save
23020 @itemx set history save on
23021 Record command history in a file, whose name may be specified with the
23022 @code{set history filename} command. By default, this option is disabled.
23024 @item set history save off
23025 Stop recording command history in a file.
23027 @cindex history size
23028 @kindex set history size
23029 @cindex @env{GDBHISTSIZE}, environment variable
23030 @item set history size @var{size}
23031 @itemx set history size unlimited
23032 Set the number of commands which @value{GDBN} keeps in its history list.
23033 This defaults to the value of the environment variable @env{GDBHISTSIZE}, or
23034 to 256 if this variable is not set. Non-numeric values of @env{GDBHISTSIZE}
23035 are ignored. If @var{size} is @code{unlimited} or if @env{GDBHISTSIZE} is
23036 either a negative number or the empty string, then the number of commands
23037 @value{GDBN} keeps in the history list is unlimited.
23039 @cindex remove duplicate history
23040 @kindex set history remove-duplicates
23041 @item set history remove-duplicates @var{count}
23042 @itemx set history remove-duplicates unlimited
23043 Control the removal of duplicate history entries in the command history list.
23044 If @var{count} is non-zero, @value{GDBN} will look back at the last @var{count}
23045 history entries and remove the first entry that is a duplicate of the current
23046 entry being added to the command history list. If @var{count} is
23047 @code{unlimited} then this lookbehind is unbounded. If @var{count} is 0, then
23048 removal of duplicate history entries is disabled.
23050 Only history entries added during the current session are considered for
23051 removal. This option is set to 0 by default.
23055 History expansion assigns special meaning to the character @kbd{!}.
23056 @ifset SYSTEM_READLINE
23057 @xref{Event Designators, , , history, GNU History Library},
23059 @ifclear SYSTEM_READLINE
23060 @xref{Event Designators},
23064 @cindex history expansion, turn on/off
23065 Since @kbd{!} is also the logical not operator in C, history expansion
23066 is off by default. If you decide to enable history expansion with the
23067 @code{set history expansion on} command, you may sometimes need to
23068 follow @kbd{!} (when it is used as logical not, in an expression) with
23069 a space or a tab to prevent it from being expanded. The readline
23070 history facilities do not attempt substitution on the strings
23071 @kbd{!=} and @kbd{!(}, even when history expansion is enabled.
23073 The commands to control history expansion are:
23076 @item set history expansion on
23077 @itemx set history expansion
23078 @kindex set history expansion
23079 Enable history expansion. History expansion is off by default.
23081 @item set history expansion off
23082 Disable history expansion.
23085 @kindex show history
23087 @itemx show history filename
23088 @itemx show history save
23089 @itemx show history size
23090 @itemx show history expansion
23091 These commands display the state of the @value{GDBN} history parameters.
23092 @code{show history} by itself displays all four states.
23097 @kindex show commands
23098 @cindex show last commands
23099 @cindex display command history
23100 @item show commands
23101 Display the last ten commands in the command history.
23103 @item show commands @var{n}
23104 Print ten commands centered on command number @var{n}.
23106 @item show commands +
23107 Print ten commands just after the commands last printed.
23111 @section Screen Size
23112 @cindex size of screen
23113 @cindex screen size
23116 @cindex pauses in output
23118 Certain commands to @value{GDBN} may produce large amounts of
23119 information output to the screen. To help you read all of it,
23120 @value{GDBN} pauses and asks you for input at the end of each page of
23121 output. Type @key{RET} when you want to continue the output, or @kbd{q}
23122 to discard the remaining output. Also, the screen width setting
23123 determines when to wrap lines of output. Depending on what is being
23124 printed, @value{GDBN} tries to break the line at a readable place,
23125 rather than simply letting it overflow onto the following line.
23127 Normally @value{GDBN} knows the size of the screen from the terminal
23128 driver software. For example, on Unix @value{GDBN} uses the termcap data base
23129 together with the value of the @code{TERM} environment variable and the
23130 @code{stty rows} and @code{stty cols} settings. If this is not correct,
23131 you can override it with the @code{set height} and @code{set
23138 @kindex show height
23139 @item set height @var{lpp}
23140 @itemx set height unlimited
23142 @itemx set width @var{cpl}
23143 @itemx set width unlimited
23145 These @code{set} commands specify a screen height of @var{lpp} lines and
23146 a screen width of @var{cpl} characters. The associated @code{show}
23147 commands display the current settings.
23149 If you specify a height of either @code{unlimited} or zero lines,
23150 @value{GDBN} does not pause during output no matter how long the
23151 output is. This is useful if output is to a file or to an editor
23154 Likewise, you can specify @samp{set width unlimited} or @samp{set
23155 width 0} to prevent @value{GDBN} from wrapping its output.
23157 @item set pagination on
23158 @itemx set pagination off
23159 @kindex set pagination
23160 Turn the output pagination on or off; the default is on. Turning
23161 pagination off is the alternative to @code{set height unlimited}. Note that
23162 running @value{GDBN} with the @option{--batch} option (@pxref{Mode
23163 Options, -batch}) also automatically disables pagination.
23165 @item show pagination
23166 @kindex show pagination
23167 Show the current pagination mode.
23172 @cindex number representation
23173 @cindex entering numbers
23175 You can always enter numbers in octal, decimal, or hexadecimal in
23176 @value{GDBN} by the usual conventions: octal numbers begin with
23177 @samp{0}, decimal numbers end with @samp{.}, and hexadecimal numbers
23178 begin with @samp{0x}. Numbers that neither begin with @samp{0} or
23179 @samp{0x}, nor end with a @samp{.} are, by default, entered in base
23180 10; likewise, the default display for numbers---when no particular
23181 format is specified---is base 10. You can change the default base for
23182 both input and output with the commands described below.
23185 @kindex set input-radix
23186 @item set input-radix @var{base}
23187 Set the default base for numeric input. Supported choices
23188 for @var{base} are decimal 8, 10, or 16. The base must itself be
23189 specified either unambiguously or using the current input radix; for
23193 set input-radix 012
23194 set input-radix 10.
23195 set input-radix 0xa
23199 sets the input base to decimal. On the other hand, @samp{set input-radix 10}
23200 leaves the input radix unchanged, no matter what it was, since
23201 @samp{10}, being without any leading or trailing signs of its base, is
23202 interpreted in the current radix. Thus, if the current radix is 16,
23203 @samp{10} is interpreted in hex, i.e.@: as 16 decimal, which doesn't
23206 @kindex set output-radix
23207 @item set output-radix @var{base}
23208 Set the default base for numeric display. Supported choices
23209 for @var{base} are decimal 8, 10, or 16. The base must itself be
23210 specified either unambiguously or using the current input radix.
23212 @kindex show input-radix
23213 @item show input-radix
23214 Display the current default base for numeric input.
23216 @kindex show output-radix
23217 @item show output-radix
23218 Display the current default base for numeric display.
23220 @item set radix @r{[}@var{base}@r{]}
23224 These commands set and show the default base for both input and output
23225 of numbers. @code{set radix} sets the radix of input and output to
23226 the same base; without an argument, it resets the radix back to its
23227 default value of 10.
23232 @section Configuring the Current ABI
23234 @value{GDBN} can determine the @dfn{ABI} (Application Binary Interface) of your
23235 application automatically. However, sometimes you need to override its
23236 conclusions. Use these commands to manage @value{GDBN}'s view of the
23242 @cindex Newlib OS ABI and its influence on the longjmp handling
23244 One @value{GDBN} configuration can debug binaries for multiple operating
23245 system targets, either via remote debugging or native emulation.
23246 @value{GDBN} will autodetect the @dfn{OS ABI} (Operating System ABI) in use,
23247 but you can override its conclusion using the @code{set osabi} command.
23248 One example where this is useful is in debugging of binaries which use
23249 an alternate C library (e.g.@: @sc{uClibc} for @sc{gnu}/Linux) which does
23250 not have the same identifying marks that the standard C library for your
23253 When @value{GDBN} is debugging the AArch64 architecture, it provides a
23254 ``Newlib'' OS ABI. This is useful for handling @code{setjmp} and
23255 @code{longjmp} when debugging binaries that use the @sc{newlib} C library.
23256 The ``Newlib'' OS ABI can be selected by @code{set osabi Newlib}.
23260 Show the OS ABI currently in use.
23263 With no argument, show the list of registered available OS ABI's.
23265 @item set osabi @var{abi}
23266 Set the current OS ABI to @var{abi}.
23269 @cindex float promotion
23271 Generally, the way that an argument of type @code{float} is passed to a
23272 function depends on whether the function is prototyped. For a prototyped
23273 (i.e.@: ANSI/ISO style) function, @code{float} arguments are passed unchanged,
23274 according to the architecture's convention for @code{float}. For unprototyped
23275 (i.e.@: K&R style) functions, @code{float} arguments are first promoted to type
23276 @code{double} and then passed.
23278 Unfortunately, some forms of debug information do not reliably indicate whether
23279 a function is prototyped. If @value{GDBN} calls a function that is not marked
23280 as prototyped, it consults @kbd{set coerce-float-to-double}.
23283 @kindex set coerce-float-to-double
23284 @item set coerce-float-to-double
23285 @itemx set coerce-float-to-double on
23286 Arguments of type @code{float} will be promoted to @code{double} when passed
23287 to an unprototyped function. This is the default setting.
23289 @item set coerce-float-to-double off
23290 Arguments of type @code{float} will be passed directly to unprototyped
23293 @kindex show coerce-float-to-double
23294 @item show coerce-float-to-double
23295 Show the current setting of promoting @code{float} to @code{double}.
23299 @kindex show cp-abi
23300 @value{GDBN} needs to know the ABI used for your program's C@t{++}
23301 objects. The correct C@t{++} ABI depends on which C@t{++} compiler was
23302 used to build your application. @value{GDBN} only fully supports
23303 programs with a single C@t{++} ABI; if your program contains code using
23304 multiple C@t{++} ABI's or if @value{GDBN} can not identify your
23305 program's ABI correctly, you can tell @value{GDBN} which ABI to use.
23306 Currently supported ABI's include ``gnu-v2'', for @code{g++} versions
23307 before 3.0, ``gnu-v3'', for @code{g++} versions 3.0 and later, and
23308 ``hpaCC'' for the HP ANSI C@t{++} compiler. Other C@t{++} compilers may
23309 use the ``gnu-v2'' or ``gnu-v3'' ABI's as well. The default setting is
23314 Show the C@t{++} ABI currently in use.
23317 With no argument, show the list of supported C@t{++} ABI's.
23319 @item set cp-abi @var{abi}
23320 @itemx set cp-abi auto
23321 Set the current C@t{++} ABI to @var{abi}, or return to automatic detection.
23325 @section Automatically loading associated files
23326 @cindex auto-loading
23328 @value{GDBN} sometimes reads files with commands and settings automatically,
23329 without being explicitly told so by the user. We call this feature
23330 @dfn{auto-loading}. While auto-loading is useful for automatically adapting
23331 @value{GDBN} to the needs of your project, it can sometimes produce unexpected
23332 results or introduce security risks (e.g., if the file comes from untrusted
23336 * Init File in the Current Directory:: @samp{set/show/info auto-load local-gdbinit}
23337 * libthread_db.so.1 file:: @samp{set/show/info auto-load libthread-db}
23339 * Auto-loading safe path:: @samp{set/show/info auto-load safe-path}
23340 * Auto-loading verbose mode:: @samp{set/show debug auto-load}
23343 There are various kinds of files @value{GDBN} can automatically load.
23344 In addition to these files, @value{GDBN} supports auto-loading code written
23345 in various extension languages. @xref{Auto-loading extensions}.
23347 Note that loading of these associated files (including the local @file{.gdbinit}
23348 file) requires accordingly configured @code{auto-load safe-path}
23349 (@pxref{Auto-loading safe path}).
23351 For these reasons, @value{GDBN} includes commands and options to let you
23352 control when to auto-load files and which files should be auto-loaded.
23355 @anchor{set auto-load off}
23356 @kindex set auto-load off
23357 @item set auto-load off
23358 Globally disable loading of all auto-loaded files.
23359 You may want to use this command with the @samp{-iex} option
23360 (@pxref{Option -init-eval-command}) such as:
23362 $ @kbd{gdb -iex "set auto-load off" untrusted-executable corefile}
23365 Be aware that system init file (@pxref{System-wide configuration})
23366 and init files from your home directory (@pxref{Home Directory Init File})
23367 still get read (as they come from generally trusted directories).
23368 To prevent @value{GDBN} from auto-loading even those init files, use the
23369 @option{-nx} option (@pxref{Mode Options}), in addition to
23370 @code{set auto-load no}.
23372 @anchor{show auto-load}
23373 @kindex show auto-load
23374 @item show auto-load
23375 Show whether auto-loading of each specific @samp{auto-load} file(s) is enabled
23379 (gdb) show auto-load
23380 gdb-scripts: Auto-loading of canned sequences of commands scripts is on.
23381 libthread-db: Auto-loading of inferior specific libthread_db is on.
23382 local-gdbinit: Auto-loading of .gdbinit script from current directory
23384 python-scripts: Auto-loading of Python scripts is on.
23385 safe-path: List of directories from which it is safe to auto-load files
23386 is $debugdir:$datadir/auto-load.
23387 scripts-directory: List of directories from which to load auto-loaded scripts
23388 is $debugdir:$datadir/auto-load.
23391 @anchor{info auto-load}
23392 @kindex info auto-load
23393 @item info auto-load
23394 Print whether each specific @samp{auto-load} file(s) have been auto-loaded or
23398 (gdb) info auto-load
23401 Yes /home/user/gdb/gdb-gdb.gdb
23402 libthread-db: No auto-loaded libthread-db.
23403 local-gdbinit: Local .gdbinit file "/home/user/gdb/.gdbinit" has been
23407 Yes /home/user/gdb/gdb-gdb.py
23411 These are @value{GDBN} control commands for the auto-loading:
23413 @multitable @columnfractions .5 .5
23414 @item @xref{set auto-load off}.
23415 @tab Disable auto-loading globally.
23416 @item @xref{show auto-load}.
23417 @tab Show setting of all kinds of files.
23418 @item @xref{info auto-load}.
23419 @tab Show state of all kinds of files.
23420 @item @xref{set auto-load gdb-scripts}.
23421 @tab Control for @value{GDBN} command scripts.
23422 @item @xref{show auto-load gdb-scripts}.
23423 @tab Show setting of @value{GDBN} command scripts.
23424 @item @xref{info auto-load gdb-scripts}.
23425 @tab Show state of @value{GDBN} command scripts.
23426 @item @xref{set auto-load python-scripts}.
23427 @tab Control for @value{GDBN} Python scripts.
23428 @item @xref{show auto-load python-scripts}.
23429 @tab Show setting of @value{GDBN} Python scripts.
23430 @item @xref{info auto-load python-scripts}.
23431 @tab Show state of @value{GDBN} Python scripts.
23432 @item @xref{set auto-load guile-scripts}.
23433 @tab Control for @value{GDBN} Guile scripts.
23434 @item @xref{show auto-load guile-scripts}.
23435 @tab Show setting of @value{GDBN} Guile scripts.
23436 @item @xref{info auto-load guile-scripts}.
23437 @tab Show state of @value{GDBN} Guile scripts.
23438 @item @xref{set auto-load scripts-directory}.
23439 @tab Control for @value{GDBN} auto-loaded scripts location.
23440 @item @xref{show auto-load scripts-directory}.
23441 @tab Show @value{GDBN} auto-loaded scripts location.
23442 @item @xref{add-auto-load-scripts-directory}.
23443 @tab Add directory for auto-loaded scripts location list.
23444 @item @xref{set auto-load local-gdbinit}.
23445 @tab Control for init file in the current directory.
23446 @item @xref{show auto-load local-gdbinit}.
23447 @tab Show setting of init file in the current directory.
23448 @item @xref{info auto-load local-gdbinit}.
23449 @tab Show state of init file in the current directory.
23450 @item @xref{set auto-load libthread-db}.
23451 @tab Control for thread debugging library.
23452 @item @xref{show auto-load libthread-db}.
23453 @tab Show setting of thread debugging library.
23454 @item @xref{info auto-load libthread-db}.
23455 @tab Show state of thread debugging library.
23456 @item @xref{set auto-load safe-path}.
23457 @tab Control directories trusted for automatic loading.
23458 @item @xref{show auto-load safe-path}.
23459 @tab Show directories trusted for automatic loading.
23460 @item @xref{add-auto-load-safe-path}.
23461 @tab Add directory trusted for automatic loading.
23464 @node Init File in the Current Directory
23465 @subsection Automatically loading init file in the current directory
23466 @cindex auto-loading init file in the current directory
23468 By default, @value{GDBN} reads and executes the canned sequences of commands
23469 from init file (if any) in the current working directory,
23470 see @ref{Init File in the Current Directory during Startup}.
23472 Note that loading of this local @file{.gdbinit} file also requires accordingly
23473 configured @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
23476 @anchor{set auto-load local-gdbinit}
23477 @kindex set auto-load local-gdbinit
23478 @item set auto-load local-gdbinit [on|off]
23479 Enable or disable the auto-loading of canned sequences of commands
23480 (@pxref{Sequences}) found in init file in the current directory.
23482 @anchor{show auto-load local-gdbinit}
23483 @kindex show auto-load local-gdbinit
23484 @item show auto-load local-gdbinit
23485 Show whether auto-loading of canned sequences of commands from init file in the
23486 current directory is enabled or disabled.
23488 @anchor{info auto-load local-gdbinit}
23489 @kindex info auto-load local-gdbinit
23490 @item info auto-load local-gdbinit
23491 Print whether canned sequences of commands from init file in the
23492 current directory have been auto-loaded.
23495 @node libthread_db.so.1 file
23496 @subsection Automatically loading thread debugging library
23497 @cindex auto-loading libthread_db.so.1
23499 This feature is currently present only on @sc{gnu}/Linux native hosts.
23501 @value{GDBN} reads in some cases thread debugging library from places specific
23502 to the inferior (@pxref{set libthread-db-search-path}).
23504 The special @samp{libthread-db-search-path} entry @samp{$sdir} is processed
23505 without checking this @samp{set auto-load libthread-db} switch as system
23506 libraries have to be trusted in general. In all other cases of
23507 @samp{libthread-db-search-path} entries @value{GDBN} checks first if @samp{set
23508 auto-load libthread-db} is enabled before trying to open such thread debugging
23511 Note that loading of this debugging library also requires accordingly configured
23512 @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
23515 @anchor{set auto-load libthread-db}
23516 @kindex set auto-load libthread-db
23517 @item set auto-load libthread-db [on|off]
23518 Enable or disable the auto-loading of inferior specific thread debugging library.
23520 @anchor{show auto-load libthread-db}
23521 @kindex show auto-load libthread-db
23522 @item show auto-load libthread-db
23523 Show whether auto-loading of inferior specific thread debugging library is
23524 enabled or disabled.
23526 @anchor{info auto-load libthread-db}
23527 @kindex info auto-load libthread-db
23528 @item info auto-load libthread-db
23529 Print the list of all loaded inferior specific thread debugging libraries and
23530 for each such library print list of inferior @var{pid}s using it.
23533 @node Auto-loading safe path
23534 @subsection Security restriction for auto-loading
23535 @cindex auto-loading safe-path
23537 As the files of inferior can come from untrusted source (such as submitted by
23538 an application user) @value{GDBN} does not always load any files automatically.
23539 @value{GDBN} provides the @samp{set auto-load safe-path} setting to list
23540 directories trusted for loading files not explicitly requested by user.
23541 Each directory can also be a shell wildcard pattern.
23543 If the path is not set properly you will see a warning and the file will not
23548 Reading symbols from /home/user/gdb/gdb...done.
23549 warning: File "/home/user/gdb/gdb-gdb.gdb" auto-loading has been
23550 declined by your `auto-load safe-path' set
23551 to "$debugdir:$datadir/auto-load".
23552 warning: File "/home/user/gdb/gdb-gdb.py" auto-loading has been
23553 declined by your `auto-load safe-path' set
23554 to "$debugdir:$datadir/auto-load".
23558 To instruct @value{GDBN} to go ahead and use the init files anyway,
23559 invoke @value{GDBN} like this:
23562 $ gdb -q -iex "set auto-load safe-path /home/user/gdb" ./gdb
23565 The list of trusted directories is controlled by the following commands:
23568 @anchor{set auto-load safe-path}
23569 @kindex set auto-load safe-path
23570 @item set auto-load safe-path @r{[}@var{directories}@r{]}
23571 Set the list of directories (and their subdirectories) trusted for automatic
23572 loading and execution of scripts. You can also enter a specific trusted file.
23573 Each directory can also be a shell wildcard pattern; wildcards do not match
23574 directory separator - see @code{FNM_PATHNAME} for system function @code{fnmatch}
23575 (@pxref{Wildcard Matching, fnmatch, , libc, GNU C Library Reference Manual}).
23576 If you omit @var{directories}, @samp{auto-load safe-path} will be reset to
23577 its default value as specified during @value{GDBN} compilation.
23579 The list of directories uses path separator (@samp{:} on GNU and Unix
23580 systems, @samp{;} on MS-Windows and MS-DOS) to separate directories, similarly
23581 to the @env{PATH} environment variable.
23583 @anchor{show auto-load safe-path}
23584 @kindex show auto-load safe-path
23585 @item show auto-load safe-path
23586 Show the list of directories trusted for automatic loading and execution of
23589 @anchor{add-auto-load-safe-path}
23590 @kindex add-auto-load-safe-path
23591 @item add-auto-load-safe-path
23592 Add an entry (or list of entries) to the list of directories trusted for
23593 automatic loading and execution of scripts. Multiple entries may be delimited
23594 by the host platform path separator in use.
23597 This variable defaults to what @code{--with-auto-load-dir} has been configured
23598 to (@pxref{with-auto-load-dir}). @file{$debugdir} and @file{$datadir}
23599 substitution applies the same as for @ref{set auto-load scripts-directory}.
23600 The default @code{set auto-load safe-path} value can be also overriden by
23601 @value{GDBN} configuration option @option{--with-auto-load-safe-path}.
23603 Setting this variable to @file{/} disables this security protection,
23604 corresponding @value{GDBN} configuration option is
23605 @option{--without-auto-load-safe-path}.
23606 This variable is supposed to be set to the system directories writable by the
23607 system superuser only. Users can add their source directories in init files in
23608 their home directories (@pxref{Home Directory Init File}). See also deprecated
23609 init file in the current directory
23610 (@pxref{Init File in the Current Directory during Startup}).
23612 To force @value{GDBN} to load the files it declined to load in the previous
23613 example, you could use one of the following ways:
23616 @item @file{~/.gdbinit}: @samp{add-auto-load-safe-path ~/src/gdb}
23617 Specify this trusted directory (or a file) as additional component of the list.
23618 You have to specify also any existing directories displayed by
23619 by @samp{show auto-load safe-path} (such as @samp{/usr:/bin} in this example).
23621 @item @kbd{gdb -iex "set auto-load safe-path /usr:/bin:~/src/gdb" @dots{}}
23622 Specify this directory as in the previous case but just for a single
23623 @value{GDBN} session.
23625 @item @kbd{gdb -iex "set auto-load safe-path /" @dots{}}
23626 Disable auto-loading safety for a single @value{GDBN} session.
23627 This assumes all the files you debug during this @value{GDBN} session will come
23628 from trusted sources.
23630 @item @kbd{./configure --without-auto-load-safe-path}
23631 During compilation of @value{GDBN} you may disable any auto-loading safety.
23632 This assumes all the files you will ever debug with this @value{GDBN} come from
23636 On the other hand you can also explicitly forbid automatic files loading which
23637 also suppresses any such warning messages:
23640 @item @kbd{gdb -iex "set auto-load no" @dots{}}
23641 You can use @value{GDBN} command-line option for a single @value{GDBN} session.
23643 @item @file{~/.gdbinit}: @samp{set auto-load no}
23644 Disable auto-loading globally for the user
23645 (@pxref{Home Directory Init File}). While it is improbable, you could also
23646 use system init file instead (@pxref{System-wide configuration}).
23649 This setting applies to the file names as entered by user. If no entry matches
23650 @value{GDBN} tries as a last resort to also resolve all the file names into
23651 their canonical form (typically resolving symbolic links) and compare the
23652 entries again. @value{GDBN} already canonicalizes most of the filenames on its
23653 own before starting the comparison so a canonical form of directories is
23654 recommended to be entered.
23656 @node Auto-loading verbose mode
23657 @subsection Displaying files tried for auto-load
23658 @cindex auto-loading verbose mode
23660 For better visibility of all the file locations where you can place scripts to
23661 be auto-loaded with inferior --- or to protect yourself against accidental
23662 execution of untrusted scripts --- @value{GDBN} provides a feature for printing
23663 all the files attempted to be loaded. Both existing and non-existing files may
23666 For example the list of directories from which it is safe to auto-load files
23667 (@pxref{Auto-loading safe path}) applies also to canonicalized filenames which
23668 may not be too obvious while setting it up.
23671 (gdb) set debug auto-load on
23672 (gdb) file ~/src/t/true
23673 auto-load: Loading canned sequences of commands script "/tmp/true-gdb.gdb"
23674 for objfile "/tmp/true".
23675 auto-load: Updating directories of "/usr:/opt".
23676 auto-load: Using directory "/usr".
23677 auto-load: Using directory "/opt".
23678 warning: File "/tmp/true-gdb.gdb" auto-loading has been declined
23679 by your `auto-load safe-path' set to "/usr:/opt".
23683 @anchor{set debug auto-load}
23684 @kindex set debug auto-load
23685 @item set debug auto-load [on|off]
23686 Set whether to print the filenames attempted to be auto-loaded.
23688 @anchor{show debug auto-load}
23689 @kindex show debug auto-load
23690 @item show debug auto-load
23691 Show whether printing of the filenames attempted to be auto-loaded is turned
23695 @node Messages/Warnings
23696 @section Optional Warnings and Messages
23698 @cindex verbose operation
23699 @cindex optional warnings
23700 By default, @value{GDBN} is silent about its inner workings. If you are
23701 running on a slow machine, you may want to use the @code{set verbose}
23702 command. This makes @value{GDBN} tell you when it does a lengthy
23703 internal operation, so you will not think it has crashed.
23705 Currently, the messages controlled by @code{set verbose} are those
23706 which announce that the symbol table for a source file is being read;
23707 see @code{symbol-file} in @ref{Files, ,Commands to Specify Files}.
23710 @kindex set verbose
23711 @item set verbose on
23712 Enables @value{GDBN} output of certain informational messages.
23714 @item set verbose off
23715 Disables @value{GDBN} output of certain informational messages.
23717 @kindex show verbose
23719 Displays whether @code{set verbose} is on or off.
23722 By default, if @value{GDBN} encounters bugs in the symbol table of an
23723 object file, it is silent; but if you are debugging a compiler, you may
23724 find this information useful (@pxref{Symbol Errors, ,Errors Reading
23729 @kindex set complaints
23730 @item set complaints @var{limit}
23731 Permits @value{GDBN} to output @var{limit} complaints about each type of
23732 unusual symbols before becoming silent about the problem. Set
23733 @var{limit} to zero to suppress all complaints; set it to a large number
23734 to prevent complaints from being suppressed.
23736 @kindex show complaints
23737 @item show complaints
23738 Displays how many symbol complaints @value{GDBN} is permitted to produce.
23742 @anchor{confirmation requests}
23743 By default, @value{GDBN} is cautious, and asks what sometimes seems to be a
23744 lot of stupid questions to confirm certain commands. For example, if
23745 you try to run a program which is already running:
23749 The program being debugged has been started already.
23750 Start it from the beginning? (y or n)
23753 If you are willing to unflinchingly face the consequences of your own
23754 commands, you can disable this ``feature'':
23758 @kindex set confirm
23760 @cindex confirmation
23761 @cindex stupid questions
23762 @item set confirm off
23763 Disables confirmation requests. Note that running @value{GDBN} with
23764 the @option{--batch} option (@pxref{Mode Options, -batch}) also
23765 automatically disables confirmation requests.
23767 @item set confirm on
23768 Enables confirmation requests (the default).
23770 @kindex show confirm
23772 Displays state of confirmation requests.
23776 @cindex command tracing
23777 If you need to debug user-defined commands or sourced files you may find it
23778 useful to enable @dfn{command tracing}. In this mode each command will be
23779 printed as it is executed, prefixed with one or more @samp{+} symbols, the
23780 quantity denoting the call depth of each command.
23783 @kindex set trace-commands
23784 @cindex command scripts, debugging
23785 @item set trace-commands on
23786 Enable command tracing.
23787 @item set trace-commands off
23788 Disable command tracing.
23789 @item show trace-commands
23790 Display the current state of command tracing.
23793 @node Debugging Output
23794 @section Optional Messages about Internal Happenings
23795 @cindex optional debugging messages
23797 @value{GDBN} has commands that enable optional debugging messages from
23798 various @value{GDBN} subsystems; normally these commands are of
23799 interest to @value{GDBN} maintainers, or when reporting a bug. This
23800 section documents those commands.
23803 @kindex set exec-done-display
23804 @item set exec-done-display
23805 Turns on or off the notification of asynchronous commands'
23806 completion. When on, @value{GDBN} will print a message when an
23807 asynchronous command finishes its execution. The default is off.
23808 @kindex show exec-done-display
23809 @item show exec-done-display
23810 Displays the current setting of asynchronous command completion
23813 @cindex ARM AArch64
23814 @item set debug aarch64
23815 Turns on or off display of debugging messages related to ARM AArch64.
23816 The default is off.
23818 @item show debug aarch64
23819 Displays the current state of displaying debugging messages related to
23821 @cindex gdbarch debugging info
23822 @cindex architecture debugging info
23823 @item set debug arch
23824 Turns on or off display of gdbarch debugging info. The default is off
23825 @item show debug arch
23826 Displays the current state of displaying gdbarch debugging info.
23827 @item set debug aix-solib
23828 @cindex AIX shared library debugging
23829 Control display of debugging messages from the AIX shared library
23830 support module. The default is off.
23831 @item show debug aix-thread
23832 Show the current state of displaying AIX shared library debugging messages.
23833 @item set debug aix-thread
23834 @cindex AIX threads
23835 Display debugging messages about inner workings of the AIX thread
23837 @item show debug aix-thread
23838 Show the current state of AIX thread debugging info display.
23839 @item set debug check-physname
23841 Check the results of the ``physname'' computation. When reading DWARF
23842 debugging information for C@t{++}, @value{GDBN} attempts to compute
23843 each entity's name. @value{GDBN} can do this computation in two
23844 different ways, depending on exactly what information is present.
23845 When enabled, this setting causes @value{GDBN} to compute the names
23846 both ways and display any discrepancies.
23847 @item show debug check-physname
23848 Show the current state of ``physname'' checking.
23849 @item set debug coff-pe-read
23850 @cindex COFF/PE exported symbols
23851 Control display of debugging messages related to reading of COFF/PE
23852 exported symbols. The default is off.
23853 @item show debug coff-pe-read
23854 Displays the current state of displaying debugging messages related to
23855 reading of COFF/PE exported symbols.
23856 @item set debug dwarf-die
23858 Dump DWARF DIEs after they are read in.
23859 The value is the number of nesting levels to print.
23860 A value of zero turns off the display.
23861 @item show debug dwarf-die
23862 Show the current state of DWARF DIE debugging.
23863 @item set debug dwarf-line
23864 @cindex DWARF Line Tables
23865 Turns on or off display of debugging messages related to reading
23866 DWARF line tables. The default is 0 (off).
23867 A value of 1 provides basic information.
23868 A value greater than 1 provides more verbose information.
23869 @item show debug dwarf-line
23870 Show the current state of DWARF line table debugging.
23871 @item set debug dwarf-read
23872 @cindex DWARF Reading
23873 Turns on or off display of debugging messages related to reading
23874 DWARF debug info. The default is 0 (off).
23875 A value of 1 provides basic information.
23876 A value greater than 1 provides more verbose information.
23877 @item show debug dwarf-read
23878 Show the current state of DWARF reader debugging.
23879 @item set debug displaced
23880 @cindex displaced stepping debugging info
23881 Turns on or off display of @value{GDBN} debugging info for the
23882 displaced stepping support. The default is off.
23883 @item show debug displaced
23884 Displays the current state of displaying @value{GDBN} debugging info
23885 related to displaced stepping.
23886 @item set debug event
23887 @cindex event debugging info
23888 Turns on or off display of @value{GDBN} event debugging info. The
23890 @item show debug event
23891 Displays the current state of displaying @value{GDBN} event debugging
23893 @item set debug expression
23894 @cindex expression debugging info
23895 Turns on or off display of debugging info about @value{GDBN}
23896 expression parsing. The default is off.
23897 @item show debug expression
23898 Displays the current state of displaying debugging info about
23899 @value{GDBN} expression parsing.
23900 @item set debug fbsd-lwp
23901 @cindex FreeBSD LWP debug messages
23902 Turns on or off debugging messages from the FreeBSD LWP debug support.
23903 @item show debug fbsd-lwp
23904 Show the current state of FreeBSD LWP debugging messages.
23905 @item set debug frame
23906 @cindex frame debugging info
23907 Turns on or off display of @value{GDBN} frame debugging info. The
23909 @item show debug frame
23910 Displays the current state of displaying @value{GDBN} frame debugging
23912 @item set debug gnu-nat
23913 @cindex @sc{gnu}/Hurd debug messages
23914 Turn on or off debugging messages from the @sc{gnu}/Hurd debug support.
23915 @item show debug gnu-nat
23916 Show the current state of @sc{gnu}/Hurd debugging messages.
23917 @item set debug infrun
23918 @cindex inferior debugging info
23919 Turns on or off display of @value{GDBN} debugging info for running the inferior.
23920 The default is off. @file{infrun.c} contains GDB's runtime state machine used
23921 for implementing operations such as single-stepping the inferior.
23922 @item show debug infrun
23923 Displays the current state of @value{GDBN} inferior debugging.
23924 @item set debug jit
23925 @cindex just-in-time compilation, debugging messages
23926 Turn on or off debugging messages from JIT debug support.
23927 @item show debug jit
23928 Displays the current state of @value{GDBN} JIT debugging.
23929 @item set debug lin-lwp
23930 @cindex @sc{gnu}/Linux LWP debug messages
23931 @cindex Linux lightweight processes
23932 Turn on or off debugging messages from the Linux LWP debug support.
23933 @item show debug lin-lwp
23934 Show the current state of Linux LWP debugging messages.
23935 @item set debug linux-namespaces
23936 @cindex @sc{gnu}/Linux namespaces debug messages
23937 Turn on or off debugging messages from the Linux namespaces debug support.
23938 @item show debug linux-namespaces
23939 Show the current state of Linux namespaces debugging messages.
23940 @item set debug mach-o
23941 @cindex Mach-O symbols processing
23942 Control display of debugging messages related to Mach-O symbols
23943 processing. The default is off.
23944 @item show debug mach-o
23945 Displays the current state of displaying debugging messages related to
23946 reading of COFF/PE exported symbols.
23947 @item set debug notification
23948 @cindex remote async notification debugging info
23949 Turn on or off debugging messages about remote async notification.
23950 The default is off.
23951 @item show debug notification
23952 Displays the current state of remote async notification debugging messages.
23953 @item set debug observer
23954 @cindex observer debugging info
23955 Turns on or off display of @value{GDBN} observer debugging. This
23956 includes info such as the notification of observable events.
23957 @item show debug observer
23958 Displays the current state of observer debugging.
23959 @item set debug overload
23960 @cindex C@t{++} overload debugging info
23961 Turns on or off display of @value{GDBN} C@t{++} overload debugging
23962 info. This includes info such as ranking of functions, etc. The default
23964 @item show debug overload
23965 Displays the current state of displaying @value{GDBN} C@t{++} overload
23967 @cindex expression parser, debugging info
23968 @cindex debug expression parser
23969 @item set debug parser
23970 Turns on or off the display of expression parser debugging output.
23971 Internally, this sets the @code{yydebug} variable in the expression
23972 parser. @xref{Tracing, , Tracing Your Parser, bison, Bison}, for
23973 details. The default is off.
23974 @item show debug parser
23975 Show the current state of expression parser debugging.
23976 @cindex packets, reporting on stdout
23977 @cindex serial connections, debugging
23978 @cindex debug remote protocol
23979 @cindex remote protocol debugging
23980 @cindex display remote packets
23981 @item set debug remote
23982 Turns on or off display of reports on all packets sent back and forth across
23983 the serial line to the remote machine. The info is printed on the
23984 @value{GDBN} standard output stream. The default is off.
23985 @item show debug remote
23986 Displays the state of display of remote packets.
23988 @item set debug separate-debug-file
23989 Turns on or off display of debug output about separate debug file search.
23990 @item show debug separate-debug-file
23991 Displays the state of separate debug file search debug output.
23993 @item set debug serial
23994 Turns on or off display of @value{GDBN} serial debugging info. The
23996 @item show debug serial
23997 Displays the current state of displaying @value{GDBN} serial debugging
23999 @item set debug solib-frv
24000 @cindex FR-V shared-library debugging
24001 Turn on or off debugging messages for FR-V shared-library code.
24002 @item show debug solib-frv
24003 Display the current state of FR-V shared-library code debugging
24005 @item set debug symbol-lookup
24006 @cindex symbol lookup
24007 Turns on or off display of debugging messages related to symbol lookup.
24008 The default is 0 (off).
24009 A value of 1 provides basic information.
24010 A value greater than 1 provides more verbose information.
24011 @item show debug symbol-lookup
24012 Show the current state of symbol lookup debugging messages.
24013 @item set debug symfile
24014 @cindex symbol file functions
24015 Turns on or off display of debugging messages related to symbol file functions.
24016 The default is off. @xref{Files}.
24017 @item show debug symfile
24018 Show the current state of symbol file debugging messages.
24019 @item set debug symtab-create
24020 @cindex symbol table creation
24021 Turns on or off display of debugging messages related to symbol table creation.
24022 The default is 0 (off).
24023 A value of 1 provides basic information.
24024 A value greater than 1 provides more verbose information.
24025 @item show debug symtab-create
24026 Show the current state of symbol table creation debugging.
24027 @item set debug target
24028 @cindex target debugging info
24029 Turns on or off display of @value{GDBN} target debugging info. This info
24030 includes what is going on at the target level of GDB, as it happens. The
24031 default is 0. Set it to 1 to track events, and to 2 to also track the
24032 value of large memory transfers.
24033 @item show debug target
24034 Displays the current state of displaying @value{GDBN} target debugging
24036 @item set debug timestamp
24037 @cindex timestampping debugging info
24038 Turns on or off display of timestamps with @value{GDBN} debugging info.
24039 When enabled, seconds and microseconds are displayed before each debugging
24041 @item show debug timestamp
24042 Displays the current state of displaying timestamps with @value{GDBN}
24044 @item set debug varobj
24045 @cindex variable object debugging info
24046 Turns on or off display of @value{GDBN} variable object debugging
24047 info. The default is off.
24048 @item show debug varobj
24049 Displays the current state of displaying @value{GDBN} variable object
24051 @item set debug xml
24052 @cindex XML parser debugging
24053 Turn on or off debugging messages for built-in XML parsers.
24054 @item show debug xml
24055 Displays the current state of XML debugging messages.
24058 @node Other Misc Settings
24059 @section Other Miscellaneous Settings
24060 @cindex miscellaneous settings
24063 @kindex set interactive-mode
24064 @item set interactive-mode
24065 If @code{on}, forces @value{GDBN} to assume that GDB was started
24066 in a terminal. In practice, this means that @value{GDBN} should wait
24067 for the user to answer queries generated by commands entered at
24068 the command prompt. If @code{off}, forces @value{GDBN} to operate
24069 in the opposite mode, and it uses the default answers to all queries.
24070 If @code{auto} (the default), @value{GDBN} tries to determine whether
24071 its standard input is a terminal, and works in interactive-mode if it
24072 is, non-interactively otherwise.
24074 In the vast majority of cases, the debugger should be able to guess
24075 correctly which mode should be used. But this setting can be useful
24076 in certain specific cases, such as running a MinGW @value{GDBN}
24077 inside a cygwin window.
24079 @kindex show interactive-mode
24080 @item show interactive-mode
24081 Displays whether the debugger is operating in interactive mode or not.
24084 @node Extending GDB
24085 @chapter Extending @value{GDBN}
24086 @cindex extending GDB
24088 @value{GDBN} provides several mechanisms for extension.
24089 @value{GDBN} also provides the ability to automatically load
24090 extensions when it reads a file for debugging. This allows the
24091 user to automatically customize @value{GDBN} for the program
24095 * Sequences:: Canned Sequences of @value{GDBN} Commands
24096 * Python:: Extending @value{GDBN} using Python
24097 * Guile:: Extending @value{GDBN} using Guile
24098 * Auto-loading extensions:: Automatically loading extensions
24099 * Multiple Extension Languages:: Working with multiple extension languages
24100 * Aliases:: Creating new spellings of existing commands
24103 To facilitate the use of extension languages, @value{GDBN} is capable
24104 of evaluating the contents of a file. When doing so, @value{GDBN}
24105 can recognize which extension language is being used by looking at
24106 the filename extension. Files with an unrecognized filename extension
24107 are always treated as a @value{GDBN} Command Files.
24108 @xref{Command Files,, Command files}.
24110 You can control how @value{GDBN} evaluates these files with the following
24114 @kindex set script-extension
24115 @kindex show script-extension
24116 @item set script-extension off
24117 All scripts are always evaluated as @value{GDBN} Command Files.
24119 @item set script-extension soft
24120 The debugger determines the scripting language based on filename
24121 extension. If this scripting language is supported, @value{GDBN}
24122 evaluates the script using that language. Otherwise, it evaluates
24123 the file as a @value{GDBN} Command File.
24125 @item set script-extension strict
24126 The debugger determines the scripting language based on filename
24127 extension, and evaluates the script using that language. If the
24128 language is not supported, then the evaluation fails.
24130 @item show script-extension
24131 Display the current value of the @code{script-extension} option.
24136 @section Canned Sequences of Commands
24138 Aside from breakpoint commands (@pxref{Break Commands, ,Breakpoint
24139 Command Lists}), @value{GDBN} provides two ways to store sequences of
24140 commands for execution as a unit: user-defined commands and command
24144 * Define:: How to define your own commands
24145 * Hooks:: Hooks for user-defined commands
24146 * Command Files:: How to write scripts of commands to be stored in a file
24147 * Output:: Commands for controlled output
24148 * Auto-loading sequences:: Controlling auto-loaded command files
24152 @subsection User-defined Commands
24154 @cindex user-defined command
24155 @cindex arguments, to user-defined commands
24156 A @dfn{user-defined command} is a sequence of @value{GDBN} commands to
24157 which you assign a new name as a command. This is done with the
24158 @code{define} command. User commands may accept an unlimited number of arguments
24159 separated by whitespace. Arguments are accessed within the user command
24160 via @code{$arg0@dots{}$argN}. A trivial example:
24164 print $arg0 + $arg1 + $arg2
24169 To execute the command use:
24176 This defines the command @code{adder}, which prints the sum of
24177 its three arguments. Note the arguments are text substitutions, so they may
24178 reference variables, use complex expressions, or even perform inferior
24181 @cindex argument count in user-defined commands
24182 @cindex how many arguments (user-defined commands)
24183 In addition, @code{$argc} may be used to find out how many arguments have
24189 print $arg0 + $arg1
24192 print $arg0 + $arg1 + $arg2
24197 Combining with the @code{eval} command (@pxref{eval}) makes it easier
24198 to process a variable number of arguments:
24205 eval "set $sum = $sum + $arg%d", $i
24215 @item define @var{commandname}
24216 Define a command named @var{commandname}. If there is already a command
24217 by that name, you are asked to confirm that you want to redefine it.
24218 The argument @var{commandname} may be a bare command name consisting of letters,
24219 numbers, dashes, and underscores. It may also start with any predefined
24220 prefix command. For example, @samp{define target my-target} creates
24221 a user-defined @samp{target my-target} command.
24223 The definition of the command is made up of other @value{GDBN} command lines,
24224 which are given following the @code{define} command. The end of these
24225 commands is marked by a line containing @code{end}.
24228 @kindex end@r{ (user-defined commands)}
24229 @item document @var{commandname}
24230 Document the user-defined command @var{commandname}, so that it can be
24231 accessed by @code{help}. The command @var{commandname} must already be
24232 defined. This command reads lines of documentation just as @code{define}
24233 reads the lines of the command definition, ending with @code{end}.
24234 After the @code{document} command is finished, @code{help} on command
24235 @var{commandname} displays the documentation you have written.
24237 You may use the @code{document} command again to change the
24238 documentation of a command. Redefining the command with @code{define}
24239 does not change the documentation.
24241 @kindex dont-repeat
24242 @cindex don't repeat command
24244 Used inside a user-defined command, this tells @value{GDBN} that this
24245 command should not be repeated when the user hits @key{RET}
24246 (@pxref{Command Syntax, repeat last command}).
24248 @kindex help user-defined
24249 @item help user-defined
24250 List all user-defined commands and all python commands defined in class
24251 COMAND_USER. The first line of the documentation or docstring is
24256 @itemx show user @var{commandname}
24257 Display the @value{GDBN} commands used to define @var{commandname} (but
24258 not its documentation). If no @var{commandname} is given, display the
24259 definitions for all user-defined commands.
24260 This does not work for user-defined python commands.
24262 @cindex infinite recursion in user-defined commands
24263 @kindex show max-user-call-depth
24264 @kindex set max-user-call-depth
24265 @item show max-user-call-depth
24266 @itemx set max-user-call-depth
24267 The value of @code{max-user-call-depth} controls how many recursion
24268 levels are allowed in user-defined commands before @value{GDBN} suspects an
24269 infinite recursion and aborts the command.
24270 This does not apply to user-defined python commands.
24273 In addition to the above commands, user-defined commands frequently
24274 use control flow commands, described in @ref{Command Files}.
24276 When user-defined commands are executed, the
24277 commands of the definition are not printed. An error in any command
24278 stops execution of the user-defined command.
24280 If used interactively, commands that would ask for confirmation proceed
24281 without asking when used inside a user-defined command. Many @value{GDBN}
24282 commands that normally print messages to say what they are doing omit the
24283 messages when used in a user-defined command.
24286 @subsection User-defined Command Hooks
24287 @cindex command hooks
24288 @cindex hooks, for commands
24289 @cindex hooks, pre-command
24292 You may define @dfn{hooks}, which are a special kind of user-defined
24293 command. Whenever you run the command @samp{foo}, if the user-defined
24294 command @samp{hook-foo} exists, it is executed (with no arguments)
24295 before that command.
24297 @cindex hooks, post-command
24299 A hook may also be defined which is run after the command you executed.
24300 Whenever you run the command @samp{foo}, if the user-defined command
24301 @samp{hookpost-foo} exists, it is executed (with no arguments) after
24302 that command. Post-execution hooks may exist simultaneously with
24303 pre-execution hooks, for the same command.
24305 It is valid for a hook to call the command which it hooks. If this
24306 occurs, the hook is not re-executed, thereby avoiding infinite recursion.
24308 @c It would be nice if hookpost could be passed a parameter indicating
24309 @c if the command it hooks executed properly or not. FIXME!
24311 @kindex stop@r{, a pseudo-command}
24312 In addition, a pseudo-command, @samp{stop} exists. Defining
24313 (@samp{hook-stop}) makes the associated commands execute every time
24314 execution stops in your program: before breakpoint commands are run,
24315 displays are printed, or the stack frame is printed.
24317 For example, to ignore @code{SIGALRM} signals while
24318 single-stepping, but treat them normally during normal execution,
24323 handle SIGALRM nopass
24327 handle SIGALRM pass
24330 define hook-continue
24331 handle SIGALRM pass
24335 As a further example, to hook at the beginning and end of the @code{echo}
24336 command, and to add extra text to the beginning and end of the message,
24344 define hookpost-echo
24348 (@value{GDBP}) echo Hello World
24349 <<<---Hello World--->>>
24354 You can define a hook for any single-word command in @value{GDBN}, but
24355 not for command aliases; you should define a hook for the basic command
24356 name, e.g.@: @code{backtrace} rather than @code{bt}.
24357 @c FIXME! So how does Joe User discover whether a command is an alias
24359 You can hook a multi-word command by adding @code{hook-} or
24360 @code{hookpost-} to the last word of the command, e.g.@:
24361 @samp{define target hook-remote} to add a hook to @samp{target remote}.
24363 If an error occurs during the execution of your hook, execution of
24364 @value{GDBN} commands stops and @value{GDBN} issues a prompt
24365 (before the command that you actually typed had a chance to run).
24367 If you try to define a hook which does not match any known command, you
24368 get a warning from the @code{define} command.
24370 @node Command Files
24371 @subsection Command Files
24373 @cindex command files
24374 @cindex scripting commands
24375 A command file for @value{GDBN} is a text file made of lines that are
24376 @value{GDBN} commands. Comments (lines starting with @kbd{#}) may
24377 also be included. An empty line in a command file does nothing; it
24378 does not mean to repeat the last command, as it would from the
24381 You can request the execution of a command file with the @code{source}
24382 command. Note that the @code{source} command is also used to evaluate
24383 scripts that are not Command Files. The exact behavior can be configured
24384 using the @code{script-extension} setting.
24385 @xref{Extending GDB,, Extending GDB}.
24389 @cindex execute commands from a file
24390 @item source [-s] [-v] @var{filename}
24391 Execute the command file @var{filename}.
24394 The lines in a command file are generally executed sequentially,
24395 unless the order of execution is changed by one of the
24396 @emph{flow-control commands} described below. The commands are not
24397 printed as they are executed. An error in any command terminates
24398 execution of the command file and control is returned to the console.
24400 @value{GDBN} first searches for @var{filename} in the current directory.
24401 If the file is not found there, and @var{filename} does not specify a
24402 directory, then @value{GDBN} also looks for the file on the source search path
24403 (specified with the @samp{directory} command);
24404 except that @file{$cdir} is not searched because the compilation directory
24405 is not relevant to scripts.
24407 If @code{-s} is specified, then @value{GDBN} searches for @var{filename}
24408 on the search path even if @var{filename} specifies a directory.
24409 The search is done by appending @var{filename} to each element of the
24410 search path. So, for example, if @var{filename} is @file{mylib/myscript}
24411 and the search path contains @file{/home/user} then @value{GDBN} will
24412 look for the script @file{/home/user/mylib/myscript}.
24413 The search is also done if @var{filename} is an absolute path.
24414 For example, if @var{filename} is @file{/tmp/myscript} and
24415 the search path contains @file{/home/user} then @value{GDBN} will
24416 look for the script @file{/home/user/tmp/myscript}.
24417 For DOS-like systems, if @var{filename} contains a drive specification,
24418 it is stripped before concatenation. For example, if @var{filename} is
24419 @file{d:myscript} and the search path contains @file{c:/tmp} then @value{GDBN}
24420 will look for the script @file{c:/tmp/myscript}.
24422 If @code{-v}, for verbose mode, is given then @value{GDBN} displays
24423 each command as it is executed. The option must be given before
24424 @var{filename}, and is interpreted as part of the filename anywhere else.
24426 Commands that would ask for confirmation if used interactively proceed
24427 without asking when used in a command file. Many @value{GDBN} commands that
24428 normally print messages to say what they are doing omit the messages
24429 when called from command files.
24431 @value{GDBN} also accepts command input from standard input. In this
24432 mode, normal output goes to standard output and error output goes to
24433 standard error. Errors in a command file supplied on standard input do
24434 not terminate execution of the command file---execution continues with
24438 gdb < cmds > log 2>&1
24441 (The syntax above will vary depending on the shell used.) This example
24442 will execute commands from the file @file{cmds}. All output and errors
24443 would be directed to @file{log}.
24445 Since commands stored on command files tend to be more general than
24446 commands typed interactively, they frequently need to deal with
24447 complicated situations, such as different or unexpected values of
24448 variables and symbols, changes in how the program being debugged is
24449 built, etc. @value{GDBN} provides a set of flow-control commands to
24450 deal with these complexities. Using these commands, you can write
24451 complex scripts that loop over data structures, execute commands
24452 conditionally, etc.
24459 This command allows to include in your script conditionally executed
24460 commands. The @code{if} command takes a single argument, which is an
24461 expression to evaluate. It is followed by a series of commands that
24462 are executed only if the expression is true (its value is nonzero).
24463 There can then optionally be an @code{else} line, followed by a series
24464 of commands that are only executed if the expression was false. The
24465 end of the list is marked by a line containing @code{end}.
24469 This command allows to write loops. Its syntax is similar to
24470 @code{if}: the command takes a single argument, which is an expression
24471 to evaluate, and must be followed by the commands to execute, one per
24472 line, terminated by an @code{end}. These commands are called the
24473 @dfn{body} of the loop. The commands in the body of @code{while} are
24474 executed repeatedly as long as the expression evaluates to true.
24478 This command exits the @code{while} loop in whose body it is included.
24479 Execution of the script continues after that @code{while}s @code{end}
24482 @kindex loop_continue
24483 @item loop_continue
24484 This command skips the execution of the rest of the body of commands
24485 in the @code{while} loop in whose body it is included. Execution
24486 branches to the beginning of the @code{while} loop, where it evaluates
24487 the controlling expression.
24489 @kindex end@r{ (if/else/while commands)}
24491 Terminate the block of commands that are the body of @code{if},
24492 @code{else}, or @code{while} flow-control commands.
24497 @subsection Commands for Controlled Output
24499 During the execution of a command file or a user-defined command, normal
24500 @value{GDBN} output is suppressed; the only output that appears is what is
24501 explicitly printed by the commands in the definition. This section
24502 describes three commands useful for generating exactly the output you
24507 @item echo @var{text}
24508 @c I do not consider backslash-space a standard C escape sequence
24509 @c because it is not in ANSI.
24510 Print @var{text}. Nonprinting characters can be included in
24511 @var{text} using C escape sequences, such as @samp{\n} to print a
24512 newline. @strong{No newline is printed unless you specify one.}
24513 In addition to the standard C escape sequences, a backslash followed
24514 by a space stands for a space. This is useful for displaying a
24515 string with spaces at the beginning or the end, since leading and
24516 trailing spaces are otherwise trimmed from all arguments.
24517 To print @samp{@w{ }and foo =@w{ }}, use the command
24518 @samp{echo \@w{ }and foo = \@w{ }}.
24520 A backslash at the end of @var{text} can be used, as in C, to continue
24521 the command onto subsequent lines. For example,
24524 echo This is some text\n\
24525 which is continued\n\
24526 onto several lines.\n
24529 produces the same output as
24532 echo This is some text\n
24533 echo which is continued\n
24534 echo onto several lines.\n
24538 @item output @var{expression}
24539 Print the value of @var{expression} and nothing but that value: no
24540 newlines, no @samp{$@var{nn} = }. The value is not entered in the
24541 value history either. @xref{Expressions, ,Expressions}, for more information
24544 @item output/@var{fmt} @var{expression}
24545 Print the value of @var{expression} in format @var{fmt}. You can use
24546 the same formats as for @code{print}. @xref{Output Formats,,Output
24547 Formats}, for more information.
24550 @item printf @var{template}, @var{expressions}@dots{}
24551 Print the values of one or more @var{expressions} under the control of
24552 the string @var{template}. To print several values, make
24553 @var{expressions} be a comma-separated list of individual expressions,
24554 which may be either numbers or pointers. Their values are printed as
24555 specified by @var{template}, exactly as a C program would do by
24556 executing the code below:
24559 printf (@var{template}, @var{expressions}@dots{});
24562 As in @code{C} @code{printf}, ordinary characters in @var{template}
24563 are printed verbatim, while @dfn{conversion specification} introduced
24564 by the @samp{%} character cause subsequent @var{expressions} to be
24565 evaluated, their values converted and formatted according to type and
24566 style information encoded in the conversion specifications, and then
24569 For example, you can print two values in hex like this:
24572 printf "foo, bar-foo = 0x%x, 0x%x\n", foo, bar-foo
24575 @code{printf} supports all the standard @code{C} conversion
24576 specifications, including the flags and modifiers between the @samp{%}
24577 character and the conversion letter, with the following exceptions:
24581 The argument-ordering modifiers, such as @samp{2$}, are not supported.
24584 The modifier @samp{*} is not supported for specifying precision or
24588 The @samp{'} flag (for separation of digits into groups according to
24589 @code{LC_NUMERIC'}) is not supported.
24592 The type modifiers @samp{hh}, @samp{j}, @samp{t}, and @samp{z} are not
24596 The conversion letter @samp{n} (as in @samp{%n}) is not supported.
24599 The conversion letters @samp{a} and @samp{A} are not supported.
24603 Note that the @samp{ll} type modifier is supported only if the
24604 underlying @code{C} implementation used to build @value{GDBN} supports
24605 the @code{long long int} type, and the @samp{L} type modifier is
24606 supported only if @code{long double} type is available.
24608 As in @code{C}, @code{printf} supports simple backslash-escape
24609 sequences, such as @code{\n}, @samp{\t}, @samp{\\}, @samp{\"},
24610 @samp{\a}, and @samp{\f}, that consist of backslash followed by a
24611 single character. Octal and hexadecimal escape sequences are not
24614 Additionally, @code{printf} supports conversion specifications for DFP
24615 (@dfn{Decimal Floating Point}) types using the following length modifiers
24616 together with a floating point specifier.
24621 @samp{H} for printing @code{Decimal32} types.
24624 @samp{D} for printing @code{Decimal64} types.
24627 @samp{DD} for printing @code{Decimal128} types.
24630 If the underlying @code{C} implementation used to build @value{GDBN} has
24631 support for the three length modifiers for DFP types, other modifiers
24632 such as width and precision will also be available for @value{GDBN} to use.
24634 In case there is no such @code{C} support, no additional modifiers will be
24635 available and the value will be printed in the standard way.
24637 Here's an example of printing DFP types using the above conversion letters:
24639 printf "D32: %Hf - D64: %Df - D128: %DDf\n",1.2345df,1.2E10dd,1.2E1dl
24644 @item eval @var{template}, @var{expressions}@dots{}
24645 Convert the values of one or more @var{expressions} under the control of
24646 the string @var{template} to a command line, and call it.
24650 @node Auto-loading sequences
24651 @subsection Controlling auto-loading native @value{GDBN} scripts
24652 @cindex native script auto-loading
24654 When a new object file is read (for example, due to the @code{file}
24655 command, or because the inferior has loaded a shared library),
24656 @value{GDBN} will look for the command file @file{@var{objfile}-gdb.gdb}.
24657 @xref{Auto-loading extensions}.
24659 Auto-loading can be enabled or disabled,
24660 and the list of auto-loaded scripts can be printed.
24663 @anchor{set auto-load gdb-scripts}
24664 @kindex set auto-load gdb-scripts
24665 @item set auto-load gdb-scripts [on|off]
24666 Enable or disable the auto-loading of canned sequences of commands scripts.
24668 @anchor{show auto-load gdb-scripts}
24669 @kindex show auto-load gdb-scripts
24670 @item show auto-load gdb-scripts
24671 Show whether auto-loading of canned sequences of commands scripts is enabled or
24674 @anchor{info auto-load gdb-scripts}
24675 @kindex info auto-load gdb-scripts
24676 @cindex print list of auto-loaded canned sequences of commands scripts
24677 @item info auto-load gdb-scripts [@var{regexp}]
24678 Print the list of all canned sequences of commands scripts that @value{GDBN}
24682 If @var{regexp} is supplied only canned sequences of commands scripts with
24683 matching names are printed.
24685 @c Python docs live in a separate file.
24686 @include python.texi
24688 @c Guile docs live in a separate file.
24689 @include guile.texi
24691 @node Auto-loading extensions
24692 @section Auto-loading extensions
24693 @cindex auto-loading extensions
24695 @value{GDBN} provides two mechanisms for automatically loading extensions
24696 when a new object file is read (for example, due to the @code{file}
24697 command, or because the inferior has loaded a shared library):
24698 @file{@var{objfile}-gdb.@var{ext}} and the @code{.debug_gdb_scripts}
24699 section of modern file formats like ELF.
24702 * objfile-gdb.ext file: objfile-gdbdotext file. The @file{@var{objfile}-gdb.@var{ext}} file
24703 * .debug_gdb_scripts section: dotdebug_gdb_scripts section. The @code{.debug_gdb_scripts} section
24704 * Which flavor to choose?::
24707 The auto-loading feature is useful for supplying application-specific
24708 debugging commands and features.
24710 Auto-loading can be enabled or disabled,
24711 and the list of auto-loaded scripts can be printed.
24712 See the @samp{auto-loading} section of each extension language
24713 for more information.
24714 For @value{GDBN} command files see @ref{Auto-loading sequences}.
24715 For Python files see @ref{Python Auto-loading}.
24717 Note that loading of this script file also requires accordingly configured
24718 @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
24720 @node objfile-gdbdotext file
24721 @subsection The @file{@var{objfile}-gdb.@var{ext}} file
24722 @cindex @file{@var{objfile}-gdb.gdb}
24723 @cindex @file{@var{objfile}-gdb.py}
24724 @cindex @file{@var{objfile}-gdb.scm}
24726 When a new object file is read, @value{GDBN} looks for a file named
24727 @file{@var{objfile}-gdb.@var{ext}} (we call it @var{script-name} below),
24728 where @var{objfile} is the object file's name and
24729 where @var{ext} is the file extension for the extension language:
24732 @item @file{@var{objfile}-gdb.gdb}
24733 GDB's own command language
24734 @item @file{@var{objfile}-gdb.py}
24736 @item @file{@var{objfile}-gdb.scm}
24740 @var{script-name} is formed by ensuring that the file name of @var{objfile}
24741 is absolute, following all symlinks, and resolving @code{.} and @code{..}
24742 components, and appending the @file{-gdb.@var{ext}} suffix.
24743 If this file exists and is readable, @value{GDBN} will evaluate it as a
24744 script in the specified extension language.
24746 If this file does not exist, then @value{GDBN} will look for
24747 @var{script-name} file in all of the directories as specified below.
24749 Note that loading of these files requires an accordingly configured
24750 @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
24752 For object files using @file{.exe} suffix @value{GDBN} tries to load first the
24753 scripts normally according to its @file{.exe} filename. But if no scripts are
24754 found @value{GDBN} also tries script filenames matching the object file without
24755 its @file{.exe} suffix. This @file{.exe} stripping is case insensitive and it
24756 is attempted on any platform. This makes the script filenames compatible
24757 between Unix and MS-Windows hosts.
24760 @anchor{set auto-load scripts-directory}
24761 @kindex set auto-load scripts-directory
24762 @item set auto-load scripts-directory @r{[}@var{directories}@r{]}
24763 Control @value{GDBN} auto-loaded scripts location. Multiple directory entries
24764 may be delimited by the host platform path separator in use
24765 (@samp{:} on Unix, @samp{;} on MS-Windows and MS-DOS).
24767 Each entry here needs to be covered also by the security setting
24768 @code{set auto-load safe-path} (@pxref{set auto-load safe-path}).
24770 @anchor{with-auto-load-dir}
24771 This variable defaults to @file{$debugdir:$datadir/auto-load}. The default
24772 @code{set auto-load safe-path} value can be also overriden by @value{GDBN}
24773 configuration option @option{--with-auto-load-dir}.
24775 Any reference to @file{$debugdir} will get replaced by
24776 @var{debug-file-directory} value (@pxref{Separate Debug Files}) and any
24777 reference to @file{$datadir} will get replaced by @var{data-directory} which is
24778 determined at @value{GDBN} startup (@pxref{Data Files}). @file{$debugdir} and
24779 @file{$datadir} must be placed as a directory component --- either alone or
24780 delimited by @file{/} or @file{\} directory separators, depending on the host
24783 The list of directories uses path separator (@samp{:} on GNU and Unix
24784 systems, @samp{;} on MS-Windows and MS-DOS) to separate directories, similarly
24785 to the @env{PATH} environment variable.
24787 @anchor{show auto-load scripts-directory}
24788 @kindex show auto-load scripts-directory
24789 @item show auto-load scripts-directory
24790 Show @value{GDBN} auto-loaded scripts location.
24792 @anchor{add-auto-load-scripts-directory}
24793 @kindex add-auto-load-scripts-directory
24794 @item add-auto-load-scripts-directory @r{[}@var{directories}@dots{}@r{]}
24795 Add an entry (or list of entries) to the list of auto-loaded scripts locations.
24796 Multiple entries may be delimited by the host platform path separator in use.
24799 @value{GDBN} does not track which files it has already auto-loaded this way.
24800 @value{GDBN} will load the associated script every time the corresponding
24801 @var{objfile} is opened.
24802 So your @file{-gdb.@var{ext}} file should be careful to avoid errors if it
24803 is evaluated more than once.
24805 @node dotdebug_gdb_scripts section
24806 @subsection The @code{.debug_gdb_scripts} section
24807 @cindex @code{.debug_gdb_scripts} section
24809 For systems using file formats like ELF and COFF,
24810 when @value{GDBN} loads a new object file
24811 it will look for a special section named @code{.debug_gdb_scripts}.
24812 If this section exists, its contents is a list of null-terminated entries
24813 specifying scripts to load. Each entry begins with a non-null prefix byte that
24814 specifies the kind of entry, typically the extension language and whether the
24815 script is in a file or inlined in @code{.debug_gdb_scripts}.
24817 The following entries are supported:
24820 @item SECTION_SCRIPT_ID_PYTHON_FILE = 1
24821 @item SECTION_SCRIPT_ID_SCHEME_FILE = 3
24822 @item SECTION_SCRIPT_ID_PYTHON_TEXT = 4
24823 @item SECTION_SCRIPT_ID_SCHEME_TEXT = 6
24826 @subsubsection Script File Entries
24828 If the entry specifies a file, @value{GDBN} will look for the file first
24829 in the current directory and then along the source search path
24830 (@pxref{Source Path, ,Specifying Source Directories}),
24831 except that @file{$cdir} is not searched, since the compilation
24832 directory is not relevant to scripts.
24834 File entries can be placed in section @code{.debug_gdb_scripts} with,
24835 for example, this GCC macro for Python scripts.
24838 /* Note: The "MS" section flags are to remove duplicates. */
24839 #define DEFINE_GDB_PY_SCRIPT(script_name) \
24841 .pushsection \".debug_gdb_scripts\", \"MS\",@@progbits,1\n\
24842 .byte 1 /* Python */\n\
24843 .asciz \"" script_name "\"\n\
24849 For Guile scripts, replace @code{.byte 1} with @code{.byte 3}.
24850 Then one can reference the macro in a header or source file like this:
24853 DEFINE_GDB_PY_SCRIPT ("my-app-scripts.py")
24856 The script name may include directories if desired.
24858 Note that loading of this script file also requires accordingly configured
24859 @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
24861 If the macro invocation is put in a header, any application or library
24862 using this header will get a reference to the specified script,
24863 and with the use of @code{"MS"} attributes on the section, the linker
24864 will remove duplicates.
24866 @subsubsection Script Text Entries
24868 Script text entries allow to put the executable script in the entry
24869 itself instead of loading it from a file.
24870 The first line of the entry, everything after the prefix byte and up to
24871 the first newline (@code{0xa}) character, is the script name, and must not
24872 contain any kind of space character, e.g., spaces or tabs.
24873 The rest of the entry, up to the trailing null byte, is the script to
24874 execute in the specified language. The name needs to be unique among
24875 all script names, as @value{GDBN} executes each script only once based
24878 Here is an example from file @file{py-section-script.c} in the @value{GDBN}
24882 #include "symcat.h"
24883 #include "gdb/section-scripts.h"
24885 ".pushsection \".debug_gdb_scripts\", \"MS\",@@progbits,1\n"
24886 ".byte " XSTRING (SECTION_SCRIPT_ID_PYTHON_TEXT) "\n"
24887 ".ascii \"gdb.inlined-script\\n\"\n"
24888 ".ascii \"class test_cmd (gdb.Command):\\n\"\n"
24889 ".ascii \" def __init__ (self):\\n\"\n"
24890 ".ascii \" super (test_cmd, self).__init__ ("
24891 "\\\"test-cmd\\\", gdb.COMMAND_OBSCURE)\\n\"\n"
24892 ".ascii \" def invoke (self, arg, from_tty):\\n\"\n"
24893 ".ascii \" print (\\\"test-cmd output, arg = %s\\\" % arg)\\n\"\n"
24894 ".ascii \"test_cmd ()\\n\"\n"
24900 Loading of inlined scripts requires a properly configured
24901 @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
24902 The path to specify in @code{auto-load safe-path} is the path of the file
24903 containing the @code{.debug_gdb_scripts} section.
24905 @node Which flavor to choose?
24906 @subsection Which flavor to choose?
24908 Given the multiple ways of auto-loading extensions, it might not always
24909 be clear which one to choose. This section provides some guidance.
24912 Benefits of the @file{-gdb.@var{ext}} way:
24916 Can be used with file formats that don't support multiple sections.
24919 Ease of finding scripts for public libraries.
24921 Scripts specified in the @code{.debug_gdb_scripts} section are searched for
24922 in the source search path.
24923 For publicly installed libraries, e.g., @file{libstdc++}, there typically
24924 isn't a source directory in which to find the script.
24927 Doesn't require source code additions.
24931 Benefits of the @code{.debug_gdb_scripts} way:
24935 Works with static linking.
24937 Scripts for libraries done the @file{-gdb.@var{ext}} way require an objfile to
24938 trigger their loading. When an application is statically linked the only
24939 objfile available is the executable, and it is cumbersome to attach all the
24940 scripts from all the input libraries to the executable's
24941 @file{-gdb.@var{ext}} script.
24944 Works with classes that are entirely inlined.
24946 Some classes can be entirely inlined, and thus there may not be an associated
24947 shared library to attach a @file{-gdb.@var{ext}} script to.
24950 Scripts needn't be copied out of the source tree.
24952 In some circumstances, apps can be built out of large collections of internal
24953 libraries, and the build infrastructure necessary to install the
24954 @file{-gdb.@var{ext}} scripts in a place where @value{GDBN} can find them is
24955 cumbersome. It may be easier to specify the scripts in the
24956 @code{.debug_gdb_scripts} section as relative paths, and add a path to the
24957 top of the source tree to the source search path.
24960 @node Multiple Extension Languages
24961 @section Multiple Extension Languages
24963 The Guile and Python extension languages do not share any state,
24964 and generally do not interfere with each other.
24965 There are some things to be aware of, however.
24967 @subsection Python comes first
24969 Python was @value{GDBN}'s first extension language, and to avoid breaking
24970 existing behaviour Python comes first. This is generally solved by the
24971 ``first one wins'' principle. @value{GDBN} maintains a list of enabled
24972 extension languages, and when it makes a call to an extension language,
24973 (say to pretty-print a value), it tries each in turn until an extension
24974 language indicates it has performed the request (e.g., has returned the
24975 pretty-printed form of a value).
24976 This extends to errors while performing such requests: If an error happens
24977 while, for example, trying to pretty-print an object then the error is
24978 reported and any following extension languages are not tried.
24981 @section Creating new spellings of existing commands
24982 @cindex aliases for commands
24984 It is often useful to define alternate spellings of existing commands.
24985 For example, if a new @value{GDBN} command defined in Python has
24986 a long name to type, it is handy to have an abbreviated version of it
24987 that involves less typing.
24989 @value{GDBN} itself uses aliases. For example @samp{s} is an alias
24990 of the @samp{step} command even though it is otherwise an ambiguous
24991 abbreviation of other commands like @samp{set} and @samp{show}.
24993 Aliases are also used to provide shortened or more common versions
24994 of multi-word commands. For example, @value{GDBN} provides the
24995 @samp{tty} alias of the @samp{set inferior-tty} command.
24997 You can define a new alias with the @samp{alias} command.
25002 @item alias [-a] [--] @var{ALIAS} = @var{COMMAND}
25006 @var{ALIAS} specifies the name of the new alias.
25007 Each word of @var{ALIAS} must consist of letters, numbers, dashes and
25010 @var{COMMAND} specifies the name of an existing command
25011 that is being aliased.
25013 The @samp{-a} option specifies that the new alias is an abbreviation
25014 of the command. Abbreviations are not shown in command
25015 lists displayed by the @samp{help} command.
25017 The @samp{--} option specifies the end of options,
25018 and is useful when @var{ALIAS} begins with a dash.
25020 Here is a simple example showing how to make an abbreviation
25021 of a command so that there is less to type.
25022 Suppose you were tired of typing @samp{disas}, the current
25023 shortest unambiguous abbreviation of the @samp{disassemble} command
25024 and you wanted an even shorter version named @samp{di}.
25025 The following will accomplish this.
25028 (gdb) alias -a di = disas
25031 Note that aliases are different from user-defined commands.
25032 With a user-defined command, you also need to write documentation
25033 for it with the @samp{document} command.
25034 An alias automatically picks up the documentation of the existing command.
25036 Here is an example where we make @samp{elms} an abbreviation of
25037 @samp{elements} in the @samp{set print elements} command.
25038 This is to show that you can make an abbreviation of any part
25042 (gdb) alias -a set print elms = set print elements
25043 (gdb) alias -a show print elms = show print elements
25044 (gdb) set p elms 20
25046 Limit on string chars or array elements to print is 200.
25049 Note that if you are defining an alias of a @samp{set} command,
25050 and you want to have an alias for the corresponding @samp{show}
25051 command, then you need to define the latter separately.
25053 Unambiguously abbreviated commands are allowed in @var{COMMAND} and
25054 @var{ALIAS}, just as they are normally.
25057 (gdb) alias -a set pr elms = set p ele
25060 Finally, here is an example showing the creation of a one word
25061 alias for a more complex command.
25062 This creates alias @samp{spe} of the command @samp{set print elements}.
25065 (gdb) alias spe = set print elements
25070 @chapter Command Interpreters
25071 @cindex command interpreters
25073 @value{GDBN} supports multiple command interpreters, and some command
25074 infrastructure to allow users or user interface writers to switch
25075 between interpreters or run commands in other interpreters.
25077 @value{GDBN} currently supports two command interpreters, the console
25078 interpreter (sometimes called the command-line interpreter or @sc{cli})
25079 and the machine interface interpreter (or @sc{gdb/mi}). This manual
25080 describes both of these interfaces in great detail.
25082 By default, @value{GDBN} will start with the console interpreter.
25083 However, the user may choose to start @value{GDBN} with another
25084 interpreter by specifying the @option{-i} or @option{--interpreter}
25085 startup options. Defined interpreters include:
25089 @cindex console interpreter
25090 The traditional console or command-line interpreter. This is the most often
25091 used interpreter with @value{GDBN}. With no interpreter specified at runtime,
25092 @value{GDBN} will use this interpreter.
25095 @cindex mi interpreter
25096 The newest @sc{gdb/mi} interface (currently @code{mi2}). Used primarily
25097 by programs wishing to use @value{GDBN} as a backend for a debugger GUI
25098 or an IDE. For more information, see @ref{GDB/MI, ,The @sc{gdb/mi}
25102 @cindex mi2 interpreter
25103 The current @sc{gdb/mi} interface.
25106 @cindex mi1 interpreter
25107 The @sc{gdb/mi} interface included in @value{GDBN} 5.1, 5.2, and 5.3.
25111 @cindex invoke another interpreter
25113 @kindex interpreter-exec
25114 You may execute commands in any interpreter from the current
25115 interpreter using the appropriate command. If you are running the
25116 console interpreter, simply use the @code{interpreter-exec} command:
25119 interpreter-exec mi "-data-list-register-names"
25122 @sc{gdb/mi} has a similar command, although it is only available in versions of
25123 @value{GDBN} which support @sc{gdb/mi} version 2 (or greater).
25125 Note that @code{interpreter-exec} only changes the interpreter for the
25126 duration of the specified command. It does not change the interpreter
25129 @cindex start a new independent interpreter
25131 Although you may only choose a single interpreter at startup, it is
25132 possible to run an independent interpreter on a specified input/output
25133 device (usually a tty).
25135 For example, consider a debugger GUI or IDE that wants to provide a
25136 @value{GDBN} console view. It may do so by embedding a terminal
25137 emulator widget in its GUI, starting @value{GDBN} in the traditional
25138 command-line mode with stdin/stdout/stderr redirected to that
25139 terminal, and then creating an MI interpreter running on a specified
25140 input/output device. The console interpreter created by @value{GDBN}
25141 at startup handles commands the user types in the terminal widget,
25142 while the GUI controls and synchronizes state with @value{GDBN} using
25143 the separate MI interpreter.
25145 To start a new secondary @dfn{user interface} running MI, use the
25146 @code{new-ui} command:
25149 @cindex new user interface
25151 new-ui @var{interpreter} @var{tty}
25154 The @var{interpreter} parameter specifies the interpreter to run.
25155 This accepts the same values as the @code{interpreter-exec} command.
25156 For example, @samp{console}, @samp{mi}, @samp{mi2}, etc. The
25157 @var{tty} parameter specifies the name of the bidirectional file the
25158 interpreter uses for input/output, usually the name of a
25159 pseudoterminal slave on Unix systems. For example:
25162 (@value{GDBP}) new-ui mi /dev/pts/9
25166 runs an MI interpreter on @file{/dev/pts/9}.
25169 @chapter @value{GDBN} Text User Interface
25171 @cindex Text User Interface
25174 * TUI Overview:: TUI overview
25175 * TUI Keys:: TUI key bindings
25176 * TUI Single Key Mode:: TUI single key mode
25177 * TUI Commands:: TUI-specific commands
25178 * TUI Configuration:: TUI configuration variables
25181 The @value{GDBN} Text User Interface (TUI) is a terminal
25182 interface which uses the @code{curses} library to show the source
25183 file, the assembly output, the program registers and @value{GDBN}
25184 commands in separate text windows. The TUI mode is supported only
25185 on platforms where a suitable version of the @code{curses} library
25188 The TUI mode is enabled by default when you invoke @value{GDBN} as
25189 @samp{@value{GDBP} -tui}.
25190 You can also switch in and out of TUI mode while @value{GDBN} runs by
25191 using various TUI commands and key bindings, such as @command{tui
25192 enable} or @kbd{C-x C-a}. @xref{TUI Commands, ,TUI Commands}, and
25193 @ref{TUI Keys, ,TUI Key Bindings}.
25196 @section TUI Overview
25198 In TUI mode, @value{GDBN} can display several text windows:
25202 This window is the @value{GDBN} command window with the @value{GDBN}
25203 prompt and the @value{GDBN} output. The @value{GDBN} input is still
25204 managed using readline.
25207 The source window shows the source file of the program. The current
25208 line and active breakpoints are displayed in this window.
25211 The assembly window shows the disassembly output of the program.
25214 This window shows the processor registers. Registers are highlighted
25215 when their values change.
25218 The source and assembly windows show the current program position
25219 by highlighting the current line and marking it with a @samp{>} marker.
25220 Breakpoints are indicated with two markers. The first marker
25221 indicates the breakpoint type:
25225 Breakpoint which was hit at least once.
25228 Breakpoint which was never hit.
25231 Hardware breakpoint which was hit at least once.
25234 Hardware breakpoint which was never hit.
25237 The second marker indicates whether the breakpoint is enabled or not:
25241 Breakpoint is enabled.
25244 Breakpoint is disabled.
25247 The source, assembly and register windows are updated when the current
25248 thread changes, when the frame changes, or when the program counter
25251 These windows are not all visible at the same time. The command
25252 window is always visible. The others can be arranged in several
25263 source and assembly,
25266 source and registers, or
25269 assembly and registers.
25272 A status line above the command window shows the following information:
25276 Indicates the current @value{GDBN} target.
25277 (@pxref{Targets, ,Specifying a Debugging Target}).
25280 Gives the current process or thread number.
25281 When no process is being debugged, this field is set to @code{No process}.
25284 Gives the current function name for the selected frame.
25285 The name is demangled if demangling is turned on (@pxref{Print Settings}).
25286 When there is no symbol corresponding to the current program counter,
25287 the string @code{??} is displayed.
25290 Indicates the current line number for the selected frame.
25291 When the current line number is not known, the string @code{??} is displayed.
25294 Indicates the current program counter address.
25298 @section TUI Key Bindings
25299 @cindex TUI key bindings
25301 The TUI installs several key bindings in the readline keymaps
25302 @ifset SYSTEM_READLINE
25303 (@pxref{Command Line Editing, , , rluserman, GNU Readline Library}).
25305 @ifclear SYSTEM_READLINE
25306 (@pxref{Command Line Editing}).
25308 The following key bindings are installed for both TUI mode and the
25309 @value{GDBN} standard mode.
25318 Enter or leave the TUI mode. When leaving the TUI mode,
25319 the curses window management stops and @value{GDBN} operates using
25320 its standard mode, writing on the terminal directly. When reentering
25321 the TUI mode, control is given back to the curses windows.
25322 The screen is then refreshed.
25326 Use a TUI layout with only one window. The layout will
25327 either be @samp{source} or @samp{assembly}. When the TUI mode
25328 is not active, it will switch to the TUI mode.
25330 Think of this key binding as the Emacs @kbd{C-x 1} binding.
25334 Use a TUI layout with at least two windows. When the current
25335 layout already has two windows, the next layout with two windows is used.
25336 When a new layout is chosen, one window will always be common to the
25337 previous layout and the new one.
25339 Think of it as the Emacs @kbd{C-x 2} binding.
25343 Change the active window. The TUI associates several key bindings
25344 (like scrolling and arrow keys) with the active window. This command
25345 gives the focus to the next TUI window.
25347 Think of it as the Emacs @kbd{C-x o} binding.
25351 Switch in and out of the TUI SingleKey mode that binds single
25352 keys to @value{GDBN} commands (@pxref{TUI Single Key Mode}).
25355 The following key bindings only work in the TUI mode:
25360 Scroll the active window one page up.
25364 Scroll the active window one page down.
25368 Scroll the active window one line up.
25372 Scroll the active window one line down.
25376 Scroll the active window one column left.
25380 Scroll the active window one column right.
25384 Refresh the screen.
25387 Because the arrow keys scroll the active window in the TUI mode, they
25388 are not available for their normal use by readline unless the command
25389 window has the focus. When another window is active, you must use
25390 other readline key bindings such as @kbd{C-p}, @kbd{C-n}, @kbd{C-b}
25391 and @kbd{C-f} to control the command window.
25393 @node TUI Single Key Mode
25394 @section TUI Single Key Mode
25395 @cindex TUI single key mode
25397 The TUI also provides a @dfn{SingleKey} mode, which binds several
25398 frequently used @value{GDBN} commands to single keys. Type @kbd{C-x s} to
25399 switch into this mode, where the following key bindings are used:
25402 @kindex c @r{(SingleKey TUI key)}
25406 @kindex d @r{(SingleKey TUI key)}
25410 @kindex f @r{(SingleKey TUI key)}
25414 @kindex n @r{(SingleKey TUI key)}
25418 @kindex q @r{(SingleKey TUI key)}
25420 exit the SingleKey mode.
25422 @kindex r @r{(SingleKey TUI key)}
25426 @kindex s @r{(SingleKey TUI key)}
25430 @kindex u @r{(SingleKey TUI key)}
25434 @kindex v @r{(SingleKey TUI key)}
25438 @kindex w @r{(SingleKey TUI key)}
25443 Other keys temporarily switch to the @value{GDBN} command prompt.
25444 The key that was pressed is inserted in the editing buffer so that
25445 it is possible to type most @value{GDBN} commands without interaction
25446 with the TUI SingleKey mode. Once the command is entered the TUI
25447 SingleKey mode is restored. The only way to permanently leave
25448 this mode is by typing @kbd{q} or @kbd{C-x s}.
25452 @section TUI-specific Commands
25453 @cindex TUI commands
25455 The TUI has specific commands to control the text windows.
25456 These commands are always available, even when @value{GDBN} is not in
25457 the TUI mode. When @value{GDBN} is in the standard mode, most
25458 of these commands will automatically switch to the TUI mode.
25460 Note that if @value{GDBN}'s @code{stdout} is not connected to a
25461 terminal, or @value{GDBN} has been started with the machine interface
25462 interpreter (@pxref{GDB/MI, ,The @sc{gdb/mi} Interface}), most of
25463 these commands will fail with an error, because it would not be
25464 possible or desirable to enable curses window management.
25469 Activate TUI mode. The last active TUI window layout will be used if
25470 TUI mode has prevsiouly been used in the current debugging session,
25471 otherwise a default layout is used.
25474 @kindex tui disable
25475 Disable TUI mode, returning to the console interpreter.
25479 List and give the size of all displayed windows.
25481 @item layout @var{name}
25483 Changes which TUI windows are displayed. In each layout the command
25484 window is always displayed, the @var{name} parameter controls which
25485 additional windows are displayed, and can be any of the following:
25489 Display the next layout.
25492 Display the previous layout.
25495 Display the source and command windows.
25498 Display the assembly and command windows.
25501 Display the source, assembly, and command windows.
25504 When in @code{src} layout display the register, source, and command
25505 windows. When in @code{asm} or @code{split} layout display the
25506 register, assembler, and command windows.
25509 @item focus @var{name}
25511 Changes which TUI window is currently active for scrolling. The
25512 @var{name} parameter can be any of the following:
25516 Make the next window active for scrolling.
25519 Make the previous window active for scrolling.
25522 Make the source window active for scrolling.
25525 Make the assembly window active for scrolling.
25528 Make the register window active for scrolling.
25531 Make the command window active for scrolling.
25536 Refresh the screen. This is similar to typing @kbd{C-L}.
25538 @item tui reg @var{group}
25540 Changes the register group displayed in the tui register window to
25541 @var{group}. If the register window is not currently displayed this
25542 command will cause the register window to be displayed. The list of
25543 register groups, as well as their order is target specific. The
25544 following groups are available on most targets:
25547 Repeatedly selecting this group will cause the display to cycle
25548 through all of the available register groups.
25551 Repeatedly selecting this group will cause the display to cycle
25552 through all of the available register groups in the reverse order to
25556 Display the general registers.
25558 Display the floating point registers.
25560 Display the system registers.
25562 Display the vector registers.
25564 Display all registers.
25569 Update the source window and the current execution point.
25571 @item winheight @var{name} +@var{count}
25572 @itemx winheight @var{name} -@var{count}
25574 Change the height of the window @var{name} by @var{count}
25575 lines. Positive counts increase the height, while negative counts
25576 decrease it. The @var{name} parameter can be one of @code{src} (the
25577 source window), @code{cmd} (the command window), @code{asm} (the
25578 disassembly window), or @code{regs} (the register display window).
25580 @item tabset @var{nchars}
25582 Set the width of tab stops to be @var{nchars} characters. This
25583 setting affects the display of TAB characters in the source and
25587 @node TUI Configuration
25588 @section TUI Configuration Variables
25589 @cindex TUI configuration variables
25591 Several configuration variables control the appearance of TUI windows.
25594 @item set tui border-kind @var{kind}
25595 @kindex set tui border-kind
25596 Select the border appearance for the source, assembly and register windows.
25597 The possible values are the following:
25600 Use a space character to draw the border.
25603 Use @sc{ascii} characters @samp{+}, @samp{-} and @samp{|} to draw the border.
25606 Use the Alternate Character Set to draw the border. The border is
25607 drawn using character line graphics if the terminal supports them.
25610 @item set tui border-mode @var{mode}
25611 @kindex set tui border-mode
25612 @itemx set tui active-border-mode @var{mode}
25613 @kindex set tui active-border-mode
25614 Select the display attributes for the borders of the inactive windows
25615 or the active window. The @var{mode} can be one of the following:
25618 Use normal attributes to display the border.
25624 Use reverse video mode.
25627 Use half bright mode.
25629 @item half-standout
25630 Use half bright and standout mode.
25633 Use extra bright or bold mode.
25635 @item bold-standout
25636 Use extra bright or bold and standout mode.
25641 @chapter Using @value{GDBN} under @sc{gnu} Emacs
25644 @cindex @sc{gnu} Emacs
25645 A special interface allows you to use @sc{gnu} Emacs to view (and
25646 edit) the source files for the program you are debugging with
25649 To use this interface, use the command @kbd{M-x gdb} in Emacs. Give the
25650 executable file you want to debug as an argument. This command starts
25651 @value{GDBN} as a subprocess of Emacs, with input and output through a newly
25652 created Emacs buffer.
25653 @c (Do not use the @code{-tui} option to run @value{GDBN} from Emacs.)
25655 Running @value{GDBN} under Emacs can be just like running @value{GDBN} normally except for two
25660 All ``terminal'' input and output goes through an Emacs buffer, called
25663 This applies both to @value{GDBN} commands and their output, and to the input
25664 and output done by the program you are debugging.
25666 This is useful because it means that you can copy the text of previous
25667 commands and input them again; you can even use parts of the output
25670 All the facilities of Emacs' Shell mode are available for interacting
25671 with your program. In particular, you can send signals the usual
25672 way---for example, @kbd{C-c C-c} for an interrupt, @kbd{C-c C-z} for a
25676 @value{GDBN} displays source code through Emacs.
25678 Each time @value{GDBN} displays a stack frame, Emacs automatically finds the
25679 source file for that frame and puts an arrow (@samp{=>}) at the
25680 left margin of the current line. Emacs uses a separate buffer for
25681 source display, and splits the screen to show both your @value{GDBN} session
25684 Explicit @value{GDBN} @code{list} or search commands still produce output as
25685 usual, but you probably have no reason to use them from Emacs.
25688 We call this @dfn{text command mode}. Emacs 22.1, and later, also uses
25689 a graphical mode, enabled by default, which provides further buffers
25690 that can control the execution and describe the state of your program.
25691 @xref{GDB Graphical Interface,,, Emacs, The @sc{gnu} Emacs Manual}.
25693 If you specify an absolute file name when prompted for the @kbd{M-x
25694 gdb} argument, then Emacs sets your current working directory to where
25695 your program resides. If you only specify the file name, then Emacs
25696 sets your current working directory to the directory associated
25697 with the previous buffer. In this case, @value{GDBN} may find your
25698 program by searching your environment's @code{PATH} variable, but on
25699 some operating systems it might not find the source. So, although the
25700 @value{GDBN} input and output session proceeds normally, the auxiliary
25701 buffer does not display the current source and line of execution.
25703 The initial working directory of @value{GDBN} is printed on the top
25704 line of the GUD buffer and this serves as a default for the commands
25705 that specify files for @value{GDBN} to operate on. @xref{Files,
25706 ,Commands to Specify Files}.
25708 By default, @kbd{M-x gdb} calls the program called @file{gdb}. If you
25709 need to call @value{GDBN} by a different name (for example, if you
25710 keep several configurations around, with different names) you can
25711 customize the Emacs variable @code{gud-gdb-command-name} to run the
25714 In the GUD buffer, you can use these special Emacs commands in
25715 addition to the standard Shell mode commands:
25719 Describe the features of Emacs' GUD Mode.
25722 Execute to another source line, like the @value{GDBN} @code{step} command; also
25723 update the display window to show the current file and location.
25726 Execute to next source line in this function, skipping all function
25727 calls, like the @value{GDBN} @code{next} command. Then update the display window
25728 to show the current file and location.
25731 Execute one instruction, like the @value{GDBN} @code{stepi} command; update
25732 display window accordingly.
25735 Execute until exit from the selected stack frame, like the @value{GDBN}
25736 @code{finish} command.
25739 Continue execution of your program, like the @value{GDBN} @code{continue}
25743 Go up the number of frames indicated by the numeric argument
25744 (@pxref{Arguments, , Numeric Arguments, Emacs, The @sc{gnu} Emacs Manual}),
25745 like the @value{GDBN} @code{up} command.
25748 Go down the number of frames indicated by the numeric argument, like the
25749 @value{GDBN} @code{down} command.
25752 In any source file, the Emacs command @kbd{C-x @key{SPC}} (@code{gud-break})
25753 tells @value{GDBN} to set a breakpoint on the source line point is on.
25755 In text command mode, if you type @kbd{M-x speedbar}, Emacs displays a
25756 separate frame which shows a backtrace when the GUD buffer is current.
25757 Move point to any frame in the stack and type @key{RET} to make it
25758 become the current frame and display the associated source in the
25759 source buffer. Alternatively, click @kbd{Mouse-2} to make the
25760 selected frame become the current one. In graphical mode, the
25761 speedbar displays watch expressions.
25763 If you accidentally delete the source-display buffer, an easy way to get
25764 it back is to type the command @code{f} in the @value{GDBN} buffer, to
25765 request a frame display; when you run under Emacs, this recreates
25766 the source buffer if necessary to show you the context of the current
25769 The source files displayed in Emacs are in ordinary Emacs buffers
25770 which are visiting the source files in the usual way. You can edit
25771 the files with these buffers if you wish; but keep in mind that @value{GDBN}
25772 communicates with Emacs in terms of line numbers. If you add or
25773 delete lines from the text, the line numbers that @value{GDBN} knows cease
25774 to correspond properly with the code.
25776 A more detailed description of Emacs' interaction with @value{GDBN} is
25777 given in the Emacs manual (@pxref{Debuggers,,, Emacs, The @sc{gnu}
25781 @chapter The @sc{gdb/mi} Interface
25783 @unnumberedsec Function and Purpose
25785 @cindex @sc{gdb/mi}, its purpose
25786 @sc{gdb/mi} is a line based machine oriented text interface to
25787 @value{GDBN} and is activated by specifying using the
25788 @option{--interpreter} command line option (@pxref{Mode Options}). It
25789 is specifically intended to support the development of systems which
25790 use the debugger as just one small component of a larger system.
25792 This chapter is a specification of the @sc{gdb/mi} interface. It is written
25793 in the form of a reference manual.
25795 Note that @sc{gdb/mi} is still under construction, so some of the
25796 features described below are incomplete and subject to change
25797 (@pxref{GDB/MI Development and Front Ends, , @sc{gdb/mi} Development and Front Ends}).
25799 @unnumberedsec Notation and Terminology
25801 @cindex notational conventions, for @sc{gdb/mi}
25802 This chapter uses the following notation:
25806 @code{|} separates two alternatives.
25809 @code{[ @var{something} ]} indicates that @var{something} is optional:
25810 it may or may not be given.
25813 @code{( @var{group} )*} means that @var{group} inside the parentheses
25814 may repeat zero or more times.
25817 @code{( @var{group} )+} means that @var{group} inside the parentheses
25818 may repeat one or more times.
25821 @code{"@var{string}"} means a literal @var{string}.
25825 @heading Dependencies
25829 * GDB/MI General Design::
25830 * GDB/MI Command Syntax::
25831 * GDB/MI Compatibility with CLI::
25832 * GDB/MI Development and Front Ends::
25833 * GDB/MI Output Records::
25834 * GDB/MI Simple Examples::
25835 * GDB/MI Command Description Format::
25836 * GDB/MI Breakpoint Commands::
25837 * GDB/MI Catchpoint Commands::
25838 * GDB/MI Program Context::
25839 * GDB/MI Thread Commands::
25840 * GDB/MI Ada Tasking Commands::
25841 * GDB/MI Program Execution::
25842 * GDB/MI Stack Manipulation::
25843 * GDB/MI Variable Objects::
25844 * GDB/MI Data Manipulation::
25845 * GDB/MI Tracepoint Commands::
25846 * GDB/MI Symbol Query::
25847 * GDB/MI File Commands::
25849 * GDB/MI Kod Commands::
25850 * GDB/MI Memory Overlay Commands::
25851 * GDB/MI Signal Handling Commands::
25853 * GDB/MI Target Manipulation::
25854 * GDB/MI File Transfer Commands::
25855 * GDB/MI Ada Exceptions Commands::
25856 * GDB/MI Support Commands::
25857 * GDB/MI Miscellaneous Commands::
25860 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
25861 @node GDB/MI General Design
25862 @section @sc{gdb/mi} General Design
25863 @cindex GDB/MI General Design
25865 Interaction of a @sc{GDB/MI} frontend with @value{GDBN} involves three
25866 parts---commands sent to @value{GDBN}, responses to those commands
25867 and notifications. Each command results in exactly one response,
25868 indicating either successful completion of the command, or an error.
25869 For the commands that do not resume the target, the response contains the
25870 requested information. For the commands that resume the target, the
25871 response only indicates whether the target was successfully resumed.
25872 Notifications is the mechanism for reporting changes in the state of the
25873 target, or in @value{GDBN} state, that cannot conveniently be associated with
25874 a command and reported as part of that command response.
25876 The important examples of notifications are:
25880 Exec notifications. These are used to report changes in
25881 target state---when a target is resumed, or stopped. It would not
25882 be feasible to include this information in response of resuming
25883 commands, because one resume commands can result in multiple events in
25884 different threads. Also, quite some time may pass before any event
25885 happens in the target, while a frontend needs to know whether the resuming
25886 command itself was successfully executed.
25889 Console output, and status notifications. Console output
25890 notifications are used to report output of CLI commands, as well as
25891 diagnostics for other commands. Status notifications are used to
25892 report the progress of a long-running operation. Naturally, including
25893 this information in command response would mean no output is produced
25894 until the command is finished, which is undesirable.
25897 General notifications. Commands may have various side effects on
25898 the @value{GDBN} or target state beyond their official purpose. For example,
25899 a command may change the selected thread. Although such changes can
25900 be included in command response, using notification allows for more
25901 orthogonal frontend design.
25905 There's no guarantee that whenever an MI command reports an error,
25906 @value{GDBN} or the target are in any specific state, and especially,
25907 the state is not reverted to the state before the MI command was
25908 processed. Therefore, whenever an MI command results in an error,
25909 we recommend that the frontend refreshes all the information shown in
25910 the user interface.
25914 * Context management::
25915 * Asynchronous and non-stop modes::
25919 @node Context management
25920 @subsection Context management
25922 @subsubsection Threads and Frames
25924 In most cases when @value{GDBN} accesses the target, this access is
25925 done in context of a specific thread and frame (@pxref{Frames}).
25926 Often, even when accessing global data, the target requires that a thread
25927 be specified. The CLI interface maintains the selected thread and frame,
25928 and supplies them to target on each command. This is convenient,
25929 because a command line user would not want to specify that information
25930 explicitly on each command, and because user interacts with
25931 @value{GDBN} via a single terminal, so no confusion is possible as
25932 to what thread and frame are the current ones.
25934 In the case of MI, the concept of selected thread and frame is less
25935 useful. First, a frontend can easily remember this information
25936 itself. Second, a graphical frontend can have more than one window,
25937 each one used for debugging a different thread, and the frontend might
25938 want to access additional threads for internal purposes. This
25939 increases the risk that by relying on implicitly selected thread, the
25940 frontend may be operating on a wrong one. Therefore, each MI command
25941 should explicitly specify which thread and frame to operate on. To
25942 make it possible, each MI command accepts the @samp{--thread} and
25943 @samp{--frame} options, the value to each is @value{GDBN} global
25944 identifier for thread and frame to operate on.
25946 Usually, each top-level window in a frontend allows the user to select
25947 a thread and a frame, and remembers the user selection for further
25948 operations. However, in some cases @value{GDBN} may suggest that the
25949 current thread or frame be changed. For example, when stopping on a
25950 breakpoint it is reasonable to switch to the thread where breakpoint is
25951 hit. For another example, if the user issues the CLI @samp{thread} or
25952 @samp{frame} commands via the frontend, it is desirable to change the
25953 frontend's selection to the one specified by user. @value{GDBN}
25954 communicates the suggestion to change current thread and frame using the
25955 @samp{=thread-selected} notification.
25957 Note that historically, MI shares the selected thread with CLI, so
25958 frontends used the @code{-thread-select} to execute commands in the
25959 right context. However, getting this to work right is cumbersome. The
25960 simplest way is for frontend to emit @code{-thread-select} command
25961 before every command. This doubles the number of commands that need
25962 to be sent. The alternative approach is to suppress @code{-thread-select}
25963 if the selected thread in @value{GDBN} is supposed to be identical to the
25964 thread the frontend wants to operate on. However, getting this
25965 optimization right can be tricky. In particular, if the frontend
25966 sends several commands to @value{GDBN}, and one of the commands changes the
25967 selected thread, then the behaviour of subsequent commands will
25968 change. So, a frontend should either wait for response from such
25969 problematic commands, or explicitly add @code{-thread-select} for
25970 all subsequent commands. No frontend is known to do this exactly
25971 right, so it is suggested to just always pass the @samp{--thread} and
25972 @samp{--frame} options.
25974 @subsubsection Language
25976 The execution of several commands depends on which language is selected.
25977 By default, the current language (@pxref{show language}) is used.
25978 But for commands known to be language-sensitive, it is recommended
25979 to use the @samp{--language} option. This option takes one argument,
25980 which is the name of the language to use while executing the command.
25984 -data-evaluate-expression --language c "sizeof (void*)"
25989 The valid language names are the same names accepted by the
25990 @samp{set language} command (@pxref{Manually}), excluding @samp{auto},
25991 @samp{local} or @samp{unknown}.
25993 @node Asynchronous and non-stop modes
25994 @subsection Asynchronous command execution and non-stop mode
25996 On some targets, @value{GDBN} is capable of processing MI commands
25997 even while the target is running. This is called @dfn{asynchronous
25998 command execution} (@pxref{Background Execution}). The frontend may
25999 specify a preferrence for asynchronous execution using the
26000 @code{-gdb-set mi-async 1} command, which should be emitted before
26001 either running the executable or attaching to the target. After the
26002 frontend has started the executable or attached to the target, it can
26003 find if asynchronous execution is enabled using the
26004 @code{-list-target-features} command.
26007 @item -gdb-set mi-async on
26008 @item -gdb-set mi-async off
26009 Set whether MI is in asynchronous mode.
26011 When @code{off}, which is the default, MI execution commands (e.g.,
26012 @code{-exec-continue}) are foreground commands, and @value{GDBN} waits
26013 for the program to stop before processing further commands.
26015 When @code{on}, MI execution commands are background execution
26016 commands (e.g., @code{-exec-continue} becomes the equivalent of the
26017 @code{c&} CLI command), and so @value{GDBN} is capable of processing
26018 MI commands even while the target is running.
26020 @item -gdb-show mi-async
26021 Show whether MI asynchronous mode is enabled.
26024 Note: In @value{GDBN} version 7.7 and earlier, this option was called
26025 @code{target-async} instead of @code{mi-async}, and it had the effect
26026 of both putting MI in asynchronous mode and making CLI background
26027 commands possible. CLI background commands are now always possible
26028 ``out of the box'' if the target supports them. The old spelling is
26029 kept as a deprecated alias for backwards compatibility.
26031 Even if @value{GDBN} can accept a command while target is running,
26032 many commands that access the target do not work when the target is
26033 running. Therefore, asynchronous command execution is most useful
26034 when combined with non-stop mode (@pxref{Non-Stop Mode}). Then,
26035 it is possible to examine the state of one thread, while other threads
26038 When a given thread is running, MI commands that try to access the
26039 target in the context of that thread may not work, or may work only on
26040 some targets. In particular, commands that try to operate on thread's
26041 stack will not work, on any target. Commands that read memory, or
26042 modify breakpoints, may work or not work, depending on the target. Note
26043 that even commands that operate on global state, such as @code{print},
26044 @code{set}, and breakpoint commands, still access the target in the
26045 context of a specific thread, so frontend should try to find a
26046 stopped thread and perform the operation on that thread (using the
26047 @samp{--thread} option).
26049 Which commands will work in the context of a running thread is
26050 highly target dependent. However, the two commands
26051 @code{-exec-interrupt}, to stop a thread, and @code{-thread-info},
26052 to find the state of a thread, will always work.
26054 @node Thread groups
26055 @subsection Thread groups
26056 @value{GDBN} may be used to debug several processes at the same time.
26057 On some platfroms, @value{GDBN} may support debugging of several
26058 hardware systems, each one having several cores with several different
26059 processes running on each core. This section describes the MI
26060 mechanism to support such debugging scenarios.
26062 The key observation is that regardless of the structure of the
26063 target, MI can have a global list of threads, because most commands that
26064 accept the @samp{--thread} option do not need to know what process that
26065 thread belongs to. Therefore, it is not necessary to introduce
26066 neither additional @samp{--process} option, nor an notion of the
26067 current process in the MI interface. The only strictly new feature
26068 that is required is the ability to find how the threads are grouped
26071 To allow the user to discover such grouping, and to support arbitrary
26072 hierarchy of machines/cores/processes, MI introduces the concept of a
26073 @dfn{thread group}. Thread group is a collection of threads and other
26074 thread groups. A thread group always has a string identifier, a type,
26075 and may have additional attributes specific to the type. A new
26076 command, @code{-list-thread-groups}, returns the list of top-level
26077 thread groups, which correspond to processes that @value{GDBN} is
26078 debugging at the moment. By passing an identifier of a thread group
26079 to the @code{-list-thread-groups} command, it is possible to obtain
26080 the members of specific thread group.
26082 To allow the user to easily discover processes, and other objects, he
26083 wishes to debug, a concept of @dfn{available thread group} is
26084 introduced. Available thread group is an thread group that
26085 @value{GDBN} is not debugging, but that can be attached to, using the
26086 @code{-target-attach} command. The list of available top-level thread
26087 groups can be obtained using @samp{-list-thread-groups --available}.
26088 In general, the content of a thread group may be only retrieved only
26089 after attaching to that thread group.
26091 Thread groups are related to inferiors (@pxref{Inferiors and
26092 Programs}). Each inferior corresponds to a thread group of a special
26093 type @samp{process}, and some additional operations are permitted on
26094 such thread groups.
26096 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
26097 @node GDB/MI Command Syntax
26098 @section @sc{gdb/mi} Command Syntax
26101 * GDB/MI Input Syntax::
26102 * GDB/MI Output Syntax::
26105 @node GDB/MI Input Syntax
26106 @subsection @sc{gdb/mi} Input Syntax
26108 @cindex input syntax for @sc{gdb/mi}
26109 @cindex @sc{gdb/mi}, input syntax
26111 @item @var{command} @expansion{}
26112 @code{@var{cli-command} | @var{mi-command}}
26114 @item @var{cli-command} @expansion{}
26115 @code{[ @var{token} ] @var{cli-command} @var{nl}}, where
26116 @var{cli-command} is any existing @value{GDBN} CLI command.
26118 @item @var{mi-command} @expansion{}
26119 @code{[ @var{token} ] "-" @var{operation} ( " " @var{option} )*
26120 @code{[} " --" @code{]} ( " " @var{parameter} )* @var{nl}}
26122 @item @var{token} @expansion{}
26123 "any sequence of digits"
26125 @item @var{option} @expansion{}
26126 @code{"-" @var{parameter} [ " " @var{parameter} ]}
26128 @item @var{parameter} @expansion{}
26129 @code{@var{non-blank-sequence} | @var{c-string}}
26131 @item @var{operation} @expansion{}
26132 @emph{any of the operations described in this chapter}
26134 @item @var{non-blank-sequence} @expansion{}
26135 @emph{anything, provided it doesn't contain special characters such as
26136 "-", @var{nl}, """ and of course " "}
26138 @item @var{c-string} @expansion{}
26139 @code{""" @var{seven-bit-iso-c-string-content} """}
26141 @item @var{nl} @expansion{}
26150 The CLI commands are still handled by the @sc{mi} interpreter; their
26151 output is described below.
26154 The @code{@var{token}}, when present, is passed back when the command
26158 Some @sc{mi} commands accept optional arguments as part of the parameter
26159 list. Each option is identified by a leading @samp{-} (dash) and may be
26160 followed by an optional argument parameter. Options occur first in the
26161 parameter list and can be delimited from normal parameters using
26162 @samp{--} (this is useful when some parameters begin with a dash).
26169 We want easy access to the existing CLI syntax (for debugging).
26172 We want it to be easy to spot a @sc{mi} operation.
26175 @node GDB/MI Output Syntax
26176 @subsection @sc{gdb/mi} Output Syntax
26178 @cindex output syntax of @sc{gdb/mi}
26179 @cindex @sc{gdb/mi}, output syntax
26180 The output from @sc{gdb/mi} consists of zero or more out-of-band records
26181 followed, optionally, by a single result record. This result record
26182 is for the most recent command. The sequence of output records is
26183 terminated by @samp{(gdb)}.
26185 If an input command was prefixed with a @code{@var{token}} then the
26186 corresponding output for that command will also be prefixed by that same
26190 @item @var{output} @expansion{}
26191 @code{( @var{out-of-band-record} )* [ @var{result-record} ] "(gdb)" @var{nl}}
26193 @item @var{result-record} @expansion{}
26194 @code{ [ @var{token} ] "^" @var{result-class} ( "," @var{result} )* @var{nl}}
26196 @item @var{out-of-band-record} @expansion{}
26197 @code{@var{async-record} | @var{stream-record}}
26199 @item @var{async-record} @expansion{}
26200 @code{@var{exec-async-output} | @var{status-async-output} | @var{notify-async-output}}
26202 @item @var{exec-async-output} @expansion{}
26203 @code{[ @var{token} ] "*" @var{async-output nl}}
26205 @item @var{status-async-output} @expansion{}
26206 @code{[ @var{token} ] "+" @var{async-output nl}}
26208 @item @var{notify-async-output} @expansion{}
26209 @code{[ @var{token} ] "=" @var{async-output nl}}
26211 @item @var{async-output} @expansion{}
26212 @code{@var{async-class} ( "," @var{result} )*}
26214 @item @var{result-class} @expansion{}
26215 @code{"done" | "running" | "connected" | "error" | "exit"}
26217 @item @var{async-class} @expansion{}
26218 @code{"stopped" | @var{others}} (where @var{others} will be added
26219 depending on the needs---this is still in development).
26221 @item @var{result} @expansion{}
26222 @code{ @var{variable} "=" @var{value}}
26224 @item @var{variable} @expansion{}
26225 @code{ @var{string} }
26227 @item @var{value} @expansion{}
26228 @code{ @var{const} | @var{tuple} | @var{list} }
26230 @item @var{const} @expansion{}
26231 @code{@var{c-string}}
26233 @item @var{tuple} @expansion{}
26234 @code{ "@{@}" | "@{" @var{result} ( "," @var{result} )* "@}" }
26236 @item @var{list} @expansion{}
26237 @code{ "[]" | "[" @var{value} ( "," @var{value} )* "]" | "["
26238 @var{result} ( "," @var{result} )* "]" }
26240 @item @var{stream-record} @expansion{}
26241 @code{@var{console-stream-output} | @var{target-stream-output} | @var{log-stream-output}}
26243 @item @var{console-stream-output} @expansion{}
26244 @code{"~" @var{c-string nl}}
26246 @item @var{target-stream-output} @expansion{}
26247 @code{"@@" @var{c-string nl}}
26249 @item @var{log-stream-output} @expansion{}
26250 @code{"&" @var{c-string nl}}
26252 @item @var{nl} @expansion{}
26255 @item @var{token} @expansion{}
26256 @emph{any sequence of digits}.
26264 All output sequences end in a single line containing a period.
26267 The @code{@var{token}} is from the corresponding request. Note that
26268 for all async output, while the token is allowed by the grammar and
26269 may be output by future versions of @value{GDBN} for select async
26270 output messages, it is generally omitted. Frontends should treat
26271 all async output as reporting general changes in the state of the
26272 target and there should be no need to associate async output to any
26276 @cindex status output in @sc{gdb/mi}
26277 @var{status-async-output} contains on-going status information about the
26278 progress of a slow operation. It can be discarded. All status output is
26279 prefixed by @samp{+}.
26282 @cindex async output in @sc{gdb/mi}
26283 @var{exec-async-output} contains asynchronous state change on the target
26284 (stopped, started, disappeared). All async output is prefixed by
26288 @cindex notify output in @sc{gdb/mi}
26289 @var{notify-async-output} contains supplementary information that the
26290 client should handle (e.g., a new breakpoint information). All notify
26291 output is prefixed by @samp{=}.
26294 @cindex console output in @sc{gdb/mi}
26295 @var{console-stream-output} is output that should be displayed as is in the
26296 console. It is the textual response to a CLI command. All the console
26297 output is prefixed by @samp{~}.
26300 @cindex target output in @sc{gdb/mi}
26301 @var{target-stream-output} is the output produced by the target program.
26302 All the target output is prefixed by @samp{@@}.
26305 @cindex log output in @sc{gdb/mi}
26306 @var{log-stream-output} is output text coming from @value{GDBN}'s internals, for
26307 instance messages that should be displayed as part of an error log. All
26308 the log output is prefixed by @samp{&}.
26311 @cindex list output in @sc{gdb/mi}
26312 New @sc{gdb/mi} commands should only output @var{lists} containing
26318 @xref{GDB/MI Stream Records, , @sc{gdb/mi} Stream Records}, for more
26319 details about the various output records.
26321 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
26322 @node GDB/MI Compatibility with CLI
26323 @section @sc{gdb/mi} Compatibility with CLI
26325 @cindex compatibility, @sc{gdb/mi} and CLI
26326 @cindex @sc{gdb/mi}, compatibility with CLI
26328 For the developers convenience CLI commands can be entered directly,
26329 but there may be some unexpected behaviour. For example, commands
26330 that query the user will behave as if the user replied yes, breakpoint
26331 command lists are not executed and some CLI commands, such as
26332 @code{if}, @code{when} and @code{define}, prompt for further input with
26333 @samp{>}, which is not valid MI output.
26335 This feature may be removed at some stage in the future and it is
26336 recommended that front ends use the @code{-interpreter-exec} command
26337 (@pxref{-interpreter-exec}).
26339 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
26340 @node GDB/MI Development and Front Ends
26341 @section @sc{gdb/mi} Development and Front Ends
26342 @cindex @sc{gdb/mi} development
26344 The application which takes the MI output and presents the state of the
26345 program being debugged to the user is called a @dfn{front end}.
26347 Although @sc{gdb/mi} is still incomplete, it is currently being used
26348 by a variety of front ends to @value{GDBN}. This makes it difficult
26349 to introduce new functionality without breaking existing usage. This
26350 section tries to minimize the problems by describing how the protocol
26353 Some changes in MI need not break a carefully designed front end, and
26354 for these the MI version will remain unchanged. The following is a
26355 list of changes that may occur within one level, so front ends should
26356 parse MI output in a way that can handle them:
26360 New MI commands may be added.
26363 New fields may be added to the output of any MI command.
26366 The range of values for fields with specified values, e.g.,
26367 @code{in_scope} (@pxref{-var-update}) may be extended.
26369 @c The format of field's content e.g type prefix, may change so parse it
26370 @c at your own risk. Yes, in general?
26372 @c The order of fields may change? Shouldn't really matter but it might
26373 @c resolve inconsistencies.
26376 If the changes are likely to break front ends, the MI version level
26377 will be increased by one. This will allow the front end to parse the
26378 output according to the MI version. Apart from mi0, new versions of
26379 @value{GDBN} will not support old versions of MI and it will be the
26380 responsibility of the front end to work with the new one.
26382 @c Starting with mi3, add a new command -mi-version that prints the MI
26385 The best way to avoid unexpected changes in MI that might break your front
26386 end is to make your project known to @value{GDBN} developers and
26387 follow development on @email{gdb@@sourceware.org} and
26388 @email{gdb-patches@@sourceware.org}.
26389 @cindex mailing lists
26391 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
26392 @node GDB/MI Output Records
26393 @section @sc{gdb/mi} Output Records
26396 * GDB/MI Result Records::
26397 * GDB/MI Stream Records::
26398 * GDB/MI Async Records::
26399 * GDB/MI Breakpoint Information::
26400 * GDB/MI Frame Information::
26401 * GDB/MI Thread Information::
26402 * GDB/MI Ada Exception Information::
26405 @node GDB/MI Result Records
26406 @subsection @sc{gdb/mi} Result Records
26408 @cindex result records in @sc{gdb/mi}
26409 @cindex @sc{gdb/mi}, result records
26410 In addition to a number of out-of-band notifications, the response to a
26411 @sc{gdb/mi} command includes one of the following result indications:
26415 @item "^done" [ "," @var{results} ]
26416 The synchronous operation was successful, @code{@var{results}} are the return
26421 This result record is equivalent to @samp{^done}. Historically, it
26422 was output instead of @samp{^done} if the command has resumed the
26423 target. This behaviour is maintained for backward compatibility, but
26424 all frontends should treat @samp{^done} and @samp{^running}
26425 identically and rely on the @samp{*running} output record to determine
26426 which threads are resumed.
26430 @value{GDBN} has connected to a remote target.
26432 @item "^error" "," "msg=" @var{c-string} [ "," "code=" @var{c-string} ]
26434 The operation failed. The @code{msg=@var{c-string}} variable contains
26435 the corresponding error message.
26437 If present, the @code{code=@var{c-string}} variable provides an error
26438 code on which consumers can rely on to detect the corresponding
26439 error condition. At present, only one error code is defined:
26442 @item "undefined-command"
26443 Indicates that the command causing the error does not exist.
26448 @value{GDBN} has terminated.
26452 @node GDB/MI Stream Records
26453 @subsection @sc{gdb/mi} Stream Records
26455 @cindex @sc{gdb/mi}, stream records
26456 @cindex stream records in @sc{gdb/mi}
26457 @value{GDBN} internally maintains a number of output streams: the console, the
26458 target, and the log. The output intended for each of these streams is
26459 funneled through the @sc{gdb/mi} interface using @dfn{stream records}.
26461 Each stream record begins with a unique @dfn{prefix character} which
26462 identifies its stream (@pxref{GDB/MI Output Syntax, , @sc{gdb/mi} Output
26463 Syntax}). In addition to the prefix, each stream record contains a
26464 @code{@var{string-output}}. This is either raw text (with an implicit new
26465 line) or a quoted C string (which does not contain an implicit newline).
26468 @item "~" @var{string-output}
26469 The console output stream contains text that should be displayed in the
26470 CLI console window. It contains the textual responses to CLI commands.
26472 @item "@@" @var{string-output}
26473 The target output stream contains any textual output from the running
26474 target. This is only present when GDB's event loop is truly
26475 asynchronous, which is currently only the case for remote targets.
26477 @item "&" @var{string-output}
26478 The log stream contains debugging messages being produced by @value{GDBN}'s
26482 @node GDB/MI Async Records
26483 @subsection @sc{gdb/mi} Async Records
26485 @cindex async records in @sc{gdb/mi}
26486 @cindex @sc{gdb/mi}, async records
26487 @dfn{Async} records are used to notify the @sc{gdb/mi} client of
26488 additional changes that have occurred. Those changes can either be a
26489 consequence of @sc{gdb/mi} commands (e.g., a breakpoint modified) or a result of
26490 target activity (e.g., target stopped).
26492 The following is the list of possible async records:
26496 @item *running,thread-id="@var{thread}"
26497 The target is now running. The @var{thread} field can be the global
26498 thread ID of the the thread that is now running, and it can be
26499 @samp{all} if all threads are running. The frontend should assume
26500 that no interaction with a running thread is possible after this
26501 notification is produced. The frontend should not assume that this
26502 notification is output only once for any command. @value{GDBN} may
26503 emit this notification several times, either for different threads,
26504 because it cannot resume all threads together, or even for a single
26505 thread, if the thread must be stepped though some code before letting
26508 @item *stopped,reason="@var{reason}",thread-id="@var{id}",stopped-threads="@var{stopped}",core="@var{core}"
26509 The target has stopped. The @var{reason} field can have one of the
26513 @item breakpoint-hit
26514 A breakpoint was reached.
26515 @item watchpoint-trigger
26516 A watchpoint was triggered.
26517 @item read-watchpoint-trigger
26518 A read watchpoint was triggered.
26519 @item access-watchpoint-trigger
26520 An access watchpoint was triggered.
26521 @item function-finished
26522 An -exec-finish or similar CLI command was accomplished.
26523 @item location-reached
26524 An -exec-until or similar CLI command was accomplished.
26525 @item watchpoint-scope
26526 A watchpoint has gone out of scope.
26527 @item end-stepping-range
26528 An -exec-next, -exec-next-instruction, -exec-step, -exec-step-instruction or
26529 similar CLI command was accomplished.
26530 @item exited-signalled
26531 The inferior exited because of a signal.
26533 The inferior exited.
26534 @item exited-normally
26535 The inferior exited normally.
26536 @item signal-received
26537 A signal was received by the inferior.
26539 The inferior has stopped due to a library being loaded or unloaded.
26540 This can happen when @code{stop-on-solib-events} (@pxref{Files}) is
26541 set or when a @code{catch load} or @code{catch unload} catchpoint is
26542 in use (@pxref{Set Catchpoints}).
26544 The inferior has forked. This is reported when @code{catch fork}
26545 (@pxref{Set Catchpoints}) has been used.
26547 The inferior has vforked. This is reported in when @code{catch vfork}
26548 (@pxref{Set Catchpoints}) has been used.
26549 @item syscall-entry
26550 The inferior entered a system call. This is reported when @code{catch
26551 syscall} (@pxref{Set Catchpoints}) has been used.
26552 @item syscall-return
26553 The inferior returned from a system call. This is reported when
26554 @code{catch syscall} (@pxref{Set Catchpoints}) has been used.
26556 The inferior called @code{exec}. This is reported when @code{catch exec}
26557 (@pxref{Set Catchpoints}) has been used.
26560 The @var{id} field identifies the global thread ID of the thread
26561 that directly caused the stop -- for example by hitting a breakpoint.
26562 Depending on whether all-stop
26563 mode is in effect (@pxref{All-Stop Mode}), @value{GDBN} may either
26564 stop all threads, or only the thread that directly triggered the stop.
26565 If all threads are stopped, the @var{stopped} field will have the
26566 value of @code{"all"}. Otherwise, the value of the @var{stopped}
26567 field will be a list of thread identifiers. Presently, this list will
26568 always include a single thread, but frontend should be prepared to see
26569 several threads in the list. The @var{core} field reports the
26570 processor core on which the stop event has happened. This field may be absent
26571 if such information is not available.
26573 @item =thread-group-added,id="@var{id}"
26574 @itemx =thread-group-removed,id="@var{id}"
26575 A thread group was either added or removed. The @var{id} field
26576 contains the @value{GDBN} identifier of the thread group. When a thread
26577 group is added, it generally might not be associated with a running
26578 process. When a thread group is removed, its id becomes invalid and
26579 cannot be used in any way.
26581 @item =thread-group-started,id="@var{id}",pid="@var{pid}"
26582 A thread group became associated with a running program,
26583 either because the program was just started or the thread group
26584 was attached to a program. The @var{id} field contains the
26585 @value{GDBN} identifier of the thread group. The @var{pid} field
26586 contains process identifier, specific to the operating system.
26588 @item =thread-group-exited,id="@var{id}"[,exit-code="@var{code}"]
26589 A thread group is no longer associated with a running program,
26590 either because the program has exited, or because it was detached
26591 from. The @var{id} field contains the @value{GDBN} identifier of the
26592 thread group. The @var{code} field is the exit code of the inferior; it exists
26593 only when the inferior exited with some code.
26595 @item =thread-created,id="@var{id}",group-id="@var{gid}"
26596 @itemx =thread-exited,id="@var{id}",group-id="@var{gid}"
26597 A thread either was created, or has exited. The @var{id} field
26598 contains the global @value{GDBN} identifier of the thread. The @var{gid}
26599 field identifies the thread group this thread belongs to.
26601 @item =thread-selected,id="@var{id}"[,frame="@var{frame}"]
26602 Informs that the selected thread or frame were changed. This notification
26603 is not emitted as result of the @code{-thread-select} or
26604 @code{-stack-select-frame} commands, but is emitted whenever an MI command
26605 that is not documented to change the selected thread and frame actually
26606 changes them. In particular, invoking, directly or indirectly
26607 (via user-defined command), the CLI @code{thread} or @code{frame} commands,
26608 will generate this notification. Changing the thread or frame from another
26609 user interface (see @ref{Interpreters}) will also generate this notification.
26611 The @var{frame} field is only present if the newly selected thread is
26612 stopped. See @ref{GDB/MI Frame Information} for the format of its value.
26614 We suggest that in response to this notification, front ends
26615 highlight the selected thread and cause subsequent commands to apply to
26618 @item =library-loaded,...
26619 Reports that a new library file was loaded by the program. This
26620 notification has 5 fields---@var{id}, @var{target-name},
26621 @var{host-name}, @var{symbols-loaded} and @var{ranges}. The @var{id} field is an
26622 opaque identifier of the library. For remote debugging case,
26623 @var{target-name} and @var{host-name} fields give the name of the
26624 library file on the target, and on the host respectively. For native
26625 debugging, both those fields have the same value. The
26626 @var{symbols-loaded} field is emitted only for backward compatibility
26627 and should not be relied on to convey any useful information. The
26628 @var{thread-group} field, if present, specifies the id of the thread
26629 group in whose context the library was loaded. If the field is
26630 absent, it means the library was loaded in the context of all present
26631 thread groups. The @var{ranges} field specifies the ranges of addresses belonging
26634 @item =library-unloaded,...
26635 Reports that a library was unloaded by the program. This notification
26636 has 3 fields---@var{id}, @var{target-name} and @var{host-name} with
26637 the same meaning as for the @code{=library-loaded} notification.
26638 The @var{thread-group} field, if present, specifies the id of the
26639 thread group in whose context the library was unloaded. If the field is
26640 absent, it means the library was unloaded in the context of all present
26643 @item =traceframe-changed,num=@var{tfnum},tracepoint=@var{tpnum}
26644 @itemx =traceframe-changed,end
26645 Reports that the trace frame was changed and its new number is
26646 @var{tfnum}. The number of the tracepoint associated with this trace
26647 frame is @var{tpnum}.
26649 @item =tsv-created,name=@var{name},initial=@var{initial}
26650 Reports that the new trace state variable @var{name} is created with
26651 initial value @var{initial}.
26653 @item =tsv-deleted,name=@var{name}
26654 @itemx =tsv-deleted
26655 Reports that the trace state variable @var{name} is deleted or all
26656 trace state variables are deleted.
26658 @item =tsv-modified,name=@var{name},initial=@var{initial}[,current=@var{current}]
26659 Reports that the trace state variable @var{name} is modified with
26660 the initial value @var{initial}. The current value @var{current} of
26661 trace state variable is optional and is reported if the current
26662 value of trace state variable is known.
26664 @item =breakpoint-created,bkpt=@{...@}
26665 @itemx =breakpoint-modified,bkpt=@{...@}
26666 @itemx =breakpoint-deleted,id=@var{number}
26667 Reports that a breakpoint was created, modified, or deleted,
26668 respectively. Only user-visible breakpoints are reported to the MI
26671 The @var{bkpt} argument is of the same form as returned by the various
26672 breakpoint commands; @xref{GDB/MI Breakpoint Commands}. The
26673 @var{number} is the ordinal number of the breakpoint.
26675 Note that if a breakpoint is emitted in the result record of a
26676 command, then it will not also be emitted in an async record.
26678 @item =record-started,thread-group="@var{id}",method="@var{method}"[,format="@var{format}"]
26679 @itemx =record-stopped,thread-group="@var{id}"
26680 Execution log recording was either started or stopped on an
26681 inferior. The @var{id} is the @value{GDBN} identifier of the thread
26682 group corresponding to the affected inferior.
26684 The @var{method} field indicates the method used to record execution. If the
26685 method in use supports multiple recording formats, @var{format} will be present
26686 and contain the currently used format. @xref{Process Record and Replay},
26687 for existing method and format values.
26689 @item =cmd-param-changed,param=@var{param},value=@var{value}
26690 Reports that a parameter of the command @code{set @var{param}} is
26691 changed to @var{value}. In the multi-word @code{set} command,
26692 the @var{param} is the whole parameter list to @code{set} command.
26693 For example, In command @code{set check type on}, @var{param}
26694 is @code{check type} and @var{value} is @code{on}.
26696 @item =memory-changed,thread-group=@var{id},addr=@var{addr},len=@var{len}[,type="code"]
26697 Reports that bytes from @var{addr} to @var{data} + @var{len} were
26698 written in an inferior. The @var{id} is the identifier of the
26699 thread group corresponding to the affected inferior. The optional
26700 @code{type="code"} part is reported if the memory written to holds
26704 @node GDB/MI Breakpoint Information
26705 @subsection @sc{gdb/mi} Breakpoint Information
26707 When @value{GDBN} reports information about a breakpoint, a
26708 tracepoint, a watchpoint, or a catchpoint, it uses a tuple with the
26713 The breakpoint number. For a breakpoint that represents one location
26714 of a multi-location breakpoint, this will be a dotted pair, like
26718 The type of the breakpoint. For ordinary breakpoints this will be
26719 @samp{breakpoint}, but many values are possible.
26722 If the type of the breakpoint is @samp{catchpoint}, then this
26723 indicates the exact type of catchpoint.
26726 This is the breakpoint disposition---either @samp{del}, meaning that
26727 the breakpoint will be deleted at the next stop, or @samp{keep},
26728 meaning that the breakpoint will not be deleted.
26731 This indicates whether the breakpoint is enabled, in which case the
26732 value is @samp{y}, or disabled, in which case the value is @samp{n}.
26733 Note that this is not the same as the field @code{enable}.
26736 The address of the breakpoint. This may be a hexidecimal number,
26737 giving the address; or the string @samp{<PENDING>}, for a pending
26738 breakpoint; or the string @samp{<MULTIPLE>}, for a breakpoint with
26739 multiple locations. This field will not be present if no address can
26740 be determined. For example, a watchpoint does not have an address.
26743 If known, the function in which the breakpoint appears.
26744 If not known, this field is not present.
26747 The name of the source file which contains this function, if known.
26748 If not known, this field is not present.
26751 The full file name of the source file which contains this function, if
26752 known. If not known, this field is not present.
26755 The line number at which this breakpoint appears, if known.
26756 If not known, this field is not present.
26759 If the source file is not known, this field may be provided. If
26760 provided, this holds the address of the breakpoint, possibly followed
26764 If this breakpoint is pending, this field is present and holds the
26765 text used to set the breakpoint, as entered by the user.
26768 Where this breakpoint's condition is evaluated, either @samp{host} or
26772 If this is a thread-specific breakpoint, then this identifies the
26773 thread in which the breakpoint can trigger.
26776 If this breakpoint is restricted to a particular Ada task, then this
26777 field will hold the task identifier.
26780 If the breakpoint is conditional, this is the condition expression.
26783 The ignore count of the breakpoint.
26786 The enable count of the breakpoint.
26788 @item traceframe-usage
26791 @item static-tracepoint-marker-string-id
26792 For a static tracepoint, the name of the static tracepoint marker.
26795 For a masked watchpoint, this is the mask.
26798 A tracepoint's pass count.
26800 @item original-location
26801 The location of the breakpoint as originally specified by the user.
26802 This field is optional.
26805 The number of times the breakpoint has been hit.
26808 This field is only given for tracepoints. This is either @samp{y},
26809 meaning that the tracepoint is installed, or @samp{n}, meaning that it
26813 Some extra data, the exact contents of which are type-dependent.
26817 For example, here is what the output of @code{-break-insert}
26818 (@pxref{GDB/MI Breakpoint Commands}) might be:
26821 -> -break-insert main
26822 <- ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
26823 enabled="y",addr="0x08048564",func="main",file="myprog.c",
26824 fullname="/home/nickrob/myprog.c",line="68",thread-groups=["i1"],
26829 @node GDB/MI Frame Information
26830 @subsection @sc{gdb/mi} Frame Information
26832 Response from many MI commands includes an information about stack
26833 frame. This information is a tuple that may have the following
26838 The level of the stack frame. The innermost frame has the level of
26839 zero. This field is always present.
26842 The name of the function corresponding to the frame. This field may
26843 be absent if @value{GDBN} is unable to determine the function name.
26846 The code address for the frame. This field is always present.
26849 The name of the source files that correspond to the frame's code
26850 address. This field may be absent.
26853 The source line corresponding to the frames' code address. This field
26857 The name of the binary file (either executable or shared library) the
26858 corresponds to the frame's code address. This field may be absent.
26862 @node GDB/MI Thread Information
26863 @subsection @sc{gdb/mi} Thread Information
26865 Whenever @value{GDBN} has to report an information about a thread, it
26866 uses a tuple with the following fields. The fields are always present unless
26871 The global numeric id assigned to the thread by @value{GDBN}.
26874 The target-specific string identifying the thread.
26877 Additional information about the thread provided by the target.
26878 It is supposed to be human-readable and not interpreted by the
26879 frontend. This field is optional.
26882 The name of the thread. If the user specified a name using the
26883 @code{thread name} command, then this name is given. Otherwise, if
26884 @value{GDBN} can extract the thread name from the target, then that
26885 name is given. If @value{GDBN} cannot find the thread name, then this
26889 The execution state of the thread, either @samp{stopped} or @samp{running},
26890 depending on whether the thread is presently running.
26893 The stack frame currently executing in the thread. This field is only present
26894 if the thread is stopped. Its format is documented in
26895 @ref{GDB/MI Frame Information}.
26898 The value of this field is an integer number of the processor core the
26899 thread was last seen on. This field is optional.
26902 @node GDB/MI Ada Exception Information
26903 @subsection @sc{gdb/mi} Ada Exception Information
26905 Whenever a @code{*stopped} record is emitted because the program
26906 stopped after hitting an exception catchpoint (@pxref{Set Catchpoints}),
26907 @value{GDBN} provides the name of the exception that was raised via
26908 the @code{exception-name} field.
26910 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
26911 @node GDB/MI Simple Examples
26912 @section Simple Examples of @sc{gdb/mi} Interaction
26913 @cindex @sc{gdb/mi}, simple examples
26915 This subsection presents several simple examples of interaction using
26916 the @sc{gdb/mi} interface. In these examples, @samp{->} means that the
26917 following line is passed to @sc{gdb/mi} as input, while @samp{<-} means
26918 the output received from @sc{gdb/mi}.
26920 Note the line breaks shown in the examples are here only for
26921 readability, they don't appear in the real output.
26923 @subheading Setting a Breakpoint
26925 Setting a breakpoint generates synchronous output which contains detailed
26926 information of the breakpoint.
26929 -> -break-insert main
26930 <- ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
26931 enabled="y",addr="0x08048564",func="main",file="myprog.c",
26932 fullname="/home/nickrob/myprog.c",line="68",thread-groups=["i1"],
26937 @subheading Program Execution
26939 Program execution generates asynchronous records and MI gives the
26940 reason that execution stopped.
26946 <- *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",thread-id="0",
26947 frame=@{addr="0x08048564",func="main",
26948 args=[@{name="argc",value="1"@},@{name="argv",value="0xbfc4d4d4"@}],
26949 file="myprog.c",fullname="/home/nickrob/myprog.c",line="68"@}
26954 <- *stopped,reason="exited-normally"
26958 @subheading Quitting @value{GDBN}
26960 Quitting @value{GDBN} just prints the result class @samp{^exit}.
26968 Please note that @samp{^exit} is printed immediately, but it might
26969 take some time for @value{GDBN} to actually exit. During that time, @value{GDBN}
26970 performs necessary cleanups, including killing programs being debugged
26971 or disconnecting from debug hardware, so the frontend should wait till
26972 @value{GDBN} exits and should only forcibly kill @value{GDBN} if it
26973 fails to exit in reasonable time.
26975 @subheading A Bad Command
26977 Here's what happens if you pass a non-existent command:
26981 <- ^error,msg="Undefined MI command: rubbish"
26986 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
26987 @node GDB/MI Command Description Format
26988 @section @sc{gdb/mi} Command Description Format
26990 The remaining sections describe blocks of commands. Each block of
26991 commands is laid out in a fashion similar to this section.
26993 @subheading Motivation
26995 The motivation for this collection of commands.
26997 @subheading Introduction
26999 A brief introduction to this collection of commands as a whole.
27001 @subheading Commands
27003 For each command in the block, the following is described:
27005 @subsubheading Synopsis
27008 -command @var{args}@dots{}
27011 @subsubheading Result
27013 @subsubheading @value{GDBN} Command
27015 The corresponding @value{GDBN} CLI command(s), if any.
27017 @subsubheading Example
27019 Example(s) formatted for readability. Some of the described commands have
27020 not been implemented yet and these are labeled N.A.@: (not available).
27023 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
27024 @node GDB/MI Breakpoint Commands
27025 @section @sc{gdb/mi} Breakpoint Commands
27027 @cindex breakpoint commands for @sc{gdb/mi}
27028 @cindex @sc{gdb/mi}, breakpoint commands
27029 This section documents @sc{gdb/mi} commands for manipulating
27032 @subheading The @code{-break-after} Command
27033 @findex -break-after
27035 @subsubheading Synopsis
27038 -break-after @var{number} @var{count}
27041 The breakpoint number @var{number} is not in effect until it has been
27042 hit @var{count} times. To see how this is reflected in the output of
27043 the @samp{-break-list} command, see the description of the
27044 @samp{-break-list} command below.
27046 @subsubheading @value{GDBN} Command
27048 The corresponding @value{GDBN} command is @samp{ignore}.
27050 @subsubheading Example
27055 ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
27056 enabled="y",addr="0x000100d0",func="main",file="hello.c",
27057 fullname="/home/foo/hello.c",line="5",thread-groups=["i1"],
27065 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
27066 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
27067 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
27068 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
27069 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
27070 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
27071 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
27072 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
27073 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
27074 line="5",thread-groups=["i1"],times="0",ignore="3"@}]@}
27079 @subheading The @code{-break-catch} Command
27080 @findex -break-catch
27083 @subheading The @code{-break-commands} Command
27084 @findex -break-commands
27086 @subsubheading Synopsis
27089 -break-commands @var{number} [ @var{command1} ... @var{commandN} ]
27092 Specifies the CLI commands that should be executed when breakpoint
27093 @var{number} is hit. The parameters @var{command1} to @var{commandN}
27094 are the commands. If no command is specified, any previously-set
27095 commands are cleared. @xref{Break Commands}. Typical use of this
27096 functionality is tracing a program, that is, printing of values of
27097 some variables whenever breakpoint is hit and then continuing.
27099 @subsubheading @value{GDBN} Command
27101 The corresponding @value{GDBN} command is @samp{commands}.
27103 @subsubheading Example
27108 ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
27109 enabled="y",addr="0x000100d0",func="main",file="hello.c",
27110 fullname="/home/foo/hello.c",line="5",thread-groups=["i1"],
27113 -break-commands 1 "print v" "continue"
27118 @subheading The @code{-break-condition} Command
27119 @findex -break-condition
27121 @subsubheading Synopsis
27124 -break-condition @var{number} @var{expr}
27127 Breakpoint @var{number} will stop the program only if the condition in
27128 @var{expr} is true. The condition becomes part of the
27129 @samp{-break-list} output (see the description of the @samp{-break-list}
27132 @subsubheading @value{GDBN} Command
27134 The corresponding @value{GDBN} command is @samp{condition}.
27136 @subsubheading Example
27140 -break-condition 1 1
27144 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
27145 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
27146 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
27147 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
27148 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
27149 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
27150 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
27151 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
27152 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
27153 line="5",cond="1",thread-groups=["i1"],times="0",ignore="3"@}]@}
27157 @subheading The @code{-break-delete} Command
27158 @findex -break-delete
27160 @subsubheading Synopsis
27163 -break-delete ( @var{breakpoint} )+
27166 Delete the breakpoint(s) whose number(s) are specified in the argument
27167 list. This is obviously reflected in the breakpoint list.
27169 @subsubheading @value{GDBN} Command
27171 The corresponding @value{GDBN} command is @samp{delete}.
27173 @subsubheading Example
27181 ^done,BreakpointTable=@{nr_rows="0",nr_cols="6",
27182 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
27183 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
27184 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
27185 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
27186 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
27187 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
27192 @subheading The @code{-break-disable} Command
27193 @findex -break-disable
27195 @subsubheading Synopsis
27198 -break-disable ( @var{breakpoint} )+
27201 Disable the named @var{breakpoint}(s). The field @samp{enabled} in the
27202 break list is now set to @samp{n} for the named @var{breakpoint}(s).
27204 @subsubheading @value{GDBN} Command
27206 The corresponding @value{GDBN} command is @samp{disable}.
27208 @subsubheading Example
27216 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
27217 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
27218 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
27219 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
27220 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
27221 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
27222 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
27223 body=[bkpt=@{number="2",type="breakpoint",disp="keep",enabled="n",
27224 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
27225 line="5",thread-groups=["i1"],times="0"@}]@}
27229 @subheading The @code{-break-enable} Command
27230 @findex -break-enable
27232 @subsubheading Synopsis
27235 -break-enable ( @var{breakpoint} )+
27238 Enable (previously disabled) @var{breakpoint}(s).
27240 @subsubheading @value{GDBN} Command
27242 The corresponding @value{GDBN} command is @samp{enable}.
27244 @subsubheading Example
27252 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
27253 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
27254 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
27255 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
27256 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
27257 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
27258 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
27259 body=[bkpt=@{number="2",type="breakpoint",disp="keep",enabled="y",
27260 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
27261 line="5",thread-groups=["i1"],times="0"@}]@}
27265 @subheading The @code{-break-info} Command
27266 @findex -break-info
27268 @subsubheading Synopsis
27271 -break-info @var{breakpoint}
27275 Get information about a single breakpoint.
27277 The result is a table of breakpoints. @xref{GDB/MI Breakpoint
27278 Information}, for details on the format of each breakpoint in the
27281 @subsubheading @value{GDBN} Command
27283 The corresponding @value{GDBN} command is @samp{info break @var{breakpoint}}.
27285 @subsubheading Example
27288 @subheading The @code{-break-insert} Command
27289 @findex -break-insert
27290 @anchor{-break-insert}
27292 @subsubheading Synopsis
27295 -break-insert [ -t ] [ -h ] [ -f ] [ -d ] [ -a ]
27296 [ -c @var{condition} ] [ -i @var{ignore-count} ]
27297 [ -p @var{thread-id} ] [ @var{location} ]
27301 If specified, @var{location}, can be one of:
27304 @item linespec location
27305 A linespec location. @xref{Linespec Locations}.
27307 @item explicit location
27308 An explicit location. @sc{gdb/mi} explicit locations are
27309 analogous to the CLI's explicit locations using the option names
27310 listed below. @xref{Explicit Locations}.
27313 @item --source @var{filename}
27314 The source file name of the location. This option requires the use
27315 of either @samp{--function} or @samp{--line}.
27317 @item --function @var{function}
27318 The name of a function or method.
27320 @item --label @var{label}
27321 The name of a label.
27323 @item --line @var{lineoffset}
27324 An absolute or relative line offset from the start of the location.
27327 @item address location
27328 An address location, *@var{address}. @xref{Address Locations}.
27332 The possible optional parameters of this command are:
27336 Insert a temporary breakpoint.
27338 Insert a hardware breakpoint.
27340 If @var{location} cannot be parsed (for example if it
27341 refers to unknown files or functions), create a pending
27342 breakpoint. Without this flag, @value{GDBN} will report
27343 an error, and won't create a breakpoint, if @var{location}
27346 Create a disabled breakpoint.
27348 Create a tracepoint. @xref{Tracepoints}. When this parameter
27349 is used together with @samp{-h}, a fast tracepoint is created.
27350 @item -c @var{condition}
27351 Make the breakpoint conditional on @var{condition}.
27352 @item -i @var{ignore-count}
27353 Initialize the @var{ignore-count}.
27354 @item -p @var{thread-id}
27355 Restrict the breakpoint to the thread with the specified global
27359 @subsubheading Result
27361 @xref{GDB/MI Breakpoint Information}, for details on the format of the
27362 resulting breakpoint.
27364 Note: this format is open to change.
27365 @c An out-of-band breakpoint instead of part of the result?
27367 @subsubheading @value{GDBN} Command
27369 The corresponding @value{GDBN} commands are @samp{break}, @samp{tbreak},
27370 @samp{hbreak}, and @samp{thbreak}. @c and @samp{rbreak}.
27372 @subsubheading Example
27377 ^done,bkpt=@{number="1",addr="0x0001072c",file="recursive2.c",
27378 fullname="/home/foo/recursive2.c,line="4",thread-groups=["i1"],
27381 -break-insert -t foo
27382 ^done,bkpt=@{number="2",addr="0x00010774",file="recursive2.c",
27383 fullname="/home/foo/recursive2.c,line="11",thread-groups=["i1"],
27387 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
27388 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
27389 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
27390 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
27391 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
27392 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
27393 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
27394 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
27395 addr="0x0001072c", func="main",file="recursive2.c",
27396 fullname="/home/foo/recursive2.c,"line="4",thread-groups=["i1"],
27398 bkpt=@{number="2",type="breakpoint",disp="del",enabled="y",
27399 addr="0x00010774",func="foo",file="recursive2.c",
27400 fullname="/home/foo/recursive2.c",line="11",thread-groups=["i1"],
27403 @c -break-insert -r foo.*
27404 @c ~int foo(int, int);
27405 @c ^done,bkpt=@{number="3",addr="0x00010774",file="recursive2.c,
27406 @c "fullname="/home/foo/recursive2.c",line="11",thread-groups=["i1"],
27411 @subheading The @code{-dprintf-insert} Command
27412 @findex -dprintf-insert
27414 @subsubheading Synopsis
27417 -dprintf-insert [ -t ] [ -f ] [ -d ]
27418 [ -c @var{condition} ] [ -i @var{ignore-count} ]
27419 [ -p @var{thread-id} ] [ @var{location} ] [ @var{format} ]
27424 If supplied, @var{location} may be specified the same way as for
27425 the @code{-break-insert} command. @xref{-break-insert}.
27427 The possible optional parameters of this command are:
27431 Insert a temporary breakpoint.
27433 If @var{location} cannot be parsed (for example, if it
27434 refers to unknown files or functions), create a pending
27435 breakpoint. Without this flag, @value{GDBN} will report
27436 an error, and won't create a breakpoint, if @var{location}
27439 Create a disabled breakpoint.
27440 @item -c @var{condition}
27441 Make the breakpoint conditional on @var{condition}.
27442 @item -i @var{ignore-count}
27443 Set the ignore count of the breakpoint (@pxref{Conditions, ignore count})
27444 to @var{ignore-count}.
27445 @item -p @var{thread-id}
27446 Restrict the breakpoint to the thread with the specified global
27450 @subsubheading Result
27452 @xref{GDB/MI Breakpoint Information}, for details on the format of the
27453 resulting breakpoint.
27455 @c An out-of-band breakpoint instead of part of the result?
27457 @subsubheading @value{GDBN} Command
27459 The corresponding @value{GDBN} command is @samp{dprintf}.
27461 @subsubheading Example
27465 4-dprintf-insert foo "At foo entry\n"
27466 4^done,bkpt=@{number="1",type="dprintf",disp="keep",enabled="y",
27467 addr="0x000000000040061b",func="foo",file="mi-dprintf.c",
27468 fullname="mi-dprintf.c",line="25",thread-groups=["i1"],
27469 times="0",script=@{"printf \"At foo entry\\n\"","continue"@},
27470 original-location="foo"@}
27472 5-dprintf-insert 26 "arg=%d, g=%d\n" arg g
27473 5^done,bkpt=@{number="2",type="dprintf",disp="keep",enabled="y",
27474 addr="0x000000000040062a",func="foo",file="mi-dprintf.c",
27475 fullname="mi-dprintf.c",line="26",thread-groups=["i1"],
27476 times="0",script=@{"printf \"arg=%d, g=%d\\n\", arg, g","continue"@},
27477 original-location="mi-dprintf.c:26"@}
27481 @subheading The @code{-break-list} Command
27482 @findex -break-list
27484 @subsubheading Synopsis
27490 Displays the list of inserted breakpoints, showing the following fields:
27494 number of the breakpoint
27496 type of the breakpoint: @samp{breakpoint} or @samp{watchpoint}
27498 should the breakpoint be deleted or disabled when it is hit: @samp{keep}
27501 is the breakpoint enabled or no: @samp{y} or @samp{n}
27503 memory location at which the breakpoint is set
27505 logical location of the breakpoint, expressed by function name, file
27507 @item Thread-groups
27508 list of thread groups to which this breakpoint applies
27510 number of times the breakpoint has been hit
27513 If there are no breakpoints or watchpoints, the @code{BreakpointTable}
27514 @code{body} field is an empty list.
27516 @subsubheading @value{GDBN} Command
27518 The corresponding @value{GDBN} command is @samp{info break}.
27520 @subsubheading Example
27525 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
27526 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
27527 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
27528 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
27529 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
27530 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
27531 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
27532 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
27533 addr="0x000100d0",func="main",file="hello.c",line="5",thread-groups=["i1"],
27535 bkpt=@{number="2",type="breakpoint",disp="keep",enabled="y",
27536 addr="0x00010114",func="foo",file="hello.c",fullname="/home/foo/hello.c",
27537 line="13",thread-groups=["i1"],times="0"@}]@}
27541 Here's an example of the result when there are no breakpoints:
27546 ^done,BreakpointTable=@{nr_rows="0",nr_cols="6",
27547 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
27548 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
27549 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
27550 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
27551 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
27552 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
27557 @subheading The @code{-break-passcount} Command
27558 @findex -break-passcount
27560 @subsubheading Synopsis
27563 -break-passcount @var{tracepoint-number} @var{passcount}
27566 Set the passcount for tracepoint @var{tracepoint-number} to
27567 @var{passcount}. If the breakpoint referred to by @var{tracepoint-number}
27568 is not a tracepoint, error is emitted. This corresponds to CLI
27569 command @samp{passcount}.
27571 @subheading The @code{-break-watch} Command
27572 @findex -break-watch
27574 @subsubheading Synopsis
27577 -break-watch [ -a | -r ]
27580 Create a watchpoint. With the @samp{-a} option it will create an
27581 @dfn{access} watchpoint, i.e., a watchpoint that triggers either on a
27582 read from or on a write to the memory location. With the @samp{-r}
27583 option, the watchpoint created is a @dfn{read} watchpoint, i.e., it will
27584 trigger only when the memory location is accessed for reading. Without
27585 either of the options, the watchpoint created is a regular watchpoint,
27586 i.e., it will trigger when the memory location is accessed for writing.
27587 @xref{Set Watchpoints, , Setting Watchpoints}.
27589 Note that @samp{-break-list} will report a single list of watchpoints and
27590 breakpoints inserted.
27592 @subsubheading @value{GDBN} Command
27594 The corresponding @value{GDBN} commands are @samp{watch}, @samp{awatch}, and
27597 @subsubheading Example
27599 Setting a watchpoint on a variable in the @code{main} function:
27604 ^done,wpt=@{number="2",exp="x"@}
27609 *stopped,reason="watchpoint-trigger",wpt=@{number="2",exp="x"@},
27610 value=@{old="-268439212",new="55"@},
27611 frame=@{func="main",args=[],file="recursive2.c",
27612 fullname="/home/foo/bar/recursive2.c",line="5"@}
27616 Setting a watchpoint on a variable local to a function. @value{GDBN} will stop
27617 the program execution twice: first for the variable changing value, then
27618 for the watchpoint going out of scope.
27623 ^done,wpt=@{number="5",exp="C"@}
27628 *stopped,reason="watchpoint-trigger",
27629 wpt=@{number="5",exp="C"@},value=@{old="-276895068",new="3"@},
27630 frame=@{func="callee4",args=[],
27631 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
27632 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="13"@}
27637 *stopped,reason="watchpoint-scope",wpnum="5",
27638 frame=@{func="callee3",args=[@{name="strarg",
27639 value="0x11940 \"A string argument.\""@}],
27640 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
27641 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="18"@}
27645 Listing breakpoints and watchpoints, at different points in the program
27646 execution. Note that once the watchpoint goes out of scope, it is
27652 ^done,wpt=@{number="2",exp="C"@}
27655 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
27656 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
27657 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
27658 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
27659 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
27660 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
27661 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
27662 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
27663 addr="0x00010734",func="callee4",
27664 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
27665 fullname="/home/foo/devo/gdb/testsuite/gdb.mi/basics.c"line="8",thread-groups=["i1"],
27667 bkpt=@{number="2",type="watchpoint",disp="keep",
27668 enabled="y",addr="",what="C",thread-groups=["i1"],times="0"@}]@}
27673 *stopped,reason="watchpoint-trigger",wpt=@{number="2",exp="C"@},
27674 value=@{old="-276895068",new="3"@},
27675 frame=@{func="callee4",args=[],
27676 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
27677 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="13"@}
27680 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
27681 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
27682 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
27683 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
27684 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
27685 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
27686 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
27687 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
27688 addr="0x00010734",func="callee4",
27689 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
27690 fullname="/home/foo/devo/gdb/testsuite/gdb.mi/basics.c",line="8",thread-groups=["i1"],
27692 bkpt=@{number="2",type="watchpoint",disp="keep",
27693 enabled="y",addr="",what="C",thread-groups=["i1"],times="-5"@}]@}
27697 ^done,reason="watchpoint-scope",wpnum="2",
27698 frame=@{func="callee3",args=[@{name="strarg",
27699 value="0x11940 \"A string argument.\""@}],
27700 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
27701 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="18"@}
27704 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
27705 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
27706 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
27707 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
27708 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
27709 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
27710 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
27711 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
27712 addr="0x00010734",func="callee4",
27713 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
27714 fullname="/home/foo/devo/gdb/testsuite/gdb.mi/basics.c",line="8",
27715 thread-groups=["i1"],times="1"@}]@}
27720 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
27721 @node GDB/MI Catchpoint Commands
27722 @section @sc{gdb/mi} Catchpoint Commands
27724 This section documents @sc{gdb/mi} commands for manipulating
27728 * Shared Library GDB/MI Catchpoint Commands::
27729 * Ada Exception GDB/MI Catchpoint Commands::
27732 @node Shared Library GDB/MI Catchpoint Commands
27733 @subsection Shared Library @sc{gdb/mi} Catchpoints
27735 @subheading The @code{-catch-load} Command
27736 @findex -catch-load
27738 @subsubheading Synopsis
27741 -catch-load [ -t ] [ -d ] @var{regexp}
27744 Add a catchpoint for library load events. If the @samp{-t} option is used,
27745 the catchpoint is a temporary one (@pxref{Set Breaks, ,Setting
27746 Breakpoints}). If the @samp{-d} option is used, the catchpoint is created
27747 in a disabled state. The @samp{regexp} argument is a regular
27748 expression used to match the name of the loaded library.
27751 @subsubheading @value{GDBN} Command
27753 The corresponding @value{GDBN} command is @samp{catch load}.
27755 @subsubheading Example
27758 -catch-load -t foo.so
27759 ^done,bkpt=@{number="1",type="catchpoint",disp="del",enabled="y",
27760 what="load of library matching foo.so",catch-type="load",times="0"@}
27765 @subheading The @code{-catch-unload} Command
27766 @findex -catch-unload
27768 @subsubheading Synopsis
27771 -catch-unload [ -t ] [ -d ] @var{regexp}
27774 Add a catchpoint for library unload events. If the @samp{-t} option is
27775 used, the catchpoint is a temporary one (@pxref{Set Breaks, ,Setting
27776 Breakpoints}). If the @samp{-d} option is used, the catchpoint is
27777 created in a disabled state. The @samp{regexp} argument is a regular
27778 expression used to match the name of the unloaded library.
27780 @subsubheading @value{GDBN} Command
27782 The corresponding @value{GDBN} command is @samp{catch unload}.
27784 @subsubheading Example
27787 -catch-unload -d bar.so
27788 ^done,bkpt=@{number="2",type="catchpoint",disp="keep",enabled="n",
27789 what="load of library matching bar.so",catch-type="unload",times="0"@}
27793 @node Ada Exception GDB/MI Catchpoint Commands
27794 @subsection Ada Exception @sc{gdb/mi} Catchpoints
27796 The following @sc{gdb/mi} commands can be used to create catchpoints
27797 that stop the execution when Ada exceptions are being raised.
27799 @subheading The @code{-catch-assert} Command
27800 @findex -catch-assert
27802 @subsubheading Synopsis
27805 -catch-assert [ -c @var{condition}] [ -d ] [ -t ]
27808 Add a catchpoint for failed Ada assertions.
27810 The possible optional parameters for this command are:
27813 @item -c @var{condition}
27814 Make the catchpoint conditional on @var{condition}.
27816 Create a disabled catchpoint.
27818 Create a temporary catchpoint.
27821 @subsubheading @value{GDBN} Command
27823 The corresponding @value{GDBN} command is @samp{catch assert}.
27825 @subsubheading Example
27829 ^done,bkptno="5",bkpt=@{number="5",type="breakpoint",disp="keep",
27830 enabled="y",addr="0x0000000000404888",what="failed Ada assertions",
27831 thread-groups=["i1"],times="0",
27832 original-location="__gnat_debug_raise_assert_failure"@}
27836 @subheading The @code{-catch-exception} Command
27837 @findex -catch-exception
27839 @subsubheading Synopsis
27842 -catch-exception [ -c @var{condition}] [ -d ] [ -e @var{exception-name} ]
27846 Add a catchpoint stopping when Ada exceptions are raised.
27847 By default, the command stops the program when any Ada exception
27848 gets raised. But it is also possible, by using some of the
27849 optional parameters described below, to create more selective
27852 The possible optional parameters for this command are:
27855 @item -c @var{condition}
27856 Make the catchpoint conditional on @var{condition}.
27858 Create a disabled catchpoint.
27859 @item -e @var{exception-name}
27860 Only stop when @var{exception-name} is raised. This option cannot
27861 be used combined with @samp{-u}.
27863 Create a temporary catchpoint.
27865 Stop only when an unhandled exception gets raised. This option
27866 cannot be used combined with @samp{-e}.
27869 @subsubheading @value{GDBN} Command
27871 The corresponding @value{GDBN} commands are @samp{catch exception}
27872 and @samp{catch exception unhandled}.
27874 @subsubheading Example
27877 -catch-exception -e Program_Error
27878 ^done,bkptno="4",bkpt=@{number="4",type="breakpoint",disp="keep",
27879 enabled="y",addr="0x0000000000404874",
27880 what="`Program_Error' Ada exception", thread-groups=["i1"],
27881 times="0",original-location="__gnat_debug_raise_exception"@}
27885 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
27886 @node GDB/MI Program Context
27887 @section @sc{gdb/mi} Program Context
27889 @subheading The @code{-exec-arguments} Command
27890 @findex -exec-arguments
27893 @subsubheading Synopsis
27896 -exec-arguments @var{args}
27899 Set the inferior program arguments, to be used in the next
27902 @subsubheading @value{GDBN} Command
27904 The corresponding @value{GDBN} command is @samp{set args}.
27906 @subsubheading Example
27910 -exec-arguments -v word
27917 @subheading The @code{-exec-show-arguments} Command
27918 @findex -exec-show-arguments
27920 @subsubheading Synopsis
27923 -exec-show-arguments
27926 Print the arguments of the program.
27928 @subsubheading @value{GDBN} Command
27930 The corresponding @value{GDBN} command is @samp{show args}.
27932 @subsubheading Example
27937 @subheading The @code{-environment-cd} Command
27938 @findex -environment-cd
27940 @subsubheading Synopsis
27943 -environment-cd @var{pathdir}
27946 Set @value{GDBN}'s working directory.
27948 @subsubheading @value{GDBN} Command
27950 The corresponding @value{GDBN} command is @samp{cd}.
27952 @subsubheading Example
27956 -environment-cd /kwikemart/marge/ezannoni/flathead-dev/devo/gdb
27962 @subheading The @code{-environment-directory} Command
27963 @findex -environment-directory
27965 @subsubheading Synopsis
27968 -environment-directory [ -r ] [ @var{pathdir} ]+
27971 Add directories @var{pathdir} to beginning of search path for source files.
27972 If the @samp{-r} option is used, the search path is reset to the default
27973 search path. If directories @var{pathdir} are supplied in addition to the
27974 @samp{-r} option, the search path is first reset and then addition
27976 Multiple directories may be specified, separated by blanks. Specifying
27977 multiple directories in a single command
27978 results in the directories added to the beginning of the
27979 search path in the same order they were presented in the command.
27980 If blanks are needed as
27981 part of a directory name, double-quotes should be used around
27982 the name. In the command output, the path will show up separated
27983 by the system directory-separator character. The directory-separator
27984 character must not be used
27985 in any directory name.
27986 If no directories are specified, the current search path is displayed.
27988 @subsubheading @value{GDBN} Command
27990 The corresponding @value{GDBN} command is @samp{dir}.
27992 @subsubheading Example
27996 -environment-directory /kwikemart/marge/ezannoni/flathead-dev/devo/gdb
27997 ^done,source-path="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb:$cdir:$cwd"
27999 -environment-directory ""
28000 ^done,source-path="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb:$cdir:$cwd"
28002 -environment-directory -r /home/jjohnstn/src/gdb /usr/src
28003 ^done,source-path="/home/jjohnstn/src/gdb:/usr/src:$cdir:$cwd"
28005 -environment-directory -r
28006 ^done,source-path="$cdir:$cwd"
28011 @subheading The @code{-environment-path} Command
28012 @findex -environment-path
28014 @subsubheading Synopsis
28017 -environment-path [ -r ] [ @var{pathdir} ]+
28020 Add directories @var{pathdir} to beginning of search path for object files.
28021 If the @samp{-r} option is used, the search path is reset to the original
28022 search path that existed at gdb start-up. If directories @var{pathdir} are
28023 supplied in addition to the
28024 @samp{-r} option, the search path is first reset and then addition
28026 Multiple directories may be specified, separated by blanks. Specifying
28027 multiple directories in a single command
28028 results in the directories added to the beginning of the
28029 search path in the same order they were presented in the command.
28030 If blanks are needed as
28031 part of a directory name, double-quotes should be used around
28032 the name. In the command output, the path will show up separated
28033 by the system directory-separator character. The directory-separator
28034 character must not be used
28035 in any directory name.
28036 If no directories are specified, the current path is displayed.
28039 @subsubheading @value{GDBN} Command
28041 The corresponding @value{GDBN} command is @samp{path}.
28043 @subsubheading Example
28048 ^done,path="/usr/bin"
28050 -environment-path /kwikemart/marge/ezannoni/flathead-dev/ppc-eabi/gdb /bin
28051 ^done,path="/kwikemart/marge/ezannoni/flathead-dev/ppc-eabi/gdb:/bin:/usr/bin"
28053 -environment-path -r /usr/local/bin
28054 ^done,path="/usr/local/bin:/usr/bin"
28059 @subheading The @code{-environment-pwd} Command
28060 @findex -environment-pwd
28062 @subsubheading Synopsis
28068 Show the current working directory.
28070 @subsubheading @value{GDBN} Command
28072 The corresponding @value{GDBN} command is @samp{pwd}.
28074 @subsubheading Example
28079 ^done,cwd="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb"
28083 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
28084 @node GDB/MI Thread Commands
28085 @section @sc{gdb/mi} Thread Commands
28088 @subheading The @code{-thread-info} Command
28089 @findex -thread-info
28091 @subsubheading Synopsis
28094 -thread-info [ @var{thread-id} ]
28097 Reports information about either a specific thread, if the
28098 @var{thread-id} parameter is present, or about all threads.
28099 @var{thread-id} is the thread's global thread ID. When printing
28100 information about all threads, also reports the global ID of the
28103 @subsubheading @value{GDBN} Command
28105 The @samp{info thread} command prints the same information
28108 @subsubheading Result
28110 The result contains the following attributes:
28114 A list of threads. The format of the elements of the list is described in
28115 @ref{GDB/MI Thread Information}.
28117 @item current-thread-id
28118 The global id of the currently selected thread. This field is omitted if there
28119 is no selected thread (for example, when the selected inferior is not running,
28120 and therefore has no threads) or if a @var{thread-id} argument was passed to
28125 @subsubheading Example
28130 @{id="2",target-id="Thread 0xb7e14b90 (LWP 21257)",
28131 frame=@{level="0",addr="0xffffe410",func="__kernel_vsyscall",
28132 args=[]@},state="running"@},
28133 @{id="1",target-id="Thread 0xb7e156b0 (LWP 21254)",
28134 frame=@{level="0",addr="0x0804891f",func="foo",
28135 args=[@{name="i",value="10"@}],
28136 file="/tmp/a.c",fullname="/tmp/a.c",line="158"@},
28137 state="running"@}],
28138 current-thread-id="1"
28142 @subheading The @code{-thread-list-ids} Command
28143 @findex -thread-list-ids
28145 @subsubheading Synopsis
28151 Produces a list of the currently known global @value{GDBN} thread ids.
28152 At the end of the list it also prints the total number of such
28155 This command is retained for historical reasons, the
28156 @code{-thread-info} command should be used instead.
28158 @subsubheading @value{GDBN} Command
28160 Part of @samp{info threads} supplies the same information.
28162 @subsubheading Example
28167 ^done,thread-ids=@{thread-id="3",thread-id="2",thread-id="1"@},
28168 current-thread-id="1",number-of-threads="3"
28173 @subheading The @code{-thread-select} Command
28174 @findex -thread-select
28176 @subsubheading Synopsis
28179 -thread-select @var{thread-id}
28182 Make thread with global thread number @var{thread-id} the current
28183 thread. It prints the number of the new current thread, and the
28184 topmost frame for that thread.
28186 This command is deprecated in favor of explicitly using the
28187 @samp{--thread} option to each command.
28189 @subsubheading @value{GDBN} Command
28191 The corresponding @value{GDBN} command is @samp{thread}.
28193 @subsubheading Example
28200 *stopped,reason="end-stepping-range",thread-id="2",line="187",
28201 file="../../../devo/gdb/testsuite/gdb.threads/linux-dp.c"
28205 thread-ids=@{thread-id="3",thread-id="2",thread-id="1"@},
28206 number-of-threads="3"
28209 ^done,new-thread-id="3",
28210 frame=@{level="0",func="vprintf",
28211 args=[@{name="format",value="0x8048e9c \"%*s%c %d %c\\n\""@},
28212 @{name="arg",value="0x2"@}],file="vprintf.c",line="31"@}
28216 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
28217 @node GDB/MI Ada Tasking Commands
28218 @section @sc{gdb/mi} Ada Tasking Commands
28220 @subheading The @code{-ada-task-info} Command
28221 @findex -ada-task-info
28223 @subsubheading Synopsis
28226 -ada-task-info [ @var{task-id} ]
28229 Reports information about either a specific Ada task, if the
28230 @var{task-id} parameter is present, or about all Ada tasks.
28232 @subsubheading @value{GDBN} Command
28234 The @samp{info tasks} command prints the same information
28235 about all Ada tasks (@pxref{Ada Tasks}).
28237 @subsubheading Result
28239 The result is a table of Ada tasks. The following columns are
28240 defined for each Ada task:
28244 This field exists only for the current thread. It has the value @samp{*}.
28247 The identifier that @value{GDBN} uses to refer to the Ada task.
28250 The identifier that the target uses to refer to the Ada task.
28253 The global thread identifier of the thread corresponding to the Ada
28256 This field should always exist, as Ada tasks are always implemented
28257 on top of a thread. But if @value{GDBN} cannot find this corresponding
28258 thread for any reason, the field is omitted.
28261 This field exists only when the task was created by another task.
28262 In this case, it provides the ID of the parent task.
28265 The base priority of the task.
28268 The current state of the task. For a detailed description of the
28269 possible states, see @ref{Ada Tasks}.
28272 The name of the task.
28276 @subsubheading Example
28280 ^done,tasks=@{nr_rows="3",nr_cols="8",
28281 hdr=[@{width="1",alignment="-1",col_name="current",colhdr=""@},
28282 @{width="3",alignment="1",col_name="id",colhdr="ID"@},
28283 @{width="9",alignment="1",col_name="task-id",colhdr="TID"@},
28284 @{width="4",alignment="1",col_name="thread-id",colhdr=""@},
28285 @{width="4",alignment="1",col_name="parent-id",colhdr="P-ID"@},
28286 @{width="3",alignment="1",col_name="priority",colhdr="Pri"@},
28287 @{width="22",alignment="-1",col_name="state",colhdr="State"@},
28288 @{width="1",alignment="2",col_name="name",colhdr="Name"@}],
28289 body=[@{current="*",id="1",task-id=" 644010",thread-id="1",priority="48",
28290 state="Child Termination Wait",name="main_task"@}]@}
28294 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
28295 @node GDB/MI Program Execution
28296 @section @sc{gdb/mi} Program Execution
28298 These are the asynchronous commands which generate the out-of-band
28299 record @samp{*stopped}. Currently @value{GDBN} only really executes
28300 asynchronously with remote targets and this interaction is mimicked in
28303 @subheading The @code{-exec-continue} Command
28304 @findex -exec-continue
28306 @subsubheading Synopsis
28309 -exec-continue [--reverse] [--all|--thread-group N]
28312 Resumes the execution of the inferior program, which will continue
28313 to execute until it reaches a debugger stop event. If the
28314 @samp{--reverse} option is specified, execution resumes in reverse until
28315 it reaches a stop event. Stop events may include
28318 breakpoints or watchpoints
28320 signals or exceptions
28322 the end of the process (or its beginning under @samp{--reverse})
28324 the end or beginning of a replay log if one is being used.
28326 In all-stop mode (@pxref{All-Stop
28327 Mode}), may resume only one thread, or all threads, depending on the
28328 value of the @samp{scheduler-locking} variable. If @samp{--all} is
28329 specified, all threads (in all inferiors) will be resumed. The @samp{--all} option is
28330 ignored in all-stop mode. If the @samp{--thread-group} options is
28331 specified, then all threads in that thread group are resumed.
28333 @subsubheading @value{GDBN} Command
28335 The corresponding @value{GDBN} corresponding is @samp{continue}.
28337 @subsubheading Example
28344 *stopped,reason="breakpoint-hit",disp="keep",bkptno="2",frame=@{
28345 func="foo",args=[],file="hello.c",fullname="/home/foo/bar/hello.c",
28351 @subheading The @code{-exec-finish} Command
28352 @findex -exec-finish
28354 @subsubheading Synopsis
28357 -exec-finish [--reverse]
28360 Resumes the execution of the inferior program until the current
28361 function is exited. Displays the results returned by the function.
28362 If the @samp{--reverse} option is specified, resumes the reverse
28363 execution of the inferior program until the point where current
28364 function was called.
28366 @subsubheading @value{GDBN} Command
28368 The corresponding @value{GDBN} command is @samp{finish}.
28370 @subsubheading Example
28372 Function returning @code{void}.
28379 *stopped,reason="function-finished",frame=@{func="main",args=[],
28380 file="hello.c",fullname="/home/foo/bar/hello.c",line="7"@}
28384 Function returning other than @code{void}. The name of the internal
28385 @value{GDBN} variable storing the result is printed, together with the
28392 *stopped,reason="function-finished",frame=@{addr="0x000107b0",func="foo",
28393 args=[@{name="a",value="1"],@{name="b",value="9"@}@},
28394 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
28395 gdb-result-var="$1",return-value="0"
28400 @subheading The @code{-exec-interrupt} Command
28401 @findex -exec-interrupt
28403 @subsubheading Synopsis
28406 -exec-interrupt [--all|--thread-group N]
28409 Interrupts the background execution of the target. Note how the token
28410 associated with the stop message is the one for the execution command
28411 that has been interrupted. The token for the interrupt itself only
28412 appears in the @samp{^done} output. If the user is trying to
28413 interrupt a non-running program, an error message will be printed.
28415 Note that when asynchronous execution is enabled, this command is
28416 asynchronous just like other execution commands. That is, first the
28417 @samp{^done} response will be printed, and the target stop will be
28418 reported after that using the @samp{*stopped} notification.
28420 In non-stop mode, only the context thread is interrupted by default.
28421 All threads (in all inferiors) will be interrupted if the
28422 @samp{--all} option is specified. If the @samp{--thread-group}
28423 option is specified, all threads in that group will be interrupted.
28425 @subsubheading @value{GDBN} Command
28427 The corresponding @value{GDBN} command is @samp{interrupt}.
28429 @subsubheading Example
28440 111*stopped,signal-name="SIGINT",signal-meaning="Interrupt",
28441 frame=@{addr="0x00010140",func="foo",args=[],file="try.c",
28442 fullname="/home/foo/bar/try.c",line="13"@}
28447 ^error,msg="mi_cmd_exec_interrupt: Inferior not executing."
28451 @subheading The @code{-exec-jump} Command
28454 @subsubheading Synopsis
28457 -exec-jump @var{location}
28460 Resumes execution of the inferior program at the location specified by
28461 parameter. @xref{Specify Location}, for a description of the
28462 different forms of @var{location}.
28464 @subsubheading @value{GDBN} Command
28466 The corresponding @value{GDBN} command is @samp{jump}.
28468 @subsubheading Example
28471 -exec-jump foo.c:10
28472 *running,thread-id="all"
28477 @subheading The @code{-exec-next} Command
28480 @subsubheading Synopsis
28483 -exec-next [--reverse]
28486 Resumes execution of the inferior program, stopping when the beginning
28487 of the next source line is reached.
28489 If the @samp{--reverse} option is specified, resumes reverse execution
28490 of the inferior program, stopping at the beginning of the previous
28491 source line. If you issue this command on the first line of a
28492 function, it will take you back to the caller of that function, to the
28493 source line where the function was called.
28496 @subsubheading @value{GDBN} Command
28498 The corresponding @value{GDBN} command is @samp{next}.
28500 @subsubheading Example
28506 *stopped,reason="end-stepping-range",line="8",file="hello.c"
28511 @subheading The @code{-exec-next-instruction} Command
28512 @findex -exec-next-instruction
28514 @subsubheading Synopsis
28517 -exec-next-instruction [--reverse]
28520 Executes one machine instruction. If the instruction is a function
28521 call, continues until the function returns. If the program stops at an
28522 instruction in the middle of a source line, the address will be
28525 If the @samp{--reverse} option is specified, resumes reverse execution
28526 of the inferior program, stopping at the previous instruction. If the
28527 previously executed instruction was a return from another function,
28528 it will continue to execute in reverse until the call to that function
28529 (from the current stack frame) is reached.
28531 @subsubheading @value{GDBN} Command
28533 The corresponding @value{GDBN} command is @samp{nexti}.
28535 @subsubheading Example
28539 -exec-next-instruction
28543 *stopped,reason="end-stepping-range",
28544 addr="0x000100d4",line="5",file="hello.c"
28549 @subheading The @code{-exec-return} Command
28550 @findex -exec-return
28552 @subsubheading Synopsis
28558 Makes current function return immediately. Doesn't execute the inferior.
28559 Displays the new current frame.
28561 @subsubheading @value{GDBN} Command
28563 The corresponding @value{GDBN} command is @samp{return}.
28565 @subsubheading Example
28569 200-break-insert callee4
28570 200^done,bkpt=@{number="1",addr="0x00010734",
28571 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",line="8"@}
28576 000*stopped,reason="breakpoint-hit",disp="keep",bkptno="1",
28577 frame=@{func="callee4",args=[],
28578 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
28579 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="8"@}
28585 111^done,frame=@{level="0",func="callee3",
28586 args=[@{name="strarg",
28587 value="0x11940 \"A string argument.\""@}],
28588 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
28589 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="18"@}
28594 @subheading The @code{-exec-run} Command
28597 @subsubheading Synopsis
28600 -exec-run [ --all | --thread-group N ] [ --start ]
28603 Starts execution of the inferior from the beginning. The inferior
28604 executes until either a breakpoint is encountered or the program
28605 exits. In the latter case the output will include an exit code, if
28606 the program has exited exceptionally.
28608 When neither the @samp{--all} nor the @samp{--thread-group} option
28609 is specified, the current inferior is started. If the
28610 @samp{--thread-group} option is specified, it should refer to a thread
28611 group of type @samp{process}, and that thread group will be started.
28612 If the @samp{--all} option is specified, then all inferiors will be started.
28614 Using the @samp{--start} option instructs the debugger to stop
28615 the execution at the start of the inferior's main subprogram,
28616 following the same behavior as the @code{start} command
28617 (@pxref{Starting}).
28619 @subsubheading @value{GDBN} Command
28621 The corresponding @value{GDBN} command is @samp{run}.
28623 @subsubheading Examples
28628 ^done,bkpt=@{number="1",addr="0x0001072c",file="recursive2.c",line="4"@}
28633 *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",
28634 frame=@{func="main",args=[],file="recursive2.c",
28635 fullname="/home/foo/bar/recursive2.c",line="4"@}
28640 Program exited normally:
28648 *stopped,reason="exited-normally"
28653 Program exited exceptionally:
28661 *stopped,reason="exited",exit-code="01"
28665 Another way the program can terminate is if it receives a signal such as
28666 @code{SIGINT}. In this case, @sc{gdb/mi} displays this:
28670 *stopped,reason="exited-signalled",signal-name="SIGINT",
28671 signal-meaning="Interrupt"
28675 @c @subheading -exec-signal
28678 @subheading The @code{-exec-step} Command
28681 @subsubheading Synopsis
28684 -exec-step [--reverse]
28687 Resumes execution of the inferior program, stopping when the beginning
28688 of the next source line is reached, if the next source line is not a
28689 function call. If it is, stop at the first instruction of the called
28690 function. If the @samp{--reverse} option is specified, resumes reverse
28691 execution of the inferior program, stopping at the beginning of the
28692 previously executed source line.
28694 @subsubheading @value{GDBN} Command
28696 The corresponding @value{GDBN} command is @samp{step}.
28698 @subsubheading Example
28700 Stepping into a function:
28706 *stopped,reason="end-stepping-range",
28707 frame=@{func="foo",args=[@{name="a",value="10"@},
28708 @{name="b",value="0"@}],file="recursive2.c",
28709 fullname="/home/foo/bar/recursive2.c",line="11"@}
28719 *stopped,reason="end-stepping-range",line="14",file="recursive2.c"
28724 @subheading The @code{-exec-step-instruction} Command
28725 @findex -exec-step-instruction
28727 @subsubheading Synopsis
28730 -exec-step-instruction [--reverse]
28733 Resumes the inferior which executes one machine instruction. If the
28734 @samp{--reverse} option is specified, resumes reverse execution of the
28735 inferior program, stopping at the previously executed instruction.
28736 The output, once @value{GDBN} has stopped, will vary depending on
28737 whether we have stopped in the middle of a source line or not. In the
28738 former case, the address at which the program stopped will be printed
28741 @subsubheading @value{GDBN} Command
28743 The corresponding @value{GDBN} command is @samp{stepi}.
28745 @subsubheading Example
28749 -exec-step-instruction
28753 *stopped,reason="end-stepping-range",
28754 frame=@{func="foo",args=[],file="try.c",
28755 fullname="/home/foo/bar/try.c",line="10"@}
28757 -exec-step-instruction
28761 *stopped,reason="end-stepping-range",
28762 frame=@{addr="0x000100f4",func="foo",args=[],file="try.c",
28763 fullname="/home/foo/bar/try.c",line="10"@}
28768 @subheading The @code{-exec-until} Command
28769 @findex -exec-until
28771 @subsubheading Synopsis
28774 -exec-until [ @var{location} ]
28777 Executes the inferior until the @var{location} specified in the
28778 argument is reached. If there is no argument, the inferior executes
28779 until a source line greater than the current one is reached. The
28780 reason for stopping in this case will be @samp{location-reached}.
28782 @subsubheading @value{GDBN} Command
28784 The corresponding @value{GDBN} command is @samp{until}.
28786 @subsubheading Example
28790 -exec-until recursive2.c:6
28794 *stopped,reason="location-reached",frame=@{func="main",args=[],
28795 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="6"@}
28800 @subheading -file-clear
28801 Is this going away????
28804 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
28805 @node GDB/MI Stack Manipulation
28806 @section @sc{gdb/mi} Stack Manipulation Commands
28808 @subheading The @code{-enable-frame-filters} Command
28809 @findex -enable-frame-filters
28812 -enable-frame-filters
28815 @value{GDBN} allows Python-based frame filters to affect the output of
28816 the MI commands relating to stack traces. As there is no way to
28817 implement this in a fully backward-compatible way, a front end must
28818 request that this functionality be enabled.
28820 Once enabled, this feature cannot be disabled.
28822 Note that if Python support has not been compiled into @value{GDBN},
28823 this command will still succeed (and do nothing).
28825 @subheading The @code{-stack-info-frame} Command
28826 @findex -stack-info-frame
28828 @subsubheading Synopsis
28834 Get info on the selected frame.
28836 @subsubheading @value{GDBN} Command
28838 The corresponding @value{GDBN} command is @samp{info frame} or @samp{frame}
28839 (without arguments).
28841 @subsubheading Example
28846 ^done,frame=@{level="1",addr="0x0001076c",func="callee3",
28847 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
28848 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="17"@}
28852 @subheading The @code{-stack-info-depth} Command
28853 @findex -stack-info-depth
28855 @subsubheading Synopsis
28858 -stack-info-depth [ @var{max-depth} ]
28861 Return the depth of the stack. If the integer argument @var{max-depth}
28862 is specified, do not count beyond @var{max-depth} frames.
28864 @subsubheading @value{GDBN} Command
28866 There's no equivalent @value{GDBN} command.
28868 @subsubheading Example
28870 For a stack with frame levels 0 through 11:
28877 -stack-info-depth 4
28880 -stack-info-depth 12
28883 -stack-info-depth 11
28886 -stack-info-depth 13
28891 @anchor{-stack-list-arguments}
28892 @subheading The @code{-stack-list-arguments} Command
28893 @findex -stack-list-arguments
28895 @subsubheading Synopsis
28898 -stack-list-arguments [ --no-frame-filters ] [ --skip-unavailable ] @var{print-values}
28899 [ @var{low-frame} @var{high-frame} ]
28902 Display a list of the arguments for the frames between @var{low-frame}
28903 and @var{high-frame} (inclusive). If @var{low-frame} and
28904 @var{high-frame} are not provided, list the arguments for the whole
28905 call stack. If the two arguments are equal, show the single frame
28906 at the corresponding level. It is an error if @var{low-frame} is
28907 larger than the actual number of frames. On the other hand,
28908 @var{high-frame} may be larger than the actual number of frames, in
28909 which case only existing frames will be returned.
28911 If @var{print-values} is 0 or @code{--no-values}, print only the names of
28912 the variables; if it is 1 or @code{--all-values}, print also their
28913 values; and if it is 2 or @code{--simple-values}, print the name,
28914 type and value for simple data types, and the name and type for arrays,
28915 structures and unions. If the option @code{--no-frame-filters} is
28916 supplied, then Python frame filters will not be executed.
28918 If the @code{--skip-unavailable} option is specified, arguments that
28919 are not available are not listed. Partially available arguments
28920 are still displayed, however.
28922 Use of this command to obtain arguments in a single frame is
28923 deprecated in favor of the @samp{-stack-list-variables} command.
28925 @subsubheading @value{GDBN} Command
28927 @value{GDBN} does not have an equivalent command. @code{gdbtk} has a
28928 @samp{gdb_get_args} command which partially overlaps with the
28929 functionality of @samp{-stack-list-arguments}.
28931 @subsubheading Example
28938 frame=@{level="0",addr="0x00010734",func="callee4",
28939 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
28940 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="8"@},
28941 frame=@{level="1",addr="0x0001076c",func="callee3",
28942 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
28943 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="17"@},
28944 frame=@{level="2",addr="0x0001078c",func="callee2",
28945 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
28946 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="22"@},
28947 frame=@{level="3",addr="0x000107b4",func="callee1",
28948 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
28949 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="27"@},
28950 frame=@{level="4",addr="0x000107e0",func="main",
28951 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
28952 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="32"@}]
28954 -stack-list-arguments 0
28957 frame=@{level="0",args=[]@},
28958 frame=@{level="1",args=[name="strarg"]@},
28959 frame=@{level="2",args=[name="intarg",name="strarg"]@},
28960 frame=@{level="3",args=[name="intarg",name="strarg",name="fltarg"]@},
28961 frame=@{level="4",args=[]@}]
28963 -stack-list-arguments 1
28966 frame=@{level="0",args=[]@},
28968 args=[@{name="strarg",value="0x11940 \"A string argument.\""@}]@},
28969 frame=@{level="2",args=[
28970 @{name="intarg",value="2"@},
28971 @{name="strarg",value="0x11940 \"A string argument.\""@}]@},
28972 @{frame=@{level="3",args=[
28973 @{name="intarg",value="2"@},
28974 @{name="strarg",value="0x11940 \"A string argument.\""@},
28975 @{name="fltarg",value="3.5"@}]@},
28976 frame=@{level="4",args=[]@}]
28978 -stack-list-arguments 0 2 2
28979 ^done,stack-args=[frame=@{level="2",args=[name="intarg",name="strarg"]@}]
28981 -stack-list-arguments 1 2 2
28982 ^done,stack-args=[frame=@{level="2",
28983 args=[@{name="intarg",value="2"@},
28984 @{name="strarg",value="0x11940 \"A string argument.\""@}]@}]
28988 @c @subheading -stack-list-exception-handlers
28991 @anchor{-stack-list-frames}
28992 @subheading The @code{-stack-list-frames} Command
28993 @findex -stack-list-frames
28995 @subsubheading Synopsis
28998 -stack-list-frames [ --no-frame-filters @var{low-frame} @var{high-frame} ]
29001 List the frames currently on the stack. For each frame it displays the
29006 The frame number, 0 being the topmost frame, i.e., the innermost function.
29008 The @code{$pc} value for that frame.
29012 File name of the source file where the function lives.
29013 @item @var{fullname}
29014 The full file name of the source file where the function lives.
29016 Line number corresponding to the @code{$pc}.
29018 The shared library where this function is defined. This is only given
29019 if the frame's function is not known.
29022 If invoked without arguments, this command prints a backtrace for the
29023 whole stack. If given two integer arguments, it shows the frames whose
29024 levels are between the two arguments (inclusive). If the two arguments
29025 are equal, it shows the single frame at the corresponding level. It is
29026 an error if @var{low-frame} is larger than the actual number of
29027 frames. On the other hand, @var{high-frame} may be larger than the
29028 actual number of frames, in which case only existing frames will be
29029 returned. If the option @code{--no-frame-filters} is supplied, then
29030 Python frame filters will not be executed.
29032 @subsubheading @value{GDBN} Command
29034 The corresponding @value{GDBN} commands are @samp{backtrace} and @samp{where}.
29036 @subsubheading Example
29038 Full stack backtrace:
29044 [frame=@{level="0",addr="0x0001076c",func="foo",
29045 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="11"@},
29046 frame=@{level="1",addr="0x000107a4",func="foo",
29047 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
29048 frame=@{level="2",addr="0x000107a4",func="foo",
29049 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
29050 frame=@{level="3",addr="0x000107a4",func="foo",
29051 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
29052 frame=@{level="4",addr="0x000107a4",func="foo",
29053 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
29054 frame=@{level="5",addr="0x000107a4",func="foo",
29055 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
29056 frame=@{level="6",addr="0x000107a4",func="foo",
29057 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
29058 frame=@{level="7",addr="0x000107a4",func="foo",
29059 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
29060 frame=@{level="8",addr="0x000107a4",func="foo",
29061 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
29062 frame=@{level="9",addr="0x000107a4",func="foo",
29063 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
29064 frame=@{level="10",addr="0x000107a4",func="foo",
29065 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
29066 frame=@{level="11",addr="0x00010738",func="main",
29067 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="4"@}]
29071 Show frames between @var{low_frame} and @var{high_frame}:
29075 -stack-list-frames 3 5
29077 [frame=@{level="3",addr="0x000107a4",func="foo",
29078 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
29079 frame=@{level="4",addr="0x000107a4",func="foo",
29080 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
29081 frame=@{level="5",addr="0x000107a4",func="foo",
29082 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@}]
29086 Show a single frame:
29090 -stack-list-frames 3 3
29092 [frame=@{level="3",addr="0x000107a4",func="foo",
29093 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@}]
29098 @subheading The @code{-stack-list-locals} Command
29099 @findex -stack-list-locals
29100 @anchor{-stack-list-locals}
29102 @subsubheading Synopsis
29105 -stack-list-locals [ --no-frame-filters ] [ --skip-unavailable ] @var{print-values}
29108 Display the local variable names for the selected frame. If
29109 @var{print-values} is 0 or @code{--no-values}, print only the names of
29110 the variables; if it is 1 or @code{--all-values}, print also their
29111 values; and if it is 2 or @code{--simple-values}, print the name,
29112 type and value for simple data types, and the name and type for arrays,
29113 structures and unions. In this last case, a frontend can immediately
29114 display the value of simple data types and create variable objects for
29115 other data types when the user wishes to explore their values in
29116 more detail. If the option @code{--no-frame-filters} is supplied, then
29117 Python frame filters will not be executed.
29119 If the @code{--skip-unavailable} option is specified, local variables
29120 that are not available are not listed. Partially available local
29121 variables are still displayed, however.
29123 This command is deprecated in favor of the
29124 @samp{-stack-list-variables} command.
29126 @subsubheading @value{GDBN} Command
29128 @samp{info locals} in @value{GDBN}, @samp{gdb_get_locals} in @code{gdbtk}.
29130 @subsubheading Example
29134 -stack-list-locals 0
29135 ^done,locals=[name="A",name="B",name="C"]
29137 -stack-list-locals --all-values
29138 ^done,locals=[@{name="A",value="1"@},@{name="B",value="2"@},
29139 @{name="C",value="@{1, 2, 3@}"@}]
29140 -stack-list-locals --simple-values
29141 ^done,locals=[@{name="A",type="int",value="1"@},
29142 @{name="B",type="int",value="2"@},@{name="C",type="int [3]"@}]
29146 @anchor{-stack-list-variables}
29147 @subheading The @code{-stack-list-variables} Command
29148 @findex -stack-list-variables
29150 @subsubheading Synopsis
29153 -stack-list-variables [ --no-frame-filters ] [ --skip-unavailable ] @var{print-values}
29156 Display the names of local variables and function arguments for the selected frame. If
29157 @var{print-values} is 0 or @code{--no-values}, print only the names of
29158 the variables; if it is 1 or @code{--all-values}, print also their
29159 values; and if it is 2 or @code{--simple-values}, print the name,
29160 type and value for simple data types, and the name and type for arrays,
29161 structures and unions. If the option @code{--no-frame-filters} is
29162 supplied, then Python frame filters will not be executed.
29164 If the @code{--skip-unavailable} option is specified, local variables
29165 and arguments that are not available are not listed. Partially
29166 available arguments and local variables are still displayed, however.
29168 @subsubheading Example
29172 -stack-list-variables --thread 1 --frame 0 --all-values
29173 ^done,variables=[@{name="x",value="11"@},@{name="s",value="@{a = 1, b = 2@}"@}]
29178 @subheading The @code{-stack-select-frame} Command
29179 @findex -stack-select-frame
29181 @subsubheading Synopsis
29184 -stack-select-frame @var{framenum}
29187 Change the selected frame. Select a different frame @var{framenum} on
29190 This command in deprecated in favor of passing the @samp{--frame}
29191 option to every command.
29193 @subsubheading @value{GDBN} Command
29195 The corresponding @value{GDBN} commands are @samp{frame}, @samp{up},
29196 @samp{down}, @samp{select-frame}, @samp{up-silent}, and @samp{down-silent}.
29198 @subsubheading Example
29202 -stack-select-frame 2
29207 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
29208 @node GDB/MI Variable Objects
29209 @section @sc{gdb/mi} Variable Objects
29213 @subheading Motivation for Variable Objects in @sc{gdb/mi}
29215 For the implementation of a variable debugger window (locals, watched
29216 expressions, etc.), we are proposing the adaptation of the existing code
29217 used by @code{Insight}.
29219 The two main reasons for that are:
29223 It has been proven in practice (it is already on its second generation).
29226 It will shorten development time (needless to say how important it is
29230 The original interface was designed to be used by Tcl code, so it was
29231 slightly changed so it could be used through @sc{gdb/mi}. This section
29232 describes the @sc{gdb/mi} operations that will be available and gives some
29233 hints about their use.
29235 @emph{Note}: In addition to the set of operations described here, we
29236 expect the @sc{gui} implementation of a variable window to require, at
29237 least, the following operations:
29240 @item @code{-gdb-show} @code{output-radix}
29241 @item @code{-stack-list-arguments}
29242 @item @code{-stack-list-locals}
29243 @item @code{-stack-select-frame}
29248 @subheading Introduction to Variable Objects
29250 @cindex variable objects in @sc{gdb/mi}
29252 Variable objects are "object-oriented" MI interface for examining and
29253 changing values of expressions. Unlike some other MI interfaces that
29254 work with expressions, variable objects are specifically designed for
29255 simple and efficient presentation in the frontend. A variable object
29256 is identified by string name. When a variable object is created, the
29257 frontend specifies the expression for that variable object. The
29258 expression can be a simple variable, or it can be an arbitrary complex
29259 expression, and can even involve CPU registers. After creating a
29260 variable object, the frontend can invoke other variable object
29261 operations---for example to obtain or change the value of a variable
29262 object, or to change display format.
29264 Variable objects have hierarchical tree structure. Any variable object
29265 that corresponds to a composite type, such as structure in C, has
29266 a number of child variable objects, for example corresponding to each
29267 element of a structure. A child variable object can itself have
29268 children, recursively. Recursion ends when we reach
29269 leaf variable objects, which always have built-in types. Child variable
29270 objects are created only by explicit request, so if a frontend
29271 is not interested in the children of a particular variable object, no
29272 child will be created.
29274 For a leaf variable object it is possible to obtain its value as a
29275 string, or set the value from a string. String value can be also
29276 obtained for a non-leaf variable object, but it's generally a string
29277 that only indicates the type of the object, and does not list its
29278 contents. Assignment to a non-leaf variable object is not allowed.
29280 A frontend does not need to read the values of all variable objects each time
29281 the program stops. Instead, MI provides an update command that lists all
29282 variable objects whose values has changed since the last update
29283 operation. This considerably reduces the amount of data that must
29284 be transferred to the frontend. As noted above, children variable
29285 objects are created on demand, and only leaf variable objects have a
29286 real value. As result, gdb will read target memory only for leaf
29287 variables that frontend has created.
29289 The automatic update is not always desirable. For example, a frontend
29290 might want to keep a value of some expression for future reference,
29291 and never update it. For another example, fetching memory is
29292 relatively slow for embedded targets, so a frontend might want
29293 to disable automatic update for the variables that are either not
29294 visible on the screen, or ``closed''. This is possible using so
29295 called ``frozen variable objects''. Such variable objects are never
29296 implicitly updated.
29298 Variable objects can be either @dfn{fixed} or @dfn{floating}. For the
29299 fixed variable object, the expression is parsed when the variable
29300 object is created, including associating identifiers to specific
29301 variables. The meaning of expression never changes. For a floating
29302 variable object the values of variables whose names appear in the
29303 expressions are re-evaluated every time in the context of the current
29304 frame. Consider this example:
29309 struct work_state state;
29316 If a fixed variable object for the @code{state} variable is created in
29317 this function, and we enter the recursive call, the variable
29318 object will report the value of @code{state} in the top-level
29319 @code{do_work} invocation. On the other hand, a floating variable
29320 object will report the value of @code{state} in the current frame.
29322 If an expression specified when creating a fixed variable object
29323 refers to a local variable, the variable object becomes bound to the
29324 thread and frame in which the variable object is created. When such
29325 variable object is updated, @value{GDBN} makes sure that the
29326 thread/frame combination the variable object is bound to still exists,
29327 and re-evaluates the variable object in context of that thread/frame.
29329 The following is the complete set of @sc{gdb/mi} operations defined to
29330 access this functionality:
29332 @multitable @columnfractions .4 .6
29333 @item @strong{Operation}
29334 @tab @strong{Description}
29336 @item @code{-enable-pretty-printing}
29337 @tab enable Python-based pretty-printing
29338 @item @code{-var-create}
29339 @tab create a variable object
29340 @item @code{-var-delete}
29341 @tab delete the variable object and/or its children
29342 @item @code{-var-set-format}
29343 @tab set the display format of this variable
29344 @item @code{-var-show-format}
29345 @tab show the display format of this variable
29346 @item @code{-var-info-num-children}
29347 @tab tells how many children this object has
29348 @item @code{-var-list-children}
29349 @tab return a list of the object's children
29350 @item @code{-var-info-type}
29351 @tab show the type of this variable object
29352 @item @code{-var-info-expression}
29353 @tab print parent-relative expression that this variable object represents
29354 @item @code{-var-info-path-expression}
29355 @tab print full expression that this variable object represents
29356 @item @code{-var-show-attributes}
29357 @tab is this variable editable? does it exist here?
29358 @item @code{-var-evaluate-expression}
29359 @tab get the value of this variable
29360 @item @code{-var-assign}
29361 @tab set the value of this variable
29362 @item @code{-var-update}
29363 @tab update the variable and its children
29364 @item @code{-var-set-frozen}
29365 @tab set frozeness attribute
29366 @item @code{-var-set-update-range}
29367 @tab set range of children to display on update
29370 In the next subsection we describe each operation in detail and suggest
29371 how it can be used.
29373 @subheading Description And Use of Operations on Variable Objects
29375 @subheading The @code{-enable-pretty-printing} Command
29376 @findex -enable-pretty-printing
29379 -enable-pretty-printing
29382 @value{GDBN} allows Python-based visualizers to affect the output of the
29383 MI variable object commands. However, because there was no way to
29384 implement this in a fully backward-compatible way, a front end must
29385 request that this functionality be enabled.
29387 Once enabled, this feature cannot be disabled.
29389 Note that if Python support has not been compiled into @value{GDBN},
29390 this command will still succeed (and do nothing).
29392 This feature is currently (as of @value{GDBN} 7.0) experimental, and
29393 may work differently in future versions of @value{GDBN}.
29395 @subheading The @code{-var-create} Command
29396 @findex -var-create
29398 @subsubheading Synopsis
29401 -var-create @{@var{name} | "-"@}
29402 @{@var{frame-addr} | "*" | "@@"@} @var{expression}
29405 This operation creates a variable object, which allows the monitoring of
29406 a variable, the result of an expression, a memory cell or a CPU
29409 The @var{name} parameter is the string by which the object can be
29410 referenced. It must be unique. If @samp{-} is specified, the varobj
29411 system will generate a string ``varNNNNNN'' automatically. It will be
29412 unique provided that one does not specify @var{name} of that format.
29413 The command fails if a duplicate name is found.
29415 The frame under which the expression should be evaluated can be
29416 specified by @var{frame-addr}. A @samp{*} indicates that the current
29417 frame should be used. A @samp{@@} indicates that a floating variable
29418 object must be created.
29420 @var{expression} is any expression valid on the current language set (must not
29421 begin with a @samp{*}), or one of the following:
29425 @samp{*@var{addr}}, where @var{addr} is the address of a memory cell
29428 @samp{*@var{addr}-@var{addr}} --- a memory address range (TBD)
29431 @samp{$@var{regname}} --- a CPU register name
29434 @cindex dynamic varobj
29435 A varobj's contents may be provided by a Python-based pretty-printer. In this
29436 case the varobj is known as a @dfn{dynamic varobj}. Dynamic varobjs
29437 have slightly different semantics in some cases. If the
29438 @code{-enable-pretty-printing} command is not sent, then @value{GDBN}
29439 will never create a dynamic varobj. This ensures backward
29440 compatibility for existing clients.
29442 @subsubheading Result
29444 This operation returns attributes of the newly-created varobj. These
29449 The name of the varobj.
29452 The number of children of the varobj. This number is not necessarily
29453 reliable for a dynamic varobj. Instead, you must examine the
29454 @samp{has_more} attribute.
29457 The varobj's scalar value. For a varobj whose type is some sort of
29458 aggregate (e.g., a @code{struct}), or for a dynamic varobj, this value
29459 will not be interesting.
29462 The varobj's type. This is a string representation of the type, as
29463 would be printed by the @value{GDBN} CLI. If @samp{print object}
29464 (@pxref{Print Settings, set print object}) is set to @code{on}, the
29465 @emph{actual} (derived) type of the object is shown rather than the
29466 @emph{declared} one.
29469 If a variable object is bound to a specific thread, then this is the
29470 thread's global identifier.
29473 For a dynamic varobj, this indicates whether there appear to be any
29474 children available. For a non-dynamic varobj, this will be 0.
29477 This attribute will be present and have the value @samp{1} if the
29478 varobj is a dynamic varobj. If the varobj is not a dynamic varobj,
29479 then this attribute will not be present.
29482 A dynamic varobj can supply a display hint to the front end. The
29483 value comes directly from the Python pretty-printer object's
29484 @code{display_hint} method. @xref{Pretty Printing API}.
29487 Typical output will look like this:
29490 name="@var{name}",numchild="@var{N}",type="@var{type}",thread-id="@var{M}",
29491 has_more="@var{has_more}"
29495 @subheading The @code{-var-delete} Command
29496 @findex -var-delete
29498 @subsubheading Synopsis
29501 -var-delete [ -c ] @var{name}
29504 Deletes a previously created variable object and all of its children.
29505 With the @samp{-c} option, just deletes the children.
29507 Returns an error if the object @var{name} is not found.
29510 @subheading The @code{-var-set-format} Command
29511 @findex -var-set-format
29513 @subsubheading Synopsis
29516 -var-set-format @var{name} @var{format-spec}
29519 Sets the output format for the value of the object @var{name} to be
29522 @anchor{-var-set-format}
29523 The syntax for the @var{format-spec} is as follows:
29526 @var{format-spec} @expansion{}
29527 @{binary | decimal | hexadecimal | octal | natural | zero-hexadecimal@}
29530 The natural format is the default format choosen automatically
29531 based on the variable type (like decimal for an @code{int}, hex
29532 for pointers, etc.).
29534 The zero-hexadecimal format has a representation similar to hexadecimal
29535 but with padding zeroes to the left of the value. For example, a 32-bit
29536 hexadecimal value of 0x1234 would be represented as 0x00001234 in the
29537 zero-hexadecimal format.
29539 For a variable with children, the format is set only on the
29540 variable itself, and the children are not affected.
29542 @subheading The @code{-var-show-format} Command
29543 @findex -var-show-format
29545 @subsubheading Synopsis
29548 -var-show-format @var{name}
29551 Returns the format used to display the value of the object @var{name}.
29554 @var{format} @expansion{}
29559 @subheading The @code{-var-info-num-children} Command
29560 @findex -var-info-num-children
29562 @subsubheading Synopsis
29565 -var-info-num-children @var{name}
29568 Returns the number of children of a variable object @var{name}:
29574 Note that this number is not completely reliable for a dynamic varobj.
29575 It will return the current number of children, but more children may
29579 @subheading The @code{-var-list-children} Command
29580 @findex -var-list-children
29582 @subsubheading Synopsis
29585 -var-list-children [@var{print-values}] @var{name} [@var{from} @var{to}]
29587 @anchor{-var-list-children}
29589 Return a list of the children of the specified variable object and
29590 create variable objects for them, if they do not already exist. With
29591 a single argument or if @var{print-values} has a value of 0 or
29592 @code{--no-values}, print only the names of the variables; if
29593 @var{print-values} is 1 or @code{--all-values}, also print their
29594 values; and if it is 2 or @code{--simple-values} print the name and
29595 value for simple data types and just the name for arrays, structures
29598 @var{from} and @var{to}, if specified, indicate the range of children
29599 to report. If @var{from} or @var{to} is less than zero, the range is
29600 reset and all children will be reported. Otherwise, children starting
29601 at @var{from} (zero-based) and up to and excluding @var{to} will be
29604 If a child range is requested, it will only affect the current call to
29605 @code{-var-list-children}, but not future calls to @code{-var-update}.
29606 For this, you must instead use @code{-var-set-update-range}. The
29607 intent of this approach is to enable a front end to implement any
29608 update approach it likes; for example, scrolling a view may cause the
29609 front end to request more children with @code{-var-list-children}, and
29610 then the front end could call @code{-var-set-update-range} with a
29611 different range to ensure that future updates are restricted to just
29614 For each child the following results are returned:
29619 Name of the variable object created for this child.
29622 The expression to be shown to the user by the front end to designate this child.
29623 For example this may be the name of a structure member.
29625 For a dynamic varobj, this value cannot be used to form an
29626 expression. There is no way to do this at all with a dynamic varobj.
29628 For C/C@t{++} structures there are several pseudo children returned to
29629 designate access qualifiers. For these pseudo children @var{exp} is
29630 @samp{public}, @samp{private}, or @samp{protected}. In this case the
29631 type and value are not present.
29633 A dynamic varobj will not report the access qualifying
29634 pseudo-children, regardless of the language. This information is not
29635 available at all with a dynamic varobj.
29638 Number of children this child has. For a dynamic varobj, this will be
29642 The type of the child. If @samp{print object}
29643 (@pxref{Print Settings, set print object}) is set to @code{on}, the
29644 @emph{actual} (derived) type of the object is shown rather than the
29645 @emph{declared} one.
29648 If values were requested, this is the value.
29651 If this variable object is associated with a thread, this is the
29652 thread's global thread id. Otherwise this result is not present.
29655 If the variable object is frozen, this variable will be present with a value of 1.
29658 A dynamic varobj can supply a display hint to the front end. The
29659 value comes directly from the Python pretty-printer object's
29660 @code{display_hint} method. @xref{Pretty Printing API}.
29663 This attribute will be present and have the value @samp{1} if the
29664 varobj is a dynamic varobj. If the varobj is not a dynamic varobj,
29665 then this attribute will not be present.
29669 The result may have its own attributes:
29673 A dynamic varobj can supply a display hint to the front end. The
29674 value comes directly from the Python pretty-printer object's
29675 @code{display_hint} method. @xref{Pretty Printing API}.
29678 This is an integer attribute which is nonzero if there are children
29679 remaining after the end of the selected range.
29682 @subsubheading Example
29686 -var-list-children n
29687 ^done,numchild=@var{n},children=[child=@{name=@var{name},exp=@var{exp},
29688 numchild=@var{n},type=@var{type}@},@r{(repeats N times)}]
29690 -var-list-children --all-values n
29691 ^done,numchild=@var{n},children=[child=@{name=@var{name},exp=@var{exp},
29692 numchild=@var{n},value=@var{value},type=@var{type}@},@r{(repeats N times)}]
29696 @subheading The @code{-var-info-type} Command
29697 @findex -var-info-type
29699 @subsubheading Synopsis
29702 -var-info-type @var{name}
29705 Returns the type of the specified variable @var{name}. The type is
29706 returned as a string in the same format as it is output by the
29710 type=@var{typename}
29714 @subheading The @code{-var-info-expression} Command
29715 @findex -var-info-expression
29717 @subsubheading Synopsis
29720 -var-info-expression @var{name}
29723 Returns a string that is suitable for presenting this
29724 variable object in user interface. The string is generally
29725 not valid expression in the current language, and cannot be evaluated.
29727 For example, if @code{a} is an array, and variable object
29728 @code{A} was created for @code{a}, then we'll get this output:
29731 (gdb) -var-info-expression A.1
29732 ^done,lang="C",exp="1"
29736 Here, the value of @code{lang} is the language name, which can be
29737 found in @ref{Supported Languages}.
29739 Note that the output of the @code{-var-list-children} command also
29740 includes those expressions, so the @code{-var-info-expression} command
29743 @subheading The @code{-var-info-path-expression} Command
29744 @findex -var-info-path-expression
29746 @subsubheading Synopsis
29749 -var-info-path-expression @var{name}
29752 Returns an expression that can be evaluated in the current
29753 context and will yield the same value that a variable object has.
29754 Compare this with the @code{-var-info-expression} command, which
29755 result can be used only for UI presentation. Typical use of
29756 the @code{-var-info-path-expression} command is creating a
29757 watchpoint from a variable object.
29759 This command is currently not valid for children of a dynamic varobj,
29760 and will give an error when invoked on one.
29762 For example, suppose @code{C} is a C@t{++} class, derived from class
29763 @code{Base}, and that the @code{Base} class has a member called
29764 @code{m_size}. Assume a variable @code{c} is has the type of
29765 @code{C} and a variable object @code{C} was created for variable
29766 @code{c}. Then, we'll get this output:
29768 (gdb) -var-info-path-expression C.Base.public.m_size
29769 ^done,path_expr=((Base)c).m_size)
29772 @subheading The @code{-var-show-attributes} Command
29773 @findex -var-show-attributes
29775 @subsubheading Synopsis
29778 -var-show-attributes @var{name}
29781 List attributes of the specified variable object @var{name}:
29784 status=@var{attr} [ ( ,@var{attr} )* ]
29788 where @var{attr} is @code{@{ @{ editable | noneditable @} | TBD @}}.
29790 @subheading The @code{-var-evaluate-expression} Command
29791 @findex -var-evaluate-expression
29793 @subsubheading Synopsis
29796 -var-evaluate-expression [-f @var{format-spec}] @var{name}
29799 Evaluates the expression that is represented by the specified variable
29800 object and returns its value as a string. The format of the string
29801 can be specified with the @samp{-f} option. The possible values of
29802 this option are the same as for @code{-var-set-format}
29803 (@pxref{-var-set-format}). If the @samp{-f} option is not specified,
29804 the current display format will be used. The current display format
29805 can be changed using the @code{-var-set-format} command.
29811 Note that one must invoke @code{-var-list-children} for a variable
29812 before the value of a child variable can be evaluated.
29814 @subheading The @code{-var-assign} Command
29815 @findex -var-assign
29817 @subsubheading Synopsis
29820 -var-assign @var{name} @var{expression}
29823 Assigns the value of @var{expression} to the variable object specified
29824 by @var{name}. The object must be @samp{editable}. If the variable's
29825 value is altered by the assign, the variable will show up in any
29826 subsequent @code{-var-update} list.
29828 @subsubheading Example
29836 ^done,changelist=[@{name="var1",in_scope="true",type_changed="false"@}]
29840 @subheading The @code{-var-update} Command
29841 @findex -var-update
29843 @subsubheading Synopsis
29846 -var-update [@var{print-values}] @{@var{name} | "*"@}
29849 Reevaluate the expressions corresponding to the variable object
29850 @var{name} and all its direct and indirect children, and return the
29851 list of variable objects whose values have changed; @var{name} must
29852 be a root variable object. Here, ``changed'' means that the result of
29853 @code{-var-evaluate-expression} before and after the
29854 @code{-var-update} is different. If @samp{*} is used as the variable
29855 object names, all existing variable objects are updated, except
29856 for frozen ones (@pxref{-var-set-frozen}). The option
29857 @var{print-values} determines whether both names and values, or just
29858 names are printed. The possible values of this option are the same
29859 as for @code{-var-list-children} (@pxref{-var-list-children}). It is
29860 recommended to use the @samp{--all-values} option, to reduce the
29861 number of MI commands needed on each program stop.
29863 With the @samp{*} parameter, if a variable object is bound to a
29864 currently running thread, it will not be updated, without any
29867 If @code{-var-set-update-range} was previously used on a varobj, then
29868 only the selected range of children will be reported.
29870 @code{-var-update} reports all the changed varobjs in a tuple named
29873 Each item in the change list is itself a tuple holding:
29877 The name of the varobj.
29880 If values were requested for this update, then this field will be
29881 present and will hold the value of the varobj.
29884 @anchor{-var-update}
29885 This field is a string which may take one of three values:
29889 The variable object's current value is valid.
29892 The variable object does not currently hold a valid value but it may
29893 hold one in the future if its associated expression comes back into
29897 The variable object no longer holds a valid value.
29898 This can occur when the executable file being debugged has changed,
29899 either through recompilation or by using the @value{GDBN} @code{file}
29900 command. The front end should normally choose to delete these variable
29904 In the future new values may be added to this list so the front should
29905 be prepared for this possibility. @xref{GDB/MI Development and Front Ends, ,@sc{GDB/MI} Development and Front Ends}.
29908 This is only present if the varobj is still valid. If the type
29909 changed, then this will be the string @samp{true}; otherwise it will
29912 When a varobj's type changes, its children are also likely to have
29913 become incorrect. Therefore, the varobj's children are automatically
29914 deleted when this attribute is @samp{true}. Also, the varobj's update
29915 range, when set using the @code{-var-set-update-range} command, is
29919 If the varobj's type changed, then this field will be present and will
29922 @item new_num_children
29923 For a dynamic varobj, if the number of children changed, or if the
29924 type changed, this will be the new number of children.
29926 The @samp{numchild} field in other varobj responses is generally not
29927 valid for a dynamic varobj -- it will show the number of children that
29928 @value{GDBN} knows about, but because dynamic varobjs lazily
29929 instantiate their children, this will not reflect the number of
29930 children which may be available.
29932 The @samp{new_num_children} attribute only reports changes to the
29933 number of children known by @value{GDBN}. This is the only way to
29934 detect whether an update has removed children (which necessarily can
29935 only happen at the end of the update range).
29938 The display hint, if any.
29941 This is an integer value, which will be 1 if there are more children
29942 available outside the varobj's update range.
29945 This attribute will be present and have the value @samp{1} if the
29946 varobj is a dynamic varobj. If the varobj is not a dynamic varobj,
29947 then this attribute will not be present.
29950 If new children were added to a dynamic varobj within the selected
29951 update range (as set by @code{-var-set-update-range}), then they will
29952 be listed in this attribute.
29955 @subsubheading Example
29962 -var-update --all-values var1
29963 ^done,changelist=[@{name="var1",value="3",in_scope="true",
29964 type_changed="false"@}]
29968 @subheading The @code{-var-set-frozen} Command
29969 @findex -var-set-frozen
29970 @anchor{-var-set-frozen}
29972 @subsubheading Synopsis
29975 -var-set-frozen @var{name} @var{flag}
29978 Set the frozenness flag on the variable object @var{name}. The
29979 @var{flag} parameter should be either @samp{1} to make the variable
29980 frozen or @samp{0} to make it unfrozen. If a variable object is
29981 frozen, then neither itself, nor any of its children, are
29982 implicitly updated by @code{-var-update} of
29983 a parent variable or by @code{-var-update *}. Only
29984 @code{-var-update} of the variable itself will update its value and
29985 values of its children. After a variable object is unfrozen, it is
29986 implicitly updated by all subsequent @code{-var-update} operations.
29987 Unfreezing a variable does not update it, only subsequent
29988 @code{-var-update} does.
29990 @subsubheading Example
29994 -var-set-frozen V 1
29999 @subheading The @code{-var-set-update-range} command
30000 @findex -var-set-update-range
30001 @anchor{-var-set-update-range}
30003 @subsubheading Synopsis
30006 -var-set-update-range @var{name} @var{from} @var{to}
30009 Set the range of children to be returned by future invocations of
30010 @code{-var-update}.
30012 @var{from} and @var{to} indicate the range of children to report. If
30013 @var{from} or @var{to} is less than zero, the range is reset and all
30014 children will be reported. Otherwise, children starting at @var{from}
30015 (zero-based) and up to and excluding @var{to} will be reported.
30017 @subsubheading Example
30021 -var-set-update-range V 1 2
30025 @subheading The @code{-var-set-visualizer} command
30026 @findex -var-set-visualizer
30027 @anchor{-var-set-visualizer}
30029 @subsubheading Synopsis
30032 -var-set-visualizer @var{name} @var{visualizer}
30035 Set a visualizer for the variable object @var{name}.
30037 @var{visualizer} is the visualizer to use. The special value
30038 @samp{None} means to disable any visualizer in use.
30040 If not @samp{None}, @var{visualizer} must be a Python expression.
30041 This expression must evaluate to a callable object which accepts a
30042 single argument. @value{GDBN} will call this object with the value of
30043 the varobj @var{name} as an argument (this is done so that the same
30044 Python pretty-printing code can be used for both the CLI and MI).
30045 When called, this object must return an object which conforms to the
30046 pretty-printing interface (@pxref{Pretty Printing API}).
30048 The pre-defined function @code{gdb.default_visualizer} may be used to
30049 select a visualizer by following the built-in process
30050 (@pxref{Selecting Pretty-Printers}). This is done automatically when
30051 a varobj is created, and so ordinarily is not needed.
30053 This feature is only available if Python support is enabled. The MI
30054 command @code{-list-features} (@pxref{GDB/MI Support Commands})
30055 can be used to check this.
30057 @subsubheading Example
30059 Resetting the visualizer:
30063 -var-set-visualizer V None
30067 Reselecting the default (type-based) visualizer:
30071 -var-set-visualizer V gdb.default_visualizer
30075 Suppose @code{SomeClass} is a visualizer class. A lambda expression
30076 can be used to instantiate this class for a varobj:
30080 -var-set-visualizer V "lambda val: SomeClass()"
30084 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
30085 @node GDB/MI Data Manipulation
30086 @section @sc{gdb/mi} Data Manipulation
30088 @cindex data manipulation, in @sc{gdb/mi}
30089 @cindex @sc{gdb/mi}, data manipulation
30090 This section describes the @sc{gdb/mi} commands that manipulate data:
30091 examine memory and registers, evaluate expressions, etc.
30093 For details about what an addressable memory unit is,
30094 @pxref{addressable memory unit}.
30096 @c REMOVED FROM THE INTERFACE.
30097 @c @subheading -data-assign
30098 @c Change the value of a program variable. Plenty of side effects.
30099 @c @subsubheading GDB Command
30101 @c @subsubheading Example
30104 @subheading The @code{-data-disassemble} Command
30105 @findex -data-disassemble
30107 @subsubheading Synopsis
30111 [ -s @var{start-addr} -e @var{end-addr} ]
30112 | [ -f @var{filename} -l @var{linenum} [ -n @var{lines} ] ]
30120 @item @var{start-addr}
30121 is the beginning address (or @code{$pc})
30122 @item @var{end-addr}
30124 @item @var{filename}
30125 is the name of the file to disassemble
30126 @item @var{linenum}
30127 is the line number to disassemble around
30129 is the number of disassembly lines to be produced. If it is -1,
30130 the whole function will be disassembled, in case no @var{end-addr} is
30131 specified. If @var{end-addr} is specified as a non-zero value, and
30132 @var{lines} is lower than the number of disassembly lines between
30133 @var{start-addr} and @var{end-addr}, only @var{lines} lines are
30134 displayed; if @var{lines} is higher than the number of lines between
30135 @var{start-addr} and @var{end-addr}, only the lines up to @var{end-addr}
30140 @item 0 disassembly only
30141 @item 1 mixed source and disassembly (deprecated)
30142 @item 2 disassembly with raw opcodes
30143 @item 3 mixed source and disassembly with raw opcodes (deprecated)
30144 @item 4 mixed source and disassembly
30145 @item 5 mixed source and disassembly with raw opcodes
30148 Modes 1 and 3 are deprecated. The output is ``source centric''
30149 which hasn't proved useful in practice.
30150 @xref{Machine Code}, for a discussion of the difference between
30151 @code{/m} and @code{/s} output of the @code{disassemble} command.
30154 @subsubheading Result
30156 The result of the @code{-data-disassemble} command will be a list named
30157 @samp{asm_insns}, the contents of this list depend on the @var{mode}
30158 used with the @code{-data-disassemble} command.
30160 For modes 0 and 2 the @samp{asm_insns} list contains tuples with the
30165 The address at which this instruction was disassembled.
30168 The name of the function this instruction is within.
30171 The decimal offset in bytes from the start of @samp{func-name}.
30174 The text disassembly for this @samp{address}.
30177 This field is only present for modes 2, 3 and 5. This contains the raw opcode
30178 bytes for the @samp{inst} field.
30182 For modes 1, 3, 4 and 5 the @samp{asm_insns} list contains tuples named
30183 @samp{src_and_asm_line}, each of which has the following fields:
30187 The line number within @samp{file}.
30190 The file name from the compilation unit. This might be an absolute
30191 file name or a relative file name depending on the compile command
30195 Absolute file name of @samp{file}. It is converted to a canonical form
30196 using the source file search path
30197 (@pxref{Source Path, ,Specifying Source Directories})
30198 and after resolving all the symbolic links.
30200 If the source file is not found this field will contain the path as
30201 present in the debug information.
30203 @item line_asm_insn
30204 This is a list of tuples containing the disassembly for @samp{line} in
30205 @samp{file}. The fields of each tuple are the same as for
30206 @code{-data-disassemble} in @var{mode} 0 and 2, so @samp{address},
30207 @samp{func-name}, @samp{offset}, @samp{inst}, and optionally
30212 Note that whatever included in the @samp{inst} field, is not
30213 manipulated directly by @sc{gdb/mi}, i.e., it is not possible to
30216 @subsubheading @value{GDBN} Command
30218 The corresponding @value{GDBN} command is @samp{disassemble}.
30220 @subsubheading Example
30222 Disassemble from the current value of @code{$pc} to @code{$pc + 20}:
30226 -data-disassemble -s $pc -e "$pc + 20" -- 0
30229 @{address="0x000107c0",func-name="main",offset="4",
30230 inst="mov 2, %o0"@},
30231 @{address="0x000107c4",func-name="main",offset="8",
30232 inst="sethi %hi(0x11800), %o2"@},
30233 @{address="0x000107c8",func-name="main",offset="12",
30234 inst="or %o2, 0x140, %o1\t! 0x11940 <_lib_version+8>"@},
30235 @{address="0x000107cc",func-name="main",offset="16",
30236 inst="sethi %hi(0x11800), %o2"@},
30237 @{address="0x000107d0",func-name="main",offset="20",
30238 inst="or %o2, 0x168, %o4\t! 0x11968 <_lib_version+48>"@}]
30242 Disassemble the whole @code{main} function. Line 32 is part of
30246 -data-disassemble -f basics.c -l 32 -- 0
30248 @{address="0x000107bc",func-name="main",offset="0",
30249 inst="save %sp, -112, %sp"@},
30250 @{address="0x000107c0",func-name="main",offset="4",
30251 inst="mov 2, %o0"@},
30252 @{address="0x000107c4",func-name="main",offset="8",
30253 inst="sethi %hi(0x11800), %o2"@},
30255 @{address="0x0001081c",func-name="main",offset="96",inst="ret "@},
30256 @{address="0x00010820",func-name="main",offset="100",inst="restore "@}]
30260 Disassemble 3 instructions from the start of @code{main}:
30264 -data-disassemble -f basics.c -l 32 -n 3 -- 0
30266 @{address="0x000107bc",func-name="main",offset="0",
30267 inst="save %sp, -112, %sp"@},
30268 @{address="0x000107c0",func-name="main",offset="4",
30269 inst="mov 2, %o0"@},
30270 @{address="0x000107c4",func-name="main",offset="8",
30271 inst="sethi %hi(0x11800), %o2"@}]
30275 Disassemble 3 instructions from the start of @code{main} in mixed mode:
30279 -data-disassemble -f basics.c -l 32 -n 3 -- 1
30281 src_and_asm_line=@{line="31",
30282 file="../../../src/gdb/testsuite/gdb.mi/basics.c",
30283 fullname="/absolute/path/to/src/gdb/testsuite/gdb.mi/basics.c",
30284 line_asm_insn=[@{address="0x000107bc",
30285 func-name="main",offset="0",inst="save %sp, -112, %sp"@}]@},
30286 src_and_asm_line=@{line="32",
30287 file="../../../src/gdb/testsuite/gdb.mi/basics.c",
30288 fullname="/absolute/path/to/src/gdb/testsuite/gdb.mi/basics.c",
30289 line_asm_insn=[@{address="0x000107c0",
30290 func-name="main",offset="4",inst="mov 2, %o0"@},
30291 @{address="0x000107c4",func-name="main",offset="8",
30292 inst="sethi %hi(0x11800), %o2"@}]@}]
30297 @subheading The @code{-data-evaluate-expression} Command
30298 @findex -data-evaluate-expression
30300 @subsubheading Synopsis
30303 -data-evaluate-expression @var{expr}
30306 Evaluate @var{expr} as an expression. The expression could contain an
30307 inferior function call. The function call will execute synchronously.
30308 If the expression contains spaces, it must be enclosed in double quotes.
30310 @subsubheading @value{GDBN} Command
30312 The corresponding @value{GDBN} commands are @samp{print}, @samp{output}, and
30313 @samp{call}. In @code{gdbtk} only, there's a corresponding
30314 @samp{gdb_eval} command.
30316 @subsubheading Example
30318 In the following example, the numbers that precede the commands are the
30319 @dfn{tokens} described in @ref{GDB/MI Command Syntax, ,@sc{gdb/mi}
30320 Command Syntax}. Notice how @sc{gdb/mi} returns the same tokens in its
30324 211-data-evaluate-expression A
30327 311-data-evaluate-expression &A
30328 311^done,value="0xefffeb7c"
30330 411-data-evaluate-expression A+3
30333 511-data-evaluate-expression "A + 3"
30339 @subheading The @code{-data-list-changed-registers} Command
30340 @findex -data-list-changed-registers
30342 @subsubheading Synopsis
30345 -data-list-changed-registers
30348 Display a list of the registers that have changed.
30350 @subsubheading @value{GDBN} Command
30352 @value{GDBN} doesn't have a direct analog for this command; @code{gdbtk}
30353 has the corresponding command @samp{gdb_changed_register_list}.
30355 @subsubheading Example
30357 On a PPC MBX board:
30365 *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",frame=@{
30366 func="main",args=[],file="try.c",fullname="/home/foo/bar/try.c",
30369 -data-list-changed-registers
30370 ^done,changed-registers=["0","1","2","4","5","6","7","8","9",
30371 "10","11","13","14","15","16","17","18","19","20","21","22","23",
30372 "24","25","26","27","28","30","31","64","65","66","67","69"]
30377 @subheading The @code{-data-list-register-names} Command
30378 @findex -data-list-register-names
30380 @subsubheading Synopsis
30383 -data-list-register-names [ ( @var{regno} )+ ]
30386 Show a list of register names for the current target. If no arguments
30387 are given, it shows a list of the names of all the registers. If
30388 integer numbers are given as arguments, it will print a list of the
30389 names of the registers corresponding to the arguments. To ensure
30390 consistency between a register name and its number, the output list may
30391 include empty register names.
30393 @subsubheading @value{GDBN} Command
30395 @value{GDBN} does not have a command which corresponds to
30396 @samp{-data-list-register-names}. In @code{gdbtk} there is a
30397 corresponding command @samp{gdb_regnames}.
30399 @subsubheading Example
30401 For the PPC MBX board:
30404 -data-list-register-names
30405 ^done,register-names=["r0","r1","r2","r3","r4","r5","r6","r7",
30406 "r8","r9","r10","r11","r12","r13","r14","r15","r16","r17","r18",
30407 "r19","r20","r21","r22","r23","r24","r25","r26","r27","r28","r29",
30408 "r30","r31","f0","f1","f2","f3","f4","f5","f6","f7","f8","f9",
30409 "f10","f11","f12","f13","f14","f15","f16","f17","f18","f19","f20",
30410 "f21","f22","f23","f24","f25","f26","f27","f28","f29","f30","f31",
30411 "", "pc","ps","cr","lr","ctr","xer"]
30413 -data-list-register-names 1 2 3
30414 ^done,register-names=["r1","r2","r3"]
30418 @subheading The @code{-data-list-register-values} Command
30419 @findex -data-list-register-values
30421 @subsubheading Synopsis
30424 -data-list-register-values
30425 [ @code{--skip-unavailable} ] @var{fmt} [ ( @var{regno} )*]
30428 Display the registers' contents. The format according to which the
30429 registers' contents are to be returned is given by @var{fmt}, followed
30430 by an optional list of numbers specifying the registers to display. A
30431 missing list of numbers indicates that the contents of all the
30432 registers must be returned. The @code{--skip-unavailable} option
30433 indicates that only the available registers are to be returned.
30435 Allowed formats for @var{fmt} are:
30452 @subsubheading @value{GDBN} Command
30454 The corresponding @value{GDBN} commands are @samp{info reg}, @samp{info
30455 all-reg}, and (in @code{gdbtk}) @samp{gdb_fetch_registers}.
30457 @subsubheading Example
30459 For a PPC MBX board (note: line breaks are for readability only, they
30460 don't appear in the actual output):
30464 -data-list-register-values r 64 65
30465 ^done,register-values=[@{number="64",value="0xfe00a300"@},
30466 @{number="65",value="0x00029002"@}]
30468 -data-list-register-values x
30469 ^done,register-values=[@{number="0",value="0xfe0043c8"@},
30470 @{number="1",value="0x3fff88"@},@{number="2",value="0xfffffffe"@},
30471 @{number="3",value="0x0"@},@{number="4",value="0xa"@},
30472 @{number="5",value="0x3fff68"@},@{number="6",value="0x3fff58"@},
30473 @{number="7",value="0xfe011e98"@},@{number="8",value="0x2"@},
30474 @{number="9",value="0xfa202820"@},@{number="10",value="0xfa202808"@},
30475 @{number="11",value="0x1"@},@{number="12",value="0x0"@},
30476 @{number="13",value="0x4544"@},@{number="14",value="0xffdfffff"@},
30477 @{number="15",value="0xffffffff"@},@{number="16",value="0xfffffeff"@},
30478 @{number="17",value="0xefffffed"@},@{number="18",value="0xfffffffe"@},
30479 @{number="19",value="0xffffffff"@},@{number="20",value="0xffffffff"@},
30480 @{number="21",value="0xffffffff"@},@{number="22",value="0xfffffff7"@},
30481 @{number="23",value="0xffffffff"@},@{number="24",value="0xffffffff"@},
30482 @{number="25",value="0xffffffff"@},@{number="26",value="0xfffffffb"@},
30483 @{number="27",value="0xffffffff"@},@{number="28",value="0xf7bfffff"@},
30484 @{number="29",value="0x0"@},@{number="30",value="0xfe010000"@},
30485 @{number="31",value="0x0"@},@{number="32",value="0x0"@},
30486 @{number="33",value="0x0"@},@{number="34",value="0x0"@},
30487 @{number="35",value="0x0"@},@{number="36",value="0x0"@},
30488 @{number="37",value="0x0"@},@{number="38",value="0x0"@},
30489 @{number="39",value="0x0"@},@{number="40",value="0x0"@},
30490 @{number="41",value="0x0"@},@{number="42",value="0x0"@},
30491 @{number="43",value="0x0"@},@{number="44",value="0x0"@},
30492 @{number="45",value="0x0"@},@{number="46",value="0x0"@},
30493 @{number="47",value="0x0"@},@{number="48",value="0x0"@},
30494 @{number="49",value="0x0"@},@{number="50",value="0x0"@},
30495 @{number="51",value="0x0"@},@{number="52",value="0x0"@},
30496 @{number="53",value="0x0"@},@{number="54",value="0x0"@},
30497 @{number="55",value="0x0"@},@{number="56",value="0x0"@},
30498 @{number="57",value="0x0"@},@{number="58",value="0x0"@},
30499 @{number="59",value="0x0"@},@{number="60",value="0x0"@},
30500 @{number="61",value="0x0"@},@{number="62",value="0x0"@},
30501 @{number="63",value="0x0"@},@{number="64",value="0xfe00a300"@},
30502 @{number="65",value="0x29002"@},@{number="66",value="0x202f04b5"@},
30503 @{number="67",value="0xfe0043b0"@},@{number="68",value="0xfe00b3e4"@},
30504 @{number="69",value="0x20002b03"@}]
30509 @subheading The @code{-data-read-memory} Command
30510 @findex -data-read-memory
30512 This command is deprecated, use @code{-data-read-memory-bytes} instead.
30514 @subsubheading Synopsis
30517 -data-read-memory [ -o @var{byte-offset} ]
30518 @var{address} @var{word-format} @var{word-size}
30519 @var{nr-rows} @var{nr-cols} [ @var{aschar} ]
30526 @item @var{address}
30527 An expression specifying the address of the first memory word to be
30528 read. Complex expressions containing embedded white space should be
30529 quoted using the C convention.
30531 @item @var{word-format}
30532 The format to be used to print the memory words. The notation is the
30533 same as for @value{GDBN}'s @code{print} command (@pxref{Output Formats,
30536 @item @var{word-size}
30537 The size of each memory word in bytes.
30539 @item @var{nr-rows}
30540 The number of rows in the output table.
30542 @item @var{nr-cols}
30543 The number of columns in the output table.
30546 If present, indicates that each row should include an @sc{ascii} dump. The
30547 value of @var{aschar} is used as a padding character when a byte is not a
30548 member of the printable @sc{ascii} character set (printable @sc{ascii}
30549 characters are those whose code is between 32 and 126, inclusively).
30551 @item @var{byte-offset}
30552 An offset to add to the @var{address} before fetching memory.
30555 This command displays memory contents as a table of @var{nr-rows} by
30556 @var{nr-cols} words, each word being @var{word-size} bytes. In total,
30557 @code{@var{nr-rows} * @var{nr-cols} * @var{word-size}} bytes are read
30558 (returned as @samp{total-bytes}). Should less than the requested number
30559 of bytes be returned by the target, the missing words are identified
30560 using @samp{N/A}. The number of bytes read from the target is returned
30561 in @samp{nr-bytes} and the starting address used to read memory in
30564 The address of the next/previous row or page is available in
30565 @samp{next-row} and @samp{prev-row}, @samp{next-page} and
30568 @subsubheading @value{GDBN} Command
30570 The corresponding @value{GDBN} command is @samp{x}. @code{gdbtk} has
30571 @samp{gdb_get_mem} memory read command.
30573 @subsubheading Example
30575 Read six bytes of memory starting at @code{bytes+6} but then offset by
30576 @code{-6} bytes. Format as three rows of two columns. One byte per
30577 word. Display each word in hex.
30581 9-data-read-memory -o -6 -- bytes+6 x 1 3 2
30582 9^done,addr="0x00001390",nr-bytes="6",total-bytes="6",
30583 next-row="0x00001396",prev-row="0x0000138e",next-page="0x00001396",
30584 prev-page="0x0000138a",memory=[
30585 @{addr="0x00001390",data=["0x00","0x01"]@},
30586 @{addr="0x00001392",data=["0x02","0x03"]@},
30587 @{addr="0x00001394",data=["0x04","0x05"]@}]
30591 Read two bytes of memory starting at address @code{shorts + 64} and
30592 display as a single word formatted in decimal.
30596 5-data-read-memory shorts+64 d 2 1 1
30597 5^done,addr="0x00001510",nr-bytes="2",total-bytes="2",
30598 next-row="0x00001512",prev-row="0x0000150e",
30599 next-page="0x00001512",prev-page="0x0000150e",memory=[
30600 @{addr="0x00001510",data=["128"]@}]
30604 Read thirty two bytes of memory starting at @code{bytes+16} and format
30605 as eight rows of four columns. Include a string encoding with @samp{x}
30606 used as the non-printable character.
30610 4-data-read-memory bytes+16 x 1 8 4 x
30611 4^done,addr="0x000013a0",nr-bytes="32",total-bytes="32",
30612 next-row="0x000013c0",prev-row="0x0000139c",
30613 next-page="0x000013c0",prev-page="0x00001380",memory=[
30614 @{addr="0x000013a0",data=["0x10","0x11","0x12","0x13"],ascii="xxxx"@},
30615 @{addr="0x000013a4",data=["0x14","0x15","0x16","0x17"],ascii="xxxx"@},
30616 @{addr="0x000013a8",data=["0x18","0x19","0x1a","0x1b"],ascii="xxxx"@},
30617 @{addr="0x000013ac",data=["0x1c","0x1d","0x1e","0x1f"],ascii="xxxx"@},
30618 @{addr="0x000013b0",data=["0x20","0x21","0x22","0x23"],ascii=" !\"#"@},
30619 @{addr="0x000013b4",data=["0x24","0x25","0x26","0x27"],ascii="$%&'"@},
30620 @{addr="0x000013b8",data=["0x28","0x29","0x2a","0x2b"],ascii="()*+"@},
30621 @{addr="0x000013bc",data=["0x2c","0x2d","0x2e","0x2f"],ascii=",-./"@}]
30625 @subheading The @code{-data-read-memory-bytes} Command
30626 @findex -data-read-memory-bytes
30628 @subsubheading Synopsis
30631 -data-read-memory-bytes [ -o @var{offset} ]
30632 @var{address} @var{count}
30639 @item @var{address}
30640 An expression specifying the address of the first addressable memory unit
30641 to be read. Complex expressions containing embedded white space should be
30642 quoted using the C convention.
30645 The number of addressable memory units to read. This should be an integer
30649 The offset relative to @var{address} at which to start reading. This
30650 should be an integer literal. This option is provided so that a frontend
30651 is not required to first evaluate address and then perform address
30652 arithmetics itself.
30656 This command attempts to read all accessible memory regions in the
30657 specified range. First, all regions marked as unreadable in the memory
30658 map (if one is defined) will be skipped. @xref{Memory Region
30659 Attributes}. Second, @value{GDBN} will attempt to read the remaining
30660 regions. For each one, if reading full region results in an errors,
30661 @value{GDBN} will try to read a subset of the region.
30663 In general, every single memory unit in the region may be readable or not,
30664 and the only way to read every readable unit is to try a read at
30665 every address, which is not practical. Therefore, @value{GDBN} will
30666 attempt to read all accessible memory units at either beginning or the end
30667 of the region, using a binary division scheme. This heuristic works
30668 well for reading accross a memory map boundary. Note that if a region
30669 has a readable range that is neither at the beginning or the end,
30670 @value{GDBN} will not read it.
30672 The result record (@pxref{GDB/MI Result Records}) that is output of
30673 the command includes a field named @samp{memory} whose content is a
30674 list of tuples. Each tuple represent a successfully read memory block
30675 and has the following fields:
30679 The start address of the memory block, as hexadecimal literal.
30682 The end address of the memory block, as hexadecimal literal.
30685 The offset of the memory block, as hexadecimal literal, relative to
30686 the start address passed to @code{-data-read-memory-bytes}.
30689 The contents of the memory block, in hex.
30695 @subsubheading @value{GDBN} Command
30697 The corresponding @value{GDBN} command is @samp{x}.
30699 @subsubheading Example
30703 -data-read-memory-bytes &a 10
30704 ^done,memory=[@{begin="0xbffff154",offset="0x00000000",
30706 contents="01000000020000000300"@}]
30711 @subheading The @code{-data-write-memory-bytes} Command
30712 @findex -data-write-memory-bytes
30714 @subsubheading Synopsis
30717 -data-write-memory-bytes @var{address} @var{contents}
30718 -data-write-memory-bytes @var{address} @var{contents} @r{[}@var{count}@r{]}
30725 @item @var{address}
30726 An expression specifying the address of the first addressable memory unit
30727 to be written. Complex expressions containing embedded white space should
30728 be quoted using the C convention.
30730 @item @var{contents}
30731 The hex-encoded data to write. It is an error if @var{contents} does
30732 not represent an integral number of addressable memory units.
30735 Optional argument indicating the number of addressable memory units to be
30736 written. If @var{count} is greater than @var{contents}' length,
30737 @value{GDBN} will repeatedly write @var{contents} until it fills
30738 @var{count} memory units.
30742 @subsubheading @value{GDBN} Command
30744 There's no corresponding @value{GDBN} command.
30746 @subsubheading Example
30750 -data-write-memory-bytes &a "aabbccdd"
30757 -data-write-memory-bytes &a "aabbccdd" 16e
30762 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
30763 @node GDB/MI Tracepoint Commands
30764 @section @sc{gdb/mi} Tracepoint Commands
30766 The commands defined in this section implement MI support for
30767 tracepoints. For detailed introduction, see @ref{Tracepoints}.
30769 @subheading The @code{-trace-find} Command
30770 @findex -trace-find
30772 @subsubheading Synopsis
30775 -trace-find @var{mode} [@var{parameters}@dots{}]
30778 Find a trace frame using criteria defined by @var{mode} and
30779 @var{parameters}. The following table lists permissible
30780 modes and their parameters. For details of operation, see @ref{tfind}.
30785 No parameters are required. Stops examining trace frames.
30788 An integer is required as parameter. Selects tracepoint frame with
30791 @item tracepoint-number
30792 An integer is required as parameter. Finds next
30793 trace frame that corresponds to tracepoint with the specified number.
30796 An address is required as parameter. Finds
30797 next trace frame that corresponds to any tracepoint at the specified
30800 @item pc-inside-range
30801 Two addresses are required as parameters. Finds next trace
30802 frame that corresponds to a tracepoint at an address inside the
30803 specified range. Both bounds are considered to be inside the range.
30805 @item pc-outside-range
30806 Two addresses are required as parameters. Finds
30807 next trace frame that corresponds to a tracepoint at an address outside
30808 the specified range. Both bounds are considered to be inside the range.
30811 Line specification is required as parameter. @xref{Specify Location}.
30812 Finds next trace frame that corresponds to a tracepoint at
30813 the specified location.
30817 If @samp{none} was passed as @var{mode}, the response does not
30818 have fields. Otherwise, the response may have the following fields:
30822 This field has either @samp{0} or @samp{1} as the value, depending
30823 on whether a matching tracepoint was found.
30826 The index of the found traceframe. This field is present iff
30827 the @samp{found} field has value of @samp{1}.
30830 The index of the found tracepoint. This field is present iff
30831 the @samp{found} field has value of @samp{1}.
30834 The information about the frame corresponding to the found trace
30835 frame. This field is present only if a trace frame was found.
30836 @xref{GDB/MI Frame Information}, for description of this field.
30840 @subsubheading @value{GDBN} Command
30842 The corresponding @value{GDBN} command is @samp{tfind}.
30844 @subheading -trace-define-variable
30845 @findex -trace-define-variable
30847 @subsubheading Synopsis
30850 -trace-define-variable @var{name} [ @var{value} ]
30853 Create trace variable @var{name} if it does not exist. If
30854 @var{value} is specified, sets the initial value of the specified
30855 trace variable to that value. Note that the @var{name} should start
30856 with the @samp{$} character.
30858 @subsubheading @value{GDBN} Command
30860 The corresponding @value{GDBN} command is @samp{tvariable}.
30862 @subheading The @code{-trace-frame-collected} Command
30863 @findex -trace-frame-collected
30865 @subsubheading Synopsis
30868 -trace-frame-collected
30869 [--var-print-values @var{var_pval}]
30870 [--comp-print-values @var{comp_pval}]
30871 [--registers-format @var{regformat}]
30872 [--memory-contents]
30875 This command returns the set of collected objects, register names,
30876 trace state variable names, memory ranges and computed expressions
30877 that have been collected at a particular trace frame. The optional
30878 parameters to the command affect the output format in different ways.
30879 See the output description table below for more details.
30881 The reported names can be used in the normal manner to create
30882 varobjs and inspect the objects themselves. The items returned by
30883 this command are categorized so that it is clear which is a variable,
30884 which is a register, which is a trace state variable, which is a
30885 memory range and which is a computed expression.
30887 For instance, if the actions were
30889 collect myVar, myArray[myIndex], myObj.field, myPtr->field, myCount + 2
30890 collect *(int*)0xaf02bef0@@40
30894 the object collected in its entirety would be @code{myVar}. The
30895 object @code{myArray} would be partially collected, because only the
30896 element at index @code{myIndex} would be collected. The remaining
30897 objects would be computed expressions.
30899 An example output would be:
30903 -trace-frame-collected
30905 explicit-variables=[@{name="myVar",value="1"@}],
30906 computed-expressions=[@{name="myArray[myIndex]",value="0"@},
30907 @{name="myObj.field",value="0"@},
30908 @{name="myPtr->field",value="1"@},
30909 @{name="myCount + 2",value="3"@},
30910 @{name="$tvar1 + 1",value="43970027"@}],
30911 registers=[@{number="0",value="0x7fe2c6e79ec8"@},
30912 @{number="1",value="0x0"@},
30913 @{number="2",value="0x4"@},
30915 @{number="125",value="0x0"@}],
30916 tvars=[@{name="$tvar1",current="43970026"@}],
30917 memory=[@{address="0x0000000000602264",length="4"@},
30918 @{address="0x0000000000615bc0",length="4"@}]
30925 @item explicit-variables
30926 The set of objects that have been collected in their entirety (as
30927 opposed to collecting just a few elements of an array or a few struct
30928 members). For each object, its name and value are printed.
30929 The @code{--var-print-values} option affects how or whether the value
30930 field is output. If @var{var_pval} is 0, then print only the names;
30931 if it is 1, print also their values; and if it is 2, print the name,
30932 type and value for simple data types, and the name and type for
30933 arrays, structures and unions.
30935 @item computed-expressions
30936 The set of computed expressions that have been collected at the
30937 current trace frame. The @code{--comp-print-values} option affects
30938 this set like the @code{--var-print-values} option affects the
30939 @code{explicit-variables} set. See above.
30942 The registers that have been collected at the current trace frame.
30943 For each register collected, the name and current value are returned.
30944 The value is formatted according to the @code{--registers-format}
30945 option. See the @command{-data-list-register-values} command for a
30946 list of the allowed formats. The default is @samp{x}.
30949 The trace state variables that have been collected at the current
30950 trace frame. For each trace state variable collected, the name and
30951 current value are returned.
30954 The set of memory ranges that have been collected at the current trace
30955 frame. Its content is a list of tuples. Each tuple represents a
30956 collected memory range and has the following fields:
30960 The start address of the memory range, as hexadecimal literal.
30963 The length of the memory range, as decimal literal.
30966 The contents of the memory block, in hex. This field is only present
30967 if the @code{--memory-contents} option is specified.
30973 @subsubheading @value{GDBN} Command
30975 There is no corresponding @value{GDBN} command.
30977 @subsubheading Example
30979 @subheading -trace-list-variables
30980 @findex -trace-list-variables
30982 @subsubheading Synopsis
30985 -trace-list-variables
30988 Return a table of all defined trace variables. Each element of the
30989 table has the following fields:
30993 The name of the trace variable. This field is always present.
30996 The initial value. This is a 64-bit signed integer. This
30997 field is always present.
31000 The value the trace variable has at the moment. This is a 64-bit
31001 signed integer. This field is absent iff current value is
31002 not defined, for example if the trace was never run, or is
31007 @subsubheading @value{GDBN} Command
31009 The corresponding @value{GDBN} command is @samp{tvariables}.
31011 @subsubheading Example
31015 -trace-list-variables
31016 ^done,trace-variables=@{nr_rows="1",nr_cols="3",
31017 hdr=[@{width="15",alignment="-1",col_name="name",colhdr="Name"@},
31018 @{width="11",alignment="-1",col_name="initial",colhdr="Initial"@},
31019 @{width="11",alignment="-1",col_name="current",colhdr="Current"@}],
31020 body=[variable=@{name="$trace_timestamp",initial="0"@}
31021 variable=@{name="$foo",initial="10",current="15"@}]@}
31025 @subheading -trace-save
31026 @findex -trace-save
31028 @subsubheading Synopsis
31031 -trace-save [ -r ] [ -ctf ] @var{filename}
31034 Saves the collected trace data to @var{filename}. Without the
31035 @samp{-r} option, the data is downloaded from the target and saved
31036 in a local file. With the @samp{-r} option the target is asked
31037 to perform the save.
31039 By default, this command will save the trace in the tfile format. You can
31040 supply the optional @samp{-ctf} argument to save it the CTF format. See
31041 @ref{Trace Files} for more information about CTF.
31043 @subsubheading @value{GDBN} Command
31045 The corresponding @value{GDBN} command is @samp{tsave}.
31048 @subheading -trace-start
31049 @findex -trace-start
31051 @subsubheading Synopsis
31057 Starts a tracing experiment. The result of this command does not
31060 @subsubheading @value{GDBN} Command
31062 The corresponding @value{GDBN} command is @samp{tstart}.
31064 @subheading -trace-status
31065 @findex -trace-status
31067 @subsubheading Synopsis
31073 Obtains the status of a tracing experiment. The result may include
31074 the following fields:
31079 May have a value of either @samp{0}, when no tracing operations are
31080 supported, @samp{1}, when all tracing operations are supported, or
31081 @samp{file} when examining trace file. In the latter case, examining
31082 of trace frame is possible but new tracing experiement cannot be
31083 started. This field is always present.
31086 May have a value of either @samp{0} or @samp{1} depending on whether
31087 tracing experiement is in progress on target. This field is present
31088 if @samp{supported} field is not @samp{0}.
31091 Report the reason why the tracing was stopped last time. This field
31092 may be absent iff tracing was never stopped on target yet. The
31093 value of @samp{request} means the tracing was stopped as result of
31094 the @code{-trace-stop} command. The value of @samp{overflow} means
31095 the tracing buffer is full. The value of @samp{disconnection} means
31096 tracing was automatically stopped when @value{GDBN} has disconnected.
31097 The value of @samp{passcount} means tracing was stopped when a
31098 tracepoint was passed a maximal number of times for that tracepoint.
31099 This field is present if @samp{supported} field is not @samp{0}.
31101 @item stopping-tracepoint
31102 The number of tracepoint whose passcount as exceeded. This field is
31103 present iff the @samp{stop-reason} field has the value of
31107 @itemx frames-created
31108 The @samp{frames} field is a count of the total number of trace frames
31109 in the trace buffer, while @samp{frames-created} is the total created
31110 during the run, including ones that were discarded, such as when a
31111 circular trace buffer filled up. Both fields are optional.
31115 These fields tell the current size of the tracing buffer and the
31116 remaining space. These fields are optional.
31119 The value of the circular trace buffer flag. @code{1} means that the
31120 trace buffer is circular and old trace frames will be discarded if
31121 necessary to make room, @code{0} means that the trace buffer is linear
31125 The value of the disconnected tracing flag. @code{1} means that
31126 tracing will continue after @value{GDBN} disconnects, @code{0} means
31127 that the trace run will stop.
31130 The filename of the trace file being examined. This field is
31131 optional, and only present when examining a trace file.
31135 @subsubheading @value{GDBN} Command
31137 The corresponding @value{GDBN} command is @samp{tstatus}.
31139 @subheading -trace-stop
31140 @findex -trace-stop
31142 @subsubheading Synopsis
31148 Stops a tracing experiment. The result of this command has the same
31149 fields as @code{-trace-status}, except that the @samp{supported} and
31150 @samp{running} fields are not output.
31152 @subsubheading @value{GDBN} Command
31154 The corresponding @value{GDBN} command is @samp{tstop}.
31157 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
31158 @node GDB/MI Symbol Query
31159 @section @sc{gdb/mi} Symbol Query Commands
31163 @subheading The @code{-symbol-info-address} Command
31164 @findex -symbol-info-address
31166 @subsubheading Synopsis
31169 -symbol-info-address @var{symbol}
31172 Describe where @var{symbol} is stored.
31174 @subsubheading @value{GDBN} Command
31176 The corresponding @value{GDBN} command is @samp{info address}.
31178 @subsubheading Example
31182 @subheading The @code{-symbol-info-file} Command
31183 @findex -symbol-info-file
31185 @subsubheading Synopsis
31191 Show the file for the symbol.
31193 @subsubheading @value{GDBN} Command
31195 There's no equivalent @value{GDBN} command. @code{gdbtk} has
31196 @samp{gdb_find_file}.
31198 @subsubheading Example
31202 @subheading The @code{-symbol-info-function} Command
31203 @findex -symbol-info-function
31205 @subsubheading Synopsis
31208 -symbol-info-function
31211 Show which function the symbol lives in.
31213 @subsubheading @value{GDBN} Command
31215 @samp{gdb_get_function} in @code{gdbtk}.
31217 @subsubheading Example
31221 @subheading The @code{-symbol-info-line} Command
31222 @findex -symbol-info-line
31224 @subsubheading Synopsis
31230 Show the core addresses of the code for a source line.
31232 @subsubheading @value{GDBN} Command
31234 The corresponding @value{GDBN} command is @samp{info line}.
31235 @code{gdbtk} has the @samp{gdb_get_line} and @samp{gdb_get_file} commands.
31237 @subsubheading Example
31241 @subheading The @code{-symbol-info-symbol} Command
31242 @findex -symbol-info-symbol
31244 @subsubheading Synopsis
31247 -symbol-info-symbol @var{addr}
31250 Describe what symbol is at location @var{addr}.
31252 @subsubheading @value{GDBN} Command
31254 The corresponding @value{GDBN} command is @samp{info symbol}.
31256 @subsubheading Example
31260 @subheading The @code{-symbol-list-functions} Command
31261 @findex -symbol-list-functions
31263 @subsubheading Synopsis
31266 -symbol-list-functions
31269 List the functions in the executable.
31271 @subsubheading @value{GDBN} Command
31273 @samp{info functions} in @value{GDBN}, @samp{gdb_listfunc} and
31274 @samp{gdb_search} in @code{gdbtk}.
31276 @subsubheading Example
31281 @subheading The @code{-symbol-list-lines} Command
31282 @findex -symbol-list-lines
31284 @subsubheading Synopsis
31287 -symbol-list-lines @var{filename}
31290 Print the list of lines that contain code and their associated program
31291 addresses for the given source filename. The entries are sorted in
31292 ascending PC order.
31294 @subsubheading @value{GDBN} Command
31296 There is no corresponding @value{GDBN} command.
31298 @subsubheading Example
31301 -symbol-list-lines basics.c
31302 ^done,lines=[@{pc="0x08048554",line="7"@},@{pc="0x0804855a",line="8"@}]
31308 @subheading The @code{-symbol-list-types} Command
31309 @findex -symbol-list-types
31311 @subsubheading Synopsis
31317 List all the type names.
31319 @subsubheading @value{GDBN} Command
31321 The corresponding commands are @samp{info types} in @value{GDBN},
31322 @samp{gdb_search} in @code{gdbtk}.
31324 @subsubheading Example
31328 @subheading The @code{-symbol-list-variables} Command
31329 @findex -symbol-list-variables
31331 @subsubheading Synopsis
31334 -symbol-list-variables
31337 List all the global and static variable names.
31339 @subsubheading @value{GDBN} Command
31341 @samp{info variables} in @value{GDBN}, @samp{gdb_search} in @code{gdbtk}.
31343 @subsubheading Example
31347 @subheading The @code{-symbol-locate} Command
31348 @findex -symbol-locate
31350 @subsubheading Synopsis
31356 @subsubheading @value{GDBN} Command
31358 @samp{gdb_loc} in @code{gdbtk}.
31360 @subsubheading Example
31364 @subheading The @code{-symbol-type} Command
31365 @findex -symbol-type
31367 @subsubheading Synopsis
31370 -symbol-type @var{variable}
31373 Show type of @var{variable}.
31375 @subsubheading @value{GDBN} Command
31377 The corresponding @value{GDBN} command is @samp{ptype}, @code{gdbtk} has
31378 @samp{gdb_obj_variable}.
31380 @subsubheading Example
31385 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
31386 @node GDB/MI File Commands
31387 @section @sc{gdb/mi} File Commands
31389 This section describes the GDB/MI commands to specify executable file names
31390 and to read in and obtain symbol table information.
31392 @subheading The @code{-file-exec-and-symbols} Command
31393 @findex -file-exec-and-symbols
31395 @subsubheading Synopsis
31398 -file-exec-and-symbols @var{file}
31401 Specify the executable file to be debugged. This file is the one from
31402 which the symbol table is also read. If no file is specified, the
31403 command clears the executable and symbol information. If breakpoints
31404 are set when using this command with no arguments, @value{GDBN} will produce
31405 error messages. Otherwise, no output is produced, except a completion
31408 @subsubheading @value{GDBN} Command
31410 The corresponding @value{GDBN} command is @samp{file}.
31412 @subsubheading Example
31416 -file-exec-and-symbols /kwikemart/marge/ezannoni/TRUNK/mbx/hello.mbx
31422 @subheading The @code{-file-exec-file} Command
31423 @findex -file-exec-file
31425 @subsubheading Synopsis
31428 -file-exec-file @var{file}
31431 Specify the executable file to be debugged. Unlike
31432 @samp{-file-exec-and-symbols}, the symbol table is @emph{not} read
31433 from this file. If used without argument, @value{GDBN} clears the information
31434 about the executable file. No output is produced, except a completion
31437 @subsubheading @value{GDBN} Command
31439 The corresponding @value{GDBN} command is @samp{exec-file}.
31441 @subsubheading Example
31445 -file-exec-file /kwikemart/marge/ezannoni/TRUNK/mbx/hello.mbx
31452 @subheading The @code{-file-list-exec-sections} Command
31453 @findex -file-list-exec-sections
31455 @subsubheading Synopsis
31458 -file-list-exec-sections
31461 List the sections of the current executable file.
31463 @subsubheading @value{GDBN} Command
31465 The @value{GDBN} command @samp{info file} shows, among the rest, the same
31466 information as this command. @code{gdbtk} has a corresponding command
31467 @samp{gdb_load_info}.
31469 @subsubheading Example
31474 @subheading The @code{-file-list-exec-source-file} Command
31475 @findex -file-list-exec-source-file
31477 @subsubheading Synopsis
31480 -file-list-exec-source-file
31483 List the line number, the current source file, and the absolute path
31484 to the current source file for the current executable. The macro
31485 information field has a value of @samp{1} or @samp{0} depending on
31486 whether or not the file includes preprocessor macro information.
31488 @subsubheading @value{GDBN} Command
31490 The @value{GDBN} equivalent is @samp{info source}
31492 @subsubheading Example
31496 123-file-list-exec-source-file
31497 123^done,line="1",file="foo.c",fullname="/home/bar/foo.c,macro-info="1"
31502 @subheading The @code{-file-list-exec-source-files} Command
31503 @findex -file-list-exec-source-files
31505 @subsubheading Synopsis
31508 -file-list-exec-source-files
31511 List the source files for the current executable.
31513 It will always output both the filename and fullname (absolute file
31514 name) of a source file.
31516 @subsubheading @value{GDBN} Command
31518 The @value{GDBN} equivalent is @samp{info sources}.
31519 @code{gdbtk} has an analogous command @samp{gdb_listfiles}.
31521 @subsubheading Example
31524 -file-list-exec-source-files
31526 @{file=foo.c,fullname=/home/foo.c@},
31527 @{file=/home/bar.c,fullname=/home/bar.c@},
31528 @{file=gdb_could_not_find_fullpath.c@}]
31532 @subheading The @code{-file-list-shared-libraries} Command
31533 @findex -file-list-shared-libraries
31535 @subsubheading Synopsis
31538 -file-list-shared-libraries [ @var{regexp} ]
31541 List the shared libraries in the program.
31542 With a regular expression @var{regexp}, only those libraries whose
31543 names match @var{regexp} are listed.
31545 @subsubheading @value{GDBN} Command
31547 The corresponding @value{GDBN} command is @samp{info shared}. The fields
31548 have a similar meaning to the @code{=library-loaded} notification.
31549 The @code{ranges} field specifies the multiple segments belonging to this
31550 library. Each range has the following fields:
31554 The address defining the inclusive lower bound of the segment.
31556 The address defining the exclusive upper bound of the segment.
31559 @subsubheading Example
31562 -file-list-exec-source-files
31563 ^done,shared-libraries=[
31564 @{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"@}]@},
31565 @{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"@}]@}]
31571 @subheading The @code{-file-list-symbol-files} Command
31572 @findex -file-list-symbol-files
31574 @subsubheading Synopsis
31577 -file-list-symbol-files
31582 @subsubheading @value{GDBN} Command
31584 The corresponding @value{GDBN} command is @samp{info file} (part of it).
31586 @subsubheading Example
31591 @subheading The @code{-file-symbol-file} Command
31592 @findex -file-symbol-file
31594 @subsubheading Synopsis
31597 -file-symbol-file @var{file}
31600 Read symbol table info from the specified @var{file} argument. When
31601 used without arguments, clears @value{GDBN}'s symbol table info. No output is
31602 produced, except for a completion notification.
31604 @subsubheading @value{GDBN} Command
31606 The corresponding @value{GDBN} command is @samp{symbol-file}.
31608 @subsubheading Example
31612 -file-symbol-file /kwikemart/marge/ezannoni/TRUNK/mbx/hello.mbx
31618 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
31619 @node GDB/MI Memory Overlay Commands
31620 @section @sc{gdb/mi} Memory Overlay Commands
31622 The memory overlay commands are not implemented.
31624 @c @subheading -overlay-auto
31626 @c @subheading -overlay-list-mapping-state
31628 @c @subheading -overlay-list-overlays
31630 @c @subheading -overlay-map
31632 @c @subheading -overlay-off
31634 @c @subheading -overlay-on
31636 @c @subheading -overlay-unmap
31638 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
31639 @node GDB/MI Signal Handling Commands
31640 @section @sc{gdb/mi} Signal Handling Commands
31642 Signal handling commands are not implemented.
31644 @c @subheading -signal-handle
31646 @c @subheading -signal-list-handle-actions
31648 @c @subheading -signal-list-signal-types
31652 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
31653 @node GDB/MI Target Manipulation
31654 @section @sc{gdb/mi} Target Manipulation Commands
31657 @subheading The @code{-target-attach} Command
31658 @findex -target-attach
31660 @subsubheading Synopsis
31663 -target-attach @var{pid} | @var{gid} | @var{file}
31666 Attach to a process @var{pid} or a file @var{file} outside of
31667 @value{GDBN}, or a thread group @var{gid}. If attaching to a thread
31668 group, the id previously returned by
31669 @samp{-list-thread-groups --available} must be used.
31671 @subsubheading @value{GDBN} Command
31673 The corresponding @value{GDBN} command is @samp{attach}.
31675 @subsubheading Example
31679 =thread-created,id="1"
31680 *stopped,thread-id="1",frame=@{addr="0xb7f7e410",func="bar",args=[]@}
31686 @subheading The @code{-target-compare-sections} Command
31687 @findex -target-compare-sections
31689 @subsubheading Synopsis
31692 -target-compare-sections [ @var{section} ]
31695 Compare data of section @var{section} on target to the exec file.
31696 Without the argument, all sections are compared.
31698 @subsubheading @value{GDBN} Command
31700 The @value{GDBN} equivalent is @samp{compare-sections}.
31702 @subsubheading Example
31707 @subheading The @code{-target-detach} Command
31708 @findex -target-detach
31710 @subsubheading Synopsis
31713 -target-detach [ @var{pid} | @var{gid} ]
31716 Detach from the remote target which normally resumes its execution.
31717 If either @var{pid} or @var{gid} is specified, detaches from either
31718 the specified process, or specified thread group. There's no output.
31720 @subsubheading @value{GDBN} Command
31722 The corresponding @value{GDBN} command is @samp{detach}.
31724 @subsubheading Example
31734 @subheading The @code{-target-disconnect} Command
31735 @findex -target-disconnect
31737 @subsubheading Synopsis
31743 Disconnect from the remote target. There's no output and the target is
31744 generally not resumed.
31746 @subsubheading @value{GDBN} Command
31748 The corresponding @value{GDBN} command is @samp{disconnect}.
31750 @subsubheading Example
31760 @subheading The @code{-target-download} Command
31761 @findex -target-download
31763 @subsubheading Synopsis
31769 Loads the executable onto the remote target.
31770 It prints out an update message every half second, which includes the fields:
31774 The name of the section.
31776 The size of what has been sent so far for that section.
31778 The size of the section.
31780 The total size of what was sent so far (the current and the previous sections).
31782 The size of the overall executable to download.
31786 Each message is sent as status record (@pxref{GDB/MI Output Syntax, ,
31787 @sc{gdb/mi} Output Syntax}).
31789 In addition, it prints the name and size of the sections, as they are
31790 downloaded. These messages include the following fields:
31794 The name of the section.
31796 The size of the section.
31798 The size of the overall executable to download.
31802 At the end, a summary is printed.
31804 @subsubheading @value{GDBN} Command
31806 The corresponding @value{GDBN} command is @samp{load}.
31808 @subsubheading Example
31810 Note: each status message appears on a single line. Here the messages
31811 have been broken down so that they can fit onto a page.
31816 +download,@{section=".text",section-size="6668",total-size="9880"@}
31817 +download,@{section=".text",section-sent="512",section-size="6668",
31818 total-sent="512",total-size="9880"@}
31819 +download,@{section=".text",section-sent="1024",section-size="6668",
31820 total-sent="1024",total-size="9880"@}
31821 +download,@{section=".text",section-sent="1536",section-size="6668",
31822 total-sent="1536",total-size="9880"@}
31823 +download,@{section=".text",section-sent="2048",section-size="6668",
31824 total-sent="2048",total-size="9880"@}
31825 +download,@{section=".text",section-sent="2560",section-size="6668",
31826 total-sent="2560",total-size="9880"@}
31827 +download,@{section=".text",section-sent="3072",section-size="6668",
31828 total-sent="3072",total-size="9880"@}
31829 +download,@{section=".text",section-sent="3584",section-size="6668",
31830 total-sent="3584",total-size="9880"@}
31831 +download,@{section=".text",section-sent="4096",section-size="6668",
31832 total-sent="4096",total-size="9880"@}
31833 +download,@{section=".text",section-sent="4608",section-size="6668",
31834 total-sent="4608",total-size="9880"@}
31835 +download,@{section=".text",section-sent="5120",section-size="6668",
31836 total-sent="5120",total-size="9880"@}
31837 +download,@{section=".text",section-sent="5632",section-size="6668",
31838 total-sent="5632",total-size="9880"@}
31839 +download,@{section=".text",section-sent="6144",section-size="6668",
31840 total-sent="6144",total-size="9880"@}
31841 +download,@{section=".text",section-sent="6656",section-size="6668",
31842 total-sent="6656",total-size="9880"@}
31843 +download,@{section=".init",section-size="28",total-size="9880"@}
31844 +download,@{section=".fini",section-size="28",total-size="9880"@}
31845 +download,@{section=".data",section-size="3156",total-size="9880"@}
31846 +download,@{section=".data",section-sent="512",section-size="3156",
31847 total-sent="7236",total-size="9880"@}
31848 +download,@{section=".data",section-sent="1024",section-size="3156",
31849 total-sent="7748",total-size="9880"@}
31850 +download,@{section=".data",section-sent="1536",section-size="3156",
31851 total-sent="8260",total-size="9880"@}
31852 +download,@{section=".data",section-sent="2048",section-size="3156",
31853 total-sent="8772",total-size="9880"@}
31854 +download,@{section=".data",section-sent="2560",section-size="3156",
31855 total-sent="9284",total-size="9880"@}
31856 +download,@{section=".data",section-sent="3072",section-size="3156",
31857 total-sent="9796",total-size="9880"@}
31858 ^done,address="0x10004",load-size="9880",transfer-rate="6586",
31865 @subheading The @code{-target-exec-status} Command
31866 @findex -target-exec-status
31868 @subsubheading Synopsis
31871 -target-exec-status
31874 Provide information on the state of the target (whether it is running or
31875 not, for instance).
31877 @subsubheading @value{GDBN} Command
31879 There's no equivalent @value{GDBN} command.
31881 @subsubheading Example
31885 @subheading The @code{-target-list-available-targets} Command
31886 @findex -target-list-available-targets
31888 @subsubheading Synopsis
31891 -target-list-available-targets
31894 List the possible targets to connect to.
31896 @subsubheading @value{GDBN} Command
31898 The corresponding @value{GDBN} command is @samp{help target}.
31900 @subsubheading Example
31904 @subheading The @code{-target-list-current-targets} Command
31905 @findex -target-list-current-targets
31907 @subsubheading Synopsis
31910 -target-list-current-targets
31913 Describe the current target.
31915 @subsubheading @value{GDBN} Command
31917 The corresponding information is printed by @samp{info file} (among
31920 @subsubheading Example
31924 @subheading The @code{-target-list-parameters} Command
31925 @findex -target-list-parameters
31927 @subsubheading Synopsis
31930 -target-list-parameters
31936 @subsubheading @value{GDBN} Command
31940 @subsubheading Example
31943 @subheading The @code{-target-flash-erase} Command
31944 @findex -target-flash-erase
31946 @subsubheading Synopsis
31949 -target-flash-erase
31952 Erases all known flash memory regions on the target.
31954 The corresponding @value{GDBN} command is @samp{flash-erase}.
31956 The output is a list of flash regions that have been erased, with starting
31957 addresses and memory region sizes.
31961 -target-flash-erase
31962 ^done,erased-regions=@{address="0x0",size="0x40000"@}
31966 @subheading The @code{-target-select} Command
31967 @findex -target-select
31969 @subsubheading Synopsis
31972 -target-select @var{type} @var{parameters @dots{}}
31975 Connect @value{GDBN} to the remote target. This command takes two args:
31979 The type of target, for instance @samp{remote}, etc.
31980 @item @var{parameters}
31981 Device names, host names and the like. @xref{Target Commands, ,
31982 Commands for Managing Targets}, for more details.
31985 The output is a connection notification, followed by the address at
31986 which the target program is, in the following form:
31989 ^connected,addr="@var{address}",func="@var{function name}",
31990 args=[@var{arg list}]
31993 @subsubheading @value{GDBN} Command
31995 The corresponding @value{GDBN} command is @samp{target}.
31997 @subsubheading Example
32001 -target-select remote /dev/ttya
32002 ^connected,addr="0xfe00a300",func="??",args=[]
32006 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
32007 @node GDB/MI File Transfer Commands
32008 @section @sc{gdb/mi} File Transfer Commands
32011 @subheading The @code{-target-file-put} Command
32012 @findex -target-file-put
32014 @subsubheading Synopsis
32017 -target-file-put @var{hostfile} @var{targetfile}
32020 Copy file @var{hostfile} from the host system (the machine running
32021 @value{GDBN}) to @var{targetfile} on the target system.
32023 @subsubheading @value{GDBN} Command
32025 The corresponding @value{GDBN} command is @samp{remote put}.
32027 @subsubheading Example
32031 -target-file-put localfile remotefile
32037 @subheading The @code{-target-file-get} Command
32038 @findex -target-file-get
32040 @subsubheading Synopsis
32043 -target-file-get @var{targetfile} @var{hostfile}
32046 Copy file @var{targetfile} from the target system to @var{hostfile}
32047 on the host system.
32049 @subsubheading @value{GDBN} Command
32051 The corresponding @value{GDBN} command is @samp{remote get}.
32053 @subsubheading Example
32057 -target-file-get remotefile localfile
32063 @subheading The @code{-target-file-delete} Command
32064 @findex -target-file-delete
32066 @subsubheading Synopsis
32069 -target-file-delete @var{targetfile}
32072 Delete @var{targetfile} from the target system.
32074 @subsubheading @value{GDBN} Command
32076 The corresponding @value{GDBN} command is @samp{remote delete}.
32078 @subsubheading Example
32082 -target-file-delete remotefile
32088 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
32089 @node GDB/MI Ada Exceptions Commands
32090 @section Ada Exceptions @sc{gdb/mi} Commands
32092 @subheading The @code{-info-ada-exceptions} Command
32093 @findex -info-ada-exceptions
32095 @subsubheading Synopsis
32098 -info-ada-exceptions [ @var{regexp}]
32101 List all Ada exceptions defined within the program being debugged.
32102 With a regular expression @var{regexp}, only those exceptions whose
32103 names match @var{regexp} are listed.
32105 @subsubheading @value{GDBN} Command
32107 The corresponding @value{GDBN} command is @samp{info exceptions}.
32109 @subsubheading Result
32111 The result is a table of Ada exceptions. The following columns are
32112 defined for each exception:
32116 The name of the exception.
32119 The address of the exception.
32123 @subsubheading Example
32126 -info-ada-exceptions aint
32127 ^done,ada-exceptions=@{nr_rows="2",nr_cols="2",
32128 hdr=[@{width="1",alignment="-1",col_name="name",colhdr="Name"@},
32129 @{width="1",alignment="-1",col_name="address",colhdr="Address"@}],
32130 body=[@{name="constraint_error",address="0x0000000000613da0"@},
32131 @{name="const.aint_global_e",address="0x0000000000613b00"@}]@}
32134 @subheading Catching Ada Exceptions
32136 The commands describing how to ask @value{GDBN} to stop when a program
32137 raises an exception are described at @ref{Ada Exception GDB/MI
32138 Catchpoint Commands}.
32141 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
32142 @node GDB/MI Support Commands
32143 @section @sc{gdb/mi} Support Commands
32145 Since new commands and features get regularly added to @sc{gdb/mi},
32146 some commands are available to help front-ends query the debugger
32147 about support for these capabilities. Similarly, it is also possible
32148 to query @value{GDBN} about target support of certain features.
32150 @subheading The @code{-info-gdb-mi-command} Command
32151 @cindex @code{-info-gdb-mi-command}
32152 @findex -info-gdb-mi-command
32154 @subsubheading Synopsis
32157 -info-gdb-mi-command @var{cmd_name}
32160 Query support for the @sc{gdb/mi} command named @var{cmd_name}.
32162 Note that the dash (@code{-}) starting all @sc{gdb/mi} commands
32163 is technically not part of the command name (@pxref{GDB/MI Input
32164 Syntax}), and thus should be omitted in @var{cmd_name}. However,
32165 for ease of use, this command also accepts the form with the leading
32168 @subsubheading @value{GDBN} Command
32170 There is no corresponding @value{GDBN} command.
32172 @subsubheading Result
32174 The result is a tuple. There is currently only one field:
32178 This field is equal to @code{"true"} if the @sc{gdb/mi} command exists,
32179 @code{"false"} otherwise.
32183 @subsubheading Example
32185 Here is an example where the @sc{gdb/mi} command does not exist:
32188 -info-gdb-mi-command unsupported-command
32189 ^done,command=@{exists="false"@}
32193 And here is an example where the @sc{gdb/mi} command is known
32197 -info-gdb-mi-command symbol-list-lines
32198 ^done,command=@{exists="true"@}
32201 @subheading The @code{-list-features} Command
32202 @findex -list-features
32203 @cindex supported @sc{gdb/mi} features, list
32205 Returns a list of particular features of the MI protocol that
32206 this version of gdb implements. A feature can be a command,
32207 or a new field in an output of some command, or even an
32208 important bugfix. While a frontend can sometimes detect presence
32209 of a feature at runtime, it is easier to perform detection at debugger
32212 The command returns a list of strings, with each string naming an
32213 available feature. Each returned string is just a name, it does not
32214 have any internal structure. The list of possible feature names
32220 (gdb) -list-features
32221 ^done,result=["feature1","feature2"]
32224 The current list of features is:
32227 @item frozen-varobjs
32228 Indicates support for the @code{-var-set-frozen} command, as well
32229 as possible presense of the @code{frozen} field in the output
32230 of @code{-varobj-create}.
32231 @item pending-breakpoints
32232 Indicates support for the @option{-f} option to the @code{-break-insert}
32235 Indicates Python scripting support, Python-based
32236 pretty-printing commands, and possible presence of the
32237 @samp{display_hint} field in the output of @code{-var-list-children}
32239 Indicates support for the @code{-thread-info} command.
32240 @item data-read-memory-bytes
32241 Indicates support for the @code{-data-read-memory-bytes} and the
32242 @code{-data-write-memory-bytes} commands.
32243 @item breakpoint-notifications
32244 Indicates that changes to breakpoints and breakpoints created via the
32245 CLI will be announced via async records.
32246 @item ada-task-info
32247 Indicates support for the @code{-ada-task-info} command.
32248 @item language-option
32249 Indicates that all @sc{gdb/mi} commands accept the @option{--language}
32250 option (@pxref{Context management}).
32251 @item info-gdb-mi-command
32252 Indicates support for the @code{-info-gdb-mi-command} command.
32253 @item undefined-command-error-code
32254 Indicates support for the "undefined-command" error code in error result
32255 records, produced when trying to execute an undefined @sc{gdb/mi} command
32256 (@pxref{GDB/MI Result Records}).
32257 @item exec-run-start-option
32258 Indicates that the @code{-exec-run} command supports the @option{--start}
32259 option (@pxref{GDB/MI Program Execution}).
32262 @subheading The @code{-list-target-features} Command
32263 @findex -list-target-features
32265 Returns a list of particular features that are supported by the
32266 target. Those features affect the permitted MI commands, but
32267 unlike the features reported by the @code{-list-features} command, the
32268 features depend on which target GDB is using at the moment. Whenever
32269 a target can change, due to commands such as @code{-target-select},
32270 @code{-target-attach} or @code{-exec-run}, the list of target features
32271 may change, and the frontend should obtain it again.
32275 (gdb) -list-target-features
32276 ^done,result=["async"]
32279 The current list of features is:
32283 Indicates that the target is capable of asynchronous command
32284 execution, which means that @value{GDBN} will accept further commands
32285 while the target is running.
32288 Indicates that the target is capable of reverse execution.
32289 @xref{Reverse Execution}, for more information.
32293 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
32294 @node GDB/MI Miscellaneous Commands
32295 @section Miscellaneous @sc{gdb/mi} Commands
32297 @c @subheading -gdb-complete
32299 @subheading The @code{-gdb-exit} Command
32302 @subsubheading Synopsis
32308 Exit @value{GDBN} immediately.
32310 @subsubheading @value{GDBN} Command
32312 Approximately corresponds to @samp{quit}.
32314 @subsubheading Example
32324 @subheading The @code{-exec-abort} Command
32325 @findex -exec-abort
32327 @subsubheading Synopsis
32333 Kill the inferior running program.
32335 @subsubheading @value{GDBN} Command
32337 The corresponding @value{GDBN} command is @samp{kill}.
32339 @subsubheading Example
32344 @subheading The @code{-gdb-set} Command
32347 @subsubheading Synopsis
32353 Set an internal @value{GDBN} variable.
32354 @c IS THIS A DOLLAR VARIABLE? OR SOMETHING LIKE ANNOTATE ?????
32356 @subsubheading @value{GDBN} Command
32358 The corresponding @value{GDBN} command is @samp{set}.
32360 @subsubheading Example
32370 @subheading The @code{-gdb-show} Command
32373 @subsubheading Synopsis
32379 Show the current value of a @value{GDBN} variable.
32381 @subsubheading @value{GDBN} Command
32383 The corresponding @value{GDBN} command is @samp{show}.
32385 @subsubheading Example
32394 @c @subheading -gdb-source
32397 @subheading The @code{-gdb-version} Command
32398 @findex -gdb-version
32400 @subsubheading Synopsis
32406 Show version information for @value{GDBN}. Used mostly in testing.
32408 @subsubheading @value{GDBN} Command
32410 The @value{GDBN} equivalent is @samp{show version}. @value{GDBN} by
32411 default shows this information when you start an interactive session.
32413 @subsubheading Example
32415 @c This example modifies the actual output from GDB to avoid overfull
32421 ~Copyright 2000 Free Software Foundation, Inc.
32422 ~GDB is free software, covered by the GNU General Public License, and
32423 ~you are welcome to change it and/or distribute copies of it under
32424 ~ certain conditions.
32425 ~Type "show copying" to see the conditions.
32426 ~There is absolutely no warranty for GDB. Type "show warranty" for
32428 ~This GDB was configured as
32429 "--host=sparc-sun-solaris2.5.1 --target=ppc-eabi".
32434 @subheading The @code{-list-thread-groups} Command
32435 @findex -list-thread-groups
32437 @subheading Synopsis
32440 -list-thread-groups [ --available ] [ --recurse 1 ] [ @var{group} ... ]
32443 Lists thread groups (@pxref{Thread groups}). When a single thread
32444 group is passed as the argument, lists the children of that group.
32445 When several thread group are passed, lists information about those
32446 thread groups. Without any parameters, lists information about all
32447 top-level thread groups.
32449 Normally, thread groups that are being debugged are reported.
32450 With the @samp{--available} option, @value{GDBN} reports thread groups
32451 available on the target.
32453 The output of this command may have either a @samp{threads} result or
32454 a @samp{groups} result. The @samp{thread} result has a list of tuples
32455 as value, with each tuple describing a thread (@pxref{GDB/MI Thread
32456 Information}). The @samp{groups} result has a list of tuples as value,
32457 each tuple describing a thread group. If top-level groups are
32458 requested (that is, no parameter is passed), or when several groups
32459 are passed, the output always has a @samp{groups} result. The format
32460 of the @samp{group} result is described below.
32462 To reduce the number of roundtrips it's possible to list thread groups
32463 together with their children, by passing the @samp{--recurse} option
32464 and the recursion depth. Presently, only recursion depth of 1 is
32465 permitted. If this option is present, then every reported thread group
32466 will also include its children, either as @samp{group} or
32467 @samp{threads} field.
32469 In general, any combination of option and parameters is permitted, with
32470 the following caveats:
32474 When a single thread group is passed, the output will typically
32475 be the @samp{threads} result. Because threads may not contain
32476 anything, the @samp{recurse} option will be ignored.
32479 When the @samp{--available} option is passed, limited information may
32480 be available. In particular, the list of threads of a process might
32481 be inaccessible. Further, specifying specific thread groups might
32482 not give any performance advantage over listing all thread groups.
32483 The frontend should assume that @samp{-list-thread-groups --available}
32484 is always an expensive operation and cache the results.
32488 The @samp{groups} result is a list of tuples, where each tuple may
32489 have the following fields:
32493 Identifier of the thread group. This field is always present.
32494 The identifier is an opaque string; frontends should not try to
32495 convert it to an integer, even though it might look like one.
32498 The type of the thread group. At present, only @samp{process} is a
32502 The target-specific process identifier. This field is only present
32503 for thread groups of type @samp{process} and only if the process exists.
32506 The exit code of this group's last exited thread, formatted in octal.
32507 This field is only present for thread groups of type @samp{process} and
32508 only if the process is not running.
32511 The number of children this thread group has. This field may be
32512 absent for an available thread group.
32515 This field has a list of tuples as value, each tuple describing a
32516 thread. It may be present if the @samp{--recurse} option is
32517 specified, and it's actually possible to obtain the threads.
32520 This field is a list of integers, each identifying a core that one
32521 thread of the group is running on. This field may be absent if
32522 such information is not available.
32525 The name of the executable file that corresponds to this thread group.
32526 The field is only present for thread groups of type @samp{process},
32527 and only if there is a corresponding executable file.
32531 @subheading Example
32535 -list-thread-groups
32536 ^done,groups=[@{id="17",type="process",pid="yyy",num_children="2"@}]
32537 -list-thread-groups 17
32538 ^done,threads=[@{id="2",target-id="Thread 0xb7e14b90 (LWP 21257)",
32539 frame=@{level="0",addr="0xffffe410",func="__kernel_vsyscall",args=[]@},state="running"@},
32540 @{id="1",target-id="Thread 0xb7e156b0 (LWP 21254)",
32541 frame=@{level="0",addr="0x0804891f",func="foo",args=[@{name="i",value="10"@}],
32542 file="/tmp/a.c",fullname="/tmp/a.c",line="158"@},state="running"@}]]
32543 -list-thread-groups --available
32544 ^done,groups=[@{id="17",type="process",pid="yyy",num_children="2",cores=[1,2]@}]
32545 -list-thread-groups --available --recurse 1
32546 ^done,groups=[@{id="17", types="process",pid="yyy",num_children="2",cores=[1,2],
32547 threads=[@{id="1",target-id="Thread 0xb7e14b90",cores=[1]@},
32548 @{id="2",target-id="Thread 0xb7e14b90",cores=[2]@}]@},..]
32549 -list-thread-groups --available --recurse 1 17 18
32550 ^done,groups=[@{id="17", types="process",pid="yyy",num_children="2",cores=[1,2],
32551 threads=[@{id="1",target-id="Thread 0xb7e14b90",cores=[1]@},
32552 @{id="2",target-id="Thread 0xb7e14b90",cores=[2]@}]@},...]
32555 @subheading The @code{-info-os} Command
32558 @subsubheading Synopsis
32561 -info-os [ @var{type} ]
32564 If no argument is supplied, the command returns a table of available
32565 operating-system-specific information types. If one of these types is
32566 supplied as an argument @var{type}, then the command returns a table
32567 of data of that type.
32569 The types of information available depend on the target operating
32572 @subsubheading @value{GDBN} Command
32574 The corresponding @value{GDBN} command is @samp{info os}.
32576 @subsubheading Example
32578 When run on a @sc{gnu}/Linux system, the output will look something
32584 ^done,OSDataTable=@{nr_rows="10",nr_cols="3",
32585 hdr=[@{width="10",alignment="-1",col_name="col0",colhdr="Type"@},
32586 @{width="10",alignment="-1",col_name="col1",colhdr="Description"@},
32587 @{width="10",alignment="-1",col_name="col2",colhdr="Title"@}],
32588 body=[item=@{col0="cpus",col1="Listing of all cpus/cores on the system",
32590 item=@{col0="files",col1="Listing of all file descriptors",
32591 col2="File descriptors"@},
32592 item=@{col0="modules",col1="Listing of all loaded kernel modules",
32593 col2="Kernel modules"@},
32594 item=@{col0="msg",col1="Listing of all message queues",
32595 col2="Message queues"@},
32596 item=@{col0="processes",col1="Listing of all processes",
32597 col2="Processes"@},
32598 item=@{col0="procgroups",col1="Listing of all process groups",
32599 col2="Process groups"@},
32600 item=@{col0="semaphores",col1="Listing of all semaphores",
32601 col2="Semaphores"@},
32602 item=@{col0="shm",col1="Listing of all shared-memory regions",
32603 col2="Shared-memory regions"@},
32604 item=@{col0="sockets",col1="Listing of all internet-domain sockets",
32606 item=@{col0="threads",col1="Listing of all threads",
32610 ^done,OSDataTable=@{nr_rows="190",nr_cols="4",
32611 hdr=[@{width="10",alignment="-1",col_name="col0",colhdr="pid"@},
32612 @{width="10",alignment="-1",col_name="col1",colhdr="user"@},
32613 @{width="10",alignment="-1",col_name="col2",colhdr="command"@},
32614 @{width="10",alignment="-1",col_name="col3",colhdr="cores"@}],
32615 body=[item=@{col0="1",col1="root",col2="/sbin/init",col3="0"@},
32616 item=@{col0="2",col1="root",col2="[kthreadd]",col3="1"@},
32617 item=@{col0="3",col1="root",col2="[ksoftirqd/0]",col3="0"@},
32619 item=@{col0="26446",col1="stan",col2="bash",col3="0"@},
32620 item=@{col0="28152",col1="stan",col2="bash",col3="1"@}]@}
32624 (Note that the MI output here includes a @code{"Title"} column that
32625 does not appear in command-line @code{info os}; this column is useful
32626 for MI clients that want to enumerate the types of data, such as in a
32627 popup menu, but is needless clutter on the command line, and
32628 @code{info os} omits it.)
32630 @subheading The @code{-add-inferior} Command
32631 @findex -add-inferior
32633 @subheading Synopsis
32639 Creates a new inferior (@pxref{Inferiors and Programs}). The created
32640 inferior is not associated with any executable. Such association may
32641 be established with the @samp{-file-exec-and-symbols} command
32642 (@pxref{GDB/MI File Commands}). The command response has a single
32643 field, @samp{inferior}, whose value is the identifier of the
32644 thread group corresponding to the new inferior.
32646 @subheading Example
32651 ^done,inferior="i3"
32654 @subheading The @code{-interpreter-exec} Command
32655 @findex -interpreter-exec
32657 @subheading Synopsis
32660 -interpreter-exec @var{interpreter} @var{command}
32662 @anchor{-interpreter-exec}
32664 Execute the specified @var{command} in the given @var{interpreter}.
32666 @subheading @value{GDBN} Command
32668 The corresponding @value{GDBN} command is @samp{interpreter-exec}.
32670 @subheading Example
32674 -interpreter-exec console "break main"
32675 &"During symbol reading, couldn't parse type; debugger out of date?.\n"
32676 &"During symbol reading, bad structure-type format.\n"
32677 ~"Breakpoint 1 at 0x8074fc6: file ../../src/gdb/main.c, line 743.\n"
32682 @subheading The @code{-inferior-tty-set} Command
32683 @findex -inferior-tty-set
32685 @subheading Synopsis
32688 -inferior-tty-set /dev/pts/1
32691 Set terminal for future runs of the program being debugged.
32693 @subheading @value{GDBN} Command
32695 The corresponding @value{GDBN} command is @samp{set inferior-tty} /dev/pts/1.
32697 @subheading Example
32701 -inferior-tty-set /dev/pts/1
32706 @subheading The @code{-inferior-tty-show} Command
32707 @findex -inferior-tty-show
32709 @subheading Synopsis
32715 Show terminal for future runs of program being debugged.
32717 @subheading @value{GDBN} Command
32719 The corresponding @value{GDBN} command is @samp{show inferior-tty}.
32721 @subheading Example
32725 -inferior-tty-set /dev/pts/1
32729 ^done,inferior_tty_terminal="/dev/pts/1"
32733 @subheading The @code{-enable-timings} Command
32734 @findex -enable-timings
32736 @subheading Synopsis
32739 -enable-timings [yes | no]
32742 Toggle the printing of the wallclock, user and system times for an MI
32743 command as a field in its output. This command is to help frontend
32744 developers optimize the performance of their code. No argument is
32745 equivalent to @samp{yes}.
32747 @subheading @value{GDBN} Command
32751 @subheading Example
32759 ^done,bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
32760 addr="0x080484ed",func="main",file="myprog.c",
32761 fullname="/home/nickrob/myprog.c",line="73",thread-groups=["i1"],
32763 time=@{wallclock="0.05185",user="0.00800",system="0.00000"@}
32771 *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",thread-id="0",
32772 frame=@{addr="0x080484ed",func="main",args=[@{name="argc",value="1"@},
32773 @{name="argv",value="0xbfb60364"@}],file="myprog.c",
32774 fullname="/home/nickrob/myprog.c",line="73"@}
32779 @chapter @value{GDBN} Annotations
32781 This chapter describes annotations in @value{GDBN}. Annotations were
32782 designed to interface @value{GDBN} to graphical user interfaces or other
32783 similar programs which want to interact with @value{GDBN} at a
32784 relatively high level.
32786 The annotation mechanism has largely been superseded by @sc{gdb/mi}
32790 This is Edition @value{EDITION}, @value{DATE}.
32794 * Annotations Overview:: What annotations are; the general syntax.
32795 * Server Prefix:: Issuing a command without affecting user state.
32796 * Prompting:: Annotations marking @value{GDBN}'s need for input.
32797 * Errors:: Annotations for error messages.
32798 * Invalidation:: Some annotations describe things now invalid.
32799 * Annotations for Running::
32800 Whether the program is running, how it stopped, etc.
32801 * Source Annotations:: Annotations describing source code.
32804 @node Annotations Overview
32805 @section What is an Annotation?
32806 @cindex annotations
32808 Annotations start with a newline character, two @samp{control-z}
32809 characters, and the name of the annotation. If there is no additional
32810 information associated with this annotation, the name of the annotation
32811 is followed immediately by a newline. If there is additional
32812 information, the name of the annotation is followed by a space, the
32813 additional information, and a newline. The additional information
32814 cannot contain newline characters.
32816 Any output not beginning with a newline and two @samp{control-z}
32817 characters denotes literal output from @value{GDBN}. Currently there is
32818 no need for @value{GDBN} to output a newline followed by two
32819 @samp{control-z} characters, but if there was such a need, the
32820 annotations could be extended with an @samp{escape} annotation which
32821 means those three characters as output.
32823 The annotation @var{level}, which is specified using the
32824 @option{--annotate} command line option (@pxref{Mode Options}), controls
32825 how much information @value{GDBN} prints together with its prompt,
32826 values of expressions, source lines, and other types of output. Level 0
32827 is for no annotations, level 1 is for use when @value{GDBN} is run as a
32828 subprocess of @sc{gnu} Emacs, level 3 is the maximum annotation suitable
32829 for programs that control @value{GDBN}, and level 2 annotations have
32830 been made obsolete (@pxref{Limitations, , Limitations of the Annotation
32831 Interface, annotate, GDB's Obsolete Annotations}).
32834 @kindex set annotate
32835 @item set annotate @var{level}
32836 The @value{GDBN} command @code{set annotate} sets the level of
32837 annotations to the specified @var{level}.
32839 @item show annotate
32840 @kindex show annotate
32841 Show the current annotation level.
32844 This chapter describes level 3 annotations.
32846 A simple example of starting up @value{GDBN} with annotations is:
32849 $ @kbd{gdb --annotate=3}
32851 Copyright 2003 Free Software Foundation, Inc.
32852 GDB is free software, covered by the GNU General Public License,
32853 and you are welcome to change it and/or distribute copies of it
32854 under certain conditions.
32855 Type "show copying" to see the conditions.
32856 There is absolutely no warranty for GDB. Type "show warranty"
32858 This GDB was configured as "i386-pc-linux-gnu"
32869 Here @samp{quit} is input to @value{GDBN}; the rest is output from
32870 @value{GDBN}. The three lines beginning @samp{^Z^Z} (where @samp{^Z}
32871 denotes a @samp{control-z} character) are annotations; the rest is
32872 output from @value{GDBN}.
32874 @node Server Prefix
32875 @section The Server Prefix
32876 @cindex server prefix
32878 If you prefix a command with @samp{server } then it will not affect
32879 the command history, nor will it affect @value{GDBN}'s notion of which
32880 command to repeat if @key{RET} is pressed on a line by itself. This
32881 means that commands can be run behind a user's back by a front-end in
32882 a transparent manner.
32884 The @code{server } prefix does not affect the recording of values into
32885 the value history; to print a value without recording it into the
32886 value history, use the @code{output} command instead of the
32887 @code{print} command.
32889 Using this prefix also disables confirmation requests
32890 (@pxref{confirmation requests}).
32893 @section Annotation for @value{GDBN} Input
32895 @cindex annotations for prompts
32896 When @value{GDBN} prompts for input, it annotates this fact so it is possible
32897 to know when to send output, when the output from a given command is
32900 Different kinds of input each have a different @dfn{input type}. Each
32901 input type has three annotations: a @code{pre-} annotation, which
32902 denotes the beginning of any prompt which is being output, a plain
32903 annotation, which denotes the end of the prompt, and then a @code{post-}
32904 annotation which denotes the end of any echo which may (or may not) be
32905 associated with the input. For example, the @code{prompt} input type
32906 features the following annotations:
32914 The input types are
32917 @findex pre-prompt annotation
32918 @findex prompt annotation
32919 @findex post-prompt annotation
32921 When @value{GDBN} is prompting for a command (the main @value{GDBN} prompt).
32923 @findex pre-commands annotation
32924 @findex commands annotation
32925 @findex post-commands annotation
32927 When @value{GDBN} prompts for a set of commands, like in the @code{commands}
32928 command. The annotations are repeated for each command which is input.
32930 @findex pre-overload-choice annotation
32931 @findex overload-choice annotation
32932 @findex post-overload-choice annotation
32933 @item overload-choice
32934 When @value{GDBN} wants the user to select between various overloaded functions.
32936 @findex pre-query annotation
32937 @findex query annotation
32938 @findex post-query annotation
32940 When @value{GDBN} wants the user to confirm a potentially dangerous operation.
32942 @findex pre-prompt-for-continue annotation
32943 @findex prompt-for-continue annotation
32944 @findex post-prompt-for-continue annotation
32945 @item prompt-for-continue
32946 When @value{GDBN} is asking the user to press return to continue. Note: Don't
32947 expect this to work well; instead use @code{set height 0} to disable
32948 prompting. This is because the counting of lines is buggy in the
32949 presence of annotations.
32954 @cindex annotations for errors, warnings and interrupts
32956 @findex quit annotation
32961 This annotation occurs right before @value{GDBN} responds to an interrupt.
32963 @findex error annotation
32968 This annotation occurs right before @value{GDBN} responds to an error.
32970 Quit and error annotations indicate that any annotations which @value{GDBN} was
32971 in the middle of may end abruptly. For example, if a
32972 @code{value-history-begin} annotation is followed by a @code{error}, one
32973 cannot expect to receive the matching @code{value-history-end}. One
32974 cannot expect not to receive it either, however; an error annotation
32975 does not necessarily mean that @value{GDBN} is immediately returning all the way
32978 @findex error-begin annotation
32979 A quit or error annotation may be preceded by
32985 Any output between that and the quit or error annotation is the error
32988 Warning messages are not yet annotated.
32989 @c If we want to change that, need to fix warning(), type_error(),
32990 @c range_error(), and possibly other places.
32993 @section Invalidation Notices
32995 @cindex annotations for invalidation messages
32996 The following annotations say that certain pieces of state may have
33000 @findex frames-invalid annotation
33001 @item ^Z^Zframes-invalid
33003 The frames (for example, output from the @code{backtrace} command) may
33006 @findex breakpoints-invalid annotation
33007 @item ^Z^Zbreakpoints-invalid
33009 The breakpoints may have changed. For example, the user just added or
33010 deleted a breakpoint.
33013 @node Annotations for Running
33014 @section Running the Program
33015 @cindex annotations for running programs
33017 @findex starting annotation
33018 @findex stopping annotation
33019 When the program starts executing due to a @value{GDBN} command such as
33020 @code{step} or @code{continue},
33026 is output. When the program stops,
33032 is output. Before the @code{stopped} annotation, a variety of
33033 annotations describe how the program stopped.
33036 @findex exited annotation
33037 @item ^Z^Zexited @var{exit-status}
33038 The program exited, and @var{exit-status} is the exit status (zero for
33039 successful exit, otherwise nonzero).
33041 @findex signalled annotation
33042 @findex signal-name annotation
33043 @findex signal-name-end annotation
33044 @findex signal-string annotation
33045 @findex signal-string-end annotation
33046 @item ^Z^Zsignalled
33047 The program exited with a signal. After the @code{^Z^Zsignalled}, the
33048 annotation continues:
33054 ^Z^Zsignal-name-end
33058 ^Z^Zsignal-string-end
33063 where @var{name} is the name of the signal, such as @code{SIGILL} or
33064 @code{SIGSEGV}, and @var{string} is the explanation of the signal, such
33065 as @code{Illegal Instruction} or @code{Segmentation fault}. The arguments
33066 @var{intro-text}, @var{middle-text}, and @var{end-text} are for the
33067 user's benefit and have no particular format.
33069 @findex signal annotation
33071 The syntax of this annotation is just like @code{signalled}, but @value{GDBN} is
33072 just saying that the program received the signal, not that it was
33073 terminated with it.
33075 @findex breakpoint annotation
33076 @item ^Z^Zbreakpoint @var{number}
33077 The program hit breakpoint number @var{number}.
33079 @findex watchpoint annotation
33080 @item ^Z^Zwatchpoint @var{number}
33081 The program hit watchpoint number @var{number}.
33084 @node Source Annotations
33085 @section Displaying Source
33086 @cindex annotations for source display
33088 @findex source annotation
33089 The following annotation is used instead of displaying source code:
33092 ^Z^Zsource @var{filename}:@var{line}:@var{character}:@var{middle}:@var{addr}
33095 where @var{filename} is an absolute file name indicating which source
33096 file, @var{line} is the line number within that file (where 1 is the
33097 first line in the file), @var{character} is the character position
33098 within the file (where 0 is the first character in the file) (for most
33099 debug formats this will necessarily point to the beginning of a line),
33100 @var{middle} is @samp{middle} if @var{addr} is in the middle of the
33101 line, or @samp{beg} if @var{addr} is at the beginning of the line, and
33102 @var{addr} is the address in the target program associated with the
33103 source which is being displayed. The @var{addr} is in the form @samp{0x}
33104 followed by one or more lowercase hex digits (note that this does not
33105 depend on the language).
33107 @node JIT Interface
33108 @chapter JIT Compilation Interface
33109 @cindex just-in-time compilation
33110 @cindex JIT compilation interface
33112 This chapter documents @value{GDBN}'s @dfn{just-in-time} (JIT) compilation
33113 interface. A JIT compiler is a program or library that generates native
33114 executable code at runtime and executes it, usually in order to achieve good
33115 performance while maintaining platform independence.
33117 Programs that use JIT compilation are normally difficult to debug because
33118 portions of their code are generated at runtime, instead of being loaded from
33119 object files, which is where @value{GDBN} normally finds the program's symbols
33120 and debug information. In order to debug programs that use JIT compilation,
33121 @value{GDBN} has an interface that allows the program to register in-memory
33122 symbol files with @value{GDBN} at runtime.
33124 If you are using @value{GDBN} to debug a program that uses this interface, then
33125 it should work transparently so long as you have not stripped the binary. If
33126 you are developing a JIT compiler, then the interface is documented in the rest
33127 of this chapter. At this time, the only known client of this interface is the
33130 Broadly speaking, the JIT interface mirrors the dynamic loader interface. The
33131 JIT compiler communicates with @value{GDBN} by writing data into a global
33132 variable and calling a fuction at a well-known symbol. When @value{GDBN}
33133 attaches, it reads a linked list of symbol files from the global variable to
33134 find existing code, and puts a breakpoint in the function so that it can find
33135 out about additional code.
33138 * Declarations:: Relevant C struct declarations
33139 * Registering Code:: Steps to register code
33140 * Unregistering Code:: Steps to unregister code
33141 * Custom Debug Info:: Emit debug information in a custom format
33145 @section JIT Declarations
33147 These are the relevant struct declarations that a C program should include to
33148 implement the interface:
33158 struct jit_code_entry
33160 struct jit_code_entry *next_entry;
33161 struct jit_code_entry *prev_entry;
33162 const char *symfile_addr;
33163 uint64_t symfile_size;
33166 struct jit_descriptor
33169 /* This type should be jit_actions_t, but we use uint32_t
33170 to be explicit about the bitwidth. */
33171 uint32_t action_flag;
33172 struct jit_code_entry *relevant_entry;
33173 struct jit_code_entry *first_entry;
33176 /* GDB puts a breakpoint in this function. */
33177 void __attribute__((noinline)) __jit_debug_register_code() @{ @};
33179 /* Make sure to specify the version statically, because the
33180 debugger may check the version before we can set it. */
33181 struct jit_descriptor __jit_debug_descriptor = @{ 1, 0, 0, 0 @};
33184 If the JIT is multi-threaded, then it is important that the JIT synchronize any
33185 modifications to this global data properly, which can easily be done by putting
33186 a global mutex around modifications to these structures.
33188 @node Registering Code
33189 @section Registering Code
33191 To register code with @value{GDBN}, the JIT should follow this protocol:
33195 Generate an object file in memory with symbols and other desired debug
33196 information. The file must include the virtual addresses of the sections.
33199 Create a code entry for the file, which gives the start and size of the symbol
33203 Add it to the linked list in the JIT descriptor.
33206 Point the relevant_entry field of the descriptor at the entry.
33209 Set @code{action_flag} to @code{JIT_REGISTER} and call
33210 @code{__jit_debug_register_code}.
33213 When @value{GDBN} is attached and the breakpoint fires, @value{GDBN} uses the
33214 @code{relevant_entry} pointer so it doesn't have to walk the list looking for
33215 new code. However, the linked list must still be maintained in order to allow
33216 @value{GDBN} to attach to a running process and still find the symbol files.
33218 @node Unregistering Code
33219 @section Unregistering Code
33221 If code is freed, then the JIT should use the following protocol:
33225 Remove the code entry corresponding to the code from the linked list.
33228 Point the @code{relevant_entry} field of the descriptor at the code entry.
33231 Set @code{action_flag} to @code{JIT_UNREGISTER} and call
33232 @code{__jit_debug_register_code}.
33235 If the JIT frees or recompiles code without unregistering it, then @value{GDBN}
33236 and the JIT will leak the memory used for the associated symbol files.
33238 @node Custom Debug Info
33239 @section Custom Debug Info
33240 @cindex custom JIT debug info
33241 @cindex JIT debug info reader
33243 Generating debug information in platform-native file formats (like ELF
33244 or COFF) may be an overkill for JIT compilers; especially if all the
33245 debug info is used for is displaying a meaningful backtrace. The
33246 issue can be resolved by having the JIT writers decide on a debug info
33247 format and also provide a reader that parses the debug info generated
33248 by the JIT compiler. This section gives a brief overview on writing
33249 such a parser. More specific details can be found in the source file
33250 @file{gdb/jit-reader.in}, which is also installed as a header at
33251 @file{@var{includedir}/gdb/jit-reader.h} for easy inclusion.
33253 The reader is implemented as a shared object (so this functionality is
33254 not available on platforms which don't allow loading shared objects at
33255 runtime). Two @value{GDBN} commands, @code{jit-reader-load} and
33256 @code{jit-reader-unload} are provided, to be used to load and unload
33257 the readers from a preconfigured directory. Once loaded, the shared
33258 object is used the parse the debug information emitted by the JIT
33262 * Using JIT Debug Info Readers:: How to use supplied readers correctly
33263 * Writing JIT Debug Info Readers:: Creating a debug-info reader
33266 @node Using JIT Debug Info Readers
33267 @subsection Using JIT Debug Info Readers
33268 @kindex jit-reader-load
33269 @kindex jit-reader-unload
33271 Readers can be loaded and unloaded using the @code{jit-reader-load}
33272 and @code{jit-reader-unload} commands.
33275 @item jit-reader-load @var{reader}
33276 Load the JIT reader named @var{reader}, which is a shared
33277 object specified as either an absolute or a relative file name. In
33278 the latter case, @value{GDBN} will try to load the reader from a
33279 pre-configured directory, usually @file{@var{libdir}/gdb/} on a UNIX
33280 system (here @var{libdir} is the system library directory, often
33281 @file{/usr/local/lib}).
33283 Only one reader can be active at a time; trying to load a second
33284 reader when one is already loaded will result in @value{GDBN}
33285 reporting an error. A new JIT reader can be loaded by first unloading
33286 the current one using @code{jit-reader-unload} and then invoking
33287 @code{jit-reader-load}.
33289 @item jit-reader-unload
33290 Unload the currently loaded JIT reader.
33294 @node Writing JIT Debug Info Readers
33295 @subsection Writing JIT Debug Info Readers
33296 @cindex writing JIT debug info readers
33298 As mentioned, a reader is essentially a shared object conforming to a
33299 certain ABI. This ABI is described in @file{jit-reader.h}.
33301 @file{jit-reader.h} defines the structures, macros and functions
33302 required to write a reader. It is installed (along with
33303 @value{GDBN}), in @file{@var{includedir}/gdb} where @var{includedir} is
33304 the system include directory.
33306 Readers need to be released under a GPL compatible license. A reader
33307 can be declared as released under such a license by placing the macro
33308 @code{GDB_DECLARE_GPL_COMPATIBLE_READER} in a source file.
33310 The entry point for readers is the symbol @code{gdb_init_reader},
33311 which is expected to be a function with the prototype
33313 @findex gdb_init_reader
33315 extern struct gdb_reader_funcs *gdb_init_reader (void);
33318 @cindex @code{struct gdb_reader_funcs}
33320 @code{struct gdb_reader_funcs} contains a set of pointers to callback
33321 functions. These functions are executed to read the debug info
33322 generated by the JIT compiler (@code{read}), to unwind stack frames
33323 (@code{unwind}) and to create canonical frame IDs
33324 (@code{get_Frame_id}). It also has a callback that is called when the
33325 reader is being unloaded (@code{destroy}). The struct looks like this
33328 struct gdb_reader_funcs
33330 /* Must be set to GDB_READER_INTERFACE_VERSION. */
33331 int reader_version;
33333 /* For use by the reader. */
33336 gdb_read_debug_info *read;
33337 gdb_unwind_frame *unwind;
33338 gdb_get_frame_id *get_frame_id;
33339 gdb_destroy_reader *destroy;
33343 @cindex @code{struct gdb_symbol_callbacks}
33344 @cindex @code{struct gdb_unwind_callbacks}
33346 The callbacks are provided with another set of callbacks by
33347 @value{GDBN} to do their job. For @code{read}, these callbacks are
33348 passed in a @code{struct gdb_symbol_callbacks} and for @code{unwind}
33349 and @code{get_frame_id}, in a @code{struct gdb_unwind_callbacks}.
33350 @code{struct gdb_symbol_callbacks} has callbacks to create new object
33351 files and new symbol tables inside those object files. @code{struct
33352 gdb_unwind_callbacks} has callbacks to read registers off the current
33353 frame and to write out the values of the registers in the previous
33354 frame. Both have a callback (@code{target_read}) to read bytes off the
33355 target's address space.
33357 @node In-Process Agent
33358 @chapter In-Process Agent
33359 @cindex debugging agent
33360 The traditional debugging model is conceptually low-speed, but works fine,
33361 because most bugs can be reproduced in debugging-mode execution. However,
33362 as multi-core or many-core processors are becoming mainstream, and
33363 multi-threaded programs become more and more popular, there should be more
33364 and more bugs that only manifest themselves at normal-mode execution, for
33365 example, thread races, because debugger's interference with the program's
33366 timing may conceal the bugs. On the other hand, in some applications,
33367 it is not feasible for the debugger to interrupt the program's execution
33368 long enough for the developer to learn anything helpful about its behavior.
33369 If the program's correctness depends on its real-time behavior, delays
33370 introduced by a debugger might cause the program to fail, even when the
33371 code itself is correct. It is useful to be able to observe the program's
33372 behavior without interrupting it.
33374 Therefore, traditional debugging model is too intrusive to reproduce
33375 some bugs. In order to reduce the interference with the program, we can
33376 reduce the number of operations performed by debugger. The
33377 @dfn{In-Process Agent}, a shared library, is running within the same
33378 process with inferior, and is able to perform some debugging operations
33379 itself. As a result, debugger is only involved when necessary, and
33380 performance of debugging can be improved accordingly. Note that
33381 interference with program can be reduced but can't be removed completely,
33382 because the in-process agent will still stop or slow down the program.
33384 The in-process agent can interpret and execute Agent Expressions
33385 (@pxref{Agent Expressions}) during performing debugging operations. The
33386 agent expressions can be used for different purposes, such as collecting
33387 data in tracepoints, and condition evaluation in breakpoints.
33389 @anchor{Control Agent}
33390 You can control whether the in-process agent is used as an aid for
33391 debugging with the following commands:
33394 @kindex set agent on
33396 Causes the in-process agent to perform some operations on behalf of the
33397 debugger. Just which operations requested by the user will be done
33398 by the in-process agent depends on the its capabilities. For example,
33399 if you request to evaluate breakpoint conditions in the in-process agent,
33400 and the in-process agent has such capability as well, then breakpoint
33401 conditions will be evaluated in the in-process agent.
33403 @kindex set agent off
33404 @item set agent off
33405 Disables execution of debugging operations by the in-process agent. All
33406 of the operations will be performed by @value{GDBN}.
33410 Display the current setting of execution of debugging operations by
33411 the in-process agent.
33415 * In-Process Agent Protocol::
33418 @node In-Process Agent Protocol
33419 @section In-Process Agent Protocol
33420 @cindex in-process agent protocol
33422 The in-process agent is able to communicate with both @value{GDBN} and
33423 GDBserver (@pxref{In-Process Agent}). This section documents the protocol
33424 used for communications between @value{GDBN} or GDBserver and the IPA.
33425 In general, @value{GDBN} or GDBserver sends commands
33426 (@pxref{IPA Protocol Commands}) and data to in-process agent, and then
33427 in-process agent replies back with the return result of the command, or
33428 some other information. The data sent to in-process agent is composed
33429 of primitive data types, such as 4-byte or 8-byte type, and composite
33430 types, which are called objects (@pxref{IPA Protocol Objects}).
33433 * IPA Protocol Objects::
33434 * IPA Protocol Commands::
33437 @node IPA Protocol Objects
33438 @subsection IPA Protocol Objects
33439 @cindex ipa protocol objects
33441 The commands sent to and results received from agent may contain some
33442 complex data types called @dfn{objects}.
33444 The in-process agent is running on the same machine with @value{GDBN}
33445 or GDBserver, so it doesn't have to handle as much differences between
33446 two ends as remote protocol (@pxref{Remote Protocol}) tries to handle.
33447 However, there are still some differences of two ends in two processes:
33451 word size. On some 64-bit machines, @value{GDBN} or GDBserver can be
33452 compiled as a 64-bit executable, while in-process agent is a 32-bit one.
33454 ABI. Some machines may have multiple types of ABI, @value{GDBN} or
33455 GDBserver is compiled with one, and in-process agent is compiled with
33459 Here are the IPA Protocol Objects:
33463 agent expression object. It represents an agent expression
33464 (@pxref{Agent Expressions}).
33465 @anchor{agent expression object}
33467 tracepoint action object. It represents a tracepoint action
33468 (@pxref{Tracepoint Actions,,Tracepoint Action Lists}) to collect registers,
33469 memory, static trace data and to evaluate expression.
33470 @anchor{tracepoint action object}
33472 tracepoint object. It represents a tracepoint (@pxref{Tracepoints}).
33473 @anchor{tracepoint object}
33477 The following table describes important attributes of each IPA protocol
33480 @multitable @columnfractions .30 .20 .50
33481 @headitem Name @tab Size @tab Description
33482 @item @emph{agent expression object} @tab @tab
33483 @item length @tab 4 @tab length of bytes code
33484 @item byte code @tab @var{length} @tab contents of byte code
33485 @item @emph{tracepoint action for collecting memory} @tab @tab
33486 @item 'M' @tab 1 @tab type of tracepoint action
33487 @item addr @tab 8 @tab if @var{basereg} is @samp{-1}, @var{addr} is the
33488 address of the lowest byte to collect, otherwise @var{addr} is the offset
33489 of @var{basereg} for memory collecting.
33490 @item len @tab 8 @tab length of memory for collecting
33491 @item basereg @tab 4 @tab the register number containing the starting
33492 memory address for collecting.
33493 @item @emph{tracepoint action for collecting registers} @tab @tab
33494 @item 'R' @tab 1 @tab type of tracepoint action
33495 @item @emph{tracepoint action for collecting static trace data} @tab @tab
33496 @item 'L' @tab 1 @tab type of tracepoint action
33497 @item @emph{tracepoint action for expression evaluation} @tab @tab
33498 @item 'X' @tab 1 @tab type of tracepoint action
33499 @item agent expression @tab length of @tab @ref{agent expression object}
33500 @item @emph{tracepoint object} @tab @tab
33501 @item number @tab 4 @tab number of tracepoint
33502 @item address @tab 8 @tab address of tracepoint inserted on
33503 @item type @tab 4 @tab type of tracepoint
33504 @item enabled @tab 1 @tab enable or disable of tracepoint
33505 @item step_count @tab 8 @tab step
33506 @item pass_count @tab 8 @tab pass
33507 @item numactions @tab 4 @tab number of tracepoint actions
33508 @item hit count @tab 8 @tab hit count
33509 @item trace frame usage @tab 8 @tab trace frame usage
33510 @item compiled_cond @tab 8 @tab compiled condition
33511 @item orig_size @tab 8 @tab orig size
33512 @item condition @tab 4 if condition is NULL otherwise length of
33513 @ref{agent expression object}
33514 @tab zero if condition is NULL, otherwise is
33515 @ref{agent expression object}
33516 @item actions @tab variable
33517 @tab numactions number of @ref{tracepoint action object}
33520 @node IPA Protocol Commands
33521 @subsection IPA Protocol Commands
33522 @cindex ipa protocol commands
33524 The spaces in each command are delimiters to ease reading this commands
33525 specification. They don't exist in real commands.
33529 @item FastTrace:@var{tracepoint_object} @var{gdb_jump_pad_head}
33530 Installs a new fast tracepoint described by @var{tracepoint_object}
33531 (@pxref{tracepoint object}). The @var{gdb_jump_pad_head}, 8-byte long, is the
33532 head of @dfn{jumppad}, which is used to jump to data collection routine
33537 @item OK @var{target_address} @var{gdb_jump_pad_head} @var{fjump_size} @var{fjump}
33538 @var{target_address} is address of tracepoint in the inferior.
33539 The @var{gdb_jump_pad_head} is updated head of jumppad. Both of
33540 @var{target_address} and @var{gdb_jump_pad_head} are 8-byte long.
33541 The @var{fjump} contains a sequence of instructions jump to jumppad entry.
33542 The @var{fjump_size}, 4-byte long, is the size of @var{fjump}.
33549 Closes the in-process agent. This command is sent when @value{GDBN} or GDBserver
33550 is about to kill inferiors.
33558 @item probe_marker_at:@var{address}
33559 Asks in-process agent to probe the marker at @var{address}.
33566 @item unprobe_marker_at:@var{address}
33567 Asks in-process agent to unprobe the marker at @var{address}.
33571 @chapter Reporting Bugs in @value{GDBN}
33572 @cindex bugs in @value{GDBN}
33573 @cindex reporting bugs in @value{GDBN}
33575 Your bug reports play an essential role in making @value{GDBN} reliable.
33577 Reporting a bug may help you by bringing a solution to your problem, or it
33578 may not. But in any case the principal function of a bug report is to help
33579 the entire community by making the next version of @value{GDBN} work better. Bug
33580 reports are your contribution to the maintenance of @value{GDBN}.
33582 In order for a bug report to serve its purpose, you must include the
33583 information that enables us to fix the bug.
33586 * Bug Criteria:: Have you found a bug?
33587 * Bug Reporting:: How to report bugs
33591 @section Have You Found a Bug?
33592 @cindex bug criteria
33594 If you are not sure whether you have found a bug, here are some guidelines:
33597 @cindex fatal signal
33598 @cindex debugger crash
33599 @cindex crash of debugger
33601 If the debugger gets a fatal signal, for any input whatever, that is a
33602 @value{GDBN} bug. Reliable debuggers never crash.
33604 @cindex error on valid input
33606 If @value{GDBN} produces an error message for valid input, that is a
33607 bug. (Note that if you're cross debugging, the problem may also be
33608 somewhere in the connection to the target.)
33610 @cindex invalid input
33612 If @value{GDBN} does not produce an error message for invalid input,
33613 that is a bug. However, you should note that your idea of
33614 ``invalid input'' might be our idea of ``an extension'' or ``support
33615 for traditional practice''.
33618 If you are an experienced user of debugging tools, your suggestions
33619 for improvement of @value{GDBN} are welcome in any case.
33622 @node Bug Reporting
33623 @section How to Report Bugs
33624 @cindex bug reports
33625 @cindex @value{GDBN} bugs, reporting
33627 A number of companies and individuals offer support for @sc{gnu} products.
33628 If you obtained @value{GDBN} from a support organization, we recommend you
33629 contact that organization first.
33631 You can find contact information for many support companies and
33632 individuals in the file @file{etc/SERVICE} in the @sc{gnu} Emacs
33634 @c should add a web page ref...
33637 @ifset BUGURL_DEFAULT
33638 In any event, we also recommend that you submit bug reports for
33639 @value{GDBN}. The preferred method is to submit them directly using
33640 @uref{http://www.gnu.org/software/gdb/bugs/, @value{GDBN}'s Bugs web
33641 page}. Alternatively, the @email{bug-gdb@@gnu.org, e-mail gateway} can
33644 @strong{Do not send bug reports to @samp{info-gdb}, or to
33645 @samp{help-gdb}, or to any newsgroups.} Most users of @value{GDBN} do
33646 not want to receive bug reports. Those that do have arranged to receive
33649 The mailing list @samp{bug-gdb} has a newsgroup @samp{gnu.gdb.bug} which
33650 serves as a repeater. The mailing list and the newsgroup carry exactly
33651 the same messages. Often people think of posting bug reports to the
33652 newsgroup instead of mailing them. This appears to work, but it has one
33653 problem which can be crucial: a newsgroup posting often lacks a mail
33654 path back to the sender. Thus, if we need to ask for more information,
33655 we may be unable to reach you. For this reason, it is better to send
33656 bug reports to the mailing list.
33658 @ifclear BUGURL_DEFAULT
33659 In any event, we also recommend that you submit bug reports for
33660 @value{GDBN} to @value{BUGURL}.
33664 The fundamental principle of reporting bugs usefully is this:
33665 @strong{report all the facts}. If you are not sure whether to state a
33666 fact or leave it out, state it!
33668 Often people omit facts because they think they know what causes the
33669 problem and assume that some details do not matter. Thus, you might
33670 assume that the name of the variable you use in an example does not matter.
33671 Well, probably it does not, but one cannot be sure. Perhaps the bug is a
33672 stray memory reference which happens to fetch from the location where that
33673 name is stored in memory; perhaps, if the name were different, the contents
33674 of that location would fool the debugger into doing the right thing despite
33675 the bug. Play it safe and give a specific, complete example. That is the
33676 easiest thing for you to do, and the most helpful.
33678 Keep in mind that the purpose of a bug report is to enable us to fix the
33679 bug. It may be that the bug has been reported previously, but neither
33680 you nor we can know that unless your bug report is complete and
33683 Sometimes people give a few sketchy facts and ask, ``Does this ring a
33684 bell?'' Those bug reports are useless, and we urge everyone to
33685 @emph{refuse to respond to them} except to chide the sender to report
33688 To enable us to fix the bug, you should include all these things:
33692 The version of @value{GDBN}. @value{GDBN} announces it if you start
33693 with no arguments; you can also print it at any time using @code{show
33696 Without this, we will not know whether there is any point in looking for
33697 the bug in the current version of @value{GDBN}.
33700 The type of machine you are using, and the operating system name and
33704 The details of the @value{GDBN} build-time configuration.
33705 @value{GDBN} shows these details if you invoke it with the
33706 @option{--configuration} command-line option, or if you type
33707 @code{show configuration} at @value{GDBN}'s prompt.
33710 What compiler (and its version) was used to compile @value{GDBN}---e.g.@:
33711 ``@value{GCC}--2.8.1''.
33714 What compiler (and its version) was used to compile the program you are
33715 debugging---e.g.@: ``@value{GCC}--2.8.1'', or ``HP92453-01 A.10.32.03 HP
33716 C Compiler''. For @value{NGCC}, you can say @kbd{@value{GCC} --version}
33717 to get this information; for other compilers, see the documentation for
33721 The command arguments you gave the compiler to compile your example and
33722 observe the bug. For example, did you use @samp{-O}? To guarantee
33723 you will not omit something important, list them all. A copy of the
33724 Makefile (or the output from make) is sufficient.
33726 If we were to try to guess the arguments, we would probably guess wrong
33727 and then we might not encounter the bug.
33730 A complete input script, and all necessary source files, that will
33734 A description of what behavior you observe that you believe is
33735 incorrect. For example, ``It gets a fatal signal.''
33737 Of course, if the bug is that @value{GDBN} gets a fatal signal, then we
33738 will certainly notice it. But if the bug is incorrect output, we might
33739 not notice unless it is glaringly wrong. You might as well not give us
33740 a chance to make a mistake.
33742 Even if the problem you experience is a fatal signal, you should still
33743 say so explicitly. Suppose something strange is going on, such as, your
33744 copy of @value{GDBN} is out of synch, or you have encountered a bug in
33745 the C library on your system. (This has happened!) Your copy might
33746 crash and ours would not. If you told us to expect a crash, then when
33747 ours fails to crash, we would know that the bug was not happening for
33748 us. If you had not told us to expect a crash, then we would not be able
33749 to draw any conclusion from our observations.
33752 @cindex recording a session script
33753 To collect all this information, you can use a session recording program
33754 such as @command{script}, which is available on many Unix systems.
33755 Just run your @value{GDBN} session inside @command{script} and then
33756 include the @file{typescript} file with your bug report.
33758 Another way to record a @value{GDBN} session is to run @value{GDBN}
33759 inside Emacs and then save the entire buffer to a file.
33762 If you wish to suggest changes to the @value{GDBN} source, send us context
33763 diffs. If you even discuss something in the @value{GDBN} source, refer to
33764 it by context, not by line number.
33766 The line numbers in our development sources will not match those in your
33767 sources. Your line numbers would convey no useful information to us.
33771 Here are some things that are not necessary:
33775 A description of the envelope of the bug.
33777 Often people who encounter a bug spend a lot of time investigating
33778 which changes to the input file will make the bug go away and which
33779 changes will not affect it.
33781 This is often time consuming and not very useful, because the way we
33782 will find the bug is by running a single example under the debugger
33783 with breakpoints, not by pure deduction from a series of examples.
33784 We recommend that you save your time for something else.
33786 Of course, if you can find a simpler example to report @emph{instead}
33787 of the original one, that is a convenience for us. Errors in the
33788 output will be easier to spot, running under the debugger will take
33789 less time, and so on.
33791 However, simplification is not vital; if you do not want to do this,
33792 report the bug anyway and send us the entire test case you used.
33795 A patch for the bug.
33797 A patch for the bug does help us if it is a good one. But do not omit
33798 the necessary information, such as the test case, on the assumption that
33799 a patch is all we need. We might see problems with your patch and decide
33800 to fix the problem another way, or we might not understand it at all.
33802 Sometimes with a program as complicated as @value{GDBN} it is very hard to
33803 construct an example that will make the program follow a certain path
33804 through the code. If you do not send us the example, we will not be able
33805 to construct one, so we will not be able to verify that the bug is fixed.
33807 And if we cannot understand what bug you are trying to fix, or why your
33808 patch should be an improvement, we will not install it. A test case will
33809 help us to understand.
33812 A guess about what the bug is or what it depends on.
33814 Such guesses are usually wrong. Even we cannot guess right about such
33815 things without first using the debugger to find the facts.
33818 @c The readline documentation is distributed with the readline code
33819 @c and consists of the two following files:
33822 @c Use -I with makeinfo to point to the appropriate directory,
33823 @c environment var TEXINPUTS with TeX.
33824 @ifclear SYSTEM_READLINE
33825 @include rluser.texi
33826 @include hsuser.texi
33830 @appendix In Memoriam
33832 The @value{GDBN} project mourns the loss of the following long-time
33837 Fred was a long-standing contributor to @value{GDBN} (1991-2006), and
33838 to Free Software in general. Outside of @value{GDBN}, he was known in
33839 the Amiga world for his series of Fish Disks, and the GeekGadget project.
33841 @item Michael Snyder
33842 Michael was one of the Global Maintainers of the @value{GDBN} project,
33843 with contributions recorded as early as 1996, until 2011. In addition
33844 to his day to day participation, he was a large driving force behind
33845 adding Reverse Debugging to @value{GDBN}.
33848 Beyond their technical contributions to the project, they were also
33849 enjoyable members of the Free Software Community. We will miss them.
33851 @node Formatting Documentation
33852 @appendix Formatting Documentation
33854 @cindex @value{GDBN} reference card
33855 @cindex reference card
33856 The @value{GDBN} 4 release includes an already-formatted reference card, ready
33857 for printing with PostScript or Ghostscript, in the @file{gdb}
33858 subdirectory of the main source directory@footnote{In
33859 @file{gdb-@value{GDBVN}/gdb/refcard.ps} of the version @value{GDBVN}
33860 release.}. If you can use PostScript or Ghostscript with your printer,
33861 you can print the reference card immediately with @file{refcard.ps}.
33863 The release also includes the source for the reference card. You
33864 can format it, using @TeX{}, by typing:
33870 The @value{GDBN} reference card is designed to print in @dfn{landscape}
33871 mode on US ``letter'' size paper;
33872 that is, on a sheet 11 inches wide by 8.5 inches
33873 high. You will need to specify this form of printing as an option to
33874 your @sc{dvi} output program.
33876 @cindex documentation
33878 All the documentation for @value{GDBN} comes as part of the machine-readable
33879 distribution. The documentation is written in Texinfo format, which is
33880 a documentation system that uses a single source file to produce both
33881 on-line information and a printed manual. You can use one of the Info
33882 formatting commands to create the on-line version of the documentation
33883 and @TeX{} (or @code{texi2roff}) to typeset the printed version.
33885 @value{GDBN} includes an already formatted copy of the on-line Info
33886 version of this manual in the @file{gdb} subdirectory. The main Info
33887 file is @file{gdb-@value{GDBVN}/gdb/gdb.info}, and it refers to
33888 subordinate files matching @samp{gdb.info*} in the same directory. If
33889 necessary, you can print out these files, or read them with any editor;
33890 but they are easier to read using the @code{info} subsystem in @sc{gnu}
33891 Emacs or the standalone @code{info} program, available as part of the
33892 @sc{gnu} Texinfo distribution.
33894 If you want to format these Info files yourself, you need one of the
33895 Info formatting programs, such as @code{texinfo-format-buffer} or
33898 If you have @code{makeinfo} installed, and are in the top level
33899 @value{GDBN} source directory (@file{gdb-@value{GDBVN}}, in the case of
33900 version @value{GDBVN}), you can make the Info file by typing:
33907 If you want to typeset and print copies of this manual, you need @TeX{},
33908 a program to print its @sc{dvi} output files, and @file{texinfo.tex}, the
33909 Texinfo definitions file.
33911 @TeX{} is a typesetting program; it does not print files directly, but
33912 produces output files called @sc{dvi} files. To print a typeset
33913 document, you need a program to print @sc{dvi} files. If your system
33914 has @TeX{} installed, chances are it has such a program. The precise
33915 command to use depends on your system; @kbd{lpr -d} is common; another
33916 (for PostScript devices) is @kbd{dvips}. The @sc{dvi} print command may
33917 require a file name without any extension or a @samp{.dvi} extension.
33919 @TeX{} also requires a macro definitions file called
33920 @file{texinfo.tex}. This file tells @TeX{} how to typeset a document
33921 written in Texinfo format. On its own, @TeX{} cannot either read or
33922 typeset a Texinfo file. @file{texinfo.tex} is distributed with GDB
33923 and is located in the @file{gdb-@var{version-number}/texinfo}
33926 If you have @TeX{} and a @sc{dvi} printer program installed, you can
33927 typeset and print this manual. First switch to the @file{gdb}
33928 subdirectory of the main source directory (for example, to
33929 @file{gdb-@value{GDBVN}/gdb}) and type:
33935 Then give @file{gdb.dvi} to your @sc{dvi} printing program.
33937 @node Installing GDB
33938 @appendix Installing @value{GDBN}
33939 @cindex installation
33942 * Requirements:: Requirements for building @value{GDBN}
33943 * Running Configure:: Invoking the @value{GDBN} @file{configure} script
33944 * Separate Objdir:: Compiling @value{GDBN} in another directory
33945 * Config Names:: Specifying names for hosts and targets
33946 * Configure Options:: Summary of options for configure
33947 * System-wide configuration:: Having a system-wide init file
33951 @section Requirements for Building @value{GDBN}
33952 @cindex building @value{GDBN}, requirements for
33954 Building @value{GDBN} requires various tools and packages to be available.
33955 Other packages will be used only if they are found.
33957 @heading Tools/Packages Necessary for Building @value{GDBN}
33959 @item ISO C90 compiler
33960 @value{GDBN} is written in ISO C90. It should be buildable with any
33961 working C90 compiler, e.g.@: GCC.
33965 @heading Tools/Packages Optional for Building @value{GDBN}
33969 @value{GDBN} can use the Expat XML parsing library. This library may be
33970 included with your operating system distribution; if it is not, you
33971 can get the latest version from @url{http://expat.sourceforge.net}.
33972 The @file{configure} script will search for this library in several
33973 standard locations; if it is installed in an unusual path, you can
33974 use the @option{--with-libexpat-prefix} option to specify its location.
33980 Remote protocol memory maps (@pxref{Memory Map Format})
33982 Target descriptions (@pxref{Target Descriptions})
33984 Remote shared library lists (@xref{Library List Format},
33985 or alternatively @pxref{Library List Format for SVR4 Targets})
33987 MS-Windows shared libraries (@pxref{Shared Libraries})
33989 Traceframe info (@pxref{Traceframe Info Format})
33991 Branch trace (@pxref{Branch Trace Format},
33992 @pxref{Branch Trace Configuration Format})
33996 @cindex compressed debug sections
33997 @value{GDBN} will use the @samp{zlib} library, if available, to read
33998 compressed debug sections. Some linkers, such as GNU gold, are capable
33999 of producing binaries with compressed debug sections. If @value{GDBN}
34000 is compiled with @samp{zlib}, it will be able to read the debug
34001 information in such binaries.
34003 The @samp{zlib} library is likely included with your operating system
34004 distribution; if it is not, you can get the latest version from
34005 @url{http://zlib.net}.
34008 @value{GDBN}'s features related to character sets (@pxref{Character
34009 Sets}) require a functioning @code{iconv} implementation. If you are
34010 on a GNU system, then this is provided by the GNU C Library. Some
34011 other systems also provide a working @code{iconv}.
34013 If @value{GDBN} is using the @code{iconv} program which is installed
34014 in a non-standard place, you will need to tell @value{GDBN} where to find it.
34015 This is done with @option{--with-iconv-bin} which specifies the
34016 directory that contains the @code{iconv} program.
34018 On systems without @code{iconv}, you can install GNU Libiconv. If you
34019 have previously installed Libiconv, you can use the
34020 @option{--with-libiconv-prefix} option to configure.
34022 @value{GDBN}'s top-level @file{configure} and @file{Makefile} will
34023 arrange to build Libiconv if a directory named @file{libiconv} appears
34024 in the top-most source directory. If Libiconv is built this way, and
34025 if the operating system does not provide a suitable @code{iconv}
34026 implementation, then the just-built library will automatically be used
34027 by @value{GDBN}. One easy way to set this up is to download GNU
34028 Libiconv, unpack it, and then rename the directory holding the
34029 Libiconv source code to @samp{libiconv}.
34032 @node Running Configure
34033 @section Invoking the @value{GDBN} @file{configure} Script
34034 @cindex configuring @value{GDBN}
34035 @value{GDBN} comes with a @file{configure} script that automates the process
34036 of preparing @value{GDBN} for installation; you can then use @code{make} to
34037 build the @code{gdb} program.
34039 @c irrelevant in info file; it's as current as the code it lives with.
34040 @footnote{If you have a more recent version of @value{GDBN} than @value{GDBVN},
34041 look at the @file{README} file in the sources; we may have improved the
34042 installation procedures since publishing this manual.}
34045 The @value{GDBN} distribution includes all the source code you need for
34046 @value{GDBN} in a single directory, whose name is usually composed by
34047 appending the version number to @samp{gdb}.
34049 For example, the @value{GDBN} version @value{GDBVN} distribution is in the
34050 @file{gdb-@value{GDBVN}} directory. That directory contains:
34053 @item gdb-@value{GDBVN}/configure @r{(and supporting files)}
34054 script for configuring @value{GDBN} and all its supporting libraries
34056 @item gdb-@value{GDBVN}/gdb
34057 the source specific to @value{GDBN} itself
34059 @item gdb-@value{GDBVN}/bfd
34060 source for the Binary File Descriptor library
34062 @item gdb-@value{GDBVN}/include
34063 @sc{gnu} include files
34065 @item gdb-@value{GDBVN}/libiberty
34066 source for the @samp{-liberty} free software library
34068 @item gdb-@value{GDBVN}/opcodes
34069 source for the library of opcode tables and disassemblers
34071 @item gdb-@value{GDBVN}/readline
34072 source for the @sc{gnu} command-line interface
34074 @item gdb-@value{GDBVN}/glob
34075 source for the @sc{gnu} filename pattern-matching subroutine
34077 @item gdb-@value{GDBVN}/mmalloc
34078 source for the @sc{gnu} memory-mapped malloc package
34081 The simplest way to configure and build @value{GDBN} is to run @file{configure}
34082 from the @file{gdb-@var{version-number}} source directory, which in
34083 this example is the @file{gdb-@value{GDBVN}} directory.
34085 First switch to the @file{gdb-@var{version-number}} source directory
34086 if you are not already in it; then run @file{configure}. Pass the
34087 identifier for the platform on which @value{GDBN} will run as an
34093 cd gdb-@value{GDBVN}
34094 ./configure @var{host}
34099 where @var{host} is an identifier such as @samp{sun4} or
34100 @samp{decstation}, that identifies the platform where @value{GDBN} will run.
34101 (You can often leave off @var{host}; @file{configure} tries to guess the
34102 correct value by examining your system.)
34104 Running @samp{configure @var{host}} and then running @code{make} builds the
34105 @file{bfd}, @file{readline}, @file{mmalloc}, and @file{libiberty}
34106 libraries, then @code{gdb} itself. The configured source files, and the
34107 binaries, are left in the corresponding source directories.
34110 @file{configure} is a Bourne-shell (@code{/bin/sh}) script; if your
34111 system does not recognize this automatically when you run a different
34112 shell, you may need to run @code{sh} on it explicitly:
34115 sh configure @var{host}
34118 If you run @file{configure} from a directory that contains source
34119 directories for multiple libraries or programs, such as the
34120 @file{gdb-@value{GDBVN}} source directory for version @value{GDBVN},
34122 creates configuration files for every directory level underneath (unless
34123 you tell it not to, with the @samp{--norecursion} option).
34125 You should run the @file{configure} script from the top directory in the
34126 source tree, the @file{gdb-@var{version-number}} directory. If you run
34127 @file{configure} from one of the subdirectories, you will configure only
34128 that subdirectory. That is usually not what you want. In particular,
34129 if you run the first @file{configure} from the @file{gdb} subdirectory
34130 of the @file{gdb-@var{version-number}} directory, you will omit the
34131 configuration of @file{bfd}, @file{readline}, and other sibling
34132 directories of the @file{gdb} subdirectory. This leads to build errors
34133 about missing include files such as @file{bfd/bfd.h}.
34135 You can install @code{@value{GDBP}} anywhere; it has no hardwired paths.
34136 However, you should make sure that the shell on your path (named by
34137 the @samp{SHELL} environment variable) is publicly readable. Remember
34138 that @value{GDBN} uses the shell to start your program---some systems refuse to
34139 let @value{GDBN} debug child processes whose programs are not readable.
34141 @node Separate Objdir
34142 @section Compiling @value{GDBN} in Another Directory
34144 If you want to run @value{GDBN} versions for several host or target machines,
34145 you need a different @code{gdb} compiled for each combination of
34146 host and target. @file{configure} is designed to make this easy by
34147 allowing you to generate each configuration in a separate subdirectory,
34148 rather than in the source directory. If your @code{make} program
34149 handles the @samp{VPATH} feature (@sc{gnu} @code{make} does), running
34150 @code{make} in each of these directories builds the @code{gdb}
34151 program specified there.
34153 To build @code{gdb} in a separate directory, run @file{configure}
34154 with the @samp{--srcdir} option to specify where to find the source.
34155 (You also need to specify a path to find @file{configure}
34156 itself from your working directory. If the path to @file{configure}
34157 would be the same as the argument to @samp{--srcdir}, you can leave out
34158 the @samp{--srcdir} option; it is assumed.)
34160 For example, with version @value{GDBVN}, you can build @value{GDBN} in a
34161 separate directory for a Sun 4 like this:
34165 cd gdb-@value{GDBVN}
34168 ../gdb-@value{GDBVN}/configure sun4
34173 When @file{configure} builds a configuration using a remote source
34174 directory, it creates a tree for the binaries with the same structure
34175 (and using the same names) as the tree under the source directory. In
34176 the example, you'd find the Sun 4 library @file{libiberty.a} in the
34177 directory @file{gdb-sun4/libiberty}, and @value{GDBN} itself in
34178 @file{gdb-sun4/gdb}.
34180 Make sure that your path to the @file{configure} script has just one
34181 instance of @file{gdb} in it. If your path to @file{configure} looks
34182 like @file{../gdb-@value{GDBVN}/gdb/configure}, you are configuring only
34183 one subdirectory of @value{GDBN}, not the whole package. This leads to
34184 build errors about missing include files such as @file{bfd/bfd.h}.
34186 One popular reason to build several @value{GDBN} configurations in separate
34187 directories is to configure @value{GDBN} for cross-compiling (where
34188 @value{GDBN} runs on one machine---the @dfn{host}---while debugging
34189 programs that run on another machine---the @dfn{target}).
34190 You specify a cross-debugging target by
34191 giving the @samp{--target=@var{target}} option to @file{configure}.
34193 When you run @code{make} to build a program or library, you must run
34194 it in a configured directory---whatever directory you were in when you
34195 called @file{configure} (or one of its subdirectories).
34197 The @code{Makefile} that @file{configure} generates in each source
34198 directory also runs recursively. If you type @code{make} in a source
34199 directory such as @file{gdb-@value{GDBVN}} (or in a separate configured
34200 directory configured with @samp{--srcdir=@var{dirname}/gdb-@value{GDBVN}}), you
34201 will build all the required libraries, and then build GDB.
34203 When you have multiple hosts or targets configured in separate
34204 directories, you can run @code{make} on them in parallel (for example,
34205 if they are NFS-mounted on each of the hosts); they will not interfere
34209 @section Specifying Names for Hosts and Targets
34211 The specifications used for hosts and targets in the @file{configure}
34212 script are based on a three-part naming scheme, but some short predefined
34213 aliases are also supported. The full naming scheme encodes three pieces
34214 of information in the following pattern:
34217 @var{architecture}-@var{vendor}-@var{os}
34220 For example, you can use the alias @code{sun4} as a @var{host} argument,
34221 or as the value for @var{target} in a @code{--target=@var{target}}
34222 option. The equivalent full name is @samp{sparc-sun-sunos4}.
34224 The @file{configure} script accompanying @value{GDBN} does not provide
34225 any query facility to list all supported host and target names or
34226 aliases. @file{configure} calls the Bourne shell script
34227 @code{config.sub} to map abbreviations to full names; you can read the
34228 script, if you wish, or you can use it to test your guesses on
34229 abbreviations---for example:
34232 % sh config.sub i386-linux
34234 % sh config.sub alpha-linux
34235 alpha-unknown-linux-gnu
34236 % sh config.sub hp9k700
34238 % sh config.sub sun4
34239 sparc-sun-sunos4.1.1
34240 % sh config.sub sun3
34241 m68k-sun-sunos4.1.1
34242 % sh config.sub i986v
34243 Invalid configuration `i986v': machine `i986v' not recognized
34247 @code{config.sub} is also distributed in the @value{GDBN} source
34248 directory (@file{gdb-@value{GDBVN}}, for version @value{GDBVN}).
34250 @node Configure Options
34251 @section @file{configure} Options
34253 Here is a summary of the @file{configure} options and arguments that
34254 are most often useful for building @value{GDBN}. @file{configure} also has
34255 several other options not listed here. @inforef{What Configure
34256 Does,,configure.info}, for a full explanation of @file{configure}.
34259 configure @r{[}--help@r{]}
34260 @r{[}--prefix=@var{dir}@r{]}
34261 @r{[}--exec-prefix=@var{dir}@r{]}
34262 @r{[}--srcdir=@var{dirname}@r{]}
34263 @r{[}--norecursion@r{]} @r{[}--rm@r{]}
34264 @r{[}--target=@var{target}@r{]}
34269 You may introduce options with a single @samp{-} rather than
34270 @samp{--} if you prefer; but you may abbreviate option names if you use
34275 Display a quick summary of how to invoke @file{configure}.
34277 @item --prefix=@var{dir}
34278 Configure the source to install programs and files under directory
34281 @item --exec-prefix=@var{dir}
34282 Configure the source to install programs under directory
34285 @c avoid splitting the warning from the explanation:
34287 @item --srcdir=@var{dirname}
34288 @strong{Warning: using this option requires @sc{gnu} @code{make}, or another
34289 @code{make} that implements the @code{VPATH} feature.}@*
34290 Use this option to make configurations in directories separate from the
34291 @value{GDBN} source directories. Among other things, you can use this to
34292 build (or maintain) several configurations simultaneously, in separate
34293 directories. @file{configure} writes configuration-specific files in
34294 the current directory, but arranges for them to use the source in the
34295 directory @var{dirname}. @file{configure} creates directories under
34296 the working directory in parallel to the source directories below
34299 @item --norecursion
34300 Configure only the directory level where @file{configure} is executed; do not
34301 propagate configuration to subdirectories.
34303 @item --target=@var{target}
34304 Configure @value{GDBN} for cross-debugging programs running on the specified
34305 @var{target}. Without this option, @value{GDBN} is configured to debug
34306 programs that run on the same machine (@var{host}) as @value{GDBN} itself.
34308 There is no convenient way to generate a list of all available targets.
34310 @item @var{host} @dots{}
34311 Configure @value{GDBN} to run on the specified @var{host}.
34313 There is no convenient way to generate a list of all available hosts.
34316 There are many other options available as well, but they are generally
34317 needed for special purposes only.
34319 @node System-wide configuration
34320 @section System-wide configuration and settings
34321 @cindex system-wide init file
34323 @value{GDBN} can be configured to have a system-wide init file;
34324 this file will be read and executed at startup (@pxref{Startup, , What
34325 @value{GDBN} does during startup}).
34327 Here is the corresponding configure option:
34330 @item --with-system-gdbinit=@var{file}
34331 Specify that the default location of the system-wide init file is
34335 If @value{GDBN} has been configured with the option @option{--prefix=$prefix},
34336 it may be subject to relocation. Two possible cases:
34340 If the default location of this init file contains @file{$prefix},
34341 it will be subject to relocation. Suppose that the configure options
34342 are @option{--prefix=$prefix --with-system-gdbinit=$prefix/etc/gdbinit};
34343 if @value{GDBN} is moved from @file{$prefix} to @file{$install}, the system
34344 init file is looked for as @file{$install/etc/gdbinit} instead of
34345 @file{$prefix/etc/gdbinit}.
34348 By contrast, if the default location does not contain the prefix,
34349 it will not be relocated. E.g.@: if @value{GDBN} has been configured with
34350 @option{--prefix=/usr/local --with-system-gdbinit=/usr/share/gdb/gdbinit},
34351 then @value{GDBN} will always look for @file{/usr/share/gdb/gdbinit},
34352 wherever @value{GDBN} is installed.
34355 If the configured location of the system-wide init file (as given by the
34356 @option{--with-system-gdbinit} option at configure time) is in the
34357 data-directory (as specified by @option{--with-gdb-datadir} at configure
34358 time) or in one of its subdirectories, then @value{GDBN} will look for the
34359 system-wide init file in the directory specified by the
34360 @option{--data-directory} command-line option.
34361 Note that the system-wide init file is only read once, during @value{GDBN}
34362 initialization. If the data-directory is changed after @value{GDBN} has
34363 started with the @code{set data-directory} command, the file will not be
34367 * System-wide Configuration Scripts:: Installed System-wide Configuration Scripts
34370 @node System-wide Configuration Scripts
34371 @subsection Installed System-wide Configuration Scripts
34372 @cindex system-wide configuration scripts
34374 The @file{system-gdbinit} directory, located inside the data-directory
34375 (as specified by @option{--with-gdb-datadir} at configure time) contains
34376 a number of scripts which can be used as system-wide init files. To
34377 automatically source those scripts at startup, @value{GDBN} should be
34378 configured with @option{--with-system-gdbinit}. Otherwise, any user
34379 should be able to source them by hand as needed.
34381 The following scripts are currently available:
34384 @item @file{elinos.py}
34386 @cindex ELinOS system-wide configuration script
34387 This script is useful when debugging a program on an ELinOS target.
34388 It takes advantage of the environment variables defined in a standard
34389 ELinOS environment in order to determine the location of the system
34390 shared libraries, and then sets the @samp{solib-absolute-prefix}
34391 and @samp{solib-search-path} variables appropriately.
34393 @item @file{wrs-linux.py}
34394 @pindex wrs-linux.py
34395 @cindex Wind River Linux system-wide configuration script
34396 This script is useful when debugging a program on a target running
34397 Wind River Linux. It expects the @env{ENV_PREFIX} to be set to
34398 the host-side sysroot used by the target system.
34402 @node Maintenance Commands
34403 @appendix Maintenance Commands
34404 @cindex maintenance commands
34405 @cindex internal commands
34407 In addition to commands intended for @value{GDBN} users, @value{GDBN}
34408 includes a number of commands intended for @value{GDBN} developers,
34409 that are not documented elsewhere in this manual. These commands are
34410 provided here for reference. (For commands that turn on debugging
34411 messages, see @ref{Debugging Output}.)
34414 @kindex maint agent
34415 @kindex maint agent-eval
34416 @item maint agent @r{[}-at @var{location}@r{,}@r{]} @var{expression}
34417 @itemx maint agent-eval @r{[}-at @var{location}@r{,}@r{]} @var{expression}
34418 Translate the given @var{expression} into remote agent bytecodes.
34419 This command is useful for debugging the Agent Expression mechanism
34420 (@pxref{Agent Expressions}). The @samp{agent} version produces an
34421 expression useful for data collection, such as by tracepoints, while
34422 @samp{maint agent-eval} produces an expression that evaluates directly
34423 to a result. For instance, a collection expression for @code{globa +
34424 globb} will include bytecodes to record four bytes of memory at each
34425 of the addresses of @code{globa} and @code{globb}, while discarding
34426 the result of the addition, while an evaluation expression will do the
34427 addition and return the sum.
34428 If @code{-at} is given, generate remote agent bytecode for @var{location}.
34429 If not, generate remote agent bytecode for current frame PC address.
34431 @kindex maint agent-printf
34432 @item maint agent-printf @var{format},@var{expr},...
34433 Translate the given format string and list of argument expressions
34434 into remote agent bytecodes and display them as a disassembled list.
34435 This command is useful for debugging the agent version of dynamic
34436 printf (@pxref{Dynamic Printf}).
34438 @kindex maint info breakpoints
34439 @item @anchor{maint info breakpoints}maint info breakpoints
34440 Using the same format as @samp{info breakpoints}, display both the
34441 breakpoints you've set explicitly, and those @value{GDBN} is using for
34442 internal purposes. Internal breakpoints are shown with negative
34443 breakpoint numbers. The type column identifies what kind of breakpoint
34448 Normal, explicitly set breakpoint.
34451 Normal, explicitly set watchpoint.
34454 Internal breakpoint, used to handle correctly stepping through
34455 @code{longjmp} calls.
34457 @item longjmp resume
34458 Internal breakpoint at the target of a @code{longjmp}.
34461 Temporary internal breakpoint used by the @value{GDBN} @code{until} command.
34464 Temporary internal breakpoint used by the @value{GDBN} @code{finish} command.
34467 Shared library events.
34471 @kindex maint info btrace
34472 @item maint info btrace
34473 Pint information about raw branch tracing data.
34475 @kindex maint btrace packet-history
34476 @item maint btrace packet-history
34477 Print the raw branch trace packets that are used to compute the
34478 execution history for the @samp{record btrace} command. Both the
34479 information and the format in which it is printed depend on the btrace
34484 For the BTS recording format, print a list of blocks of sequential
34485 code. For each block, the following information is printed:
34489 Newer blocks have higher numbers. The oldest block has number zero.
34490 @item Lowest @samp{PC}
34491 @item Highest @samp{PC}
34495 For the Intel Processor Trace recording format, print a list of
34496 Intel Processor Trace packets. For each packet, the following
34497 information is printed:
34500 @item Packet number
34501 Newer packets have higher numbers. The oldest packet has number zero.
34503 The packet's offset in the trace stream.
34504 @item Packet opcode and payload
34508 @kindex maint btrace clear-packet-history
34509 @item maint btrace clear-packet-history
34510 Discards the cached packet history printed by the @samp{maint btrace
34511 packet-history} command. The history will be computed again when
34514 @kindex maint btrace clear
34515 @item maint btrace clear
34516 Discard the branch trace data. The data will be fetched anew and the
34517 branch trace will be recomputed when needed.
34519 This implicitly truncates the branch trace to a single branch trace
34520 buffer. When updating branch trace incrementally, the branch trace
34521 available to @value{GDBN} may be bigger than a single branch trace
34524 @kindex maint set btrace pt skip-pad
34525 @item maint set btrace pt skip-pad
34526 @kindex maint show btrace pt skip-pad
34527 @item maint show btrace pt skip-pad
34528 Control whether @value{GDBN} will skip PAD packets when computing the
34531 @kindex set displaced-stepping
34532 @kindex show displaced-stepping
34533 @cindex displaced stepping support
34534 @cindex out-of-line single-stepping
34535 @item set displaced-stepping
34536 @itemx show displaced-stepping
34537 Control whether or not @value{GDBN} will do @dfn{displaced stepping}
34538 if the target supports it. Displaced stepping is a way to single-step
34539 over breakpoints without removing them from the inferior, by executing
34540 an out-of-line copy of the instruction that was originally at the
34541 breakpoint location. It is also known as out-of-line single-stepping.
34544 @item set displaced-stepping on
34545 If the target architecture supports it, @value{GDBN} will use
34546 displaced stepping to step over breakpoints.
34548 @item set displaced-stepping off
34549 @value{GDBN} will not use displaced stepping to step over breakpoints,
34550 even if such is supported by the target architecture.
34552 @cindex non-stop mode, and @samp{set displaced-stepping}
34553 @item set displaced-stepping auto
34554 This is the default mode. @value{GDBN} will use displaced stepping
34555 only if non-stop mode is active (@pxref{Non-Stop Mode}) and the target
34556 architecture supports displaced stepping.
34559 @kindex maint check-psymtabs
34560 @item maint check-psymtabs
34561 Check the consistency of currently expanded psymtabs versus symtabs.
34562 Use this to check, for example, whether a symbol is in one but not the other.
34564 @kindex maint check-symtabs
34565 @item maint check-symtabs
34566 Check the consistency of currently expanded symtabs.
34568 @kindex maint expand-symtabs
34569 @item maint expand-symtabs [@var{regexp}]
34570 Expand symbol tables.
34571 If @var{regexp} is specified, only expand symbol tables for file
34572 names matching @var{regexp}.
34574 @kindex maint set catch-demangler-crashes
34575 @kindex maint show catch-demangler-crashes
34576 @cindex demangler crashes
34577 @item maint set catch-demangler-crashes [on|off]
34578 @itemx maint show catch-demangler-crashes
34579 Control whether @value{GDBN} should attempt to catch crashes in the
34580 symbol name demangler. The default is to attempt to catch crashes.
34581 If enabled, the first time a crash is caught, a core file is created,
34582 the offending symbol is displayed and the user is presented with the
34583 option to terminate the current session.
34585 @kindex maint cplus first_component
34586 @item maint cplus first_component @var{name}
34587 Print the first C@t{++} class/namespace component of @var{name}.
34589 @kindex maint cplus namespace
34590 @item maint cplus namespace
34591 Print the list of possible C@t{++} namespaces.
34593 @kindex maint deprecate
34594 @kindex maint undeprecate
34595 @cindex deprecated commands
34596 @item maint deprecate @var{command} @r{[}@var{replacement}@r{]}
34597 @itemx maint undeprecate @var{command}
34598 Deprecate or undeprecate the named @var{command}. Deprecated commands
34599 cause @value{GDBN} to issue a warning when you use them. The optional
34600 argument @var{replacement} says which newer command should be used in
34601 favor of the deprecated one; if it is given, @value{GDBN} will mention
34602 the replacement as part of the warning.
34604 @kindex maint dump-me
34605 @item maint dump-me
34606 @cindex @code{SIGQUIT} signal, dump core of @value{GDBN}
34607 Cause a fatal signal in the debugger and force it to dump its core.
34608 This is supported only on systems which support aborting a program
34609 with the @code{SIGQUIT} signal.
34611 @kindex maint internal-error
34612 @kindex maint internal-warning
34613 @kindex maint demangler-warning
34614 @cindex demangler crashes
34615 @item maint internal-error @r{[}@var{message-text}@r{]}
34616 @itemx maint internal-warning @r{[}@var{message-text}@r{]}
34617 @itemx maint demangler-warning @r{[}@var{message-text}@r{]}
34619 Cause @value{GDBN} to call the internal function @code{internal_error},
34620 @code{internal_warning} or @code{demangler_warning} and hence behave
34621 as though an internal problem has been detected. In addition to
34622 reporting the internal problem, these functions give the user the
34623 opportunity to either quit @value{GDBN} or (for @code{internal_error}
34624 and @code{internal_warning}) create a core file of the current
34625 @value{GDBN} session.
34627 These commands take an optional parameter @var{message-text} that is
34628 used as the text of the error or warning message.
34630 Here's an example of using @code{internal-error}:
34633 (@value{GDBP}) @kbd{maint internal-error testing, 1, 2}
34634 @dots{}/maint.c:121: internal-error: testing, 1, 2
34635 A problem internal to GDB has been detected. Further
34636 debugging may prove unreliable.
34637 Quit this debugging session? (y or n) @kbd{n}
34638 Create a core file? (y or n) @kbd{n}
34642 @cindex @value{GDBN} internal error
34643 @cindex internal errors, control of @value{GDBN} behavior
34644 @cindex demangler crashes
34646 @kindex maint set internal-error
34647 @kindex maint show internal-error
34648 @kindex maint set internal-warning
34649 @kindex maint show internal-warning
34650 @kindex maint set demangler-warning
34651 @kindex maint show demangler-warning
34652 @item maint set internal-error @var{action} [ask|yes|no]
34653 @itemx maint show internal-error @var{action}
34654 @itemx maint set internal-warning @var{action} [ask|yes|no]
34655 @itemx maint show internal-warning @var{action}
34656 @itemx maint set demangler-warning @var{action} [ask|yes|no]
34657 @itemx maint show demangler-warning @var{action}
34658 When @value{GDBN} reports an internal problem (error or warning) it
34659 gives the user the opportunity to both quit @value{GDBN} and create a
34660 core file of the current @value{GDBN} session. These commands let you
34661 override the default behaviour for each particular @var{action},
34662 described in the table below.
34666 You can specify that @value{GDBN} should always (yes) or never (no)
34667 quit. The default is to ask the user what to do.
34670 You can specify that @value{GDBN} should always (yes) or never (no)
34671 create a core file. The default is to ask the user what to do. Note
34672 that there is no @code{corefile} option for @code{demangler-warning}:
34673 demangler warnings always create a core file and this cannot be
34677 @kindex maint packet
34678 @item maint packet @var{text}
34679 If @value{GDBN} is talking to an inferior via the serial protocol,
34680 then this command sends the string @var{text} to the inferior, and
34681 displays the response packet. @value{GDBN} supplies the initial
34682 @samp{$} character, the terminating @samp{#} character, and the
34685 @kindex maint print architecture
34686 @item maint print architecture @r{[}@var{file}@r{]}
34687 Print the entire architecture configuration. The optional argument
34688 @var{file} names the file where the output goes.
34690 @kindex maint print c-tdesc @r{[}@var{file}@r{]}
34691 @item maint print c-tdesc
34692 Print the target description (@pxref{Target Descriptions}) as
34693 a C source file. By default, the target description is for the current
34694 target, but if the optional argument @var{file} is provided, that file
34695 is used to produce the description. The @var{file} should be an XML
34696 document, of the form described in @ref{Target Description Format}.
34697 The created source file is built into @value{GDBN} when @value{GDBN} is
34698 built again. This command is used by developers after they add or
34699 modify XML target descriptions.
34701 @kindex maint print dummy-frames
34702 @item maint print dummy-frames
34703 Prints the contents of @value{GDBN}'s internal dummy-frame stack.
34706 (@value{GDBP}) @kbd{b add}
34708 (@value{GDBP}) @kbd{print add(2,3)}
34709 Breakpoint 2, add (a=2, b=3) at @dots{}
34711 The program being debugged stopped while in a function called from GDB.
34713 (@value{GDBP}) @kbd{maint print dummy-frames}
34714 0xa8206d8: id=@{stack=0xbfffe734,code=0xbfffe73f,!special@}, ptid=process 9353
34718 Takes an optional file parameter.
34720 @kindex maint print registers
34721 @kindex maint print raw-registers
34722 @kindex maint print cooked-registers
34723 @kindex maint print register-groups
34724 @kindex maint print remote-registers
34725 @item maint print registers @r{[}@var{file}@r{]}
34726 @itemx maint print raw-registers @r{[}@var{file}@r{]}
34727 @itemx maint print cooked-registers @r{[}@var{file}@r{]}
34728 @itemx maint print register-groups @r{[}@var{file}@r{]}
34729 @itemx maint print remote-registers @r{[}@var{file}@r{]}
34730 Print @value{GDBN}'s internal register data structures.
34732 The command @code{maint print raw-registers} includes the contents of
34733 the raw register cache; the command @code{maint print
34734 cooked-registers} includes the (cooked) value of all registers,
34735 including registers which aren't available on the target nor visible
34736 to user; the command @code{maint print register-groups} includes the
34737 groups that each register is a member of; and the command @code{maint
34738 print remote-registers} includes the remote target's register numbers
34739 and offsets in the `G' packets.
34741 These commands take an optional parameter, a file name to which to
34742 write the information.
34744 @kindex maint print reggroups
34745 @item maint print reggroups @r{[}@var{file}@r{]}
34746 Print @value{GDBN}'s internal register group data structures. The
34747 optional argument @var{file} tells to what file to write the
34750 The register groups info looks like this:
34753 (@value{GDBP}) @kbd{maint print reggroups}
34766 This command forces @value{GDBN} to flush its internal register cache.
34768 @kindex maint print objfiles
34769 @cindex info for known object files
34770 @item maint print objfiles @r{[}@var{regexp}@r{]}
34771 Print a dump of all known object files.
34772 If @var{regexp} is specified, only print object files whose names
34773 match @var{regexp}. For each object file, this command prints its name,
34774 address in memory, and all of its psymtabs and symtabs.
34776 @kindex maint print user-registers
34777 @cindex user registers
34778 @item maint print user-registers
34779 List all currently available @dfn{user registers}. User registers
34780 typically provide alternate names for actual hardware registers. They
34781 include the four ``standard'' registers @code{$fp}, @code{$pc},
34782 @code{$sp}, and @code{$ps}. @xref{standard registers}. User
34783 registers can be used in expressions in the same way as the canonical
34784 register names, but only the latter are listed by the @code{info
34785 registers} and @code{maint print registers} commands.
34787 @kindex maint print section-scripts
34788 @cindex info for known .debug_gdb_scripts-loaded scripts
34789 @item maint print section-scripts [@var{regexp}]
34790 Print a dump of scripts specified in the @code{.debug_gdb_section} section.
34791 If @var{regexp} is specified, only print scripts loaded by object files
34792 matching @var{regexp}.
34793 For each script, this command prints its name as specified in the objfile,
34794 and the full path if known.
34795 @xref{dotdebug_gdb_scripts section}.
34797 @kindex maint print statistics
34798 @cindex bcache statistics
34799 @item maint print statistics
34800 This command prints, for each object file in the program, various data
34801 about that object file followed by the byte cache (@dfn{bcache})
34802 statistics for the object file. The objfile data includes the number
34803 of minimal, partial, full, and stabs symbols, the number of types
34804 defined by the objfile, the number of as yet unexpanded psym tables,
34805 the number of line tables and string tables, and the amount of memory
34806 used by the various tables. The bcache statistics include the counts,
34807 sizes, and counts of duplicates of all and unique objects, max,
34808 average, and median entry size, total memory used and its overhead and
34809 savings, and various measures of the hash table size and chain
34812 @kindex maint print target-stack
34813 @cindex target stack description
34814 @item maint print target-stack
34815 A @dfn{target} is an interface between the debugger and a particular
34816 kind of file or process. Targets can be stacked in @dfn{strata},
34817 so that more than one target can potentially respond to a request.
34818 In particular, memory accesses will walk down the stack of targets
34819 until they find a target that is interested in handling that particular
34822 This command prints a short description of each layer that was pushed on
34823 the @dfn{target stack}, starting from the top layer down to the bottom one.
34825 @kindex maint print type
34826 @cindex type chain of a data type
34827 @item maint print type @var{expr}
34828 Print the type chain for a type specified by @var{expr}. The argument
34829 can be either a type name or a symbol. If it is a symbol, the type of
34830 that symbol is described. The type chain produced by this command is
34831 a recursive definition of the data type as stored in @value{GDBN}'s
34832 data structures, including its flags and contained types.
34834 @kindex maint selftest
34836 Run any self tests that were compiled in to @value{GDBN}. This will
34837 print a message showing how many tests were run, and how many failed.
34839 @kindex maint set dwarf always-disassemble
34840 @kindex maint show dwarf always-disassemble
34841 @item maint set dwarf always-disassemble
34842 @item maint show dwarf always-disassemble
34843 Control the behavior of @code{info address} when using DWARF debugging
34846 The default is @code{off}, which means that @value{GDBN} should try to
34847 describe a variable's location in an easily readable format. When
34848 @code{on}, @value{GDBN} will instead display the DWARF location
34849 expression in an assembly-like format. Note that some locations are
34850 too complex for @value{GDBN} to describe simply; in this case you will
34851 always see the disassembly form.
34853 Here is an example of the resulting disassembly:
34856 (gdb) info addr argc
34857 Symbol "argc" is a complex DWARF expression:
34861 For more information on these expressions, see
34862 @uref{http://www.dwarfstd.org/, the DWARF standard}.
34864 @kindex maint set dwarf max-cache-age
34865 @kindex maint show dwarf max-cache-age
34866 @item maint set dwarf max-cache-age
34867 @itemx maint show dwarf max-cache-age
34868 Control the DWARF compilation unit cache.
34870 @cindex DWARF compilation units cache
34871 In object files with inter-compilation-unit references, such as those
34872 produced by the GCC option @samp{-feliminate-dwarf2-dups}, the DWARF
34873 reader needs to frequently refer to previously read compilation units.
34874 This setting controls how long a compilation unit will remain in the
34875 cache if it is not referenced. A higher limit means that cached
34876 compilation units will be stored in memory longer, and more total
34877 memory will be used. Setting it to zero disables caching, which will
34878 slow down @value{GDBN} startup, but reduce memory consumption.
34880 @kindex maint set profile
34881 @kindex maint show profile
34882 @cindex profiling GDB
34883 @item maint set profile
34884 @itemx maint show profile
34885 Control profiling of @value{GDBN}.
34887 Profiling will be disabled until you use the @samp{maint set profile}
34888 command to enable it. When you enable profiling, the system will begin
34889 collecting timing and execution count data; when you disable profiling or
34890 exit @value{GDBN}, the results will be written to a log file. Remember that
34891 if you use profiling, @value{GDBN} will overwrite the profiling log file
34892 (often called @file{gmon.out}). If you have a record of important profiling
34893 data in a @file{gmon.out} file, be sure to move it to a safe location.
34895 Configuring with @samp{--enable-profiling} arranges for @value{GDBN} to be
34896 compiled with the @samp{-pg} compiler option.
34898 @kindex maint set show-debug-regs
34899 @kindex maint show show-debug-regs
34900 @cindex hardware debug registers
34901 @item maint set show-debug-regs
34902 @itemx maint show show-debug-regs
34903 Control whether to show variables that mirror the hardware debug
34904 registers. Use @code{on} to enable, @code{off} to disable. If
34905 enabled, the debug registers values are shown when @value{GDBN} inserts or
34906 removes a hardware breakpoint or watchpoint, and when the inferior
34907 triggers a hardware-assisted breakpoint or watchpoint.
34909 @kindex maint set show-all-tib
34910 @kindex maint show show-all-tib
34911 @item maint set show-all-tib
34912 @itemx maint show show-all-tib
34913 Control whether to show all non zero areas within a 1k block starting
34914 at thread local base, when using the @samp{info w32 thread-information-block}
34917 @kindex maint set target-async
34918 @kindex maint show target-async
34919 @item maint set target-async
34920 @itemx maint show target-async
34921 This controls whether @value{GDBN} targets operate in synchronous or
34922 asynchronous mode (@pxref{Background Execution}). Normally the
34923 default is asynchronous, if it is available; but this can be changed
34924 to more easily debug problems occurring only in synchronous mode.
34926 @kindex maint set target-non-stop @var{mode} [on|off|auto]
34927 @kindex maint show target-non-stop
34928 @item maint set target-non-stop
34929 @itemx maint show target-non-stop
34931 This controls whether @value{GDBN} targets always operate in non-stop
34932 mode even if @code{set non-stop} is @code{off} (@pxref{Non-Stop
34933 Mode}). The default is @code{auto}, meaning non-stop mode is enabled
34934 if supported by the target.
34937 @item maint set target-non-stop auto
34938 This is the default mode. @value{GDBN} controls the target in
34939 non-stop mode if the target supports it.
34941 @item maint set target-non-stop on
34942 @value{GDBN} controls the target in non-stop mode even if the target
34943 does not indicate support.
34945 @item maint set target-non-stop off
34946 @value{GDBN} does not control the target in non-stop mode even if the
34947 target supports it.
34950 @kindex maint set per-command
34951 @kindex maint show per-command
34952 @item maint set per-command
34953 @itemx maint show per-command
34954 @cindex resources used by commands
34956 @value{GDBN} can display the resources used by each command.
34957 This is useful in debugging performance problems.
34960 @item maint set per-command space [on|off]
34961 @itemx maint show per-command space
34962 Enable or disable the printing of the memory used by GDB for each command.
34963 If enabled, @value{GDBN} will display how much memory each command
34964 took, following the command's own output.
34965 This can also be requested by invoking @value{GDBN} with the
34966 @option{--statistics} command-line switch (@pxref{Mode Options}).
34968 @item maint set per-command time [on|off]
34969 @itemx maint show per-command time
34970 Enable or disable the printing of the execution time of @value{GDBN}
34972 If enabled, @value{GDBN} will display how much time it
34973 took to execute each command, following the command's own output.
34974 Both CPU time and wallclock time are printed.
34975 Printing both is useful when trying to determine whether the cost is
34976 CPU or, e.g., disk/network latency.
34977 Note that the CPU time printed is for @value{GDBN} only, it does not include
34978 the execution time of the inferior because there's no mechanism currently
34979 to compute how much time was spent by @value{GDBN} and how much time was
34980 spent by the program been debugged.
34981 This can also be requested by invoking @value{GDBN} with the
34982 @option{--statistics} command-line switch (@pxref{Mode Options}).
34984 @item maint set per-command symtab [on|off]
34985 @itemx maint show per-command symtab
34986 Enable or disable the printing of basic symbol table statistics
34988 If enabled, @value{GDBN} will display the following information:
34992 number of symbol tables
34994 number of primary symbol tables
34996 number of blocks in the blockvector
35000 @kindex maint space
35001 @cindex memory used by commands
35002 @item maint space @var{value}
35003 An alias for @code{maint set per-command space}.
35004 A non-zero value enables it, zero disables it.
35007 @cindex time of command execution
35008 @item maint time @var{value}
35009 An alias for @code{maint set per-command time}.
35010 A non-zero value enables it, zero disables it.
35012 @kindex maint translate-address
35013 @item maint translate-address @r{[}@var{section}@r{]} @var{addr}
35014 Find the symbol stored at the location specified by the address
35015 @var{addr} and an optional section name @var{section}. If found,
35016 @value{GDBN} prints the name of the closest symbol and an offset from
35017 the symbol's location to the specified address. This is similar to
35018 the @code{info address} command (@pxref{Symbols}), except that this
35019 command also allows to find symbols in other sections.
35021 If section was not specified, the section in which the symbol was found
35022 is also printed. For dynamically linked executables, the name of
35023 executable or shared library containing the symbol is printed as well.
35027 The following command is useful for non-interactive invocations of
35028 @value{GDBN}, such as in the test suite.
35031 @item set watchdog @var{nsec}
35032 @kindex set watchdog
35033 @cindex watchdog timer
35034 @cindex timeout for commands
35035 Set the maximum number of seconds @value{GDBN} will wait for the
35036 target operation to finish. If this time expires, @value{GDBN}
35037 reports and error and the command is aborted.
35039 @item show watchdog
35040 Show the current setting of the target wait timeout.
35043 @node Remote Protocol
35044 @appendix @value{GDBN} Remote Serial Protocol
35049 * Stop Reply Packets::
35050 * General Query Packets::
35051 * Architecture-Specific Protocol Details::
35052 * Tracepoint Packets::
35053 * Host I/O Packets::
35055 * Notification Packets::
35056 * Remote Non-Stop::
35057 * Packet Acknowledgment::
35059 * File-I/O Remote Protocol Extension::
35060 * Library List Format::
35061 * Library List Format for SVR4 Targets::
35062 * Memory Map Format::
35063 * Thread List Format::
35064 * Traceframe Info Format::
35065 * Branch Trace Format::
35066 * Branch Trace Configuration Format::
35072 There may be occasions when you need to know something about the
35073 protocol---for example, if there is only one serial port to your target
35074 machine, you might want your program to do something special if it
35075 recognizes a packet meant for @value{GDBN}.
35077 In the examples below, @samp{->} and @samp{<-} are used to indicate
35078 transmitted and received data, respectively.
35080 @cindex protocol, @value{GDBN} remote serial
35081 @cindex serial protocol, @value{GDBN} remote
35082 @cindex remote serial protocol
35083 All @value{GDBN} commands and responses (other than acknowledgments
35084 and notifications, see @ref{Notification Packets}) are sent as a
35085 @var{packet}. A @var{packet} is introduced with the character
35086 @samp{$}, the actual @var{packet-data}, and the terminating character
35087 @samp{#} followed by a two-digit @var{checksum}:
35090 @code{$}@var{packet-data}@code{#}@var{checksum}
35094 @cindex checksum, for @value{GDBN} remote
35096 The two-digit @var{checksum} is computed as the modulo 256 sum of all
35097 characters between the leading @samp{$} and the trailing @samp{#} (an
35098 eight bit unsigned checksum).
35100 Implementors should note that prior to @value{GDBN} 5.0 the protocol
35101 specification also included an optional two-digit @var{sequence-id}:
35104 @code{$}@var{sequence-id}@code{:}@var{packet-data}@code{#}@var{checksum}
35107 @cindex sequence-id, for @value{GDBN} remote
35109 That @var{sequence-id} was appended to the acknowledgment. @value{GDBN}
35110 has never output @var{sequence-id}s. Stubs that handle packets added
35111 since @value{GDBN} 5.0 must not accept @var{sequence-id}.
35113 When either the host or the target machine receives a packet, the first
35114 response expected is an acknowledgment: either @samp{+} (to indicate
35115 the package was received correctly) or @samp{-} (to request
35119 -> @code{$}@var{packet-data}@code{#}@var{checksum}
35124 The @samp{+}/@samp{-} acknowledgments can be disabled
35125 once a connection is established.
35126 @xref{Packet Acknowledgment}, for details.
35128 The host (@value{GDBN}) sends @var{command}s, and the target (the
35129 debugging stub incorporated in your program) sends a @var{response}. In
35130 the case of step and continue @var{command}s, the response is only sent
35131 when the operation has completed, and the target has again stopped all
35132 threads in all attached processes. This is the default all-stop mode
35133 behavior, but the remote protocol also supports @value{GDBN}'s non-stop
35134 execution mode; see @ref{Remote Non-Stop}, for details.
35136 @var{packet-data} consists of a sequence of characters with the
35137 exception of @samp{#} and @samp{$} (see @samp{X} packet for additional
35140 @cindex remote protocol, field separator
35141 Fields within the packet should be separated using @samp{,} @samp{;} or
35142 @samp{:}. Except where otherwise noted all numbers are represented in
35143 @sc{hex} with leading zeros suppressed.
35145 Implementors should note that prior to @value{GDBN} 5.0, the character
35146 @samp{:} could not appear as the third character in a packet (as it
35147 would potentially conflict with the @var{sequence-id}).
35149 @cindex remote protocol, binary data
35150 @anchor{Binary Data}
35151 Binary data in most packets is encoded either as two hexadecimal
35152 digits per byte of binary data. This allowed the traditional remote
35153 protocol to work over connections which were only seven-bit clean.
35154 Some packets designed more recently assume an eight-bit clean
35155 connection, and use a more efficient encoding to send and receive
35158 The binary data representation uses @code{7d} (@sc{ascii} @samp{@}})
35159 as an escape character. Any escaped byte is transmitted as the escape
35160 character followed by the original character XORed with @code{0x20}.
35161 For example, the byte @code{0x7d} would be transmitted as the two
35162 bytes @code{0x7d 0x5d}. The bytes @code{0x23} (@sc{ascii} @samp{#}),
35163 @code{0x24} (@sc{ascii} @samp{$}), and @code{0x7d} (@sc{ascii}
35164 @samp{@}}) must always be escaped. Responses sent by the stub
35165 must also escape @code{0x2a} (@sc{ascii} @samp{*}), so that it
35166 is not interpreted as the start of a run-length encoded sequence
35169 Response @var{data} can be run-length encoded to save space.
35170 Run-length encoding replaces runs of identical characters with one
35171 instance of the repeated character, followed by a @samp{*} and a
35172 repeat count. The repeat count is itself sent encoded, to avoid
35173 binary characters in @var{data}: a value of @var{n} is sent as
35174 @code{@var{n}+29}. For a repeat count greater or equal to 3, this
35175 produces a printable @sc{ascii} character, e.g.@: a space (@sc{ascii}
35176 code 32) for a repeat count of 3. (This is because run-length
35177 encoding starts to win for counts 3 or more.) Thus, for example,
35178 @samp{0* } is a run-length encoding of ``0000'': the space character
35179 after @samp{*} means repeat the leading @code{0} @w{@code{32 - 29 =
35182 The printable characters @samp{#} and @samp{$} or with a numeric value
35183 greater than 126 must not be used. Runs of six repeats (@samp{#}) or
35184 seven repeats (@samp{$}) can be expanded using a repeat count of only
35185 five (@samp{"}). For example, @samp{00000000} can be encoded as
35188 The error response returned for some packets includes a two character
35189 error number. That number is not well defined.
35191 @cindex empty response, for unsupported packets
35192 For any @var{command} not supported by the stub, an empty response
35193 (@samp{$#00}) should be returned. That way it is possible to extend the
35194 protocol. A newer @value{GDBN} can tell if a packet is supported based
35197 At a minimum, a stub is required to support the @samp{g} and @samp{G}
35198 commands for register access, and the @samp{m} and @samp{M} commands
35199 for memory access. Stubs that only control single-threaded targets
35200 can implement run control with the @samp{c} (continue), and @samp{s}
35201 (step) commands. Stubs that support multi-threading targets should
35202 support the @samp{vCont} command. All other commands are optional.
35207 The following table provides a complete list of all currently defined
35208 @var{command}s and their corresponding response @var{data}.
35209 @xref{File-I/O Remote Protocol Extension}, for details about the File
35210 I/O extension of the remote protocol.
35212 Each packet's description has a template showing the packet's overall
35213 syntax, followed by an explanation of the packet's meaning. We
35214 include spaces in some of the templates for clarity; these are not
35215 part of the packet's syntax. No @value{GDBN} packet uses spaces to
35216 separate its components. For example, a template like @samp{foo
35217 @var{bar} @var{baz}} describes a packet beginning with the three ASCII
35218 bytes @samp{foo}, followed by a @var{bar}, followed directly by a
35219 @var{baz}. @value{GDBN} does not transmit a space character between the
35220 @samp{foo} and the @var{bar}, or between the @var{bar} and the
35223 @cindex @var{thread-id}, in remote protocol
35224 @anchor{thread-id syntax}
35225 Several packets and replies include a @var{thread-id} field to identify
35226 a thread. Normally these are positive numbers with a target-specific
35227 interpretation, formatted as big-endian hex strings. A @var{thread-id}
35228 can also be a literal @samp{-1} to indicate all threads, or @samp{0} to
35231 In addition, the remote protocol supports a multiprocess feature in
35232 which the @var{thread-id} syntax is extended to optionally include both
35233 process and thread ID fields, as @samp{p@var{pid}.@var{tid}}.
35234 The @var{pid} (process) and @var{tid} (thread) components each have the
35235 format described above: a positive number with target-specific
35236 interpretation formatted as a big-endian hex string, literal @samp{-1}
35237 to indicate all processes or threads (respectively), or @samp{0} to
35238 indicate an arbitrary process or thread. Specifying just a process, as
35239 @samp{p@var{pid}}, is equivalent to @samp{p@var{pid}.-1}. It is an
35240 error to specify all processes but a specific thread, such as
35241 @samp{p-1.@var{tid}}. Note that the @samp{p} prefix is @emph{not} used
35242 for those packets and replies explicitly documented to include a process
35243 ID, rather than a @var{thread-id}.
35245 The multiprocess @var{thread-id} syntax extensions are only used if both
35246 @value{GDBN} and the stub report support for the @samp{multiprocess}
35247 feature using @samp{qSupported}. @xref{multiprocess extensions}, for
35250 Note that all packet forms beginning with an upper- or lower-case
35251 letter, other than those described here, are reserved for future use.
35253 Here are the packet descriptions.
35258 @cindex @samp{!} packet
35259 @anchor{extended mode}
35260 Enable extended mode. In extended mode, the remote server is made
35261 persistent. The @samp{R} packet is used to restart the program being
35267 The remote target both supports and has enabled extended mode.
35271 @cindex @samp{?} packet
35273 Indicate the reason the target halted. The reply is the same as for
35274 step and continue. This packet has a special interpretation when the
35275 target is in non-stop mode; see @ref{Remote Non-Stop}.
35278 @xref{Stop Reply Packets}, for the reply specifications.
35280 @item A @var{arglen},@var{argnum},@var{arg},@dots{}
35281 @cindex @samp{A} packet
35282 Initialized @code{argv[]} array passed into program. @var{arglen}
35283 specifies the number of bytes in the hex encoded byte stream
35284 @var{arg}. See @code{gdbserver} for more details.
35289 The arguments were set.
35295 @cindex @samp{b} packet
35296 (Don't use this packet; its behavior is not well-defined.)
35297 Change the serial line speed to @var{baud}.
35299 JTC: @emph{When does the transport layer state change? When it's
35300 received, or after the ACK is transmitted. In either case, there are
35301 problems if the command or the acknowledgment packet is dropped.}
35303 Stan: @emph{If people really wanted to add something like this, and get
35304 it working for the first time, they ought to modify ser-unix.c to send
35305 some kind of out-of-band message to a specially-setup stub and have the
35306 switch happen "in between" packets, so that from remote protocol's point
35307 of view, nothing actually happened.}
35309 @item B @var{addr},@var{mode}
35310 @cindex @samp{B} packet
35311 Set (@var{mode} is @samp{S}) or clear (@var{mode} is @samp{C}) a
35312 breakpoint at @var{addr}.
35314 Don't use this packet. Use the @samp{Z} and @samp{z} packets instead
35315 (@pxref{insert breakpoint or watchpoint packet}).
35317 @cindex @samp{bc} packet
35320 Backward continue. Execute the target system in reverse. No parameter.
35321 @xref{Reverse Execution}, for more information.
35324 @xref{Stop Reply Packets}, for the reply specifications.
35326 @cindex @samp{bs} packet
35329 Backward single step. Execute one instruction in reverse. No parameter.
35330 @xref{Reverse Execution}, for more information.
35333 @xref{Stop Reply Packets}, for the reply specifications.
35335 @item c @r{[}@var{addr}@r{]}
35336 @cindex @samp{c} packet
35337 Continue at @var{addr}, which is the address to resume. If @var{addr}
35338 is omitted, resume at current address.
35340 This packet is deprecated for multi-threading support. @xref{vCont
35344 @xref{Stop Reply Packets}, for the reply specifications.
35346 @item C @var{sig}@r{[};@var{addr}@r{]}
35347 @cindex @samp{C} packet
35348 Continue with signal @var{sig} (hex signal number). If
35349 @samp{;@var{addr}} is omitted, resume at same address.
35351 This packet is deprecated for multi-threading support. @xref{vCont
35355 @xref{Stop Reply Packets}, for the reply specifications.
35358 @cindex @samp{d} packet
35361 Don't use this packet; instead, define a general set packet
35362 (@pxref{General Query Packets}).
35366 @cindex @samp{D} packet
35367 The first form of the packet is used to detach @value{GDBN} from the
35368 remote system. It is sent to the remote target
35369 before @value{GDBN} disconnects via the @code{detach} command.
35371 The second form, including a process ID, is used when multiprocess
35372 protocol extensions are enabled (@pxref{multiprocess extensions}), to
35373 detach only a specific process. The @var{pid} is specified as a
35374 big-endian hex string.
35384 @item F @var{RC},@var{EE},@var{CF};@var{XX}
35385 @cindex @samp{F} packet
35386 A reply from @value{GDBN} to an @samp{F} packet sent by the target.
35387 This is part of the File-I/O protocol extension. @xref{File-I/O
35388 Remote Protocol Extension}, for the specification.
35391 @anchor{read registers packet}
35392 @cindex @samp{g} packet
35393 Read general registers.
35397 @item @var{XX@dots{}}
35398 Each byte of register data is described by two hex digits. The bytes
35399 with the register are transmitted in target byte order. The size of
35400 each register and their position within the @samp{g} packet are
35401 determined by the @value{GDBN} internal gdbarch functions
35402 @code{DEPRECATED_REGISTER_RAW_SIZE} and @code{gdbarch_register_name}.
35404 When reading registers from a trace frame (@pxref{Analyze Collected
35405 Data,,Using the Collected Data}), the stub may also return a string of
35406 literal @samp{x}'s in place of the register data digits, to indicate
35407 that the corresponding register has not been collected, thus its value
35408 is unavailable. For example, for an architecture with 4 registers of
35409 4 bytes each, the following reply indicates to @value{GDBN} that
35410 registers 0 and 2 have not been collected, while registers 1 and 3
35411 have been collected, and both have zero value:
35415 <- @code{xxxxxxxx00000000xxxxxxxx00000000}
35422 @item G @var{XX@dots{}}
35423 @cindex @samp{G} packet
35424 Write general registers. @xref{read registers packet}, for a
35425 description of the @var{XX@dots{}} data.
35435 @item H @var{op} @var{thread-id}
35436 @cindex @samp{H} packet
35437 Set thread for subsequent operations (@samp{m}, @samp{M}, @samp{g},
35438 @samp{G}, et.al.). Depending on the operation to be performed, @var{op}
35439 should be @samp{c} for step and continue operations (note that this
35440 is deprecated, supporting the @samp{vCont} command is a better
35441 option), and @samp{g} for other operations. The thread designator
35442 @var{thread-id} has the format and interpretation described in
35443 @ref{thread-id syntax}.
35454 @c 'H': How restrictive (or permissive) is the thread model. If a
35455 @c thread is selected and stopped, are other threads allowed
35456 @c to continue to execute? As I mentioned above, I think the
35457 @c semantics of each command when a thread is selected must be
35458 @c described. For example:
35460 @c 'g': If the stub supports threads and a specific thread is
35461 @c selected, returns the register block from that thread;
35462 @c otherwise returns current registers.
35464 @c 'G' If the stub supports threads and a specific thread is
35465 @c selected, sets the registers of the register block of
35466 @c that thread; otherwise sets current registers.
35468 @item i @r{[}@var{addr}@r{[},@var{nnn}@r{]]}
35469 @anchor{cycle step packet}
35470 @cindex @samp{i} packet
35471 Step the remote target by a single clock cycle. If @samp{,@var{nnn}} is
35472 present, cycle step @var{nnn} cycles. If @var{addr} is present, cycle
35473 step starting at that address.
35476 @cindex @samp{I} packet
35477 Signal, then cycle step. @xref{step with signal packet}. @xref{cycle
35481 @cindex @samp{k} packet
35484 The exact effect of this packet is not specified.
35486 For a bare-metal target, it may power cycle or reset the target
35487 system. For that reason, the @samp{k} packet has no reply.
35489 For a single-process target, it may kill that process if possible.
35491 A multiple-process target may choose to kill just one process, or all
35492 that are under @value{GDBN}'s control. For more precise control, use
35493 the vKill packet (@pxref{vKill packet}).
35495 If the target system immediately closes the connection in response to
35496 @samp{k}, @value{GDBN} does not consider the lack of packet
35497 acknowledgment to be an error, and assumes the kill was successful.
35499 If connected using @kbd{target extended-remote}, and the target does
35500 not close the connection in response to a kill request, @value{GDBN}
35501 probes the target state as if a new connection was opened
35502 (@pxref{? packet}).
35504 @item m @var{addr},@var{length}
35505 @cindex @samp{m} packet
35506 Read @var{length} addressable memory units starting at address @var{addr}
35507 (@pxref{addressable memory unit}). Note that @var{addr} may not be aligned to
35508 any particular boundary.
35510 The stub need not use any particular size or alignment when gathering
35511 data from memory for the response; even if @var{addr} is word-aligned
35512 and @var{length} is a multiple of the word size, the stub is free to
35513 use byte accesses, or not. For this reason, this packet may not be
35514 suitable for accessing memory-mapped I/O devices.
35515 @cindex alignment of remote memory accesses
35516 @cindex size of remote memory accesses
35517 @cindex memory, alignment and size of remote accesses
35521 @item @var{XX@dots{}}
35522 Memory contents; each byte is transmitted as a two-digit hexadecimal number.
35523 The reply may contain fewer addressable memory units than requested if the
35524 server was able to read only part of the region of memory.
35529 @item M @var{addr},@var{length}:@var{XX@dots{}}
35530 @cindex @samp{M} packet
35531 Write @var{length} addressable memory units starting at address @var{addr}
35532 (@pxref{addressable memory unit}). The data is given by @var{XX@dots{}}; each
35533 byte is transmitted as a two-digit hexadecimal number.
35540 for an error (this includes the case where only part of the data was
35545 @cindex @samp{p} packet
35546 Read the value of register @var{n}; @var{n} is in hex.
35547 @xref{read registers packet}, for a description of how the returned
35548 register value is encoded.
35552 @item @var{XX@dots{}}
35553 the register's value
35557 Indicating an unrecognized @var{query}.
35560 @item P @var{n@dots{}}=@var{r@dots{}}
35561 @anchor{write register packet}
35562 @cindex @samp{P} packet
35563 Write register @var{n@dots{}} with value @var{r@dots{}}. The register
35564 number @var{n} is in hexadecimal, and @var{r@dots{}} contains two hex
35565 digits for each byte in the register (target byte order).
35575 @item q @var{name} @var{params}@dots{}
35576 @itemx Q @var{name} @var{params}@dots{}
35577 @cindex @samp{q} packet
35578 @cindex @samp{Q} packet
35579 General query (@samp{q}) and set (@samp{Q}). These packets are
35580 described fully in @ref{General Query Packets}.
35583 @cindex @samp{r} packet
35584 Reset the entire system.
35586 Don't use this packet; use the @samp{R} packet instead.
35589 @cindex @samp{R} packet
35590 Restart the program being debugged. The @var{XX}, while needed, is ignored.
35591 This packet is only available in extended mode (@pxref{extended mode}).
35593 The @samp{R} packet has no reply.
35595 @item s @r{[}@var{addr}@r{]}
35596 @cindex @samp{s} packet
35597 Single step, resuming at @var{addr}. If
35598 @var{addr} is omitted, resume at same address.
35600 This packet is deprecated for multi-threading support. @xref{vCont
35604 @xref{Stop Reply Packets}, for the reply specifications.
35606 @item S @var{sig}@r{[};@var{addr}@r{]}
35607 @anchor{step with signal packet}
35608 @cindex @samp{S} packet
35609 Step with signal. This is analogous to the @samp{C} packet, but
35610 requests a single-step, rather than a normal resumption of execution.
35612 This packet is deprecated for multi-threading support. @xref{vCont
35616 @xref{Stop Reply Packets}, for the reply specifications.
35618 @item t @var{addr}:@var{PP},@var{MM}
35619 @cindex @samp{t} packet
35620 Search backwards starting at address @var{addr} for a match with pattern
35621 @var{PP} and mask @var{MM}, both of which are are 4 byte long.
35622 There must be at least 3 digits in @var{addr}.
35624 @item T @var{thread-id}
35625 @cindex @samp{T} packet
35626 Find out if the thread @var{thread-id} is alive. @xref{thread-id syntax}.
35631 thread is still alive
35637 Packets starting with @samp{v} are identified by a multi-letter name,
35638 up to the first @samp{;} or @samp{?} (or the end of the packet).
35640 @item vAttach;@var{pid}
35641 @cindex @samp{vAttach} packet
35642 Attach to a new process with the specified process ID @var{pid}.
35643 The process ID is a
35644 hexadecimal integer identifying the process. In all-stop mode, all
35645 threads in the attached process are stopped; in non-stop mode, it may be
35646 attached without being stopped if that is supported by the target.
35648 @c In non-stop mode, on a successful vAttach, the stub should set the
35649 @c current thread to a thread of the newly-attached process. After
35650 @c attaching, GDB queries for the attached process's thread ID with qC.
35651 @c Also note that, from a user perspective, whether or not the
35652 @c target is stopped on attach in non-stop mode depends on whether you
35653 @c use the foreground or background version of the attach command, not
35654 @c on what vAttach does; GDB does the right thing with respect to either
35655 @c stopping or restarting threads.
35657 This packet is only available in extended mode (@pxref{extended mode}).
35663 @item @r{Any stop packet}
35664 for success in all-stop mode (@pxref{Stop Reply Packets})
35666 for success in non-stop mode (@pxref{Remote Non-Stop})
35669 @item vCont@r{[};@var{action}@r{[}:@var{thread-id}@r{]]}@dots{}
35670 @cindex @samp{vCont} packet
35671 @anchor{vCont packet}
35672 Resume the inferior, specifying different actions for each thread.
35674 For each inferior thread, the leftmost action with a matching
35675 @var{thread-id} is applied. Threads that don't match any action
35676 remain in their current state. Thread IDs are specified using the
35677 syntax described in @ref{thread-id syntax}. If multiprocess
35678 extensions (@pxref{multiprocess extensions}) are supported, actions
35679 can be specified to match all threads in a process by using the
35680 @samp{p@var{pid}.-1} form of the @var{thread-id}. An action with no
35681 @var{thread-id} matches all threads. Specifying no actions is an
35684 Currently supported actions are:
35690 Continue with signal @var{sig}. The signal @var{sig} should be two hex digits.
35694 Step with signal @var{sig}. The signal @var{sig} should be two hex digits.
35697 @item r @var{start},@var{end}
35698 Step once, and then keep stepping as long as the thread stops at
35699 addresses between @var{start} (inclusive) and @var{end} (exclusive).
35700 The remote stub reports a stop reply when either the thread goes out
35701 of the range or is stopped due to an unrelated reason, such as hitting
35702 a breakpoint. @xref{range stepping}.
35704 If the range is empty (@var{start} == @var{end}), then the action
35705 becomes equivalent to the @samp{s} action. In other words,
35706 single-step once, and report the stop (even if the stepped instruction
35707 jumps to @var{start}).
35709 (A stop reply may be sent at any point even if the PC is still within
35710 the stepping range; for example, it is valid to implement this packet
35711 in a degenerate way as a single instruction step operation.)
35715 The optional argument @var{addr} normally associated with the
35716 @samp{c}, @samp{C}, @samp{s}, and @samp{S} packets is
35717 not supported in @samp{vCont}.
35719 The @samp{t} action is only relevant in non-stop mode
35720 (@pxref{Remote Non-Stop}) and may be ignored by the stub otherwise.
35721 A stop reply should be generated for any affected thread not already stopped.
35722 When a thread is stopped by means of a @samp{t} action,
35723 the corresponding stop reply should indicate that the thread has stopped with
35724 signal @samp{0}, regardless of whether the target uses some other signal
35725 as an implementation detail.
35727 The server must ignore @samp{c}, @samp{C}, @samp{s}, @samp{S}, and
35728 @samp{r} actions for threads that are already running. Conversely,
35729 the server must ignore @samp{t} actions for threads that are already
35732 @emph{Note:} In non-stop mode, a thread is considered running until
35733 @value{GDBN} acknowleges an asynchronous stop notification for it with
35734 the @samp{vStopped} packet (@pxref{Remote Non-Stop}).
35736 The stub must support @samp{vCont} if it reports support for
35737 multiprocess extensions (@pxref{multiprocess extensions}).
35740 @xref{Stop Reply Packets}, for the reply specifications.
35743 @cindex @samp{vCont?} packet
35744 Request a list of actions supported by the @samp{vCont} packet.
35748 @item vCont@r{[};@var{action}@dots{}@r{]}
35749 The @samp{vCont} packet is supported. Each @var{action} is a supported
35750 command in the @samp{vCont} packet.
35752 The @samp{vCont} packet is not supported.
35755 @anchor{vCtrlC packet}
35757 @cindex @samp{vCtrlC} packet
35758 Interrupt remote target as if a control-C was pressed on the remote
35759 terminal. This is the equivalent to reacting to the @code{^C}
35760 (@samp{\003}, the control-C character) character in all-stop mode
35761 while the target is running, except this works in non-stop mode.
35762 @xref{interrupting remote targets}, for more info on the all-stop
35773 @item vFile:@var{operation}:@var{parameter}@dots{}
35774 @cindex @samp{vFile} packet
35775 Perform a file operation on the target system. For details,
35776 see @ref{Host I/O Packets}.
35778 @item vFlashErase:@var{addr},@var{length}
35779 @cindex @samp{vFlashErase} packet
35780 Direct the stub to erase @var{length} bytes of flash starting at
35781 @var{addr}. The region may enclose any number of flash blocks, but
35782 its start and end must fall on block boundaries, as indicated by the
35783 flash block size appearing in the memory map (@pxref{Memory Map
35784 Format}). @value{GDBN} groups flash memory programming operations
35785 together, and sends a @samp{vFlashDone} request after each group; the
35786 stub is allowed to delay erase operation until the @samp{vFlashDone}
35787 packet is received.
35797 @item vFlashWrite:@var{addr}:@var{XX@dots{}}
35798 @cindex @samp{vFlashWrite} packet
35799 Direct the stub to write data to flash address @var{addr}. The data
35800 is passed in binary form using the same encoding as for the @samp{X}
35801 packet (@pxref{Binary Data}). The memory ranges specified by
35802 @samp{vFlashWrite} packets preceding a @samp{vFlashDone} packet must
35803 not overlap, and must appear in order of increasing addresses
35804 (although @samp{vFlashErase} packets for higher addresses may already
35805 have been received; the ordering is guaranteed only between
35806 @samp{vFlashWrite} packets). If a packet writes to an address that was
35807 neither erased by a preceding @samp{vFlashErase} packet nor by some other
35808 target-specific method, the results are unpredictable.
35816 for vFlashWrite addressing non-flash memory
35822 @cindex @samp{vFlashDone} packet
35823 Indicate to the stub that flash programming operation is finished.
35824 The stub is permitted to delay or batch the effects of a group of
35825 @samp{vFlashErase} and @samp{vFlashWrite} packets until a
35826 @samp{vFlashDone} packet is received. The contents of the affected
35827 regions of flash memory are unpredictable until the @samp{vFlashDone}
35828 request is completed.
35830 @item vKill;@var{pid}
35831 @cindex @samp{vKill} packet
35832 @anchor{vKill packet}
35833 Kill the process with the specified process ID @var{pid}, which is a
35834 hexadecimal integer identifying the process. This packet is used in
35835 preference to @samp{k} when multiprocess protocol extensions are
35836 supported; see @ref{multiprocess extensions}.
35846 @item vMustReplyEmpty
35847 @cindex @samp{vMustReplyEmpty} packet
35848 The correct reply to an unknown @samp{v} packet is to return the empty
35849 string, however, some older versions of @command{gdbserver} would
35850 incorrectly return @samp{OK} for unknown @samp{v} packets.
35852 The @samp{vMustReplyEmpty} is used as a feature test to check how
35853 @command{gdbserver} handles unknown packets, it is important that this
35854 packet be handled in the same way as other unknown @samp{v} packets.
35855 If this packet is handled differently to other unknown @samp{v}
35856 packets then it is possile that @value{GDBN} may run into problems in
35857 other areas, specifically around use of @samp{vFile:setfs:}.
35859 @item vRun;@var{filename}@r{[};@var{argument}@r{]}@dots{}
35860 @cindex @samp{vRun} packet
35861 Run the program @var{filename}, passing it each @var{argument} on its
35862 command line. The file and arguments are hex-encoded strings. If
35863 @var{filename} is an empty string, the stub may use a default program
35864 (e.g.@: the last program run). The program is created in the stopped
35867 @c FIXME: What about non-stop mode?
35869 This packet is only available in extended mode (@pxref{extended mode}).
35875 @item @r{Any stop packet}
35876 for success (@pxref{Stop Reply Packets})
35880 @cindex @samp{vStopped} packet
35881 @xref{Notification Packets}.
35883 @item X @var{addr},@var{length}:@var{XX@dots{}}
35885 @cindex @samp{X} packet
35886 Write data to memory, where the data is transmitted in binary.
35887 Memory is specified by its address @var{addr} and number of addressable memory
35888 units @var{length} (@pxref{addressable memory unit});
35889 @samp{@var{XX}@dots{}} is binary data (@pxref{Binary Data}).
35899 @item z @var{type},@var{addr},@var{kind}
35900 @itemx Z @var{type},@var{addr},@var{kind}
35901 @anchor{insert breakpoint or watchpoint packet}
35902 @cindex @samp{z} packet
35903 @cindex @samp{Z} packets
35904 Insert (@samp{Z}) or remove (@samp{z}) a @var{type} breakpoint or
35905 watchpoint starting at address @var{address} of kind @var{kind}.
35907 Each breakpoint and watchpoint packet @var{type} is documented
35910 @emph{Implementation notes: A remote target shall return an empty string
35911 for an unrecognized breakpoint or watchpoint packet @var{type}. A
35912 remote target shall support either both or neither of a given
35913 @samp{Z@var{type}@dots{}} and @samp{z@var{type}@dots{}} packet pair. To
35914 avoid potential problems with duplicate packets, the operations should
35915 be implemented in an idempotent way.}
35917 @item z0,@var{addr},@var{kind}
35918 @itemx Z0,@var{addr},@var{kind}@r{[};@var{cond_list}@dots{}@r{]}@r{[};cmds:@var{persist},@var{cmd_list}@dots{}@r{]}
35919 @cindex @samp{z0} packet
35920 @cindex @samp{Z0} packet
35921 Insert (@samp{Z0}) or remove (@samp{z0}) a software breakpoint at address
35922 @var{addr} of type @var{kind}.
35924 A software breakpoint is implemented by replacing the instruction at
35925 @var{addr} with a software breakpoint or trap instruction. The
35926 @var{kind} is target-specific and typically indicates the size of the
35927 breakpoint in bytes that should be inserted. E.g., the @sc{arm} and
35928 @sc{mips} can insert either a 2 or 4 byte breakpoint. Some
35929 architectures have additional meanings for @var{kind}
35930 (@pxref{Architecture-Specific Protocol Details}); if no
35931 architecture-specific value is being used, it should be @samp{0}.
35932 @var{kind} is hex-encoded. @var{cond_list} is an optional list of
35933 conditional expressions in bytecode form that should be evaluated on
35934 the target's side. These are the conditions that should be taken into
35935 consideration when deciding if the breakpoint trigger should be
35936 reported back to @value{GDBN}.
35938 See also the @samp{swbreak} stop reason (@pxref{swbreak stop reason})
35939 for how to best report a software breakpoint event to @value{GDBN}.
35941 The @var{cond_list} parameter is comprised of a series of expressions,
35942 concatenated without separators. Each expression has the following form:
35946 @item X @var{len},@var{expr}
35947 @var{len} is the length of the bytecode expression and @var{expr} is the
35948 actual conditional expression in bytecode form.
35952 The optional @var{cmd_list} parameter introduces commands that may be
35953 run on the target, rather than being reported back to @value{GDBN}.
35954 The parameter starts with a numeric flag @var{persist}; if the flag is
35955 nonzero, then the breakpoint may remain active and the commands
35956 continue to be run even when @value{GDBN} disconnects from the target.
35957 Following this flag is a series of expressions concatenated with no
35958 separators. Each expression has the following form:
35962 @item X @var{len},@var{expr}
35963 @var{len} is the length of the bytecode expression and @var{expr} is the
35964 actual conditional expression in bytecode form.
35968 @emph{Implementation note: It is possible for a target to copy or move
35969 code that contains software breakpoints (e.g., when implementing
35970 overlays). The behavior of this packet, in the presence of such a
35971 target, is not defined.}
35983 @item z1,@var{addr},@var{kind}
35984 @itemx Z1,@var{addr},@var{kind}@r{[};@var{cond_list}@dots{}@r{]}@r{[};cmds:@var{persist},@var{cmd_list}@dots{}@r{]}
35985 @cindex @samp{z1} packet
35986 @cindex @samp{Z1} packet
35987 Insert (@samp{Z1}) or remove (@samp{z1}) a hardware breakpoint at
35988 address @var{addr}.
35990 A hardware breakpoint is implemented using a mechanism that is not
35991 dependent on being able to modify the target's memory. The
35992 @var{kind}, @var{cond_list}, and @var{cmd_list} arguments have the
35993 same meaning as in @samp{Z0} packets.
35995 @emph{Implementation note: A hardware breakpoint is not affected by code
36008 @item z2,@var{addr},@var{kind}
36009 @itemx Z2,@var{addr},@var{kind}
36010 @cindex @samp{z2} packet
36011 @cindex @samp{Z2} packet
36012 Insert (@samp{Z2}) or remove (@samp{z2}) a write watchpoint at @var{addr}.
36013 The number of bytes to watch is specified by @var{kind}.
36025 @item z3,@var{addr},@var{kind}
36026 @itemx Z3,@var{addr},@var{kind}
36027 @cindex @samp{z3} packet
36028 @cindex @samp{Z3} packet
36029 Insert (@samp{Z3}) or remove (@samp{z3}) a read watchpoint at @var{addr}.
36030 The number of bytes to watch is specified by @var{kind}.
36042 @item z4,@var{addr},@var{kind}
36043 @itemx Z4,@var{addr},@var{kind}
36044 @cindex @samp{z4} packet
36045 @cindex @samp{Z4} packet
36046 Insert (@samp{Z4}) or remove (@samp{z4}) an access watchpoint at @var{addr}.
36047 The number of bytes to watch is specified by @var{kind}.
36061 @node Stop Reply Packets
36062 @section Stop Reply Packets
36063 @cindex stop reply packets
36065 The @samp{C}, @samp{c}, @samp{S}, @samp{s}, @samp{vCont},
36066 @samp{vAttach}, @samp{vRun}, @samp{vStopped}, and @samp{?} packets can
36067 receive any of the below as a reply. Except for @samp{?}
36068 and @samp{vStopped}, that reply is only returned
36069 when the target halts. In the below the exact meaning of @dfn{signal
36070 number} is defined by the header @file{include/gdb/signals.h} in the
36071 @value{GDBN} source code.
36073 In non-stop mode, the server will simply reply @samp{OK} to commands
36074 such as @samp{vCont}; any stop will be the subject of a future
36075 notification. @xref{Remote Non-Stop}.
36077 As in the description of request packets, we include spaces in the
36078 reply templates for clarity; these are not part of the reply packet's
36079 syntax. No @value{GDBN} stop reply packet uses spaces to separate its
36085 The program received signal number @var{AA} (a two-digit hexadecimal
36086 number). This is equivalent to a @samp{T} response with no
36087 @var{n}:@var{r} pairs.
36089 @item T @var{AA} @var{n1}:@var{r1};@var{n2}:@var{r2};@dots{}
36090 @cindex @samp{T} packet reply
36091 The program received signal number @var{AA} (a two-digit hexadecimal
36092 number). This is equivalent to an @samp{S} response, except that the
36093 @samp{@var{n}:@var{r}} pairs can carry values of important registers
36094 and other information directly in the stop reply packet, reducing
36095 round-trip latency. Single-step and breakpoint traps are reported
36096 this way. Each @samp{@var{n}:@var{r}} pair is interpreted as follows:
36100 If @var{n} is a hexadecimal number, it is a register number, and the
36101 corresponding @var{r} gives that register's value. The data @var{r} is a
36102 series of bytes in target byte order, with each byte given by a
36103 two-digit hex number.
36106 If @var{n} is @samp{thread}, then @var{r} is the @var{thread-id} of
36107 the stopped thread, as specified in @ref{thread-id syntax}.
36110 If @var{n} is @samp{core}, then @var{r} is the hexadecimal number of
36111 the core on which the stop event was detected.
36114 If @var{n} is a recognized @dfn{stop reason}, it describes a more
36115 specific event that stopped the target. The currently defined stop
36116 reasons are listed below. The @var{aa} should be @samp{05}, the trap
36117 signal. At most one stop reason should be present.
36120 Otherwise, @value{GDBN} should ignore this @samp{@var{n}:@var{r}} pair
36121 and go on to the next; this allows us to extend the protocol in the
36125 The currently defined stop reasons are:
36131 The packet indicates a watchpoint hit, and @var{r} is the data address, in
36134 @item syscall_entry
36135 @itemx syscall_return
36136 The packet indicates a syscall entry or return, and @var{r} is the
36137 syscall number, in hex.
36139 @cindex shared library events, remote reply
36141 The packet indicates that the loaded libraries have changed.
36142 @value{GDBN} should use @samp{qXfer:libraries:read} to fetch a new
36143 list of loaded libraries. The @var{r} part is ignored.
36145 @cindex replay log events, remote reply
36147 The packet indicates that the target cannot continue replaying
36148 logged execution events, because it has reached the end (or the
36149 beginning when executing backward) of the log. The value of @var{r}
36150 will be either @samp{begin} or @samp{end}. @xref{Reverse Execution},
36151 for more information.
36154 @anchor{swbreak stop reason}
36155 The packet indicates a software breakpoint instruction was executed,
36156 irrespective of whether it was @value{GDBN} that planted the
36157 breakpoint or the breakpoint is hardcoded in the program. The @var{r}
36158 part must be left empty.
36160 On some architectures, such as x86, at the architecture level, when a
36161 breakpoint instruction executes the program counter points at the
36162 breakpoint address plus an offset. On such targets, the stub is
36163 responsible for adjusting the PC to point back at the breakpoint
36166 This packet should not be sent by default; older @value{GDBN} versions
36167 did not support it. @value{GDBN} requests it, by supplying an
36168 appropriate @samp{qSupported} feature (@pxref{qSupported}). The
36169 remote stub must also supply the appropriate @samp{qSupported} feature
36170 indicating support.
36172 This packet is required for correct non-stop mode operation.
36175 The packet indicates the target stopped for a hardware breakpoint.
36176 The @var{r} part must be left empty.
36178 The same remarks about @samp{qSupported} and non-stop mode above
36181 @cindex fork events, remote reply
36183 The packet indicates that @code{fork} was called, and @var{r}
36184 is the thread ID of the new child process. Refer to
36185 @ref{thread-id syntax} for the format of the @var{thread-id}
36186 field. This packet is only applicable to targets that support
36189 This packet should not be sent by default; older @value{GDBN} versions
36190 did not support it. @value{GDBN} requests it, by supplying an
36191 appropriate @samp{qSupported} feature (@pxref{qSupported}). The
36192 remote stub must also supply the appropriate @samp{qSupported} feature
36193 indicating support.
36195 @cindex vfork events, remote reply
36197 The packet indicates that @code{vfork} was called, and @var{r}
36198 is the thread ID of the new child process. Refer to
36199 @ref{thread-id syntax} for the format of the @var{thread-id}
36200 field. This packet is only applicable to targets that support
36203 This packet should not be sent by default; older @value{GDBN} versions
36204 did not support it. @value{GDBN} requests it, by supplying an
36205 appropriate @samp{qSupported} feature (@pxref{qSupported}). The
36206 remote stub must also supply the appropriate @samp{qSupported} feature
36207 indicating support.
36209 @cindex vforkdone events, remote reply
36211 The packet indicates that a child process created by a vfork
36212 has either called @code{exec} or terminated, so that the
36213 address spaces of the parent and child process are no longer
36214 shared. The @var{r} part is ignored. This packet is only
36215 applicable to targets that support vforkdone events.
36217 This packet should not be sent by default; older @value{GDBN} versions
36218 did not support it. @value{GDBN} requests it, by supplying an
36219 appropriate @samp{qSupported} feature (@pxref{qSupported}). The
36220 remote stub must also supply the appropriate @samp{qSupported} feature
36221 indicating support.
36223 @cindex exec events, remote reply
36225 The packet indicates that @code{execve} was called, and @var{r}
36226 is the absolute pathname of the file that was executed, in hex.
36227 This packet is only applicable to targets that support exec events.
36229 This packet should not be sent by default; older @value{GDBN} versions
36230 did not support it. @value{GDBN} requests it, by supplying an
36231 appropriate @samp{qSupported} feature (@pxref{qSupported}). The
36232 remote stub must also supply the appropriate @samp{qSupported} feature
36233 indicating support.
36235 @cindex thread create event, remote reply
36236 @anchor{thread create event}
36238 The packet indicates that the thread was just created. The new thread
36239 is stopped until @value{GDBN} sets it running with a resumption packet
36240 (@pxref{vCont packet}). This packet should not be sent by default;
36241 @value{GDBN} requests it with the @ref{QThreadEvents} packet. See
36242 also the @samp{w} (@pxref{thread exit event}) remote reply below. The
36243 @var{r} part is ignored.
36248 @itemx W @var{AA} ; process:@var{pid}
36249 The process exited, and @var{AA} is the exit status. This is only
36250 applicable to certain targets.
36252 The second form of the response, including the process ID of the
36253 exited process, can be used only when @value{GDBN} has reported
36254 support for multiprocess protocol extensions; see @ref{multiprocess
36255 extensions}. Both @var{AA} and @var{pid} are formatted as big-endian
36259 @itemx X @var{AA} ; process:@var{pid}
36260 The process terminated with signal @var{AA}.
36262 The second form of the response, including the process ID of the
36263 terminated process, can be used only when @value{GDBN} has reported
36264 support for multiprocess protocol extensions; see @ref{multiprocess
36265 extensions}. Both @var{AA} and @var{pid} are formatted as big-endian
36268 @anchor{thread exit event}
36269 @cindex thread exit event, remote reply
36270 @item w @var{AA} ; @var{tid}
36272 The thread exited, and @var{AA} is the exit status. This response
36273 should not be sent by default; @value{GDBN} requests it with the
36274 @ref{QThreadEvents} packet. See also @ref{thread create event} above.
36275 @var{AA} is formatted as a big-endian hex string.
36278 There are no resumed threads left in the target. In other words, even
36279 though the process is alive, the last resumed thread has exited. For
36280 example, say the target process has two threads: thread 1 and thread
36281 2. The client leaves thread 1 stopped, and resumes thread 2, which
36282 subsequently exits. At this point, even though the process is still
36283 alive, and thus no @samp{W} stop reply is sent, no thread is actually
36284 executing either. The @samp{N} stop reply thus informs the client
36285 that it can stop waiting for stop replies. This packet should not be
36286 sent by default; older @value{GDBN} versions did not support it.
36287 @value{GDBN} requests it, by supplying an appropriate
36288 @samp{qSupported} feature (@pxref{qSupported}). The remote stub must
36289 also supply the appropriate @samp{qSupported} feature indicating
36292 @item O @var{XX}@dots{}
36293 @samp{@var{XX}@dots{}} is hex encoding of @sc{ascii} data, to be
36294 written as the program's console output. This can happen at any time
36295 while the program is running and the debugger should continue to wait
36296 for @samp{W}, @samp{T}, etc. This reply is not permitted in non-stop mode.
36298 @item F @var{call-id},@var{parameter}@dots{}
36299 @var{call-id} is the identifier which says which host system call should
36300 be called. This is just the name of the function. Translation into the
36301 correct system call is only applicable as it's defined in @value{GDBN}.
36302 @xref{File-I/O Remote Protocol Extension}, for a list of implemented
36305 @samp{@var{parameter}@dots{}} is a list of parameters as defined for
36306 this very system call.
36308 The target replies with this packet when it expects @value{GDBN} to
36309 call a host system call on behalf of the target. @value{GDBN} replies
36310 with an appropriate @samp{F} packet and keeps up waiting for the next
36311 reply packet from the target. The latest @samp{C}, @samp{c}, @samp{S}
36312 or @samp{s} action is expected to be continued. @xref{File-I/O Remote
36313 Protocol Extension}, for more details.
36317 @node General Query Packets
36318 @section General Query Packets
36319 @cindex remote query requests
36321 Packets starting with @samp{q} are @dfn{general query packets};
36322 packets starting with @samp{Q} are @dfn{general set packets}. General
36323 query and set packets are a semi-unified form for retrieving and
36324 sending information to and from the stub.
36326 The initial letter of a query or set packet is followed by a name
36327 indicating what sort of thing the packet applies to. For example,
36328 @value{GDBN} may use a @samp{qSymbol} packet to exchange symbol
36329 definitions with the stub. These packet names follow some
36334 The name must not contain commas, colons or semicolons.
36336 Most @value{GDBN} query and set packets have a leading upper case
36339 The names of custom vendor packets should use a company prefix, in
36340 lower case, followed by a period. For example, packets designed at
36341 the Acme Corporation might begin with @samp{qacme.foo} (for querying
36342 foos) or @samp{Qacme.bar} (for setting bars).
36345 The name of a query or set packet should be separated from any
36346 parameters by a @samp{:}; the parameters themselves should be
36347 separated by @samp{,} or @samp{;}. Stubs must be careful to match the
36348 full packet name, and check for a separator or the end of the packet,
36349 in case two packet names share a common prefix. New packets should not begin
36350 with @samp{qC}, @samp{qP}, or @samp{qL}@footnote{The @samp{qP} and @samp{qL}
36351 packets predate these conventions, and have arguments without any terminator
36352 for the packet name; we suspect they are in widespread use in places that
36353 are difficult to upgrade. The @samp{qC} packet has no arguments, but some
36354 existing stubs (e.g.@: RedBoot) are known to not check for the end of the
36357 Like the descriptions of the other packets, each description here
36358 has a template showing the packet's overall syntax, followed by an
36359 explanation of the packet's meaning. We include spaces in some of the
36360 templates for clarity; these are not part of the packet's syntax. No
36361 @value{GDBN} packet uses spaces to separate its components.
36363 Here are the currently defined query and set packets:
36369 Turn on or off the agent as a helper to perform some debugging operations
36370 delegated from @value{GDBN} (@pxref{Control Agent}).
36372 @item QAllow:@var{op}:@var{val}@dots{}
36373 @cindex @samp{QAllow} packet
36374 Specify which operations @value{GDBN} expects to request of the
36375 target, as a semicolon-separated list of operation name and value
36376 pairs. Possible values for @var{op} include @samp{WriteReg},
36377 @samp{WriteMem}, @samp{InsertBreak}, @samp{InsertTrace},
36378 @samp{InsertFastTrace}, and @samp{Stop}. @var{val} is either 0,
36379 indicating that @value{GDBN} will not request the operation, or 1,
36380 indicating that it may. (The target can then use this to set up its
36381 own internals optimally, for instance if the debugger never expects to
36382 insert breakpoints, it may not need to install its own trap handler.)
36385 @cindex current thread, remote request
36386 @cindex @samp{qC} packet
36387 Return the current thread ID.
36391 @item QC @var{thread-id}
36392 Where @var{thread-id} is a thread ID as documented in
36393 @ref{thread-id syntax}.
36394 @item @r{(anything else)}
36395 Any other reply implies the old thread ID.
36398 @item qCRC:@var{addr},@var{length}
36399 @cindex CRC of memory block, remote request
36400 @cindex @samp{qCRC} packet
36401 @anchor{qCRC packet}
36402 Compute the CRC checksum of a block of memory using CRC-32 defined in
36403 IEEE 802.3. The CRC is computed byte at a time, taking the most
36404 significant bit of each byte first. The initial pattern code
36405 @code{0xffffffff} is used to ensure leading zeros affect the CRC.
36407 @emph{Note:} This is the same CRC used in validating separate debug
36408 files (@pxref{Separate Debug Files, , Debugging Information in Separate
36409 Files}). However the algorithm is slightly different. When validating
36410 separate debug files, the CRC is computed taking the @emph{least}
36411 significant bit of each byte first, and the final result is inverted to
36412 detect trailing zeros.
36417 An error (such as memory fault)
36418 @item C @var{crc32}
36419 The specified memory region's checksum is @var{crc32}.
36422 @item QDisableRandomization:@var{value}
36423 @cindex disable address space randomization, remote request
36424 @cindex @samp{QDisableRandomization} packet
36425 Some target operating systems will randomize the virtual address space
36426 of the inferior process as a security feature, but provide a feature
36427 to disable such randomization, e.g.@: to allow for a more deterministic
36428 debugging experience. On such systems, this packet with a @var{value}
36429 of 1 directs the target to disable address space randomization for
36430 processes subsequently started via @samp{vRun} packets, while a packet
36431 with a @var{value} of 0 tells the target to enable address space
36434 This packet is only available in extended mode (@pxref{extended mode}).
36439 The request succeeded.
36442 An error occurred. The error number @var{nn} is given as hex digits.
36445 An empty reply indicates that @samp{QDisableRandomization} is not supported
36449 This packet is not probed by default; the remote stub must request it,
36450 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
36451 This should only be done on targets that actually support disabling
36452 address space randomization.
36454 @item QStartupWithShell:@var{value}
36455 @cindex startup with shell, remote request
36456 @cindex @samp{QStartupWithShell} packet
36457 On UNIX-like targets, it is possible to start the inferior using a
36458 shell program. This is the default behavior on both @value{GDBN} and
36459 @command{gdbserver} (@pxref{set startup-with-shell}). This packet is
36460 used to inform @command{gdbserver} whether it should start the
36461 inferior using a shell or not.
36463 If @var{value} is @samp{0}, @command{gdbserver} will not use a shell
36464 to start the inferior. If @var{value} is @samp{1},
36465 @command{gdbserver} will use a shell to start the inferior. All other
36466 values are considered an error.
36468 This packet is only available in extended mode (@pxref{extended
36474 The request succeeded.
36477 An error occurred. The error number @var{nn} is given as hex digits.
36480 This packet is not probed by default; the remote stub must request it,
36481 by supplying an appropriate @samp{qSupported} response
36482 (@pxref{qSupported}). This should only be done on targets that
36483 actually support starting the inferior using a shell.
36485 Use of this packet is controlled by the @code{set startup-with-shell}
36486 command; @pxref{set startup-with-shell}.
36489 @itemx qsThreadInfo
36490 @cindex list active threads, remote request
36491 @cindex @samp{qfThreadInfo} packet
36492 @cindex @samp{qsThreadInfo} packet
36493 Obtain a list of all active thread IDs from the target (OS). Since there
36494 may be too many active threads to fit into one reply packet, this query
36495 works iteratively: it may require more than one query/reply sequence to
36496 obtain the entire list of threads. The first query of the sequence will
36497 be the @samp{qfThreadInfo} query; subsequent queries in the
36498 sequence will be the @samp{qsThreadInfo} query.
36500 NOTE: This packet replaces the @samp{qL} query (see below).
36504 @item m @var{thread-id}
36506 @item m @var{thread-id},@var{thread-id}@dots{}
36507 a comma-separated list of thread IDs
36509 (lower case letter @samp{L}) denotes end of list.
36512 In response to each query, the target will reply with a list of one or
36513 more thread IDs, separated by commas.
36514 @value{GDBN} will respond to each reply with a request for more thread
36515 ids (using the @samp{qs} form of the query), until the target responds
36516 with @samp{l} (lower-case ell, for @dfn{last}).
36517 Refer to @ref{thread-id syntax}, for the format of the @var{thread-id}
36520 @emph{Note: @value{GDBN} will send the @code{qfThreadInfo} query during the
36521 initial connection with the remote target, and the very first thread ID
36522 mentioned in the reply will be stopped by @value{GDBN} in a subsequent
36523 message. Therefore, the stub should ensure that the first thread ID in
36524 the @code{qfThreadInfo} reply is suitable for being stopped by @value{GDBN}.}
36526 @item qGetTLSAddr:@var{thread-id},@var{offset},@var{lm}
36527 @cindex get thread-local storage address, remote request
36528 @cindex @samp{qGetTLSAddr} packet
36529 Fetch the address associated with thread local storage specified
36530 by @var{thread-id}, @var{offset}, and @var{lm}.
36532 @var{thread-id} is the thread ID associated with the
36533 thread for which to fetch the TLS address. @xref{thread-id syntax}.
36535 @var{offset} is the (big endian, hex encoded) offset associated with the
36536 thread local variable. (This offset is obtained from the debug
36537 information associated with the variable.)
36539 @var{lm} is the (big endian, hex encoded) OS/ABI-specific encoding of the
36540 load module associated with the thread local storage. For example,
36541 a @sc{gnu}/Linux system will pass the link map address of the shared
36542 object associated with the thread local storage under consideration.
36543 Other operating environments may choose to represent the load module
36544 differently, so the precise meaning of this parameter will vary.
36548 @item @var{XX}@dots{}
36549 Hex encoded (big endian) bytes representing the address of the thread
36550 local storage requested.
36553 An error occurred. The error number @var{nn} is given as hex digits.
36556 An empty reply indicates that @samp{qGetTLSAddr} is not supported by the stub.
36559 @item qGetTIBAddr:@var{thread-id}
36560 @cindex get thread information block address
36561 @cindex @samp{qGetTIBAddr} packet
36562 Fetch address of the Windows OS specific Thread Information Block.
36564 @var{thread-id} is the thread ID associated with the thread.
36568 @item @var{XX}@dots{}
36569 Hex encoded (big endian) bytes representing the linear address of the
36570 thread information block.
36573 An error occured. This means that either the thread was not found, or the
36574 address could not be retrieved.
36577 An empty reply indicates that @samp{qGetTIBAddr} is not supported by the stub.
36580 @item qL @var{startflag} @var{threadcount} @var{nextthread}
36581 Obtain thread information from RTOS. Where: @var{startflag} (one hex
36582 digit) is one to indicate the first query and zero to indicate a
36583 subsequent query; @var{threadcount} (two hex digits) is the maximum
36584 number of threads the response packet can contain; and @var{nextthread}
36585 (eight hex digits), for subsequent queries (@var{startflag} is zero), is
36586 returned in the response as @var{argthread}.
36588 Don't use this packet; use the @samp{qfThreadInfo} query instead (see above).
36592 @item qM @var{count} @var{done} @var{argthread} @var{thread}@dots{}
36593 Where: @var{count} (two hex digits) is the number of threads being
36594 returned; @var{done} (one hex digit) is zero to indicate more threads
36595 and one indicates no further threads; @var{argthreadid} (eight hex
36596 digits) is @var{nextthread} from the request packet; @var{thread}@dots{}
36597 is a sequence of thread IDs, @var{threadid} (eight hex
36598 digits), from the target. See @code{remote.c:parse_threadlist_response()}.
36602 @cindex section offsets, remote request
36603 @cindex @samp{qOffsets} packet
36604 Get section offsets that the target used when relocating the downloaded
36609 @item Text=@var{xxx};Data=@var{yyy}@r{[};Bss=@var{zzz}@r{]}
36610 Relocate the @code{Text} section by @var{xxx} from its original address.
36611 Relocate the @code{Data} section by @var{yyy} from its original address.
36612 If the object file format provides segment information (e.g.@: @sc{elf}
36613 @samp{PT_LOAD} program headers), @value{GDBN} will relocate entire
36614 segments by the supplied offsets.
36616 @emph{Note: while a @code{Bss} offset may be included in the response,
36617 @value{GDBN} ignores this and instead applies the @code{Data} offset
36618 to the @code{Bss} section.}
36620 @item TextSeg=@var{xxx}@r{[};DataSeg=@var{yyy}@r{]}
36621 Relocate the first segment of the object file, which conventionally
36622 contains program code, to a starting address of @var{xxx}. If
36623 @samp{DataSeg} is specified, relocate the second segment, which
36624 conventionally contains modifiable data, to a starting address of
36625 @var{yyy}. @value{GDBN} will report an error if the object file
36626 does not contain segment information, or does not contain at least
36627 as many segments as mentioned in the reply. Extra segments are
36628 kept at fixed offsets relative to the last relocated segment.
36631 @item qP @var{mode} @var{thread-id}
36632 @cindex thread information, remote request
36633 @cindex @samp{qP} packet
36634 Returns information on @var{thread-id}. Where: @var{mode} is a hex
36635 encoded 32 bit mode; @var{thread-id} is a thread ID
36636 (@pxref{thread-id syntax}).
36638 Don't use this packet; use the @samp{qThreadExtraInfo} query instead
36641 Reply: see @code{remote.c:remote_unpack_thread_info_response()}.
36645 @cindex non-stop mode, remote request
36646 @cindex @samp{QNonStop} packet
36648 Enter non-stop (@samp{QNonStop:1}) or all-stop (@samp{QNonStop:0}) mode.
36649 @xref{Remote Non-Stop}, for more information.
36654 The request succeeded.
36657 An error occurred. The error number @var{nn} is given as hex digits.
36660 An empty reply indicates that @samp{QNonStop} is not supported by
36664 This packet is not probed by default; the remote stub must request it,
36665 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
36666 Use of this packet is controlled by the @code{set non-stop} command;
36667 @pxref{Non-Stop Mode}.
36669 @item QCatchSyscalls:1 @r{[};@var{sysno}@r{]}@dots{}
36670 @itemx QCatchSyscalls:0
36671 @cindex catch syscalls from inferior, remote request
36672 @cindex @samp{QCatchSyscalls} packet
36673 @anchor{QCatchSyscalls}
36674 Enable (@samp{QCatchSyscalls:1}) or disable (@samp{QCatchSyscalls:0})
36675 catching syscalls from the inferior process.
36677 For @samp{QCatchSyscalls:1}, each listed syscall @var{sysno} (encoded
36678 in hex) should be reported to @value{GDBN}. If no syscall @var{sysno}
36679 is listed, every system call should be reported.
36681 Note that if a syscall not in the list is reported, @value{GDBN} will
36682 still filter the event according to its own list from all corresponding
36683 @code{catch syscall} commands. However, it is more efficient to only
36684 report the requested syscalls.
36686 Multiple @samp{QCatchSyscalls:1} packets do not combine; any earlier
36687 @samp{QCatchSyscalls:1} list is completely replaced by the new list.
36689 If the inferior process execs, the state of @samp{QCatchSyscalls} is
36690 kept for the new process too. On targets where exec may affect syscall
36691 numbers, for example with exec between 32 and 64-bit processes, the
36692 client should send a new packet with the new syscall list.
36697 The request succeeded.
36700 An error occurred. @var{nn} are hex digits.
36703 An empty reply indicates that @samp{QCatchSyscalls} is not supported by
36707 Use of this packet is controlled by the @code{set remote catch-syscalls}
36708 command (@pxref{Remote Configuration, set remote catch-syscalls}).
36709 This packet is not probed by default; the remote stub must request it,
36710 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
36712 @item QPassSignals: @var{signal} @r{[};@var{signal}@r{]}@dots{}
36713 @cindex pass signals to inferior, remote request
36714 @cindex @samp{QPassSignals} packet
36715 @anchor{QPassSignals}
36716 Each listed @var{signal} should be passed directly to the inferior process.
36717 Signals are numbered identically to continue packets and stop replies
36718 (@pxref{Stop Reply Packets}). Each @var{signal} list item should be
36719 strictly greater than the previous item. These signals do not need to stop
36720 the inferior, or be reported to @value{GDBN}. All other signals should be
36721 reported to @value{GDBN}. Multiple @samp{QPassSignals} packets do not
36722 combine; any earlier @samp{QPassSignals} list is completely replaced by the
36723 new list. This packet improves performance when using @samp{handle
36724 @var{signal} nostop noprint pass}.
36729 The request succeeded.
36732 An error occurred. The error number @var{nn} is given as hex digits.
36735 An empty reply indicates that @samp{QPassSignals} is not supported by
36739 Use of this packet is controlled by the @code{set remote pass-signals}
36740 command (@pxref{Remote Configuration, set remote pass-signals}).
36741 This packet is not probed by default; the remote stub must request it,
36742 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
36744 @item QProgramSignals: @var{signal} @r{[};@var{signal}@r{]}@dots{}
36745 @cindex signals the inferior may see, remote request
36746 @cindex @samp{QProgramSignals} packet
36747 @anchor{QProgramSignals}
36748 Each listed @var{signal} may be delivered to the inferior process.
36749 Others should be silently discarded.
36751 In some cases, the remote stub may need to decide whether to deliver a
36752 signal to the program or not without @value{GDBN} involvement. One
36753 example of that is while detaching --- the program's threads may have
36754 stopped for signals that haven't yet had a chance of being reported to
36755 @value{GDBN}, and so the remote stub can use the signal list specified
36756 by this packet to know whether to deliver or ignore those pending
36759 This does not influence whether to deliver a signal as requested by a
36760 resumption packet (@pxref{vCont packet}).
36762 Signals are numbered identically to continue packets and stop replies
36763 (@pxref{Stop Reply Packets}). Each @var{signal} list item should be
36764 strictly greater than the previous item. Multiple
36765 @samp{QProgramSignals} packets do not combine; any earlier
36766 @samp{QProgramSignals} list is completely replaced by the new list.
36771 The request succeeded.
36774 An error occurred. The error number @var{nn} is given as hex digits.
36777 An empty reply indicates that @samp{QProgramSignals} is not supported
36781 Use of this packet is controlled by the @code{set remote program-signals}
36782 command (@pxref{Remote Configuration, set remote program-signals}).
36783 This packet is not probed by default; the remote stub must request it,
36784 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
36786 @anchor{QThreadEvents}
36787 @item QThreadEvents:1
36788 @itemx QThreadEvents:0
36789 @cindex thread create/exit events, remote request
36790 @cindex @samp{QThreadEvents} packet
36792 Enable (@samp{QThreadEvents:1}) or disable (@samp{QThreadEvents:0})
36793 reporting of thread create and exit events. @xref{thread create
36794 event}, for the reply specifications. For example, this is used in
36795 non-stop mode when @value{GDBN} stops a set of threads and
36796 synchronously waits for the their corresponding stop replies. Without
36797 exit events, if one of the threads exits, @value{GDBN} would hang
36798 forever not knowing that it should no longer expect a stop for that
36799 same thread. @value{GDBN} does not enable this feature unless the
36800 stub reports that it supports it by including @samp{QThreadEvents+} in
36801 its @samp{qSupported} reply.
36806 The request succeeded.
36809 An error occurred. The error number @var{nn} is given as hex digits.
36812 An empty reply indicates that @samp{QThreadEvents} is not supported by
36816 Use of this packet is controlled by the @code{set remote thread-events}
36817 command (@pxref{Remote Configuration, set remote thread-events}).
36819 @item qRcmd,@var{command}
36820 @cindex execute remote command, remote request
36821 @cindex @samp{qRcmd} packet
36822 @var{command} (hex encoded) is passed to the local interpreter for
36823 execution. Invalid commands should be reported using the output
36824 string. Before the final result packet, the target may also respond
36825 with a number of intermediate @samp{O@var{output}} console output
36826 packets. @emph{Implementors should note that providing access to a
36827 stubs's interpreter may have security implications}.
36832 A command response with no output.
36834 A command response with the hex encoded output string @var{OUTPUT}.
36836 Indicate a badly formed request.
36838 An empty reply indicates that @samp{qRcmd} is not recognized.
36841 (Note that the @code{qRcmd} packet's name is separated from the
36842 command by a @samp{,}, not a @samp{:}, contrary to the naming
36843 conventions above. Please don't use this packet as a model for new
36846 @item qSearch:memory:@var{address};@var{length};@var{search-pattern}
36847 @cindex searching memory, in remote debugging
36849 @cindex @samp{qSearch:memory} packet
36851 @cindex @samp{qSearch memory} packet
36852 @anchor{qSearch memory}
36853 Search @var{length} bytes at @var{address} for @var{search-pattern}.
36854 Both @var{address} and @var{length} are encoded in hex;
36855 @var{search-pattern} is a sequence of bytes, also hex encoded.
36860 The pattern was not found.
36862 The pattern was found at @var{address}.
36864 A badly formed request or an error was encountered while searching memory.
36866 An empty reply indicates that @samp{qSearch:memory} is not recognized.
36869 @item QStartNoAckMode
36870 @cindex @samp{QStartNoAckMode} packet
36871 @anchor{QStartNoAckMode}
36872 Request that the remote stub disable the normal @samp{+}/@samp{-}
36873 protocol acknowledgments (@pxref{Packet Acknowledgment}).
36878 The stub has switched to no-acknowledgment mode.
36879 @value{GDBN} acknowledges this reponse,
36880 but neither the stub nor @value{GDBN} shall send or expect further
36881 @samp{+}/@samp{-} acknowledgments in the current connection.
36883 An empty reply indicates that the stub does not support no-acknowledgment mode.
36886 @item qSupported @r{[}:@var{gdbfeature} @r{[};@var{gdbfeature}@r{]}@dots{} @r{]}
36887 @cindex supported packets, remote query
36888 @cindex features of the remote protocol
36889 @cindex @samp{qSupported} packet
36890 @anchor{qSupported}
36891 Tell the remote stub about features supported by @value{GDBN}, and
36892 query the stub for features it supports. This packet allows
36893 @value{GDBN} and the remote stub to take advantage of each others'
36894 features. @samp{qSupported} also consolidates multiple feature probes
36895 at startup, to improve @value{GDBN} performance---a single larger
36896 packet performs better than multiple smaller probe packets on
36897 high-latency links. Some features may enable behavior which must not
36898 be on by default, e.g.@: because it would confuse older clients or
36899 stubs. Other features may describe packets which could be
36900 automatically probed for, but are not. These features must be
36901 reported before @value{GDBN} will use them. This ``default
36902 unsupported'' behavior is not appropriate for all packets, but it
36903 helps to keep the initial connection time under control with new
36904 versions of @value{GDBN} which support increasing numbers of packets.
36908 @item @var{stubfeature} @r{[};@var{stubfeature}@r{]}@dots{}
36909 The stub supports or does not support each returned @var{stubfeature},
36910 depending on the form of each @var{stubfeature} (see below for the
36913 An empty reply indicates that @samp{qSupported} is not recognized,
36914 or that no features needed to be reported to @value{GDBN}.
36917 The allowed forms for each feature (either a @var{gdbfeature} in the
36918 @samp{qSupported} packet, or a @var{stubfeature} in the response)
36922 @item @var{name}=@var{value}
36923 The remote protocol feature @var{name} is supported, and associated
36924 with the specified @var{value}. The format of @var{value} depends
36925 on the feature, but it must not include a semicolon.
36927 The remote protocol feature @var{name} is supported, and does not
36928 need an associated value.
36930 The remote protocol feature @var{name} is not supported.
36932 The remote protocol feature @var{name} may be supported, and
36933 @value{GDBN} should auto-detect support in some other way when it is
36934 needed. This form will not be used for @var{gdbfeature} notifications,
36935 but may be used for @var{stubfeature} responses.
36938 Whenever the stub receives a @samp{qSupported} request, the
36939 supplied set of @value{GDBN} features should override any previous
36940 request. This allows @value{GDBN} to put the stub in a known
36941 state, even if the stub had previously been communicating with
36942 a different version of @value{GDBN}.
36944 The following values of @var{gdbfeature} (for the packet sent by @value{GDBN})
36949 This feature indicates whether @value{GDBN} supports multiprocess
36950 extensions to the remote protocol. @value{GDBN} does not use such
36951 extensions unless the stub also reports that it supports them by
36952 including @samp{multiprocess+} in its @samp{qSupported} reply.
36953 @xref{multiprocess extensions}, for details.
36956 This feature indicates that @value{GDBN} supports the XML target
36957 description. If the stub sees @samp{xmlRegisters=} with target
36958 specific strings separated by a comma, it will report register
36962 This feature indicates whether @value{GDBN} supports the
36963 @samp{qRelocInsn} packet (@pxref{Tracepoint Packets,,Relocate
36964 instruction reply packet}).
36967 This feature indicates whether @value{GDBN} supports the swbreak stop
36968 reason in stop replies. @xref{swbreak stop reason}, for details.
36971 This feature indicates whether @value{GDBN} supports the hwbreak stop
36972 reason in stop replies. @xref{swbreak stop reason}, for details.
36975 This feature indicates whether @value{GDBN} supports fork event
36976 extensions to the remote protocol. @value{GDBN} does not use such
36977 extensions unless the stub also reports that it supports them by
36978 including @samp{fork-events+} in its @samp{qSupported} reply.
36981 This feature indicates whether @value{GDBN} supports vfork event
36982 extensions to the remote protocol. @value{GDBN} does not use such
36983 extensions unless the stub also reports that it supports them by
36984 including @samp{vfork-events+} in its @samp{qSupported} reply.
36987 This feature indicates whether @value{GDBN} supports exec event
36988 extensions to the remote protocol. @value{GDBN} does not use such
36989 extensions unless the stub also reports that it supports them by
36990 including @samp{exec-events+} in its @samp{qSupported} reply.
36992 @item vContSupported
36993 This feature indicates whether @value{GDBN} wants to know the
36994 supported actions in the reply to @samp{vCont?} packet.
36997 Stubs should ignore any unknown values for
36998 @var{gdbfeature}. Any @value{GDBN} which sends a @samp{qSupported}
36999 packet supports receiving packets of unlimited length (earlier
37000 versions of @value{GDBN} may reject overly long responses). Additional values
37001 for @var{gdbfeature} may be defined in the future to let the stub take
37002 advantage of new features in @value{GDBN}, e.g.@: incompatible
37003 improvements in the remote protocol---the @samp{multiprocess} feature is
37004 an example of such a feature. The stub's reply should be independent
37005 of the @var{gdbfeature} entries sent by @value{GDBN}; first @value{GDBN}
37006 describes all the features it supports, and then the stub replies with
37007 all the features it supports.
37009 Similarly, @value{GDBN} will silently ignore unrecognized stub feature
37010 responses, as long as each response uses one of the standard forms.
37012 Some features are flags. A stub which supports a flag feature
37013 should respond with a @samp{+} form response. Other features
37014 require values, and the stub should respond with an @samp{=}
37017 Each feature has a default value, which @value{GDBN} will use if
37018 @samp{qSupported} is not available or if the feature is not mentioned
37019 in the @samp{qSupported} response. The default values are fixed; a
37020 stub is free to omit any feature responses that match the defaults.
37022 Not all features can be probed, but for those which can, the probing
37023 mechanism is useful: in some cases, a stub's internal
37024 architecture may not allow the protocol layer to know some information
37025 about the underlying target in advance. This is especially common in
37026 stubs which may be configured for multiple targets.
37028 These are the currently defined stub features and their properties:
37030 @multitable @columnfractions 0.35 0.2 0.12 0.2
37031 @c NOTE: The first row should be @headitem, but we do not yet require
37032 @c a new enough version of Texinfo (4.7) to use @headitem.
37034 @tab Value Required
37038 @item @samp{PacketSize}
37043 @item @samp{qXfer:auxv:read}
37048 @item @samp{qXfer:btrace:read}
37053 @item @samp{qXfer:btrace-conf:read}
37058 @item @samp{qXfer:exec-file:read}
37063 @item @samp{qXfer:features:read}
37068 @item @samp{qXfer:libraries:read}
37073 @item @samp{qXfer:libraries-svr4:read}
37078 @item @samp{augmented-libraries-svr4-read}
37083 @item @samp{qXfer:memory-map:read}
37088 @item @samp{qXfer:sdata:read}
37093 @item @samp{qXfer:spu:read}
37098 @item @samp{qXfer:spu:write}
37103 @item @samp{qXfer:siginfo:read}
37108 @item @samp{qXfer:siginfo:write}
37113 @item @samp{qXfer:threads:read}
37118 @item @samp{qXfer:traceframe-info:read}
37123 @item @samp{qXfer:uib:read}
37128 @item @samp{qXfer:fdpic:read}
37133 @item @samp{Qbtrace:off}
37138 @item @samp{Qbtrace:bts}
37143 @item @samp{Qbtrace:pt}
37148 @item @samp{Qbtrace-conf:bts:size}
37153 @item @samp{Qbtrace-conf:pt:size}
37158 @item @samp{QNonStop}
37163 @item @samp{QCatchSyscalls}
37168 @item @samp{QPassSignals}
37173 @item @samp{QStartNoAckMode}
37178 @item @samp{multiprocess}
37183 @item @samp{ConditionalBreakpoints}
37188 @item @samp{ConditionalTracepoints}
37193 @item @samp{ReverseContinue}
37198 @item @samp{ReverseStep}
37203 @item @samp{TracepointSource}
37208 @item @samp{QAgent}
37213 @item @samp{QAllow}
37218 @item @samp{QDisableRandomization}
37223 @item @samp{EnableDisableTracepoints}
37228 @item @samp{QTBuffer:size}
37233 @item @samp{tracenz}
37238 @item @samp{BreakpointCommands}
37243 @item @samp{swbreak}
37248 @item @samp{hwbreak}
37253 @item @samp{fork-events}
37258 @item @samp{vfork-events}
37263 @item @samp{exec-events}
37268 @item @samp{QThreadEvents}
37273 @item @samp{no-resumed}
37280 These are the currently defined stub features, in more detail:
37283 @cindex packet size, remote protocol
37284 @item PacketSize=@var{bytes}
37285 The remote stub can accept packets up to at least @var{bytes} in
37286 length. @value{GDBN} will send packets up to this size for bulk
37287 transfers, and will never send larger packets. This is a limit on the
37288 data characters in the packet, including the frame and checksum.
37289 There is no trailing NUL byte in a remote protocol packet; if the stub
37290 stores packets in a NUL-terminated format, it should allow an extra
37291 byte in its buffer for the NUL. If this stub feature is not supported,
37292 @value{GDBN} guesses based on the size of the @samp{g} packet response.
37294 @item qXfer:auxv:read
37295 The remote stub understands the @samp{qXfer:auxv:read} packet
37296 (@pxref{qXfer auxiliary vector read}).
37298 @item qXfer:btrace:read
37299 The remote stub understands the @samp{qXfer:btrace:read}
37300 packet (@pxref{qXfer btrace read}).
37302 @item qXfer:btrace-conf:read
37303 The remote stub understands the @samp{qXfer:btrace-conf:read}
37304 packet (@pxref{qXfer btrace-conf read}).
37306 @item qXfer:exec-file:read
37307 The remote stub understands the @samp{qXfer:exec-file:read} packet
37308 (@pxref{qXfer executable filename read}).
37310 @item qXfer:features:read
37311 The remote stub understands the @samp{qXfer:features:read} packet
37312 (@pxref{qXfer target description read}).
37314 @item qXfer:libraries:read
37315 The remote stub understands the @samp{qXfer:libraries:read} packet
37316 (@pxref{qXfer library list read}).
37318 @item qXfer:libraries-svr4:read
37319 The remote stub understands the @samp{qXfer:libraries-svr4:read} packet
37320 (@pxref{qXfer svr4 library list read}).
37322 @item augmented-libraries-svr4-read
37323 The remote stub understands the augmented form of the
37324 @samp{qXfer:libraries-svr4:read} packet
37325 (@pxref{qXfer svr4 library list read}).
37327 @item qXfer:memory-map:read
37328 The remote stub understands the @samp{qXfer:memory-map:read} packet
37329 (@pxref{qXfer memory map read}).
37331 @item qXfer:sdata:read
37332 The remote stub understands the @samp{qXfer:sdata:read} packet
37333 (@pxref{qXfer sdata read}).
37335 @item qXfer:spu:read
37336 The remote stub understands the @samp{qXfer:spu:read} packet
37337 (@pxref{qXfer spu read}).
37339 @item qXfer:spu:write
37340 The remote stub understands the @samp{qXfer:spu:write} packet
37341 (@pxref{qXfer spu write}).
37343 @item qXfer:siginfo:read
37344 The remote stub understands the @samp{qXfer:siginfo:read} packet
37345 (@pxref{qXfer siginfo read}).
37347 @item qXfer:siginfo:write
37348 The remote stub understands the @samp{qXfer:siginfo:write} packet
37349 (@pxref{qXfer siginfo write}).
37351 @item qXfer:threads:read
37352 The remote stub understands the @samp{qXfer:threads:read} packet
37353 (@pxref{qXfer threads read}).
37355 @item qXfer:traceframe-info:read
37356 The remote stub understands the @samp{qXfer:traceframe-info:read}
37357 packet (@pxref{qXfer traceframe info read}).
37359 @item qXfer:uib:read
37360 The remote stub understands the @samp{qXfer:uib:read}
37361 packet (@pxref{qXfer unwind info block}).
37363 @item qXfer:fdpic:read
37364 The remote stub understands the @samp{qXfer:fdpic:read}
37365 packet (@pxref{qXfer fdpic loadmap read}).
37368 The remote stub understands the @samp{QNonStop} packet
37369 (@pxref{QNonStop}).
37371 @item QCatchSyscalls
37372 The remote stub understands the @samp{QCatchSyscalls} packet
37373 (@pxref{QCatchSyscalls}).
37376 The remote stub understands the @samp{QPassSignals} packet
37377 (@pxref{QPassSignals}).
37379 @item QStartNoAckMode
37380 The remote stub understands the @samp{QStartNoAckMode} packet and
37381 prefers to operate in no-acknowledgment mode. @xref{Packet Acknowledgment}.
37384 @anchor{multiprocess extensions}
37385 @cindex multiprocess extensions, in remote protocol
37386 The remote stub understands the multiprocess extensions to the remote
37387 protocol syntax. The multiprocess extensions affect the syntax of
37388 thread IDs in both packets and replies (@pxref{thread-id syntax}), and
37389 add process IDs to the @samp{D} packet and @samp{W} and @samp{X}
37390 replies. Note that reporting this feature indicates support for the
37391 syntactic extensions only, not that the stub necessarily supports
37392 debugging of more than one process at a time. The stub must not use
37393 multiprocess extensions in packet replies unless @value{GDBN} has also
37394 indicated it supports them in its @samp{qSupported} request.
37396 @item qXfer:osdata:read
37397 The remote stub understands the @samp{qXfer:osdata:read} packet
37398 ((@pxref{qXfer osdata read}).
37400 @item ConditionalBreakpoints
37401 The target accepts and implements evaluation of conditional expressions
37402 defined for breakpoints. The target will only report breakpoint triggers
37403 when such conditions are true (@pxref{Conditions, ,Break Conditions}).
37405 @item ConditionalTracepoints
37406 The remote stub accepts and implements conditional expressions defined
37407 for tracepoints (@pxref{Tracepoint Conditions}).
37409 @item ReverseContinue
37410 The remote stub accepts and implements the reverse continue packet
37414 The remote stub accepts and implements the reverse step packet
37417 @item TracepointSource
37418 The remote stub understands the @samp{QTDPsrc} packet that supplies
37419 the source form of tracepoint definitions.
37422 The remote stub understands the @samp{QAgent} packet.
37425 The remote stub understands the @samp{QAllow} packet.
37427 @item QDisableRandomization
37428 The remote stub understands the @samp{QDisableRandomization} packet.
37430 @item StaticTracepoint
37431 @cindex static tracepoints, in remote protocol
37432 The remote stub supports static tracepoints.
37434 @item InstallInTrace
37435 @anchor{install tracepoint in tracing}
37436 The remote stub supports installing tracepoint in tracing.
37438 @item EnableDisableTracepoints
37439 The remote stub supports the @samp{QTEnable} (@pxref{QTEnable}) and
37440 @samp{QTDisable} (@pxref{QTDisable}) packets that allow tracepoints
37441 to be enabled and disabled while a trace experiment is running.
37443 @item QTBuffer:size
37444 The remote stub supports the @samp{QTBuffer:size} (@pxref{QTBuffer-size})
37445 packet that allows to change the size of the trace buffer.
37448 @cindex string tracing, in remote protocol
37449 The remote stub supports the @samp{tracenz} bytecode for collecting strings.
37450 See @ref{Bytecode Descriptions} for details about the bytecode.
37452 @item BreakpointCommands
37453 @cindex breakpoint commands, in remote protocol
37454 The remote stub supports running a breakpoint's command list itself,
37455 rather than reporting the hit to @value{GDBN}.
37458 The remote stub understands the @samp{Qbtrace:off} packet.
37461 The remote stub understands the @samp{Qbtrace:bts} packet.
37464 The remote stub understands the @samp{Qbtrace:pt} packet.
37466 @item Qbtrace-conf:bts:size
37467 The remote stub understands the @samp{Qbtrace-conf:bts:size} packet.
37469 @item Qbtrace-conf:pt:size
37470 The remote stub understands the @samp{Qbtrace-conf:pt:size} packet.
37473 The remote stub reports the @samp{swbreak} stop reason for memory
37477 The remote stub reports the @samp{hwbreak} stop reason for hardware
37481 The remote stub reports the @samp{fork} stop reason for fork events.
37484 The remote stub reports the @samp{vfork} stop reason for vfork events
37485 and vforkdone events.
37488 The remote stub reports the @samp{exec} stop reason for exec events.
37490 @item vContSupported
37491 The remote stub reports the supported actions in the reply to
37492 @samp{vCont?} packet.
37494 @item QThreadEvents
37495 The remote stub understands the @samp{QThreadEvents} packet.
37498 The remote stub reports the @samp{N} stop reply.
37503 @cindex symbol lookup, remote request
37504 @cindex @samp{qSymbol} packet
37505 Notify the target that @value{GDBN} is prepared to serve symbol lookup
37506 requests. Accept requests from the target for the values of symbols.
37511 The target does not need to look up any (more) symbols.
37512 @item qSymbol:@var{sym_name}
37513 The target requests the value of symbol @var{sym_name} (hex encoded).
37514 @value{GDBN} may provide the value by using the
37515 @samp{qSymbol:@var{sym_value}:@var{sym_name}} message, described
37519 @item qSymbol:@var{sym_value}:@var{sym_name}
37520 Set the value of @var{sym_name} to @var{sym_value}.
37522 @var{sym_name} (hex encoded) is the name of a symbol whose value the
37523 target has previously requested.
37525 @var{sym_value} (hex) is the value for symbol @var{sym_name}. If
37526 @value{GDBN} cannot supply a value for @var{sym_name}, then this field
37532 The target does not need to look up any (more) symbols.
37533 @item qSymbol:@var{sym_name}
37534 The target requests the value of a new symbol @var{sym_name} (hex
37535 encoded). @value{GDBN} will continue to supply the values of symbols
37536 (if available), until the target ceases to request them.
37541 @itemx QTDisconnected
37548 @itemx qTMinFTPILen
37550 @xref{Tracepoint Packets}.
37552 @item qThreadExtraInfo,@var{thread-id}
37553 @cindex thread attributes info, remote request
37554 @cindex @samp{qThreadExtraInfo} packet
37555 Obtain from the target OS a printable string description of thread
37556 attributes for the thread @var{thread-id}; see @ref{thread-id syntax},
37557 for the forms of @var{thread-id}. This
37558 string may contain anything that the target OS thinks is interesting
37559 for @value{GDBN} to tell the user about the thread. The string is
37560 displayed in @value{GDBN}'s @code{info threads} display. Some
37561 examples of possible thread extra info strings are @samp{Runnable}, or
37562 @samp{Blocked on Mutex}.
37566 @item @var{XX}@dots{}
37567 Where @samp{@var{XX}@dots{}} is a hex encoding of @sc{ascii} data,
37568 comprising the printable string containing the extra information about
37569 the thread's attributes.
37572 (Note that the @code{qThreadExtraInfo} packet's name is separated from
37573 the command by a @samp{,}, not a @samp{:}, contrary to the naming
37574 conventions above. Please don't use this packet as a model for new
37593 @xref{Tracepoint Packets}.
37595 @item qXfer:@var{object}:read:@var{annex}:@var{offset},@var{length}
37596 @cindex read special object, remote request
37597 @cindex @samp{qXfer} packet
37598 @anchor{qXfer read}
37599 Read uninterpreted bytes from the target's special data area
37600 identified by the keyword @var{object}. Request @var{length} bytes
37601 starting at @var{offset} bytes into the data. The content and
37602 encoding of @var{annex} is specific to @var{object}; it can supply
37603 additional details about what data to access.
37608 Data @var{data} (@pxref{Binary Data}) has been read from the
37609 target. There may be more data at a higher address (although
37610 it is permitted to return @samp{m} even for the last valid
37611 block of data, as long as at least one byte of data was read).
37612 It is possible for @var{data} to have fewer bytes than the @var{length} in the
37616 Data @var{data} (@pxref{Binary Data}) has been read from the target.
37617 There is no more data to be read. It is possible for @var{data} to
37618 have fewer bytes than the @var{length} in the request.
37621 The @var{offset} in the request is at the end of the data.
37622 There is no more data to be read.
37625 The request was malformed, or @var{annex} was invalid.
37628 The offset was invalid, or there was an error encountered reading the data.
37629 The @var{nn} part is a hex-encoded @code{errno} value.
37632 An empty reply indicates the @var{object} string was not recognized by
37633 the stub, or that the object does not support reading.
37636 Here are the specific requests of this form defined so far. All the
37637 @samp{qXfer:@var{object}:read:@dots{}} requests use the same reply
37638 formats, listed above.
37641 @item qXfer:auxv:read::@var{offset},@var{length}
37642 @anchor{qXfer auxiliary vector read}
37643 Access the target's @dfn{auxiliary vector}. @xref{OS Information,
37644 auxiliary vector}. Note @var{annex} must be empty.
37646 This packet is not probed by default; the remote stub must request it,
37647 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
37649 @item qXfer:btrace:read:@var{annex}:@var{offset},@var{length}
37650 @anchor{qXfer btrace read}
37652 Return a description of the current branch trace.
37653 @xref{Branch Trace Format}. The annex part of the generic @samp{qXfer}
37654 packet may have one of the following values:
37658 Returns all available branch trace.
37661 Returns all available branch trace if the branch trace changed since
37662 the last read request.
37665 Returns the new branch trace since the last read request. Adds a new
37666 block to the end of the trace that begins at zero and ends at the source
37667 location of the first branch in the trace buffer. This extra block is
37668 used to stitch traces together.
37670 If the trace buffer overflowed, returns an error indicating the overflow.
37673 This packet is not probed by default; the remote stub must request it
37674 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
37676 @item qXfer:btrace-conf:read::@var{offset},@var{length}
37677 @anchor{qXfer btrace-conf read}
37679 Return a description of the current branch trace configuration.
37680 @xref{Branch Trace Configuration Format}.
37682 This packet is not probed by default; the remote stub must request it
37683 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
37685 @item qXfer:exec-file:read:@var{annex}:@var{offset},@var{length}
37686 @anchor{qXfer executable filename read}
37687 Return the full absolute name of the file that was executed to create
37688 a process running on the remote system. The annex specifies the
37689 numeric process ID of the process to query, encoded as a hexadecimal
37690 number. If the annex part is empty the remote stub should return the
37691 filename corresponding to the currently executing process.
37693 This packet is not probed by default; the remote stub must request it,
37694 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
37696 @item qXfer:features:read:@var{annex}:@var{offset},@var{length}
37697 @anchor{qXfer target description read}
37698 Access the @dfn{target description}. @xref{Target Descriptions}. The
37699 annex specifies which XML document to access. The main description is
37700 always loaded from the @samp{target.xml} annex.
37702 This packet is not probed by default; the remote stub must request it,
37703 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
37705 @item qXfer:libraries:read:@var{annex}:@var{offset},@var{length}
37706 @anchor{qXfer library list read}
37707 Access the target's list of loaded libraries. @xref{Library List Format}.
37708 The annex part of the generic @samp{qXfer} packet must be empty
37709 (@pxref{qXfer read}).
37711 Targets which maintain a list of libraries in the program's memory do
37712 not need to implement this packet; it is designed for platforms where
37713 the operating system manages the list of loaded libraries.
37715 This packet is not probed by default; the remote stub must request it,
37716 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
37718 @item qXfer:libraries-svr4:read:@var{annex}:@var{offset},@var{length}
37719 @anchor{qXfer svr4 library list read}
37720 Access the target's list of loaded libraries when the target is an SVR4
37721 platform. @xref{Library List Format for SVR4 Targets}. The annex part
37722 of the generic @samp{qXfer} packet must be empty unless the remote
37723 stub indicated it supports the augmented form of this packet
37724 by supplying an appropriate @samp{qSupported} response
37725 (@pxref{qXfer read}, @ref{qSupported}).
37727 This packet is optional for better performance on SVR4 targets.
37728 @value{GDBN} uses memory read packets to read the SVR4 library list otherwise.
37730 This packet is not probed by default; the remote stub must request it,
37731 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
37733 If the remote stub indicates it supports the augmented form of this
37734 packet then the annex part of the generic @samp{qXfer} packet may
37735 contain a semicolon-separated list of @samp{@var{name}=@var{value}}
37736 arguments. The currently supported arguments are:
37739 @item start=@var{address}
37740 A hexadecimal number specifying the address of the @samp{struct
37741 link_map} to start reading the library list from. If unset or zero
37742 then the first @samp{struct link_map} in the library list will be
37743 chosen as the starting point.
37745 @item prev=@var{address}
37746 A hexadecimal number specifying the address of the @samp{struct
37747 link_map} immediately preceding the @samp{struct link_map}
37748 specified by the @samp{start} argument. If unset or zero then
37749 the remote stub will expect that no @samp{struct link_map}
37750 exists prior to the starting point.
37754 Arguments that are not understood by the remote stub will be silently
37757 @item qXfer:memory-map:read::@var{offset},@var{length}
37758 @anchor{qXfer memory map read}
37759 Access the target's @dfn{memory-map}. @xref{Memory Map Format}. The
37760 annex part of the generic @samp{qXfer} packet must be empty
37761 (@pxref{qXfer read}).
37763 This packet is not probed by default; the remote stub must request it,
37764 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
37766 @item qXfer:sdata:read::@var{offset},@var{length}
37767 @anchor{qXfer sdata read}
37769 Read contents of the extra collected static tracepoint marker
37770 information. The annex part of the generic @samp{qXfer} packet must
37771 be empty (@pxref{qXfer read}). @xref{Tracepoint Actions,,Tracepoint
37774 This packet is not probed by default; the remote stub must request it,
37775 by supplying an appropriate @samp{qSupported} response
37776 (@pxref{qSupported}).
37778 @item qXfer:siginfo:read::@var{offset},@var{length}
37779 @anchor{qXfer siginfo read}
37780 Read contents of the extra signal information on the target
37781 system. The annex part of the generic @samp{qXfer} packet must be
37782 empty (@pxref{qXfer read}).
37784 This packet is not probed by default; the remote stub must request it,
37785 by supplying an appropriate @samp{qSupported} response
37786 (@pxref{qSupported}).
37788 @item qXfer:spu:read:@var{annex}:@var{offset},@var{length}
37789 @anchor{qXfer spu read}
37790 Read contents of an @code{spufs} file on the target system. The
37791 annex specifies which file to read; it must be of the form
37792 @file{@var{id}/@var{name}}, where @var{id} specifies an SPU context ID
37793 in the target process, and @var{name} identifes the @code{spufs} file
37794 in that context to be accessed.
37796 This packet is not probed by default; the remote stub must request it,
37797 by supplying an appropriate @samp{qSupported} response
37798 (@pxref{qSupported}).
37800 @item qXfer:threads:read::@var{offset},@var{length}
37801 @anchor{qXfer threads read}
37802 Access the list of threads on target. @xref{Thread List Format}. The
37803 annex part of the generic @samp{qXfer} packet must be empty
37804 (@pxref{qXfer read}).
37806 This packet is not probed by default; the remote stub must request it,
37807 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
37809 @item qXfer:traceframe-info:read::@var{offset},@var{length}
37810 @anchor{qXfer traceframe info read}
37812 Return a description of the current traceframe's contents.
37813 @xref{Traceframe Info Format}. The annex part of the generic
37814 @samp{qXfer} packet must be empty (@pxref{qXfer read}).
37816 This packet is not probed by default; the remote stub must request it,
37817 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
37819 @item qXfer:uib:read:@var{pc}:@var{offset},@var{length}
37820 @anchor{qXfer unwind info block}
37822 Return the unwind information block for @var{pc}. This packet is used
37823 on OpenVMS/ia64 to ask the kernel unwind information.
37825 This packet is not probed by default.
37827 @item qXfer:fdpic:read:@var{annex}:@var{offset},@var{length}
37828 @anchor{qXfer fdpic loadmap read}
37829 Read contents of @code{loadmap}s on the target system. The
37830 annex, either @samp{exec} or @samp{interp}, specifies which @code{loadmap},
37831 executable @code{loadmap} or interpreter @code{loadmap} to read.
37833 This packet is not probed by default; the remote stub must request it,
37834 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
37836 @item qXfer:osdata:read::@var{offset},@var{length}
37837 @anchor{qXfer osdata read}
37838 Access the target's @dfn{operating system information}.
37839 @xref{Operating System Information}.
37843 @item qXfer:@var{object}:write:@var{annex}:@var{offset}:@var{data}@dots{}
37844 @cindex write data into object, remote request
37845 @anchor{qXfer write}
37846 Write uninterpreted bytes into the target's special data area
37847 identified by the keyword @var{object}, starting at @var{offset} bytes
37848 into the data. The binary-encoded data (@pxref{Binary Data}) to be
37849 written is given by @var{data}@dots{}. The content and encoding of @var{annex}
37850 is specific to @var{object}; it can supply additional details about what data
37856 @var{nn} (hex encoded) is the number of bytes written.
37857 This may be fewer bytes than supplied in the request.
37860 The request was malformed, or @var{annex} was invalid.
37863 The offset was invalid, or there was an error encountered writing the data.
37864 The @var{nn} part is a hex-encoded @code{errno} value.
37867 An empty reply indicates the @var{object} string was not
37868 recognized by the stub, or that the object does not support writing.
37871 Here are the specific requests of this form defined so far. All the
37872 @samp{qXfer:@var{object}:write:@dots{}} requests use the same reply
37873 formats, listed above.
37876 @item qXfer:siginfo:write::@var{offset}:@var{data}@dots{}
37877 @anchor{qXfer siginfo write}
37878 Write @var{data} to the extra signal information on the target system.
37879 The annex part of the generic @samp{qXfer} packet must be
37880 empty (@pxref{qXfer write}).
37882 This packet is not probed by default; the remote stub must request it,
37883 by supplying an appropriate @samp{qSupported} response
37884 (@pxref{qSupported}).
37886 @item qXfer:spu:write:@var{annex}:@var{offset}:@var{data}@dots{}
37887 @anchor{qXfer spu write}
37888 Write @var{data} to an @code{spufs} file on the target system. The
37889 annex specifies which file to write; it must be of the form
37890 @file{@var{id}/@var{name}}, where @var{id} specifies an SPU context ID
37891 in the target process, and @var{name} identifes the @code{spufs} file
37892 in that context to be accessed.
37894 This packet is not probed by default; the remote stub must request it,
37895 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
37898 @item qXfer:@var{object}:@var{operation}:@dots{}
37899 Requests of this form may be added in the future. When a stub does
37900 not recognize the @var{object} keyword, or its support for
37901 @var{object} does not recognize the @var{operation} keyword, the stub
37902 must respond with an empty packet.
37904 @item qAttached:@var{pid}
37905 @cindex query attached, remote request
37906 @cindex @samp{qAttached} packet
37907 Return an indication of whether the remote server attached to an
37908 existing process or created a new process. When the multiprocess
37909 protocol extensions are supported (@pxref{multiprocess extensions}),
37910 @var{pid} is an integer in hexadecimal format identifying the target
37911 process. Otherwise, @value{GDBN} will omit the @var{pid} field and
37912 the query packet will be simplified as @samp{qAttached}.
37914 This query is used, for example, to know whether the remote process
37915 should be detached or killed when a @value{GDBN} session is ended with
37916 the @code{quit} command.
37921 The remote server attached to an existing process.
37923 The remote server created a new process.
37925 A badly formed request or an error was encountered.
37929 Enable branch tracing for the current thread using Branch Trace Store.
37934 Branch tracing has been enabled.
37936 A badly formed request or an error was encountered.
37940 Enable branch tracing for the current thread using Intel Processor Trace.
37945 Branch tracing has been enabled.
37947 A badly formed request or an error was encountered.
37951 Disable branch tracing for the current thread.
37956 Branch tracing has been disabled.
37958 A badly formed request or an error was encountered.
37961 @item Qbtrace-conf:bts:size=@var{value}
37962 Set the requested ring buffer size for new threads that use the
37963 btrace recording method in bts format.
37968 The ring buffer size has been set.
37970 A badly formed request or an error was encountered.
37973 @item Qbtrace-conf:pt:size=@var{value}
37974 Set the requested ring buffer size for new threads that use the
37975 btrace recording method in pt format.
37980 The ring buffer size has been set.
37982 A badly formed request or an error was encountered.
37987 @node Architecture-Specific Protocol Details
37988 @section Architecture-Specific Protocol Details
37990 This section describes how the remote protocol is applied to specific
37991 target architectures. Also see @ref{Standard Target Features}, for
37992 details of XML target descriptions for each architecture.
37995 * ARM-Specific Protocol Details::
37996 * MIPS-Specific Protocol Details::
37999 @node ARM-Specific Protocol Details
38000 @subsection @acronym{ARM}-specific Protocol Details
38003 * ARM Breakpoint Kinds::
38006 @node ARM Breakpoint Kinds
38007 @subsubsection @acronym{ARM} Breakpoint Kinds
38008 @cindex breakpoint kinds, @acronym{ARM}
38010 These breakpoint kinds are defined for the @samp{Z0} and @samp{Z1} packets.
38015 16-bit Thumb mode breakpoint.
38018 32-bit Thumb mode (Thumb-2) breakpoint.
38021 32-bit @acronym{ARM} mode breakpoint.
38025 @node MIPS-Specific Protocol Details
38026 @subsection @acronym{MIPS}-specific Protocol Details
38029 * MIPS Register packet Format::
38030 * MIPS Breakpoint Kinds::
38033 @node MIPS Register packet Format
38034 @subsubsection @acronym{MIPS} Register Packet Format
38035 @cindex register packet format, @acronym{MIPS}
38037 The following @code{g}/@code{G} packets have previously been defined.
38038 In the below, some thirty-two bit registers are transferred as
38039 sixty-four bits. Those registers should be zero/sign extended (which?)
38040 to fill the space allocated. Register bytes are transferred in target
38041 byte order. The two nibbles within a register byte are transferred
38042 most-significant -- least-significant.
38047 All registers are transferred as thirty-two bit quantities in the order:
38048 32 general-purpose; sr; lo; hi; bad; cause; pc; 32 floating-point
38049 registers; fsr; fir; fp.
38052 All registers are transferred as sixty-four bit quantities (including
38053 thirty-two bit registers such as @code{sr}). The ordering is the same
38058 @node MIPS Breakpoint Kinds
38059 @subsubsection @acronym{MIPS} Breakpoint Kinds
38060 @cindex breakpoint kinds, @acronym{MIPS}
38062 These breakpoint kinds are defined for the @samp{Z0} and @samp{Z1} packets.
38067 16-bit @acronym{MIPS16} mode breakpoint.
38070 16-bit @acronym{microMIPS} mode breakpoint.
38073 32-bit standard @acronym{MIPS} mode breakpoint.
38076 32-bit @acronym{microMIPS} mode breakpoint.
38080 @node Tracepoint Packets
38081 @section Tracepoint Packets
38082 @cindex tracepoint packets
38083 @cindex packets, tracepoint
38085 Here we describe the packets @value{GDBN} uses to implement
38086 tracepoints (@pxref{Tracepoints}).
38090 @item QTDP:@var{n}:@var{addr}:@var{ena}:@var{step}:@var{pass}[:F@var{flen}][:X@var{len},@var{bytes}]@r{[}-@r{]}
38091 @cindex @samp{QTDP} packet
38092 Create a new tracepoint, number @var{n}, at @var{addr}. If @var{ena}
38093 is @samp{E}, then the tracepoint is enabled; if it is @samp{D}, then
38094 the tracepoint is disabled. The @var{step} gives the tracepoint's step
38095 count, and @var{pass} gives its pass count. If an @samp{F} is present,
38096 then the tracepoint is to be a fast tracepoint, and the @var{flen} is
38097 the number of bytes that the target should copy elsewhere to make room
38098 for the tracepoint. If an @samp{X} is present, it introduces a
38099 tracepoint condition, which consists of a hexadecimal length, followed
38100 by a comma and hex-encoded bytes, in a manner similar to action
38101 encodings as described below. If the trailing @samp{-} is present,
38102 further @samp{QTDP} packets will follow to specify this tracepoint's
38108 The packet was understood and carried out.
38110 @xref{Tracepoint Packets,,Relocate instruction reply packet}.
38112 The packet was not recognized.
38115 @item QTDP:-@var{n}:@var{addr}:@r{[}S@r{]}@var{action}@dots{}@r{[}-@r{]}
38116 Define actions to be taken when a tracepoint is hit. The @var{n} and
38117 @var{addr} must be the same as in the initial @samp{QTDP} packet for
38118 this tracepoint. This packet may only be sent immediately after
38119 another @samp{QTDP} packet that ended with a @samp{-}. If the
38120 trailing @samp{-} is present, further @samp{QTDP} packets will follow,
38121 specifying more actions for this tracepoint.
38123 In the series of action packets for a given tracepoint, at most one
38124 can have an @samp{S} before its first @var{action}. If such a packet
38125 is sent, it and the following packets define ``while-stepping''
38126 actions. Any prior packets define ordinary actions --- that is, those
38127 taken when the tracepoint is first hit. If no action packet has an
38128 @samp{S}, then all the packets in the series specify ordinary
38129 tracepoint actions.
38131 The @samp{@var{action}@dots{}} portion of the packet is a series of
38132 actions, concatenated without separators. Each action has one of the
38138 Collect the registers whose bits are set in @var{mask},
38139 a hexadecimal number whose @var{i}'th bit is set if register number
38140 @var{i} should be collected. (The least significant bit is numbered
38141 zero.) Note that @var{mask} may be any number of digits long; it may
38142 not fit in a 32-bit word.
38144 @item M @var{basereg},@var{offset},@var{len}
38145 Collect @var{len} bytes of memory starting at the address in register
38146 number @var{basereg}, plus @var{offset}. If @var{basereg} is
38147 @samp{-1}, then the range has a fixed address: @var{offset} is the
38148 address of the lowest byte to collect. The @var{basereg},
38149 @var{offset}, and @var{len} parameters are all unsigned hexadecimal
38150 values (the @samp{-1} value for @var{basereg} is a special case).
38152 @item X @var{len},@var{expr}
38153 Evaluate @var{expr}, whose length is @var{len}, and collect memory as
38154 it directs. The agent expression @var{expr} is as described in
38155 @ref{Agent Expressions}. Each byte of the expression is encoded as a
38156 two-digit hex number in the packet; @var{len} is the number of bytes
38157 in the expression (and thus one-half the number of hex digits in the
38162 Any number of actions may be packed together in a single @samp{QTDP}
38163 packet, as long as the packet does not exceed the maximum packet
38164 length (400 bytes, for many stubs). There may be only one @samp{R}
38165 action per tracepoint, and it must precede any @samp{M} or @samp{X}
38166 actions. Any registers referred to by @samp{M} and @samp{X} actions
38167 must be collected by a preceding @samp{R} action. (The
38168 ``while-stepping'' actions are treated as if they were attached to a
38169 separate tracepoint, as far as these restrictions are concerned.)
38174 The packet was understood and carried out.
38176 @xref{Tracepoint Packets,,Relocate instruction reply packet}.
38178 The packet was not recognized.
38181 @item QTDPsrc:@var{n}:@var{addr}:@var{type}:@var{start}:@var{slen}:@var{bytes}
38182 @cindex @samp{QTDPsrc} packet
38183 Specify a source string of tracepoint @var{n} at address @var{addr}.
38184 This is useful to get accurate reproduction of the tracepoints
38185 originally downloaded at the beginning of the trace run. The @var{type}
38186 is the name of the tracepoint part, such as @samp{cond} for the
38187 tracepoint's conditional expression (see below for a list of types), while
38188 @var{bytes} is the string, encoded in hexadecimal.
38190 @var{start} is the offset of the @var{bytes} within the overall source
38191 string, while @var{slen} is the total length of the source string.
38192 This is intended for handling source strings that are longer than will
38193 fit in a single packet.
38194 @c Add detailed example when this info is moved into a dedicated
38195 @c tracepoint descriptions section.
38197 The available string types are @samp{at} for the location,
38198 @samp{cond} for the conditional, and @samp{cmd} for an action command.
38199 @value{GDBN} sends a separate packet for each command in the action
38200 list, in the same order in which the commands are stored in the list.
38202 The target does not need to do anything with source strings except
38203 report them back as part of the replies to the @samp{qTfP}/@samp{qTsP}
38206 Although this packet is optional, and @value{GDBN} will only send it
38207 if the target replies with @samp{TracepointSource} @xref{General
38208 Query Packets}, it makes both disconnected tracing and trace files
38209 much easier to use. Otherwise the user must be careful that the
38210 tracepoints in effect while looking at trace frames are identical to
38211 the ones in effect during the trace run; even a small discrepancy
38212 could cause @samp{tdump} not to work, or a particular trace frame not
38215 @item QTDV:@var{n}:@var{value}:@var{builtin}:@var{name}
38216 @cindex define trace state variable, remote request
38217 @cindex @samp{QTDV} packet
38218 Create a new trace state variable, number @var{n}, with an initial
38219 value of @var{value}, which is a 64-bit signed integer. Both @var{n}
38220 and @var{value} are encoded as hexadecimal values. @value{GDBN} has
38221 the option of not using this packet for initial values of zero; the
38222 target should simply create the trace state variables as they are
38223 mentioned in expressions. The value @var{builtin} should be 1 (one)
38224 if the trace state variable is builtin and 0 (zero) if it is not builtin.
38225 @value{GDBN} only sets @var{builtin} to 1 if a previous @samp{qTfV} or
38226 @samp{qTsV} packet had it set. The contents of @var{name} is the
38227 hex-encoded name (without the leading @samp{$}) of the trace state
38230 @item QTFrame:@var{n}
38231 @cindex @samp{QTFrame} packet
38232 Select the @var{n}'th tracepoint frame from the buffer, and use the
38233 register and memory contents recorded there to answer subsequent
38234 request packets from @value{GDBN}.
38236 A successful reply from the stub indicates that the stub has found the
38237 requested frame. The response is a series of parts, concatenated
38238 without separators, describing the frame we selected. Each part has
38239 one of the following forms:
38243 The selected frame is number @var{n} in the trace frame buffer;
38244 @var{f} is a hexadecimal number. If @var{f} is @samp{-1}, then there
38245 was no frame matching the criteria in the request packet.
38248 The selected trace frame records a hit of tracepoint number @var{t};
38249 @var{t} is a hexadecimal number.
38253 @item QTFrame:pc:@var{addr}
38254 Like @samp{QTFrame:@var{n}}, but select the first tracepoint frame after the
38255 currently selected frame whose PC is @var{addr};
38256 @var{addr} is a hexadecimal number.
38258 @item QTFrame:tdp:@var{t}
38259 Like @samp{QTFrame:@var{n}}, but select the first tracepoint frame after the
38260 currently selected frame that is a hit of tracepoint @var{t}; @var{t}
38261 is a hexadecimal number.
38263 @item QTFrame:range:@var{start}:@var{end}
38264 Like @samp{QTFrame:@var{n}}, but select the first tracepoint frame after the
38265 currently selected frame whose PC is between @var{start} (inclusive)
38266 and @var{end} (inclusive); @var{start} and @var{end} are hexadecimal
38269 @item QTFrame:outside:@var{start}:@var{end}
38270 Like @samp{QTFrame:range:@var{start}:@var{end}}, but select the first
38271 frame @emph{outside} the given range of addresses (exclusive).
38274 @cindex @samp{qTMinFTPILen} packet
38275 This packet requests the minimum length of instruction at which a fast
38276 tracepoint (@pxref{Set Tracepoints}) may be placed. For instance, on
38277 the 32-bit x86 architecture, it is possible to use a 4-byte jump, but
38278 it depends on the target system being able to create trampolines in
38279 the first 64K of memory, which might or might not be possible for that
38280 system. So the reply to this packet will be 4 if it is able to
38287 The minimum instruction length is currently unknown.
38289 The minimum instruction length is @var{length}, where @var{length}
38290 is a hexadecimal number greater or equal to 1. A reply
38291 of 1 means that a fast tracepoint may be placed on any instruction
38292 regardless of size.
38294 An error has occurred.
38296 An empty reply indicates that the request is not supported by the stub.
38300 @cindex @samp{QTStart} packet
38301 Begin the tracepoint experiment. Begin collecting data from
38302 tracepoint hits in the trace frame buffer. This packet supports the
38303 @samp{qRelocInsn} reply (@pxref{Tracepoint Packets,,Relocate
38304 instruction reply packet}).
38307 @cindex @samp{QTStop} packet
38308 End the tracepoint experiment. Stop collecting trace frames.
38310 @item QTEnable:@var{n}:@var{addr}
38312 @cindex @samp{QTEnable} packet
38313 Enable tracepoint @var{n} at address @var{addr} in a started tracepoint
38314 experiment. If the tracepoint was previously disabled, then collection
38315 of data from it will resume.
38317 @item QTDisable:@var{n}:@var{addr}
38319 @cindex @samp{QTDisable} packet
38320 Disable tracepoint @var{n} at address @var{addr} in a started tracepoint
38321 experiment. No more data will be collected from the tracepoint unless
38322 @samp{QTEnable:@var{n}:@var{addr}} is subsequently issued.
38325 @cindex @samp{QTinit} packet
38326 Clear the table of tracepoints, and empty the trace frame buffer.
38328 @item QTro:@var{start1},@var{end1}:@var{start2},@var{end2}:@dots{}
38329 @cindex @samp{QTro} packet
38330 Establish the given ranges of memory as ``transparent''. The stub
38331 will answer requests for these ranges from memory's current contents,
38332 if they were not collected as part of the tracepoint hit.
38334 @value{GDBN} uses this to mark read-only regions of memory, like those
38335 containing program code. Since these areas never change, they should
38336 still have the same contents they did when the tracepoint was hit, so
38337 there's no reason for the stub to refuse to provide their contents.
38339 @item QTDisconnected:@var{value}
38340 @cindex @samp{QTDisconnected} packet
38341 Set the choice to what to do with the tracing run when @value{GDBN}
38342 disconnects from the target. A @var{value} of 1 directs the target to
38343 continue the tracing run, while 0 tells the target to stop tracing if
38344 @value{GDBN} is no longer in the picture.
38347 @cindex @samp{qTStatus} packet
38348 Ask the stub if there is a trace experiment running right now.
38350 The reply has the form:
38354 @item T@var{running}@r{[};@var{field}@r{]}@dots{}
38355 @var{running} is a single digit @code{1} if the trace is presently
38356 running, or @code{0} if not. It is followed by semicolon-separated
38357 optional fields that an agent may use to report additional status.
38361 If the trace is not running, the agent may report any of several
38362 explanations as one of the optional fields:
38367 No trace has been run yet.
38369 @item tstop[:@var{text}]:0
38370 The trace was stopped by a user-originated stop command. The optional
38371 @var{text} field is a user-supplied string supplied as part of the
38372 stop command (for instance, an explanation of why the trace was
38373 stopped manually). It is hex-encoded.
38376 The trace stopped because the trace buffer filled up.
38378 @item tdisconnected:0
38379 The trace stopped because @value{GDBN} disconnected from the target.
38381 @item tpasscount:@var{tpnum}
38382 The trace stopped because tracepoint @var{tpnum} exceeded its pass count.
38384 @item terror:@var{text}:@var{tpnum}
38385 The trace stopped because tracepoint @var{tpnum} had an error. The
38386 string @var{text} is available to describe the nature of the error
38387 (for instance, a divide by zero in the condition expression); it
38391 The trace stopped for some other reason.
38395 Additional optional fields supply statistical and other information.
38396 Although not required, they are extremely useful for users monitoring
38397 the progress of a trace run. If a trace has stopped, and these
38398 numbers are reported, they must reflect the state of the just-stopped
38403 @item tframes:@var{n}
38404 The number of trace frames in the buffer.
38406 @item tcreated:@var{n}
38407 The total number of trace frames created during the run. This may
38408 be larger than the trace frame count, if the buffer is circular.
38410 @item tsize:@var{n}
38411 The total size of the trace buffer, in bytes.
38413 @item tfree:@var{n}
38414 The number of bytes still unused in the buffer.
38416 @item circular:@var{n}
38417 The value of the circular trace buffer flag. @code{1} means that the
38418 trace buffer is circular and old trace frames will be discarded if
38419 necessary to make room, @code{0} means that the trace buffer is linear
38422 @item disconn:@var{n}
38423 The value of the disconnected tracing flag. @code{1} means that
38424 tracing will continue after @value{GDBN} disconnects, @code{0} means
38425 that the trace run will stop.
38429 @item qTP:@var{tp}:@var{addr}
38430 @cindex tracepoint status, remote request
38431 @cindex @samp{qTP} packet
38432 Ask the stub for the current state of tracepoint number @var{tp} at
38433 address @var{addr}.
38437 @item V@var{hits}:@var{usage}
38438 The tracepoint has been hit @var{hits} times so far during the trace
38439 run, and accounts for @var{usage} in the trace buffer. Note that
38440 @code{while-stepping} steps are not counted as separate hits, but the
38441 steps' space consumption is added into the usage number.
38445 @item qTV:@var{var}
38446 @cindex trace state variable value, remote request
38447 @cindex @samp{qTV} packet
38448 Ask the stub for the value of the trace state variable number @var{var}.
38453 The value of the variable is @var{value}. This will be the current
38454 value of the variable if the user is examining a running target, or a
38455 saved value if the variable was collected in the trace frame that the
38456 user is looking at. Note that multiple requests may result in
38457 different reply values, such as when requesting values while the
38458 program is running.
38461 The value of the variable is unknown. This would occur, for example,
38462 if the user is examining a trace frame in which the requested variable
38467 @cindex @samp{qTfP} packet
38469 @cindex @samp{qTsP} packet
38470 These packets request data about tracepoints that are being used by
38471 the target. @value{GDBN} sends @code{qTfP} to get the first piece
38472 of data, and multiple @code{qTsP} to get additional pieces. Replies
38473 to these packets generally take the form of the @code{QTDP} packets
38474 that define tracepoints. (FIXME add detailed syntax)
38477 @cindex @samp{qTfV} packet
38479 @cindex @samp{qTsV} packet
38480 These packets request data about trace state variables that are on the
38481 target. @value{GDBN} sends @code{qTfV} to get the first vari of data,
38482 and multiple @code{qTsV} to get additional variables. Replies to
38483 these packets follow the syntax of the @code{QTDV} packets that define
38484 trace state variables.
38490 @cindex @samp{qTfSTM} packet
38491 @cindex @samp{qTsSTM} packet
38492 These packets request data about static tracepoint markers that exist
38493 in the target program. @value{GDBN} sends @code{qTfSTM} to get the
38494 first piece of data, and multiple @code{qTsSTM} to get additional
38495 pieces. Replies to these packets take the following form:
38499 @item m @var{address}:@var{id}:@var{extra}
38501 @item m @var{address}:@var{id}:@var{extra},@var{address}:@var{id}:@var{extra}@dots{}
38502 a comma-separated list of markers
38504 (lower case letter @samp{L}) denotes end of list.
38506 An error occurred. The error number @var{nn} is given as hex digits.
38508 An empty reply indicates that the request is not supported by the
38512 The @var{address} is encoded in hex;
38513 @var{id} and @var{extra} are strings encoded in hex.
38515 In response to each query, the target will reply with a list of one or
38516 more markers, separated by commas. @value{GDBN} will respond to each
38517 reply with a request for more markers (using the @samp{qs} form of the
38518 query), until the target responds with @samp{l} (lower-case ell, for
38521 @item qTSTMat:@var{address}
38523 @cindex @samp{qTSTMat} packet
38524 This packets requests data about static tracepoint markers in the
38525 target program at @var{address}. Replies to this packet follow the
38526 syntax of the @samp{qTfSTM} and @code{qTsSTM} packets that list static
38527 tracepoint markers.
38529 @item QTSave:@var{filename}
38530 @cindex @samp{QTSave} packet
38531 This packet directs the target to save trace data to the file name
38532 @var{filename} in the target's filesystem. The @var{filename} is encoded
38533 as a hex string; the interpretation of the file name (relative vs
38534 absolute, wild cards, etc) is up to the target.
38536 @item qTBuffer:@var{offset},@var{len}
38537 @cindex @samp{qTBuffer} packet
38538 Return up to @var{len} bytes of the current contents of trace buffer,
38539 starting at @var{offset}. The trace buffer is treated as if it were
38540 a contiguous collection of traceframes, as per the trace file format.
38541 The reply consists as many hex-encoded bytes as the target can deliver
38542 in a packet; it is not an error to return fewer than were asked for.
38543 A reply consisting of just @code{l} indicates that no bytes are
38546 @item QTBuffer:circular:@var{value}
38547 This packet directs the target to use a circular trace buffer if
38548 @var{value} is 1, or a linear buffer if the value is 0.
38550 @item QTBuffer:size:@var{size}
38551 @anchor{QTBuffer-size}
38552 @cindex @samp{QTBuffer size} packet
38553 This packet directs the target to make the trace buffer be of size
38554 @var{size} if possible. A value of @code{-1} tells the target to
38555 use whatever size it prefers.
38557 @item QTNotes:@r{[}@var{type}:@var{text}@r{]}@r{[};@var{type}:@var{text}@r{]}@dots{}
38558 @cindex @samp{QTNotes} packet
38559 This packet adds optional textual notes to the trace run. Allowable
38560 types include @code{user}, @code{notes}, and @code{tstop}, the
38561 @var{text} fields are arbitrary strings, hex-encoded.
38565 @subsection Relocate instruction reply packet
38566 When installing fast tracepoints in memory, the target may need to
38567 relocate the instruction currently at the tracepoint address to a
38568 different address in memory. For most instructions, a simple copy is
38569 enough, but, for example, call instructions that implicitly push the
38570 return address on the stack, and relative branches or other
38571 PC-relative instructions require offset adjustment, so that the effect
38572 of executing the instruction at a different address is the same as if
38573 it had executed in the original location.
38575 In response to several of the tracepoint packets, the target may also
38576 respond with a number of intermediate @samp{qRelocInsn} request
38577 packets before the final result packet, to have @value{GDBN} handle
38578 this relocation operation. If a packet supports this mechanism, its
38579 documentation will explicitly say so. See for example the above
38580 descriptions for the @samp{QTStart} and @samp{QTDP} packets. The
38581 format of the request is:
38584 @item qRelocInsn:@var{from};@var{to}
38586 This requests @value{GDBN} to copy instruction at address @var{from}
38587 to address @var{to}, possibly adjusted so that executing the
38588 instruction at @var{to} has the same effect as executing it at
38589 @var{from}. @value{GDBN} writes the adjusted instruction to target
38590 memory starting at @var{to}.
38595 @item qRelocInsn:@var{adjusted_size}
38596 Informs the stub the relocation is complete. The @var{adjusted_size} is
38597 the length in bytes of resulting relocated instruction sequence.
38599 A badly formed request was detected, or an error was encountered while
38600 relocating the instruction.
38603 @node Host I/O Packets
38604 @section Host I/O Packets
38605 @cindex Host I/O, remote protocol
38606 @cindex file transfer, remote protocol
38608 The @dfn{Host I/O} packets allow @value{GDBN} to perform I/O
38609 operations on the far side of a remote link. For example, Host I/O is
38610 used to upload and download files to a remote target with its own
38611 filesystem. Host I/O uses the same constant values and data structure
38612 layout as the target-initiated File-I/O protocol. However, the
38613 Host I/O packets are structured differently. The target-initiated
38614 protocol relies on target memory to store parameters and buffers.
38615 Host I/O requests are initiated by @value{GDBN}, and the
38616 target's memory is not involved. @xref{File-I/O Remote Protocol
38617 Extension}, for more details on the target-initiated protocol.
38619 The Host I/O request packets all encode a single operation along with
38620 its arguments. They have this format:
38624 @item vFile:@var{operation}: @var{parameter}@dots{}
38625 @var{operation} is the name of the particular request; the target
38626 should compare the entire packet name up to the second colon when checking
38627 for a supported operation. The format of @var{parameter} depends on
38628 the operation. Numbers are always passed in hexadecimal. Negative
38629 numbers have an explicit minus sign (i.e.@: two's complement is not
38630 used). Strings (e.g.@: filenames) are encoded as a series of
38631 hexadecimal bytes. The last argument to a system call may be a
38632 buffer of escaped binary data (@pxref{Binary Data}).
38636 The valid responses to Host I/O packets are:
38640 @item F @var{result} [, @var{errno}] [; @var{attachment}]
38641 @var{result} is the integer value returned by this operation, usually
38642 non-negative for success and -1 for errors. If an error has occured,
38643 @var{errno} will be included in the result specifying a
38644 value defined by the File-I/O protocol (@pxref{Errno Values}). For
38645 operations which return data, @var{attachment} supplies the data as a
38646 binary buffer. Binary buffers in response packets are escaped in the
38647 normal way (@pxref{Binary Data}). See the individual packet
38648 documentation for the interpretation of @var{result} and
38652 An empty response indicates that this operation is not recognized.
38656 These are the supported Host I/O operations:
38659 @item vFile:open: @var{filename}, @var{flags}, @var{mode}
38660 Open a file at @var{filename} and return a file descriptor for it, or
38661 return -1 if an error occurs. The @var{filename} is a string,
38662 @var{flags} is an integer indicating a mask of open flags
38663 (@pxref{Open Flags}), and @var{mode} is an integer indicating a mask
38664 of mode bits to use if the file is created (@pxref{mode_t Values}).
38665 @xref{open}, for details of the open flags and mode values.
38667 @item vFile:close: @var{fd}
38668 Close the open file corresponding to @var{fd} and return 0, or
38669 -1 if an error occurs.
38671 @item vFile:pread: @var{fd}, @var{count}, @var{offset}
38672 Read data from the open file corresponding to @var{fd}. Up to
38673 @var{count} bytes will be read from the file, starting at @var{offset}
38674 relative to the start of the file. The target may read fewer bytes;
38675 common reasons include packet size limits and an end-of-file
38676 condition. The number of bytes read is returned. Zero should only be
38677 returned for a successful read at the end of the file, or if
38678 @var{count} was zero.
38680 The data read should be returned as a binary attachment on success.
38681 If zero bytes were read, the response should include an empty binary
38682 attachment (i.e.@: a trailing semicolon). The return value is the
38683 number of target bytes read; the binary attachment may be longer if
38684 some characters were escaped.
38686 @item vFile:pwrite: @var{fd}, @var{offset}, @var{data}
38687 Write @var{data} (a binary buffer) to the open file corresponding
38688 to @var{fd}. Start the write at @var{offset} from the start of the
38689 file. Unlike many @code{write} system calls, there is no
38690 separate @var{count} argument; the length of @var{data} in the
38691 packet is used. @samp{vFile:write} returns the number of bytes written,
38692 which may be shorter than the length of @var{data}, or -1 if an
38695 @item vFile:fstat: @var{fd}
38696 Get information about the open file corresponding to @var{fd}.
38697 On success the information is returned as a binary attachment
38698 and the return value is the size of this attachment in bytes.
38699 If an error occurs the return value is -1. The format of the
38700 returned binary attachment is as described in @ref{struct stat}.
38702 @item vFile:unlink: @var{filename}
38703 Delete the file at @var{filename} on the target. Return 0,
38704 or -1 if an error occurs. The @var{filename} is a string.
38706 @item vFile:readlink: @var{filename}
38707 Read value of symbolic link @var{filename} on the target. Return
38708 the number of bytes read, or -1 if an error occurs.
38710 The data read should be returned as a binary attachment on success.
38711 If zero bytes were read, the response should include an empty binary
38712 attachment (i.e.@: a trailing semicolon). The return value is the
38713 number of target bytes read; the binary attachment may be longer if
38714 some characters were escaped.
38716 @item vFile:setfs: @var{pid}
38717 Select the filesystem on which @code{vFile} operations with
38718 @var{filename} arguments will operate. This is required for
38719 @value{GDBN} to be able to access files on remote targets where
38720 the remote stub does not share a common filesystem with the
38723 If @var{pid} is nonzero, select the filesystem as seen by process
38724 @var{pid}. If @var{pid} is zero, select the filesystem as seen by
38725 the remote stub. Return 0 on success, or -1 if an error occurs.
38726 If @code{vFile:setfs:} indicates success, the selected filesystem
38727 remains selected until the next successful @code{vFile:setfs:}
38733 @section Interrupts
38734 @cindex interrupts (remote protocol)
38735 @anchor{interrupting remote targets}
38737 In all-stop mode, when a program on the remote target is running,
38738 @value{GDBN} may attempt to interrupt it by sending a @samp{Ctrl-C},
38739 @code{BREAK} or a @code{BREAK} followed by @code{g}, control of which
38740 is specified via @value{GDBN}'s @samp{interrupt-sequence}.
38742 The precise meaning of @code{BREAK} is defined by the transport
38743 mechanism and may, in fact, be undefined. @value{GDBN} does not
38744 currently define a @code{BREAK} mechanism for any of the network
38745 interfaces except for TCP, in which case @value{GDBN} sends the
38746 @code{telnet} BREAK sequence.
38748 @samp{Ctrl-C}, on the other hand, is defined and implemented for all
38749 transport mechanisms. It is represented by sending the single byte
38750 @code{0x03} without any of the usual packet overhead described in
38751 the Overview section (@pxref{Overview}). When a @code{0x03} byte is
38752 transmitted as part of a packet, it is considered to be packet data
38753 and does @emph{not} represent an interrupt. E.g., an @samp{X} packet
38754 (@pxref{X packet}), used for binary downloads, may include an unescaped
38755 @code{0x03} as part of its packet.
38757 @code{BREAK} followed by @code{g} is also known as Magic SysRq g.
38758 When Linux kernel receives this sequence from serial port,
38759 it stops execution and connects to gdb.
38761 In non-stop mode, because packet resumptions are asynchronous
38762 (@pxref{vCont packet}), @value{GDBN} is always free to send a remote
38763 command to the remote stub, even when the target is running. For that
38764 reason, @value{GDBN} instead sends a regular packet (@pxref{vCtrlC
38765 packet}) with the usual packet framing instead of the single byte
38768 Stubs are not required to recognize these interrupt mechanisms and the
38769 precise meaning associated with receipt of the interrupt is
38770 implementation defined. If the target supports debugging of multiple
38771 threads and/or processes, it should attempt to interrupt all
38772 currently-executing threads and processes.
38773 If the stub is successful at interrupting the
38774 running program, it should send one of the stop
38775 reply packets (@pxref{Stop Reply Packets}) to @value{GDBN} as a result
38776 of successfully stopping the program in all-stop mode, and a stop reply
38777 for each stopped thread in non-stop mode.
38778 Interrupts received while the
38779 program is stopped are queued and the program will be interrupted when
38780 it is resumed next time.
38782 @node Notification Packets
38783 @section Notification Packets
38784 @cindex notification packets
38785 @cindex packets, notification
38787 The @value{GDBN} remote serial protocol includes @dfn{notifications},
38788 packets that require no acknowledgment. Both the GDB and the stub
38789 may send notifications (although the only notifications defined at
38790 present are sent by the stub). Notifications carry information
38791 without incurring the round-trip latency of an acknowledgment, and so
38792 are useful for low-impact communications where occasional packet loss
38795 A notification packet has the form @samp{% @var{data} #
38796 @var{checksum}}, where @var{data} is the content of the notification,
38797 and @var{checksum} is a checksum of @var{data}, computed and formatted
38798 as for ordinary @value{GDBN} packets. A notification's @var{data}
38799 never contains @samp{$}, @samp{%} or @samp{#} characters. Upon
38800 receiving a notification, the recipient sends no @samp{+} or @samp{-}
38801 to acknowledge the notification's receipt or to report its corruption.
38803 Every notification's @var{data} begins with a name, which contains no
38804 colon characters, followed by a colon character.
38806 Recipients should silently ignore corrupted notifications and
38807 notifications they do not understand. Recipients should restart
38808 timeout periods on receipt of a well-formed notification, whether or
38809 not they understand it.
38811 Senders should only send the notifications described here when this
38812 protocol description specifies that they are permitted. In the
38813 future, we may extend the protocol to permit existing notifications in
38814 new contexts; this rule helps older senders avoid confusing newer
38817 (Older versions of @value{GDBN} ignore bytes received until they see
38818 the @samp{$} byte that begins an ordinary packet, so new stubs may
38819 transmit notifications without fear of confusing older clients. There
38820 are no notifications defined for @value{GDBN} to send at the moment, but we
38821 assume that most older stubs would ignore them, as well.)
38823 Each notification is comprised of three parts:
38825 @item @var{name}:@var{event}
38826 The notification packet is sent by the side that initiates the
38827 exchange (currently, only the stub does that), with @var{event}
38828 carrying the specific information about the notification, and
38829 @var{name} specifying the name of the notification.
38831 The acknowledge sent by the other side, usually @value{GDBN}, to
38832 acknowledge the exchange and request the event.
38835 The purpose of an asynchronous notification mechanism is to report to
38836 @value{GDBN} that something interesting happened in the remote stub.
38838 The remote stub may send notification @var{name}:@var{event}
38839 at any time, but @value{GDBN} acknowledges the notification when
38840 appropriate. The notification event is pending before @value{GDBN}
38841 acknowledges. Only one notification at a time may be pending; if
38842 additional events occur before @value{GDBN} has acknowledged the
38843 previous notification, they must be queued by the stub for later
38844 synchronous transmission in response to @var{ack} packets from
38845 @value{GDBN}. Because the notification mechanism is unreliable,
38846 the stub is permitted to resend a notification if it believes
38847 @value{GDBN} may not have received it.
38849 Specifically, notifications may appear when @value{GDBN} is not
38850 otherwise reading input from the stub, or when @value{GDBN} is
38851 expecting to read a normal synchronous response or a
38852 @samp{+}/@samp{-} acknowledgment to a packet it has sent.
38853 Notification packets are distinct from any other communication from
38854 the stub so there is no ambiguity.
38856 After receiving a notification, @value{GDBN} shall acknowledge it by
38857 sending a @var{ack} packet as a regular, synchronous request to the
38858 stub. Such acknowledgment is not required to happen immediately, as
38859 @value{GDBN} is permitted to send other, unrelated packets to the
38860 stub first, which the stub should process normally.
38862 Upon receiving a @var{ack} packet, if the stub has other queued
38863 events to report to @value{GDBN}, it shall respond by sending a
38864 normal @var{event}. @value{GDBN} shall then send another @var{ack}
38865 packet to solicit further responses; again, it is permitted to send
38866 other, unrelated packets as well which the stub should process
38869 If the stub receives a @var{ack} packet and there are no additional
38870 @var{event} to report, the stub shall return an @samp{OK} response.
38871 At this point, @value{GDBN} has finished processing a notification
38872 and the stub has completed sending any queued events. @value{GDBN}
38873 won't accept any new notifications until the final @samp{OK} is
38874 received . If further notification events occur, the stub shall send
38875 a new notification, @value{GDBN} shall accept the notification, and
38876 the process shall be repeated.
38878 The process of asynchronous notification can be illustrated by the
38881 <- @code{%Stop:T0505:98e7ffbf;04:4ce6ffbf;08:b1b6e54c;thread:p7526.7526;core:0;}
38884 <- @code{T0505:68f37db7;04:40f37db7;08:63850408;thread:p7526.7528;core:0;}
38886 <- @code{T0505:68e3fdb6;04:40e3fdb6;08:63850408;thread:p7526.7529;core:0;}
38891 The following notifications are defined:
38892 @multitable @columnfractions 0.12 0.12 0.38 0.38
38901 @tab @var{reply}. The @var{reply} has the form of a stop reply, as
38902 described in @ref{Stop Reply Packets}. Refer to @ref{Remote Non-Stop},
38903 for information on how these notifications are acknowledged by
38905 @tab Report an asynchronous stop event in non-stop mode.
38909 @node Remote Non-Stop
38910 @section Remote Protocol Support for Non-Stop Mode
38912 @value{GDBN}'s remote protocol supports non-stop debugging of
38913 multi-threaded programs, as described in @ref{Non-Stop Mode}. If the stub
38914 supports non-stop mode, it should report that to @value{GDBN} by including
38915 @samp{QNonStop+} in its @samp{qSupported} response (@pxref{qSupported}).
38917 @value{GDBN} typically sends a @samp{QNonStop} packet only when
38918 establishing a new connection with the stub. Entering non-stop mode
38919 does not alter the state of any currently-running threads, but targets
38920 must stop all threads in any already-attached processes when entering
38921 all-stop mode. @value{GDBN} uses the @samp{?} packet as necessary to
38922 probe the target state after a mode change.
38924 In non-stop mode, when an attached process encounters an event that
38925 would otherwise be reported with a stop reply, it uses the
38926 asynchronous notification mechanism (@pxref{Notification Packets}) to
38927 inform @value{GDBN}. In contrast to all-stop mode, where all threads
38928 in all processes are stopped when a stop reply is sent, in non-stop
38929 mode only the thread reporting the stop event is stopped. That is,
38930 when reporting a @samp{S} or @samp{T} response to indicate completion
38931 of a step operation, hitting a breakpoint, or a fault, only the
38932 affected thread is stopped; any other still-running threads continue
38933 to run. When reporting a @samp{W} or @samp{X} response, all running
38934 threads belonging to other attached processes continue to run.
38936 In non-stop mode, the target shall respond to the @samp{?} packet as
38937 follows. First, any incomplete stop reply notification/@samp{vStopped}
38938 sequence in progress is abandoned. The target must begin a new
38939 sequence reporting stop events for all stopped threads, whether or not
38940 it has previously reported those events to @value{GDBN}. The first
38941 stop reply is sent as a synchronous reply to the @samp{?} packet, and
38942 subsequent stop replies are sent as responses to @samp{vStopped} packets
38943 using the mechanism described above. The target must not send
38944 asynchronous stop reply notifications until the sequence is complete.
38945 If all threads are running when the target receives the @samp{?} packet,
38946 or if the target is not attached to any process, it shall respond
38949 If the stub supports non-stop mode, it should also support the
38950 @samp{swbreak} stop reason if software breakpoints are supported, and
38951 the @samp{hwbreak} stop reason if hardware breakpoints are supported
38952 (@pxref{swbreak stop reason}). This is because given the asynchronous
38953 nature of non-stop mode, between the time a thread hits a breakpoint
38954 and the time the event is finally processed by @value{GDBN}, the
38955 breakpoint may have already been removed from the target. Due to
38956 this, @value{GDBN} needs to be able to tell whether a trap stop was
38957 caused by a delayed breakpoint event, which should be ignored, as
38958 opposed to a random trap signal, which should be reported to the user.
38959 Note the @samp{swbreak} feature implies that the target is responsible
38960 for adjusting the PC when a software breakpoint triggers, if
38961 necessary, such as on the x86 architecture.
38963 @node Packet Acknowledgment
38964 @section Packet Acknowledgment
38966 @cindex acknowledgment, for @value{GDBN} remote
38967 @cindex packet acknowledgment, for @value{GDBN} remote
38968 By default, when either the host or the target machine receives a packet,
38969 the first response expected is an acknowledgment: either @samp{+} (to indicate
38970 the package was received correctly) or @samp{-} (to request retransmission).
38971 This mechanism allows the @value{GDBN} remote protocol to operate over
38972 unreliable transport mechanisms, such as a serial line.
38974 In cases where the transport mechanism is itself reliable (such as a pipe or
38975 TCP connection), the @samp{+}/@samp{-} acknowledgments are redundant.
38976 It may be desirable to disable them in that case to reduce communication
38977 overhead, or for other reasons. This can be accomplished by means of the
38978 @samp{QStartNoAckMode} packet; @pxref{QStartNoAckMode}.
38980 When in no-acknowledgment mode, neither the stub nor @value{GDBN} shall send or
38981 expect @samp{+}/@samp{-} protocol acknowledgments. The packet
38982 and response format still includes the normal checksum, as described in
38983 @ref{Overview}, but the checksum may be ignored by the receiver.
38985 If the stub supports @samp{QStartNoAckMode} and prefers to operate in
38986 no-acknowledgment mode, it should report that to @value{GDBN}
38987 by including @samp{QStartNoAckMode+} in its response to @samp{qSupported};
38988 @pxref{qSupported}.
38989 If @value{GDBN} also supports @samp{QStartNoAckMode} and it has not been
38990 disabled via the @code{set remote noack-packet off} command
38991 (@pxref{Remote Configuration}),
38992 @value{GDBN} may then send a @samp{QStartNoAckMode} packet to the stub.
38993 Only then may the stub actually turn off packet acknowledgments.
38994 @value{GDBN} sends a final @samp{+} acknowledgment of the stub's @samp{OK}
38995 response, which can be safely ignored by the stub.
38997 Note that @code{set remote noack-packet} command only affects negotiation
38998 between @value{GDBN} and the stub when subsequent connections are made;
38999 it does not affect the protocol acknowledgment state for any current
39001 Since @samp{+}/@samp{-} acknowledgments are enabled by default when a
39002 new connection is established,
39003 there is also no protocol request to re-enable the acknowledgments
39004 for the current connection, once disabled.
39009 Example sequence of a target being re-started. Notice how the restart
39010 does not get any direct output:
39015 @emph{target restarts}
39018 <- @code{T001:1234123412341234}
39022 Example sequence of a target being stepped by a single instruction:
39025 -> @code{G1445@dots{}}
39030 <- @code{T001:1234123412341234}
39034 <- @code{1455@dots{}}
39038 @node File-I/O Remote Protocol Extension
39039 @section File-I/O Remote Protocol Extension
39040 @cindex File-I/O remote protocol extension
39043 * File-I/O Overview::
39044 * Protocol Basics::
39045 * The F Request Packet::
39046 * The F Reply Packet::
39047 * The Ctrl-C Message::
39049 * List of Supported Calls::
39050 * Protocol-specific Representation of Datatypes::
39052 * File-I/O Examples::
39055 @node File-I/O Overview
39056 @subsection File-I/O Overview
39057 @cindex file-i/o overview
39059 The @dfn{File I/O remote protocol extension} (short: File-I/O) allows the
39060 target to use the host's file system and console I/O to perform various
39061 system calls. System calls on the target system are translated into a
39062 remote protocol packet to the host system, which then performs the needed
39063 actions and returns a response packet to the target system.
39064 This simulates file system operations even on targets that lack file systems.
39066 The protocol is defined to be independent of both the host and target systems.
39067 It uses its own internal representation of datatypes and values. Both
39068 @value{GDBN} and the target's @value{GDBN} stub are responsible for
39069 translating the system-dependent value representations into the internal
39070 protocol representations when data is transmitted.
39072 The communication is synchronous. A system call is possible only when
39073 @value{GDBN} is waiting for a response from the @samp{C}, @samp{c}, @samp{S}
39074 or @samp{s} packets. While @value{GDBN} handles the request for a system call,
39075 the target is stopped to allow deterministic access to the target's
39076 memory. Therefore File-I/O is not interruptible by target signals. On
39077 the other hand, it is possible to interrupt File-I/O by a user interrupt
39078 (@samp{Ctrl-C}) within @value{GDBN}.
39080 The target's request to perform a host system call does not finish
39081 the latest @samp{C}, @samp{c}, @samp{S} or @samp{s} action. That means,
39082 after finishing the system call, the target returns to continuing the
39083 previous activity (continue, step). No additional continue or step
39084 request from @value{GDBN} is required.
39087 (@value{GDBP}) continue
39088 <- target requests 'system call X'
39089 target is stopped, @value{GDBN} executes system call
39090 -> @value{GDBN} returns result
39091 ... target continues, @value{GDBN} returns to wait for the target
39092 <- target hits breakpoint and sends a Txx packet
39095 The protocol only supports I/O on the console and to regular files on
39096 the host file system. Character or block special devices, pipes,
39097 named pipes, sockets or any other communication method on the host
39098 system are not supported by this protocol.
39100 File I/O is not supported in non-stop mode.
39102 @node Protocol Basics
39103 @subsection Protocol Basics
39104 @cindex protocol basics, file-i/o
39106 The File-I/O protocol uses the @code{F} packet as the request as well
39107 as reply packet. Since a File-I/O system call can only occur when
39108 @value{GDBN} is waiting for a response from the continuing or stepping target,
39109 the File-I/O request is a reply that @value{GDBN} has to expect as a result
39110 of a previous @samp{C}, @samp{c}, @samp{S} or @samp{s} packet.
39111 This @code{F} packet contains all information needed to allow @value{GDBN}
39112 to call the appropriate host system call:
39116 A unique identifier for the requested system call.
39119 All parameters to the system call. Pointers are given as addresses
39120 in the target memory address space. Pointers to strings are given as
39121 pointer/length pair. Numerical values are given as they are.
39122 Numerical control flags are given in a protocol-specific representation.
39126 At this point, @value{GDBN} has to perform the following actions.
39130 If the parameters include pointer values to data needed as input to a
39131 system call, @value{GDBN} requests this data from the target with a
39132 standard @code{m} packet request. This additional communication has to be
39133 expected by the target implementation and is handled as any other @code{m}
39137 @value{GDBN} translates all value from protocol representation to host
39138 representation as needed. Datatypes are coerced into the host types.
39141 @value{GDBN} calls the system call.
39144 It then coerces datatypes back to protocol representation.
39147 If the system call is expected to return data in buffer space specified
39148 by pointer parameters to the call, the data is transmitted to the
39149 target using a @code{M} or @code{X} packet. This packet has to be expected
39150 by the target implementation and is handled as any other @code{M} or @code{X}
39155 Eventually @value{GDBN} replies with another @code{F} packet which contains all
39156 necessary information for the target to continue. This at least contains
39163 @code{errno}, if has been changed by the system call.
39170 After having done the needed type and value coercion, the target continues
39171 the latest continue or step action.
39173 @node The F Request Packet
39174 @subsection The @code{F} Request Packet
39175 @cindex file-i/o request packet
39176 @cindex @code{F} request packet
39178 The @code{F} request packet has the following format:
39181 @item F@var{call-id},@var{parameter@dots{}}
39183 @var{call-id} is the identifier to indicate the host system call to be called.
39184 This is just the name of the function.
39186 @var{parameter@dots{}} are the parameters to the system call.
39187 Parameters are hexadecimal integer values, either the actual values in case
39188 of scalar datatypes, pointers to target buffer space in case of compound
39189 datatypes and unspecified memory areas, or pointer/length pairs in case
39190 of string parameters. These are appended to the @var{call-id} as a
39191 comma-delimited list. All values are transmitted in ASCII
39192 string representation, pointer/length pairs separated by a slash.
39198 @node The F Reply Packet
39199 @subsection The @code{F} Reply Packet
39200 @cindex file-i/o reply packet
39201 @cindex @code{F} reply packet
39203 The @code{F} reply packet has the following format:
39207 @item F@var{retcode},@var{errno},@var{Ctrl-C flag};@var{call-specific attachment}
39209 @var{retcode} is the return code of the system call as hexadecimal value.
39211 @var{errno} is the @code{errno} set by the call, in protocol-specific
39213 This parameter can be omitted if the call was successful.
39215 @var{Ctrl-C flag} is only sent if the user requested a break. In this
39216 case, @var{errno} must be sent as well, even if the call was successful.
39217 The @var{Ctrl-C flag} itself consists of the character @samp{C}:
39224 or, if the call was interrupted before the host call has been performed:
39231 assuming 4 is the protocol-specific representation of @code{EINTR}.
39236 @node The Ctrl-C Message
39237 @subsection The @samp{Ctrl-C} Message
39238 @cindex ctrl-c message, in file-i/o protocol
39240 If the @samp{Ctrl-C} flag is set in the @value{GDBN}
39241 reply packet (@pxref{The F Reply Packet}),
39242 the target should behave as if it had
39243 gotten a break message. The meaning for the target is ``system call
39244 interrupted by @code{SIGINT}''. Consequentially, the target should actually stop
39245 (as with a break message) and return to @value{GDBN} with a @code{T02}
39248 It's important for the target to know in which
39249 state the system call was interrupted. There are two possible cases:
39253 The system call hasn't been performed on the host yet.
39256 The system call on the host has been finished.
39260 These two states can be distinguished by the target by the value of the
39261 returned @code{errno}. If it's the protocol representation of @code{EINTR}, the system
39262 call hasn't been performed. This is equivalent to the @code{EINTR} handling
39263 on POSIX systems. In any other case, the target may presume that the
39264 system call has been finished --- successfully or not --- and should behave
39265 as if the break message arrived right after the system call.
39267 @value{GDBN} must behave reliably. If the system call has not been called
39268 yet, @value{GDBN} may send the @code{F} reply immediately, setting @code{EINTR} as
39269 @code{errno} in the packet. If the system call on the host has been finished
39270 before the user requests a break, the full action must be finished by
39271 @value{GDBN}. This requires sending @code{M} or @code{X} packets as necessary.
39272 The @code{F} packet may only be sent when either nothing has happened
39273 or the full action has been completed.
39276 @subsection Console I/O
39277 @cindex console i/o as part of file-i/o
39279 By default and if not explicitly closed by the target system, the file
39280 descriptors 0, 1 and 2 are connected to the @value{GDBN} console. Output
39281 on the @value{GDBN} console is handled as any other file output operation
39282 (@code{write(1, @dots{})} or @code{write(2, @dots{})}). Console input is handled
39283 by @value{GDBN} so that after the target read request from file descriptor
39284 0 all following typing is buffered until either one of the following
39289 The user types @kbd{Ctrl-c}. The behaviour is as explained above, and the
39291 system call is treated as finished.
39294 The user presses @key{RET}. This is treated as end of input with a trailing
39298 The user types @kbd{Ctrl-d}. This is treated as end of input. No trailing
39299 character (neither newline nor @samp{Ctrl-D}) is appended to the input.
39303 If the user has typed more characters than fit in the buffer given to
39304 the @code{read} call, the trailing characters are buffered in @value{GDBN} until
39305 either another @code{read(0, @dots{})} is requested by the target, or debugging
39306 is stopped at the user's request.
39309 @node List of Supported Calls
39310 @subsection List of Supported Calls
39311 @cindex list of supported file-i/o calls
39328 @unnumberedsubsubsec open
39329 @cindex open, file-i/o system call
39334 int open(const char *pathname, int flags);
39335 int open(const char *pathname, int flags, mode_t mode);
39339 @samp{Fopen,@var{pathptr}/@var{len},@var{flags},@var{mode}}
39342 @var{flags} is the bitwise @code{OR} of the following values:
39346 If the file does not exist it will be created. The host
39347 rules apply as far as file ownership and time stamps
39351 When used with @code{O_CREAT}, if the file already exists it is
39352 an error and open() fails.
39355 If the file already exists and the open mode allows
39356 writing (@code{O_RDWR} or @code{O_WRONLY} is given) it will be
39357 truncated to zero length.
39360 The file is opened in append mode.
39363 The file is opened for reading only.
39366 The file is opened for writing only.
39369 The file is opened for reading and writing.
39373 Other bits are silently ignored.
39377 @var{mode} is the bitwise @code{OR} of the following values:
39381 User has read permission.
39384 User has write permission.
39387 Group has read permission.
39390 Group has write permission.
39393 Others have read permission.
39396 Others have write permission.
39400 Other bits are silently ignored.
39403 @item Return value:
39404 @code{open} returns the new file descriptor or -1 if an error
39411 @var{pathname} already exists and @code{O_CREAT} and @code{O_EXCL} were used.
39414 @var{pathname} refers to a directory.
39417 The requested access is not allowed.
39420 @var{pathname} was too long.
39423 A directory component in @var{pathname} does not exist.
39426 @var{pathname} refers to a device, pipe, named pipe or socket.
39429 @var{pathname} refers to a file on a read-only filesystem and
39430 write access was requested.
39433 @var{pathname} is an invalid pointer value.
39436 No space on device to create the file.
39439 The process already has the maximum number of files open.
39442 The limit on the total number of files open on the system
39446 The call was interrupted by the user.
39452 @unnumberedsubsubsec close
39453 @cindex close, file-i/o system call
39462 @samp{Fclose,@var{fd}}
39464 @item Return value:
39465 @code{close} returns zero on success, or -1 if an error occurred.
39471 @var{fd} isn't a valid open file descriptor.
39474 The call was interrupted by the user.
39480 @unnumberedsubsubsec read
39481 @cindex read, file-i/o system call
39486 int read(int fd, void *buf, unsigned int count);
39490 @samp{Fread,@var{fd},@var{bufptr},@var{count}}
39492 @item Return value:
39493 On success, the number of bytes read is returned.
39494 Zero indicates end of file. If count is zero, read
39495 returns zero as well. On error, -1 is returned.
39501 @var{fd} is not a valid file descriptor or is not open for
39505 @var{bufptr} is an invalid pointer value.
39508 The call was interrupted by the user.
39514 @unnumberedsubsubsec write
39515 @cindex write, file-i/o system call
39520 int write(int fd, const void *buf, unsigned int count);
39524 @samp{Fwrite,@var{fd},@var{bufptr},@var{count}}
39526 @item Return value:
39527 On success, the number of bytes written are returned.
39528 Zero indicates nothing was written. On error, -1
39535 @var{fd} is not a valid file descriptor or is not open for
39539 @var{bufptr} is an invalid pointer value.
39542 An attempt was made to write a file that exceeds the
39543 host-specific maximum file size allowed.
39546 No space on device to write the data.
39549 The call was interrupted by the user.
39555 @unnumberedsubsubsec lseek
39556 @cindex lseek, file-i/o system call
39561 long lseek (int fd, long offset, int flag);
39565 @samp{Flseek,@var{fd},@var{offset},@var{flag}}
39567 @var{flag} is one of:
39571 The offset is set to @var{offset} bytes.
39574 The offset is set to its current location plus @var{offset}
39578 The offset is set to the size of the file plus @var{offset}
39582 @item Return value:
39583 On success, the resulting unsigned offset in bytes from
39584 the beginning of the file is returned. Otherwise, a
39585 value of -1 is returned.
39591 @var{fd} is not a valid open file descriptor.
39594 @var{fd} is associated with the @value{GDBN} console.
39597 @var{flag} is not a proper value.
39600 The call was interrupted by the user.
39606 @unnumberedsubsubsec rename
39607 @cindex rename, file-i/o system call
39612 int rename(const char *oldpath, const char *newpath);
39616 @samp{Frename,@var{oldpathptr}/@var{len},@var{newpathptr}/@var{len}}
39618 @item Return value:
39619 On success, zero is returned. On error, -1 is returned.
39625 @var{newpath} is an existing directory, but @var{oldpath} is not a
39629 @var{newpath} is a non-empty directory.
39632 @var{oldpath} or @var{newpath} is a directory that is in use by some
39636 An attempt was made to make a directory a subdirectory
39640 A component used as a directory in @var{oldpath} or new
39641 path is not a directory. Or @var{oldpath} is a directory
39642 and @var{newpath} exists but is not a directory.
39645 @var{oldpathptr} or @var{newpathptr} are invalid pointer values.
39648 No access to the file or the path of the file.
39652 @var{oldpath} or @var{newpath} was too long.
39655 A directory component in @var{oldpath} or @var{newpath} does not exist.
39658 The file is on a read-only filesystem.
39661 The device containing the file has no room for the new
39665 The call was interrupted by the user.
39671 @unnumberedsubsubsec unlink
39672 @cindex unlink, file-i/o system call
39677 int unlink(const char *pathname);
39681 @samp{Funlink,@var{pathnameptr}/@var{len}}
39683 @item Return value:
39684 On success, zero is returned. On error, -1 is returned.
39690 No access to the file or the path of the file.
39693 The system does not allow unlinking of directories.
39696 The file @var{pathname} cannot be unlinked because it's
39697 being used by another process.
39700 @var{pathnameptr} is an invalid pointer value.
39703 @var{pathname} was too long.
39706 A directory component in @var{pathname} does not exist.
39709 A component of the path is not a directory.
39712 The file is on a read-only filesystem.
39715 The call was interrupted by the user.
39721 @unnumberedsubsubsec stat/fstat
39722 @cindex fstat, file-i/o system call
39723 @cindex stat, file-i/o system call
39728 int stat(const char *pathname, struct stat *buf);
39729 int fstat(int fd, struct stat *buf);
39733 @samp{Fstat,@var{pathnameptr}/@var{len},@var{bufptr}}@*
39734 @samp{Ffstat,@var{fd},@var{bufptr}}
39736 @item Return value:
39737 On success, zero is returned. On error, -1 is returned.
39743 @var{fd} is not a valid open file.
39746 A directory component in @var{pathname} does not exist or the
39747 path is an empty string.
39750 A component of the path is not a directory.
39753 @var{pathnameptr} is an invalid pointer value.
39756 No access to the file or the path of the file.
39759 @var{pathname} was too long.
39762 The call was interrupted by the user.
39768 @unnumberedsubsubsec gettimeofday
39769 @cindex gettimeofday, file-i/o system call
39774 int gettimeofday(struct timeval *tv, void *tz);
39778 @samp{Fgettimeofday,@var{tvptr},@var{tzptr}}
39780 @item Return value:
39781 On success, 0 is returned, -1 otherwise.
39787 @var{tz} is a non-NULL pointer.
39790 @var{tvptr} and/or @var{tzptr} is an invalid pointer value.
39796 @unnumberedsubsubsec isatty
39797 @cindex isatty, file-i/o system call
39802 int isatty(int fd);
39806 @samp{Fisatty,@var{fd}}
39808 @item Return value:
39809 Returns 1 if @var{fd} refers to the @value{GDBN} console, 0 otherwise.
39815 The call was interrupted by the user.
39820 Note that the @code{isatty} call is treated as a special case: it returns
39821 1 to the target if the file descriptor is attached
39822 to the @value{GDBN} console, 0 otherwise. Implementing through system calls
39823 would require implementing @code{ioctl} and would be more complex than
39828 @unnumberedsubsubsec system
39829 @cindex system, file-i/o system call
39834 int system(const char *command);
39838 @samp{Fsystem,@var{commandptr}/@var{len}}
39840 @item Return value:
39841 If @var{len} is zero, the return value indicates whether a shell is
39842 available. A zero return value indicates a shell is not available.
39843 For non-zero @var{len}, the value returned is -1 on error and the
39844 return status of the command otherwise. Only the exit status of the
39845 command is returned, which is extracted from the host's @code{system}
39846 return value by calling @code{WEXITSTATUS(retval)}. In case
39847 @file{/bin/sh} could not be executed, 127 is returned.
39853 The call was interrupted by the user.
39858 @value{GDBN} takes over the full task of calling the necessary host calls
39859 to perform the @code{system} call. The return value of @code{system} on
39860 the host is simplified before it's returned
39861 to the target. Any termination signal information from the child process
39862 is discarded, and the return value consists
39863 entirely of the exit status of the called command.
39865 Due to security concerns, the @code{system} call is by default refused
39866 by @value{GDBN}. The user has to allow this call explicitly with the
39867 @code{set remote system-call-allowed 1} command.
39870 @item set remote system-call-allowed
39871 @kindex set remote system-call-allowed
39872 Control whether to allow the @code{system} calls in the File I/O
39873 protocol for the remote target. The default is zero (disabled).
39875 @item show remote system-call-allowed
39876 @kindex show remote system-call-allowed
39877 Show whether the @code{system} calls are allowed in the File I/O
39881 @node Protocol-specific Representation of Datatypes
39882 @subsection Protocol-specific Representation of Datatypes
39883 @cindex protocol-specific representation of datatypes, in file-i/o protocol
39886 * Integral Datatypes::
39888 * Memory Transfer::
39893 @node Integral Datatypes
39894 @unnumberedsubsubsec Integral Datatypes
39895 @cindex integral datatypes, in file-i/o protocol
39897 The integral datatypes used in the system calls are @code{int},
39898 @code{unsigned int}, @code{long}, @code{unsigned long},
39899 @code{mode_t}, and @code{time_t}.
39901 @code{int}, @code{unsigned int}, @code{mode_t} and @code{time_t} are
39902 implemented as 32 bit values in this protocol.
39904 @code{long} and @code{unsigned long} are implemented as 64 bit types.
39906 @xref{Limits}, for corresponding MIN and MAX values (similar to those
39907 in @file{limits.h}) to allow range checking on host and target.
39909 @code{time_t} datatypes are defined as seconds since the Epoch.
39911 All integral datatypes transferred as part of a memory read or write of a
39912 structured datatype e.g.@: a @code{struct stat} have to be given in big endian
39915 @node Pointer Values
39916 @unnumberedsubsubsec Pointer Values
39917 @cindex pointer values, in file-i/o protocol
39919 Pointers to target data are transmitted as they are. An exception
39920 is made for pointers to buffers for which the length isn't
39921 transmitted as part of the function call, namely strings. Strings
39922 are transmitted as a pointer/length pair, both as hex values, e.g.@:
39929 which is a pointer to data of length 18 bytes at position 0x1aaf.
39930 The length is defined as the full string length in bytes, including
39931 the trailing null byte. For example, the string @code{"hello world"}
39932 at address 0x123456 is transmitted as
39938 @node Memory Transfer
39939 @unnumberedsubsubsec Memory Transfer
39940 @cindex memory transfer, in file-i/o protocol
39942 Structured data which is transferred using a memory read or write (for
39943 example, a @code{struct stat}) is expected to be in a protocol-specific format
39944 with all scalar multibyte datatypes being big endian. Translation to
39945 this representation needs to be done both by the target before the @code{F}
39946 packet is sent, and by @value{GDBN} before
39947 it transfers memory to the target. Transferred pointers to structured
39948 data should point to the already-coerced data at any time.
39952 @unnumberedsubsubsec struct stat
39953 @cindex struct stat, in file-i/o protocol
39955 The buffer of type @code{struct stat} used by the target and @value{GDBN}
39956 is defined as follows:
39960 unsigned int st_dev; /* device */
39961 unsigned int st_ino; /* inode */
39962 mode_t st_mode; /* protection */
39963 unsigned int st_nlink; /* number of hard links */
39964 unsigned int st_uid; /* user ID of owner */
39965 unsigned int st_gid; /* group ID of owner */
39966 unsigned int st_rdev; /* device type (if inode device) */
39967 unsigned long st_size; /* total size, in bytes */
39968 unsigned long st_blksize; /* blocksize for filesystem I/O */
39969 unsigned long st_blocks; /* number of blocks allocated */
39970 time_t st_atime; /* time of last access */
39971 time_t st_mtime; /* time of last modification */
39972 time_t st_ctime; /* time of last change */
39976 The integral datatypes conform to the definitions given in the
39977 appropriate section (see @ref{Integral Datatypes}, for details) so this
39978 structure is of size 64 bytes.
39980 The values of several fields have a restricted meaning and/or
39986 A value of 0 represents a file, 1 the console.
39989 No valid meaning for the target. Transmitted unchanged.
39992 Valid mode bits are described in @ref{Constants}. Any other
39993 bits have currently no meaning for the target.
39998 No valid meaning for the target. Transmitted unchanged.
40003 These values have a host and file system dependent
40004 accuracy. Especially on Windows hosts, the file system may not
40005 support exact timing values.
40008 The target gets a @code{struct stat} of the above representation and is
40009 responsible for coercing it to the target representation before
40012 Note that due to size differences between the host, target, and protocol
40013 representations of @code{struct stat} members, these members could eventually
40014 get truncated on the target.
40016 @node struct timeval
40017 @unnumberedsubsubsec struct timeval
40018 @cindex struct timeval, in file-i/o protocol
40020 The buffer of type @code{struct timeval} used by the File-I/O protocol
40021 is defined as follows:
40025 time_t tv_sec; /* second */
40026 long tv_usec; /* microsecond */
40030 The integral datatypes conform to the definitions given in the
40031 appropriate section (see @ref{Integral Datatypes}, for details) so this
40032 structure is of size 8 bytes.
40035 @subsection Constants
40036 @cindex constants, in file-i/o protocol
40038 The following values are used for the constants inside of the
40039 protocol. @value{GDBN} and target are responsible for translating these
40040 values before and after the call as needed.
40051 @unnumberedsubsubsec Open Flags
40052 @cindex open flags, in file-i/o protocol
40054 All values are given in hexadecimal representation.
40066 @node mode_t Values
40067 @unnumberedsubsubsec mode_t Values
40068 @cindex mode_t values, in file-i/o protocol
40070 All values are given in octal representation.
40087 @unnumberedsubsubsec Errno Values
40088 @cindex errno values, in file-i/o protocol
40090 All values are given in decimal representation.
40115 @code{EUNKNOWN} is used as a fallback error value if a host system returns
40116 any error value not in the list of supported error numbers.
40119 @unnumberedsubsubsec Lseek Flags
40120 @cindex lseek flags, in file-i/o protocol
40129 @unnumberedsubsubsec Limits
40130 @cindex limits, in file-i/o protocol
40132 All values are given in decimal representation.
40135 INT_MIN -2147483648
40137 UINT_MAX 4294967295
40138 LONG_MIN -9223372036854775808
40139 LONG_MAX 9223372036854775807
40140 ULONG_MAX 18446744073709551615
40143 @node File-I/O Examples
40144 @subsection File-I/O Examples
40145 @cindex file-i/o examples
40147 Example sequence of a write call, file descriptor 3, buffer is at target
40148 address 0x1234, 6 bytes should be written:
40151 <- @code{Fwrite,3,1234,6}
40152 @emph{request memory read from target}
40155 @emph{return "6 bytes written"}
40159 Example sequence of a read call, file descriptor 3, buffer is at target
40160 address 0x1234, 6 bytes should be read:
40163 <- @code{Fread,3,1234,6}
40164 @emph{request memory write to target}
40165 -> @code{X1234,6:XXXXXX}
40166 @emph{return "6 bytes read"}
40170 Example sequence of a read call, call fails on the host due to invalid
40171 file descriptor (@code{EBADF}):
40174 <- @code{Fread,3,1234,6}
40178 Example sequence of a read call, user presses @kbd{Ctrl-c} before syscall on
40182 <- @code{Fread,3,1234,6}
40187 Example sequence of a read call, user presses @kbd{Ctrl-c} after syscall on
40191 <- @code{Fread,3,1234,6}
40192 -> @code{X1234,6:XXXXXX}
40196 @node Library List Format
40197 @section Library List Format
40198 @cindex library list format, remote protocol
40200 On some platforms, a dynamic loader (e.g.@: @file{ld.so}) runs in the
40201 same process as your application to manage libraries. In this case,
40202 @value{GDBN} can use the loader's symbol table and normal memory
40203 operations to maintain a list of shared libraries. On other
40204 platforms, the operating system manages loaded libraries.
40205 @value{GDBN} can not retrieve the list of currently loaded libraries
40206 through memory operations, so it uses the @samp{qXfer:libraries:read}
40207 packet (@pxref{qXfer library list read}) instead. The remote stub
40208 queries the target's operating system and reports which libraries
40211 The @samp{qXfer:libraries:read} packet returns an XML document which
40212 lists loaded libraries and their offsets. Each library has an
40213 associated name and one or more segment or section base addresses,
40214 which report where the library was loaded in memory.
40216 For the common case of libraries that are fully linked binaries, the
40217 library should have a list of segments. If the target supports
40218 dynamic linking of a relocatable object file, its library XML element
40219 should instead include a list of allocated sections. The segment or
40220 section bases are start addresses, not relocation offsets; they do not
40221 depend on the library's link-time base addresses.
40223 @value{GDBN} must be linked with the Expat library to support XML
40224 library lists. @xref{Expat}.
40226 A simple memory map, with one loaded library relocated by a single
40227 offset, looks like this:
40231 <library name="/lib/libc.so.6">
40232 <segment address="0x10000000"/>
40237 Another simple memory map, with one loaded library with three
40238 allocated sections (.text, .data, .bss), looks like this:
40242 <library name="sharedlib.o">
40243 <section address="0x10000000"/>
40244 <section address="0x20000000"/>
40245 <section address="0x30000000"/>
40250 The format of a library list is described by this DTD:
40253 <!-- library-list: Root element with versioning -->
40254 <!ELEMENT library-list (library)*>
40255 <!ATTLIST library-list version CDATA #FIXED "1.0">
40256 <!ELEMENT library (segment*, section*)>
40257 <!ATTLIST library name CDATA #REQUIRED>
40258 <!ELEMENT segment EMPTY>
40259 <!ATTLIST segment address CDATA #REQUIRED>
40260 <!ELEMENT section EMPTY>
40261 <!ATTLIST section address CDATA #REQUIRED>
40264 In addition, segments and section descriptors cannot be mixed within a
40265 single library element, and you must supply at least one segment or
40266 section for each library.
40268 @node Library List Format for SVR4 Targets
40269 @section Library List Format for SVR4 Targets
40270 @cindex library list format, remote protocol
40272 On SVR4 platforms @value{GDBN} can use the symbol table of a dynamic loader
40273 (e.g.@: @file{ld.so}) and normal memory operations to maintain a list of
40274 shared libraries. Still a special library list provided by this packet is
40275 more efficient for the @value{GDBN} remote protocol.
40277 The @samp{qXfer:libraries-svr4:read} packet returns an XML document which lists
40278 loaded libraries and their SVR4 linker parameters. For each library on SVR4
40279 target, the following parameters are reported:
40283 @code{name}, the absolute file name from the @code{l_name} field of
40284 @code{struct link_map}.
40286 @code{lm} with address of @code{struct link_map} used for TLS
40287 (Thread Local Storage) access.
40289 @code{l_addr}, the displacement as read from the field @code{l_addr} of
40290 @code{struct link_map}. For prelinked libraries this is not an absolute
40291 memory address. It is a displacement of absolute memory address against
40292 address the file was prelinked to during the library load.
40294 @code{l_ld}, which is memory address of the @code{PT_DYNAMIC} segment
40297 Additionally the single @code{main-lm} attribute specifies address of
40298 @code{struct link_map} used for the main executable. This parameter is used
40299 for TLS access and its presence is optional.
40301 @value{GDBN} must be linked with the Expat library to support XML
40302 SVR4 library lists. @xref{Expat}.
40304 A simple memory map, with two loaded libraries (which do not use prelink),
40308 <library-list-svr4 version="1.0" main-lm="0xe4f8f8">
40309 <library name="/lib/ld-linux.so.2" lm="0xe4f51c" l_addr="0xe2d000"
40311 <library name="/lib/libc.so.6" lm="0xe4fbe8" l_addr="0x154000"
40313 </library-list-svr>
40316 The format of an SVR4 library list is described by this DTD:
40319 <!-- library-list-svr4: Root element with versioning -->
40320 <!ELEMENT library-list-svr4 (library)*>
40321 <!ATTLIST library-list-svr4 version CDATA #FIXED "1.0">
40322 <!ATTLIST library-list-svr4 main-lm CDATA #IMPLIED>
40323 <!ELEMENT library EMPTY>
40324 <!ATTLIST library name CDATA #REQUIRED>
40325 <!ATTLIST library lm CDATA #REQUIRED>
40326 <!ATTLIST library l_addr CDATA #REQUIRED>
40327 <!ATTLIST library l_ld CDATA #REQUIRED>
40330 @node Memory Map Format
40331 @section Memory Map Format
40332 @cindex memory map format
40334 To be able to write into flash memory, @value{GDBN} needs to obtain a
40335 memory map from the target. This section describes the format of the
40338 The memory map is obtained using the @samp{qXfer:memory-map:read}
40339 (@pxref{qXfer memory map read}) packet and is an XML document that
40340 lists memory regions.
40342 @value{GDBN} must be linked with the Expat library to support XML
40343 memory maps. @xref{Expat}.
40345 The top-level structure of the document is shown below:
40348 <?xml version="1.0"?>
40349 <!DOCTYPE memory-map
40350 PUBLIC "+//IDN gnu.org//DTD GDB Memory Map V1.0//EN"
40351 "http://sourceware.org/gdb/gdb-memory-map.dtd">
40357 Each region can be either:
40362 A region of RAM starting at @var{addr} and extending for @var{length}
40366 <memory type="ram" start="@var{addr}" length="@var{length}"/>
40371 A region of read-only memory:
40374 <memory type="rom" start="@var{addr}" length="@var{length}"/>
40379 A region of flash memory, with erasure blocks @var{blocksize}
40383 <memory type="flash" start="@var{addr}" length="@var{length}">
40384 <property name="blocksize">@var{blocksize}</property>
40390 Regions must not overlap. @value{GDBN} assumes that areas of memory not covered
40391 by the memory map are RAM, and uses the ordinary @samp{M} and @samp{X}
40392 packets to write to addresses in such ranges.
40394 The formal DTD for memory map format is given below:
40397 <!-- ................................................... -->
40398 <!-- Memory Map XML DTD ................................ -->
40399 <!-- File: memory-map.dtd .............................. -->
40400 <!-- .................................... .............. -->
40401 <!-- memory-map.dtd -->
40402 <!-- memory-map: Root element with versioning -->
40403 <!ELEMENT memory-map (memory | property)>
40404 <!ATTLIST memory-map version CDATA #FIXED "1.0.0">
40405 <!ELEMENT memory (property)>
40406 <!-- memory: Specifies a memory region,
40407 and its type, or device. -->
40408 <!ATTLIST memory type CDATA #REQUIRED
40409 start CDATA #REQUIRED
40410 length CDATA #REQUIRED
40411 device CDATA #IMPLIED>
40412 <!-- property: Generic attribute tag -->
40413 <!ELEMENT property (#PCDATA | property)*>
40414 <!ATTLIST property name CDATA #REQUIRED>
40417 @node Thread List Format
40418 @section Thread List Format
40419 @cindex thread list format
40421 To efficiently update the list of threads and their attributes,
40422 @value{GDBN} issues the @samp{qXfer:threads:read} packet
40423 (@pxref{qXfer threads read}) and obtains the XML document with
40424 the following structure:
40427 <?xml version="1.0"?>
40429 <thread id="id" core="0" name="name">
40430 ... description ...
40435 Each @samp{thread} element must have the @samp{id} attribute that
40436 identifies the thread (@pxref{thread-id syntax}). The
40437 @samp{core} attribute, if present, specifies which processor core
40438 the thread was last executing on. The @samp{name} attribute, if
40439 present, specifies the human-readable name of the thread. The content
40440 of the of @samp{thread} element is interpreted as human-readable
40441 auxiliary information.
40443 @node Traceframe Info Format
40444 @section Traceframe Info Format
40445 @cindex traceframe info format
40447 To be able to know which objects in the inferior can be examined when
40448 inspecting a tracepoint hit, @value{GDBN} needs to obtain the list of
40449 memory ranges, registers and trace state variables that have been
40450 collected in a traceframe.
40452 This list is obtained using the @samp{qXfer:traceframe-info:read}
40453 (@pxref{qXfer traceframe info read}) packet and is an XML document.
40455 @value{GDBN} must be linked with the Expat library to support XML
40456 traceframe info discovery. @xref{Expat}.
40458 The top-level structure of the document is shown below:
40461 <?xml version="1.0"?>
40462 <!DOCTYPE traceframe-info
40463 PUBLIC "+//IDN gnu.org//DTD GDB Memory Map V1.0//EN"
40464 "http://sourceware.org/gdb/gdb-traceframe-info.dtd">
40470 Each traceframe block can be either:
40475 A region of collected memory starting at @var{addr} and extending for
40476 @var{length} bytes from there:
40479 <memory start="@var{addr}" length="@var{length}"/>
40483 A block indicating trace state variable numbered @var{number} has been
40487 <tvar id="@var{number}"/>
40492 The formal DTD for the traceframe info format is given below:
40495 <!ELEMENT traceframe-info (memory | tvar)* >
40496 <!ATTLIST traceframe-info version CDATA #FIXED "1.0">
40498 <!ELEMENT memory EMPTY>
40499 <!ATTLIST memory start CDATA #REQUIRED
40500 length CDATA #REQUIRED>
40502 <!ATTLIST tvar id CDATA #REQUIRED>
40505 @node Branch Trace Format
40506 @section Branch Trace Format
40507 @cindex branch trace format
40509 In order to display the branch trace of an inferior thread,
40510 @value{GDBN} needs to obtain the list of branches. This list is
40511 represented as list of sequential code blocks that are connected via
40512 branches. The code in each block has been executed sequentially.
40514 This list is obtained using the @samp{qXfer:btrace:read}
40515 (@pxref{qXfer btrace read}) packet and is an XML document.
40517 @value{GDBN} must be linked with the Expat library to support XML
40518 traceframe info discovery. @xref{Expat}.
40520 The top-level structure of the document is shown below:
40523 <?xml version="1.0"?>
40525 PUBLIC "+//IDN gnu.org//DTD GDB Branch Trace V1.0//EN"
40526 "http://sourceware.org/gdb/gdb-btrace.dtd">
40535 A block of sequentially executed instructions starting at @var{begin}
40536 and ending at @var{end}:
40539 <block begin="@var{begin}" end="@var{end}"/>
40544 The formal DTD for the branch trace format is given below:
40547 <!ELEMENT btrace (block* | pt) >
40548 <!ATTLIST btrace version CDATA #FIXED "1.0">
40550 <!ELEMENT block EMPTY>
40551 <!ATTLIST block begin CDATA #REQUIRED
40552 end CDATA #REQUIRED>
40554 <!ELEMENT pt (pt-config?, raw?)>
40556 <!ELEMENT pt-config (cpu?)>
40558 <!ELEMENT cpu EMPTY>
40559 <!ATTLIST cpu vendor CDATA #REQUIRED
40560 family CDATA #REQUIRED
40561 model CDATA #REQUIRED
40562 stepping CDATA #REQUIRED>
40564 <!ELEMENT raw (#PCDATA)>
40567 @node Branch Trace Configuration Format
40568 @section Branch Trace Configuration Format
40569 @cindex branch trace configuration format
40571 For each inferior thread, @value{GDBN} can obtain the branch trace
40572 configuration using the @samp{qXfer:btrace-conf:read}
40573 (@pxref{qXfer btrace-conf read}) packet.
40575 The configuration describes the branch trace format and configuration
40576 settings for that format. The following information is described:
40580 This thread uses the @dfn{Branch Trace Store} (@acronym{BTS}) format.
40583 The size of the @acronym{BTS} ring buffer in bytes.
40586 This thread uses the @dfn{Intel Processor Trace} (@acronym{Intel
40590 The size of the @acronym{Intel PT} ring buffer in bytes.
40594 @value{GDBN} must be linked with the Expat library to support XML
40595 branch trace configuration discovery. @xref{Expat}.
40597 The formal DTD for the branch trace configuration format is given below:
40600 <!ELEMENT btrace-conf (bts?, pt?)>
40601 <!ATTLIST btrace-conf version CDATA #FIXED "1.0">
40603 <!ELEMENT bts EMPTY>
40604 <!ATTLIST bts size CDATA #IMPLIED>
40606 <!ELEMENT pt EMPTY>
40607 <!ATTLIST pt size CDATA #IMPLIED>
40610 @include agentexpr.texi
40612 @node Target Descriptions
40613 @appendix Target Descriptions
40614 @cindex target descriptions
40616 One of the challenges of using @value{GDBN} to debug embedded systems
40617 is that there are so many minor variants of each processor
40618 architecture in use. It is common practice for vendors to start with
40619 a standard processor core --- ARM, PowerPC, or @acronym{MIPS}, for example ---
40620 and then make changes to adapt it to a particular market niche. Some
40621 architectures have hundreds of variants, available from dozens of
40622 vendors. This leads to a number of problems:
40626 With so many different customized processors, it is difficult for
40627 the @value{GDBN} maintainers to keep up with the changes.
40629 Since individual variants may have short lifetimes or limited
40630 audiences, it may not be worthwhile to carry information about every
40631 variant in the @value{GDBN} source tree.
40633 When @value{GDBN} does support the architecture of the embedded system
40634 at hand, the task of finding the correct architecture name to give the
40635 @command{set architecture} command can be error-prone.
40638 To address these problems, the @value{GDBN} remote protocol allows a
40639 target system to not only identify itself to @value{GDBN}, but to
40640 actually describe its own features. This lets @value{GDBN} support
40641 processor variants it has never seen before --- to the extent that the
40642 descriptions are accurate, and that @value{GDBN} understands them.
40644 @value{GDBN} must be linked with the Expat library to support XML
40645 target descriptions. @xref{Expat}.
40648 * Retrieving Descriptions:: How descriptions are fetched from a target.
40649 * Target Description Format:: The contents of a target description.
40650 * Predefined Target Types:: Standard types available for target
40652 * Enum Target Types:: How to define enum target types.
40653 * Standard Target Features:: Features @value{GDBN} knows about.
40656 @node Retrieving Descriptions
40657 @section Retrieving Descriptions
40659 Target descriptions can be read from the target automatically, or
40660 specified by the user manually. The default behavior is to read the
40661 description from the target. @value{GDBN} retrieves it via the remote
40662 protocol using @samp{qXfer} requests (@pxref{General Query Packets,
40663 qXfer}). The @var{annex} in the @samp{qXfer} packet will be
40664 @samp{target.xml}. The contents of the @samp{target.xml} annex are an
40665 XML document, of the form described in @ref{Target Description
40668 Alternatively, you can specify a file to read for the target description.
40669 If a file is set, the target will not be queried. The commands to
40670 specify a file are:
40673 @cindex set tdesc filename
40674 @item set tdesc filename @var{path}
40675 Read the target description from @var{path}.
40677 @cindex unset tdesc filename
40678 @item unset tdesc filename
40679 Do not read the XML target description from a file. @value{GDBN}
40680 will use the description supplied by the current target.
40682 @cindex show tdesc filename
40683 @item show tdesc filename
40684 Show the filename to read for a target description, if any.
40688 @node Target Description Format
40689 @section Target Description Format
40690 @cindex target descriptions, XML format
40692 A target description annex is an @uref{http://www.w3.org/XML/, XML}
40693 document which complies with the Document Type Definition provided in
40694 the @value{GDBN} sources in @file{gdb/features/gdb-target.dtd}. This
40695 means you can use generally available tools like @command{xmllint} to
40696 check that your feature descriptions are well-formed and valid.
40697 However, to help people unfamiliar with XML write descriptions for
40698 their targets, we also describe the grammar here.
40700 Target descriptions can identify the architecture of the remote target
40701 and (for some architectures) provide information about custom register
40702 sets. They can also identify the OS ABI of the remote target.
40703 @value{GDBN} can use this information to autoconfigure for your
40704 target, or to warn you if you connect to an unsupported target.
40706 Here is a simple target description:
40709 <target version="1.0">
40710 <architecture>i386:x86-64</architecture>
40715 This minimal description only says that the target uses
40716 the x86-64 architecture.
40718 A target description has the following overall form, with [ ] marking
40719 optional elements and @dots{} marking repeatable elements. The elements
40720 are explained further below.
40723 <?xml version="1.0"?>
40724 <!DOCTYPE target SYSTEM "gdb-target.dtd">
40725 <target version="1.0">
40726 @r{[}@var{architecture}@r{]}
40727 @r{[}@var{osabi}@r{]}
40728 @r{[}@var{compatible}@r{]}
40729 @r{[}@var{feature}@dots{}@r{]}
40734 The description is generally insensitive to whitespace and line
40735 breaks, under the usual common-sense rules. The XML version
40736 declaration and document type declaration can generally be omitted
40737 (@value{GDBN} does not require them), but specifying them may be
40738 useful for XML validation tools. The @samp{version} attribute for
40739 @samp{<target>} may also be omitted, but we recommend
40740 including it; if future versions of @value{GDBN} use an incompatible
40741 revision of @file{gdb-target.dtd}, they will detect and report
40742 the version mismatch.
40744 @subsection Inclusion
40745 @cindex target descriptions, inclusion
40748 @cindex <xi:include>
40751 It can sometimes be valuable to split a target description up into
40752 several different annexes, either for organizational purposes, or to
40753 share files between different possible target descriptions. You can
40754 divide a description into multiple files by replacing any element of
40755 the target description with an inclusion directive of the form:
40758 <xi:include href="@var{document}"/>
40762 When @value{GDBN} encounters an element of this form, it will retrieve
40763 the named XML @var{document}, and replace the inclusion directive with
40764 the contents of that document. If the current description was read
40765 using @samp{qXfer}, then so will be the included document;
40766 @var{document} will be interpreted as the name of an annex. If the
40767 current description was read from a file, @value{GDBN} will look for
40768 @var{document} as a file in the same directory where it found the
40769 original description.
40771 @subsection Architecture
40772 @cindex <architecture>
40774 An @samp{<architecture>} element has this form:
40777 <architecture>@var{arch}</architecture>
40780 @var{arch} is one of the architectures from the set accepted by
40781 @code{set architecture} (@pxref{Targets, ,Specifying a Debugging Target}).
40784 @cindex @code{<osabi>}
40786 This optional field was introduced in @value{GDBN} version 7.0.
40787 Previous versions of @value{GDBN} ignore it.
40789 An @samp{<osabi>} element has this form:
40792 <osabi>@var{abi-name}</osabi>
40795 @var{abi-name} is an OS ABI name from the same selection accepted by
40796 @w{@code{set osabi}} (@pxref{ABI, ,Configuring the Current ABI}).
40798 @subsection Compatible Architecture
40799 @cindex @code{<compatible>}
40801 This optional field was introduced in @value{GDBN} version 7.0.
40802 Previous versions of @value{GDBN} ignore it.
40804 A @samp{<compatible>} element has this form:
40807 <compatible>@var{arch}</compatible>
40810 @var{arch} is one of the architectures from the set accepted by
40811 @code{set architecture} (@pxref{Targets, ,Specifying a Debugging Target}).
40813 A @samp{<compatible>} element is used to specify that the target
40814 is able to run binaries in some other than the main target architecture
40815 given by the @samp{<architecture>} element. For example, on the
40816 Cell Broadband Engine, the main architecture is @code{powerpc:common}
40817 or @code{powerpc:common64}, but the system is able to run binaries
40818 in the @code{spu} architecture as well. The way to describe this
40819 capability with @samp{<compatible>} is as follows:
40822 <architecture>powerpc:common</architecture>
40823 <compatible>spu</compatible>
40826 @subsection Features
40829 Each @samp{<feature>} describes some logical portion of the target
40830 system. Features are currently used to describe available CPU
40831 registers and the types of their contents. A @samp{<feature>} element
40835 <feature name="@var{name}">
40836 @r{[}@var{type}@dots{}@r{]}
40842 Each feature's name should be unique within the description. The name
40843 of a feature does not matter unless @value{GDBN} has some special
40844 knowledge of the contents of that feature; if it does, the feature
40845 should have its standard name. @xref{Standard Target Features}.
40849 Any register's value is a collection of bits which @value{GDBN} must
40850 interpret. The default interpretation is a two's complement integer,
40851 but other types can be requested by name in the register description.
40852 Some predefined types are provided by @value{GDBN} (@pxref{Predefined
40853 Target Types}), and the description can define additional composite
40856 Each type element must have an @samp{id} attribute, which gives
40857 a unique (within the containing @samp{<feature>}) name to the type.
40858 Types must be defined before they are used.
40861 Some targets offer vector registers, which can be treated as arrays
40862 of scalar elements. These types are written as @samp{<vector>} elements,
40863 specifying the array element type, @var{type}, and the number of elements,
40867 <vector id="@var{id}" type="@var{type}" count="@var{count}"/>
40871 If a register's value is usefully viewed in multiple ways, define it
40872 with a union type containing the useful representations. The
40873 @samp{<union>} element contains one or more @samp{<field>} elements,
40874 each of which has a @var{name} and a @var{type}:
40877 <union id="@var{id}">
40878 <field name="@var{name}" type="@var{type}"/>
40885 If a register's value is composed from several separate values, define
40886 it with either a structure type or a flags type.
40887 A flags type may only contain bitfields.
40888 A structure type may either contain only bitfields or contain no bitfields.
40889 If the value contains only bitfields, its total size in bytes must be
40892 Non-bitfield values have a @var{name} and @var{type}.
40895 <struct id="@var{id}">
40896 <field name="@var{name}" type="@var{type}"/>
40901 Both @var{name} and @var{type} values are required.
40902 No implicit padding is added.
40904 Bitfield values have a @var{name}, @var{start}, @var{end} and @var{type}.
40907 <struct id="@var{id}" size="@var{size}">
40908 <field name="@var{name}" start="@var{start}" end="@var{end}" type="@var{type}"/>
40914 <flags id="@var{id}" size="@var{size}">
40915 <field name="@var{name}" start="@var{start}" end="@var{end}" type="@var{type}"/>
40920 The @var{name} value is required.
40921 Bitfield values may be named with the empty string, @samp{""},
40922 in which case the field is ``filler'' and its value is not printed.
40923 Not all bits need to be specified, so ``filler'' fields are optional.
40925 The @var{start} and @var{end} values are required, and @var{type}
40927 The field's @var{start} must be less than or equal to its @var{end},
40928 and zero represents the least significant bit.
40930 The default value of @var{type} is @code{bool} for single bit fields,
40931 and an unsigned integer otherwise.
40933 Which to choose? Structures or flags?
40935 Registers defined with @samp{flags} have these advantages over
40936 defining them with @samp{struct}:
40940 Arithmetic may be performed on them as if they were integers.
40942 They are printed in a more readable fashion.
40945 Registers defined with @samp{struct} have one advantage over
40946 defining them with @samp{flags}:
40950 One can fetch individual fields like in @samp{C}.
40953 (gdb) print $my_struct_reg.field3
40959 @subsection Registers
40962 Each register is represented as an element with this form:
40965 <reg name="@var{name}"
40966 bitsize="@var{size}"
40967 @r{[}regnum="@var{num}"@r{]}
40968 @r{[}save-restore="@var{save-restore}"@r{]}
40969 @r{[}type="@var{type}"@r{]}
40970 @r{[}group="@var{group}"@r{]}/>
40974 The components are as follows:
40979 The register's name; it must be unique within the target description.
40982 The register's size, in bits.
40985 The register's number. If omitted, a register's number is one greater
40986 than that of the previous register (either in the current feature or in
40987 a preceding feature); the first register in the target description
40988 defaults to zero. This register number is used to read or write
40989 the register; e.g.@: it is used in the remote @code{p} and @code{P}
40990 packets, and registers appear in the @code{g} and @code{G} packets
40991 in order of increasing register number.
40994 Whether the register should be preserved across inferior function
40995 calls; this must be either @code{yes} or @code{no}. The default is
40996 @code{yes}, which is appropriate for most registers except for
40997 some system control registers; this is not related to the target's
41001 The type of the register. It may be a predefined type, a type
41002 defined in the current feature, or one of the special types @code{int}
41003 and @code{float}. @code{int} is an integer type of the correct size
41004 for @var{bitsize}, and @code{float} is a floating point type (in the
41005 architecture's normal floating point format) of the correct size for
41006 @var{bitsize}. The default is @code{int}.
41009 The register group to which this register belongs. It must
41010 be either @code{general}, @code{float}, or @code{vector}. If no
41011 @var{group} is specified, @value{GDBN} will not display the register
41012 in @code{info registers}.
41016 @node Predefined Target Types
41017 @section Predefined Target Types
41018 @cindex target descriptions, predefined types
41020 Type definitions in the self-description can build up composite types
41021 from basic building blocks, but can not define fundamental types. Instead,
41022 standard identifiers are provided by @value{GDBN} for the fundamental
41023 types. The currently supported types are:
41028 Boolean type, occupying a single bit.
41035 Signed integer types holding the specified number of bits.
41042 Unsigned integer types holding the specified number of bits.
41046 Pointers to unspecified code and data. The program counter and
41047 any dedicated return address register may be marked as code
41048 pointers; printing a code pointer converts it into a symbolic
41049 address. The stack pointer and any dedicated address registers
41050 may be marked as data pointers.
41053 Single precision IEEE floating point.
41056 Double precision IEEE floating point.
41059 The 12-byte extended precision format used by ARM FPA registers.
41062 The 10-byte extended precision format used by x87 registers.
41065 32bit @sc{eflags} register used by x86.
41068 32bit @sc{mxcsr} register used by x86.
41072 @node Enum Target Types
41073 @section Enum Target Types
41074 @cindex target descriptions, enum types
41076 Enum target types are useful in @samp{struct} and @samp{flags}
41077 register descriptions. @xref{Target Description Format}.
41079 Enum types have a name, size and a list of name/value pairs.
41082 <enum id="@var{id}" size="@var{size}">
41083 <evalue name="@var{name}" value="@var{value}"/>
41088 Enums must be defined before they are used.
41091 <enum id="levels_type" size="4">
41092 <evalue name="low" value="0"/>
41093 <evalue name="high" value="1"/>
41095 <flags id="flags_type" size="4">
41096 <field name="X" start="0"/>
41097 <field name="LEVEL" start="1" end="1" type="levels_type"/>
41099 <reg name="flags" bitsize="32" type="flags_type"/>
41102 Given that description, a value of 3 for the @samp{flags} register
41103 would be printed as:
41106 (gdb) info register flags
41107 flags 0x3 [ X LEVEL=high ]
41110 @node Standard Target Features
41111 @section Standard Target Features
41112 @cindex target descriptions, standard features
41114 A target description must contain either no registers or all the
41115 target's registers. If the description contains no registers, then
41116 @value{GDBN} will assume a default register layout, selected based on
41117 the architecture. If the description contains any registers, the
41118 default layout will not be used; the standard registers must be
41119 described in the target description, in such a way that @value{GDBN}
41120 can recognize them.
41122 This is accomplished by giving specific names to feature elements
41123 which contain standard registers. @value{GDBN} will look for features
41124 with those names and verify that they contain the expected registers;
41125 if any known feature is missing required registers, or if any required
41126 feature is missing, @value{GDBN} will reject the target
41127 description. You can add additional registers to any of the
41128 standard features --- @value{GDBN} will display them just as if
41129 they were added to an unrecognized feature.
41131 This section lists the known features and their expected contents.
41132 Sample XML documents for these features are included in the
41133 @value{GDBN} source tree, in the directory @file{gdb/features}.
41135 Names recognized by @value{GDBN} should include the name of the
41136 company or organization which selected the name, and the overall
41137 architecture to which the feature applies; so e.g.@: the feature
41138 containing ARM core registers is named @samp{org.gnu.gdb.arm.core}.
41140 The names of registers are not case sensitive for the purpose
41141 of recognizing standard features, but @value{GDBN} will only display
41142 registers using the capitalization used in the description.
41145 * AArch64 Features::
41149 * MicroBlaze Features::
41153 * Nios II Features::
41154 * PowerPC Features::
41155 * S/390 and System z Features::
41161 @node AArch64 Features
41162 @subsection AArch64 Features
41163 @cindex target descriptions, AArch64 features
41165 The @samp{org.gnu.gdb.aarch64.core} feature is required for AArch64
41166 targets. It should contain registers @samp{x0} through @samp{x30},
41167 @samp{sp}, @samp{pc}, and @samp{cpsr}.
41169 The @samp{org.gnu.gdb.aarch64.fpu} feature is optional. If present,
41170 it should contain registers @samp{v0} through @samp{v31}, @samp{fpsr},
41174 @subsection ARC Features
41175 @cindex target descriptions, ARC Features
41177 ARC processors are highly configurable, so even core registers and their number
41178 are not completely predetermined. In addition flags and PC registers which are
41179 important to @value{GDBN} are not ``core'' registers in ARC. It is required
41180 that one of the core registers features is present.
41181 @samp{org.gnu.gdb.arc.aux-minimal} feature is mandatory.
41183 The @samp{org.gnu.gdb.arc.core.v2} feature is required for ARC EM and ARC HS
41184 targets with a normal register file. It should contain registers @samp{r0}
41185 through @samp{r25}, @samp{gp}, @samp{fp}, @samp{sp}, @samp{r30}, @samp{blink},
41186 @samp{lp_count} and @samp{pcl}. This feature may contain register @samp{ilink}
41187 and any of extension core registers @samp{r32} through @samp{r59/acch}.
41188 @samp{ilink} and extension core registers are not available to read/write, when
41189 debugging GNU/Linux applications, thus @samp{ilink} is made optional.
41191 The @samp{org.gnu.gdb.arc.core-reduced.v2} feature is required for ARC EM and
41192 ARC HS targets with a reduced register file. It should contain registers
41193 @samp{r0} through @samp{r3}, @samp{r10} through @samp{r15}, @samp{gp},
41194 @samp{fp}, @samp{sp}, @samp{r30}, @samp{blink}, @samp{lp_count} and @samp{pcl}.
41195 This feature may contain register @samp{ilink} and any of extension core
41196 registers @samp{r32} through @samp{r59/acch}.
41198 The @samp{org.gnu.gdb.arc.core.arcompact} feature is required for ARCompact
41199 targets with a normal register file. It should contain registers @samp{r0}
41200 through @samp{r25}, @samp{gp}, @samp{fp}, @samp{sp}, @samp{r30}, @samp{blink},
41201 @samp{lp_count} and @samp{pcl}. This feature may contain registers
41202 @samp{ilink1}, @samp{ilink2} and any of extension core registers @samp{r32}
41203 through @samp{r59/acch}. @samp{ilink1} and @samp{ilink2} and extension core
41204 registers are not available when debugging GNU/Linux applications. The only
41205 difference with @samp{org.gnu.gdb.arc.core.v2} feature is in the names of
41206 @samp{ilink1} and @samp{ilink2} registers and that @samp{r30} is mandatory in
41207 ARC v2, but @samp{ilink2} is optional on ARCompact.
41209 The @samp{org.gnu.gdb.arc.aux-minimal} feature is required for all ARC
41210 targets. It should contain registers @samp{pc} and @samp{status32}.
41213 @subsection ARM Features
41214 @cindex target descriptions, ARM features
41216 The @samp{org.gnu.gdb.arm.core} feature is required for non-M-profile
41218 It should contain registers @samp{r0} through @samp{r13}, @samp{sp},
41219 @samp{lr}, @samp{pc}, and @samp{cpsr}.
41221 For M-profile targets (e.g. Cortex-M3), the @samp{org.gnu.gdb.arm.core}
41222 feature is replaced by @samp{org.gnu.gdb.arm.m-profile}. It should contain
41223 registers @samp{r0} through @samp{r13}, @samp{sp}, @samp{lr}, @samp{pc},
41226 The @samp{org.gnu.gdb.arm.fpa} feature is optional. If present, it
41227 should contain registers @samp{f0} through @samp{f7} and @samp{fps}.
41229 The @samp{org.gnu.gdb.xscale.iwmmxt} feature is optional. If present,
41230 it should contain at least registers @samp{wR0} through @samp{wR15} and
41231 @samp{wCGR0} through @samp{wCGR3}. The @samp{wCID}, @samp{wCon},
41232 @samp{wCSSF}, and @samp{wCASF} registers are optional.
41234 The @samp{org.gnu.gdb.arm.vfp} feature is optional. If present, it
41235 should contain at least registers @samp{d0} through @samp{d15}. If
41236 they are present, @samp{d16} through @samp{d31} should also be included.
41237 @value{GDBN} will synthesize the single-precision registers from
41238 halves of the double-precision registers.
41240 The @samp{org.gnu.gdb.arm.neon} feature is optional. It does not
41241 need to contain registers; it instructs @value{GDBN} to display the
41242 VFP double-precision registers as vectors and to synthesize the
41243 quad-precision registers from pairs of double-precision registers.
41244 If this feature is present, @samp{org.gnu.gdb.arm.vfp} must also
41245 be present and include 32 double-precision registers.
41247 @node i386 Features
41248 @subsection i386 Features
41249 @cindex target descriptions, i386 features
41251 The @samp{org.gnu.gdb.i386.core} feature is required for i386/amd64
41252 targets. It should describe the following registers:
41256 @samp{eax} through @samp{edi} plus @samp{eip} for i386
41258 @samp{rax} through @samp{r15} plus @samp{rip} for amd64
41260 @samp{eflags}, @samp{cs}, @samp{ss}, @samp{ds}, @samp{es},
41261 @samp{fs}, @samp{gs}
41263 @samp{st0} through @samp{st7}
41265 @samp{fctrl}, @samp{fstat}, @samp{ftag}, @samp{fiseg}, @samp{fioff},
41266 @samp{foseg}, @samp{fooff} and @samp{fop}
41269 The register sets may be different, depending on the target.
41271 The @samp{org.gnu.gdb.i386.sse} feature is optional. It should
41272 describe registers:
41276 @samp{xmm0} through @samp{xmm7} for i386
41278 @samp{xmm0} through @samp{xmm15} for amd64
41283 The @samp{org.gnu.gdb.i386.avx} feature is optional and requires the
41284 @samp{org.gnu.gdb.i386.sse} feature. It should
41285 describe the upper 128 bits of @sc{ymm} registers:
41289 @samp{ymm0h} through @samp{ymm7h} for i386
41291 @samp{ymm0h} through @samp{ymm15h} for amd64
41294 The @samp{org.gnu.gdb.i386.mpx} is an optional feature representing Intel
41295 Memory Protection Extension (MPX). It should describe the following registers:
41299 @samp{bnd0raw} through @samp{bnd3raw} for i386 and amd64.
41301 @samp{bndcfgu} and @samp{bndstatus} for i386 and amd64.
41304 The @samp{org.gnu.gdb.i386.linux} feature is optional. It should
41305 describe a single register, @samp{orig_eax}.
41307 The @samp{org.gnu.gdb.i386.segments} feature is optional. It should
41308 describe two system registers: @samp{fs_base} and @samp{gs_base}.
41310 The @samp{org.gnu.gdb.i386.avx512} feature is optional and requires the
41311 @samp{org.gnu.gdb.i386.avx} feature. It should
41312 describe additional @sc{xmm} registers:
41316 @samp{xmm16h} through @samp{xmm31h}, only valid for amd64.
41319 It should describe the upper 128 bits of additional @sc{ymm} registers:
41323 @samp{ymm16h} through @samp{ymm31h}, only valid for amd64.
41327 describe the upper 256 bits of @sc{zmm} registers:
41331 @samp{zmm0h} through @samp{zmm7h} for i386.
41333 @samp{zmm0h} through @samp{zmm15h} for amd64.
41337 describe the additional @sc{zmm} registers:
41341 @samp{zmm16h} through @samp{zmm31h}, only valid for amd64.
41344 The @samp{org.gnu.gdb.i386.pkeys} feature is optional. It should
41345 describe a single register, @samp{pkru}. It is a 32-bit register
41346 valid for i386 and amd64.
41348 @node MicroBlaze Features
41349 @subsection MicroBlaze Features
41350 @cindex target descriptions, MicroBlaze features
41352 The @samp{org.gnu.gdb.microblaze.core} feature is required for MicroBlaze
41353 targets. It should contain registers @samp{r0} through @samp{r31},
41354 @samp{rpc}, @samp{rmsr}, @samp{rear}, @samp{resr}, @samp{rfsr}, @samp{rbtr},
41355 @samp{rpvr}, @samp{rpvr1} through @samp{rpvr11}, @samp{redr}, @samp{rpid},
41356 @samp{rzpr}, @samp{rtlbx}, @samp{rtlbsx}, @samp{rtlblo}, and @samp{rtlbhi}.
41358 The @samp{org.gnu.gdb.microblaze.stack-protect} feature is optional.
41359 If present, it should contain registers @samp{rshr} and @samp{rslr}
41361 @node MIPS Features
41362 @subsection @acronym{MIPS} Features
41363 @cindex target descriptions, @acronym{MIPS} features
41365 The @samp{org.gnu.gdb.mips.cpu} feature is required for @acronym{MIPS} targets.
41366 It should contain registers @samp{r0} through @samp{r31}, @samp{lo},
41367 @samp{hi}, and @samp{pc}. They may be 32-bit or 64-bit depending
41370 The @samp{org.gnu.gdb.mips.cp0} feature is also required. It should
41371 contain at least the @samp{status}, @samp{badvaddr}, and @samp{cause}
41372 registers. They may be 32-bit or 64-bit depending on the target.
41374 The @samp{org.gnu.gdb.mips.fpu} feature is currently required, though
41375 it may be optional in a future version of @value{GDBN}. It should
41376 contain registers @samp{f0} through @samp{f31}, @samp{fcsr}, and
41377 @samp{fir}. They may be 32-bit or 64-bit depending on the target.
41379 The @samp{org.gnu.gdb.mips.dsp} feature is optional. It should
41380 contain registers @samp{hi1} through @samp{hi3}, @samp{lo1} through
41381 @samp{lo3}, and @samp{dspctl}. The @samp{dspctl} register should
41382 be 32-bit and the rest may be 32-bit or 64-bit depending on the target.
41384 The @samp{org.gnu.gdb.mips.linux} feature is optional. It should
41385 contain a single register, @samp{restart}, which is used by the
41386 Linux kernel to control restartable syscalls.
41388 @node M68K Features
41389 @subsection M68K Features
41390 @cindex target descriptions, M68K features
41393 @item @samp{org.gnu.gdb.m68k.core}
41394 @itemx @samp{org.gnu.gdb.coldfire.core}
41395 @itemx @samp{org.gnu.gdb.fido.core}
41396 One of those features must be always present.
41397 The feature that is present determines which flavor of m68k is
41398 used. The feature that is present should contain registers
41399 @samp{d0} through @samp{d7}, @samp{a0} through @samp{a5}, @samp{fp},
41400 @samp{sp}, @samp{ps} and @samp{pc}.
41402 @item @samp{org.gnu.gdb.coldfire.fp}
41403 This feature is optional. If present, it should contain registers
41404 @samp{fp0} through @samp{fp7}, @samp{fpcontrol}, @samp{fpstatus} and
41408 @node NDS32 Features
41409 @subsection NDS32 Features
41410 @cindex target descriptions, NDS32 features
41412 The @samp{org.gnu.gdb.nds32.core} feature is required for NDS32
41413 targets. It should contain at least registers @samp{r0} through
41414 @samp{r10}, @samp{r15}, @samp{fp}, @samp{gp}, @samp{lp}, @samp{sp},
41417 The @samp{org.gnu.gdb.nds32.fpu} feature is optional. If present,
41418 it should contain 64-bit double-precision floating-point registers
41419 @samp{fd0} through @emph{fdN}, which should be @samp{fd3}, @samp{fd7},
41420 @samp{fd15}, or @samp{fd31} based on the FPU configuration implemented.
41422 @emph{Note:} The first sixteen 64-bit double-precision floating-point
41423 registers are overlapped with the thirty-two 32-bit single-precision
41424 floating-point registers. The 32-bit single-precision registers, if
41425 not being listed explicitly, will be synthesized from halves of the
41426 overlapping 64-bit double-precision registers. Listing 32-bit
41427 single-precision registers explicitly is deprecated, and the
41428 support to it could be totally removed some day.
41430 @node Nios II Features
41431 @subsection Nios II Features
41432 @cindex target descriptions, Nios II features
41434 The @samp{org.gnu.gdb.nios2.cpu} feature is required for Nios II
41435 targets. It should contain the 32 core registers (@samp{zero},
41436 @samp{at}, @samp{r2} through @samp{r23}, @samp{et} through @samp{ra}),
41437 @samp{pc}, and the 16 control registers (@samp{status} through
41440 @node PowerPC Features
41441 @subsection PowerPC Features
41442 @cindex target descriptions, PowerPC features
41444 The @samp{org.gnu.gdb.power.core} feature is required for PowerPC
41445 targets. It should contain registers @samp{r0} through @samp{r31},
41446 @samp{pc}, @samp{msr}, @samp{cr}, @samp{lr}, @samp{ctr}, and
41447 @samp{xer}. They may be 32-bit or 64-bit depending on the target.
41449 The @samp{org.gnu.gdb.power.fpu} feature is optional. It should
41450 contain registers @samp{f0} through @samp{f31} and @samp{fpscr}.
41452 The @samp{org.gnu.gdb.power.altivec} feature is optional. It should
41453 contain registers @samp{vr0} through @samp{vr31}, @samp{vscr},
41456 The @samp{org.gnu.gdb.power.vsx} feature is optional. It should
41457 contain registers @samp{vs0h} through @samp{vs31h}. @value{GDBN}
41458 will combine these registers with the floating point registers
41459 (@samp{f0} through @samp{f31}) and the altivec registers (@samp{vr0}
41460 through @samp{vr31}) to present the 128-bit wide registers @samp{vs0}
41461 through @samp{vs63}, the set of vector registers for POWER7.
41463 The @samp{org.gnu.gdb.power.spe} feature is optional. It should
41464 contain registers @samp{ev0h} through @samp{ev31h}, @samp{acc}, and
41465 @samp{spefscr}. SPE targets should provide 32-bit registers in
41466 @samp{org.gnu.gdb.power.core} and provide the upper halves in
41467 @samp{ev0h} through @samp{ev31h}. @value{GDBN} will combine
41468 these to present registers @samp{ev0} through @samp{ev31} to the
41471 @node S/390 and System z Features
41472 @subsection S/390 and System z Features
41473 @cindex target descriptions, S/390 features
41474 @cindex target descriptions, System z features
41476 The @samp{org.gnu.gdb.s390.core} feature is required for S/390 and
41477 System z targets. It should contain the PSW and the 16 general
41478 registers. In particular, System z targets should provide the 64-bit
41479 registers @samp{pswm}, @samp{pswa}, and @samp{r0} through @samp{r15}.
41480 S/390 targets should provide the 32-bit versions of these registers.
41481 A System z target that runs in 31-bit addressing mode should provide
41482 32-bit versions of @samp{pswm} and @samp{pswa}, as well as the general
41483 register's upper halves @samp{r0h} through @samp{r15h}, and their
41484 lower halves @samp{r0l} through @samp{r15l}.
41486 The @samp{org.gnu.gdb.s390.fpr} feature is required. It should
41487 contain the 64-bit registers @samp{f0} through @samp{f15}, and
41490 The @samp{org.gnu.gdb.s390.acr} feature is required. It should
41491 contain the 32-bit registers @samp{acr0} through @samp{acr15}.
41493 The @samp{org.gnu.gdb.s390.linux} feature is optional. It should
41494 contain the register @samp{orig_r2}, which is 64-bit wide on System z
41495 targets and 32-bit otherwise. In addition, the feature may contain
41496 the @samp{last_break} register, whose width depends on the addressing
41497 mode, as well as the @samp{system_call} register, which is always
41500 The @samp{org.gnu.gdb.s390.tdb} feature is optional. It should
41501 contain the 64-bit registers @samp{tdb0}, @samp{tac}, @samp{tct},
41502 @samp{atia}, and @samp{tr0} through @samp{tr15}.
41504 The @samp{org.gnu.gdb.s390.vx} feature is optional. It should contain
41505 64-bit wide registers @samp{v0l} through @samp{v15l}, which will be
41506 combined by @value{GDBN} with the floating point registers @samp{f0}
41507 through @samp{f15} to present the 128-bit wide vector registers
41508 @samp{v0} through @samp{v15}. In addition, this feature should
41509 contain the 128-bit wide vector registers @samp{v16} through
41512 @node Sparc Features
41513 @subsection Sparc Features
41514 @cindex target descriptions, sparc32 features
41515 @cindex target descriptions, sparc64 features
41516 The @samp{org.gnu.gdb.sparc.cpu} feature is required for sparc32/sparc64
41517 targets. It should describe the following registers:
41521 @samp{g0} through @samp{g7}
41523 @samp{o0} through @samp{o7}
41525 @samp{l0} through @samp{l7}
41527 @samp{i0} through @samp{i7}
41530 They may be 32-bit or 64-bit depending on the target.
41532 Also the @samp{org.gnu.gdb.sparc.fpu} feature is required for sparc32/sparc64
41533 targets. It should describe the following registers:
41537 @samp{f0} through @samp{f31}
41539 @samp{f32} through @samp{f62} for sparc64
41542 The @samp{org.gnu.gdb.sparc.cp0} feature is required for sparc32/sparc64
41543 targets. It should describe the following registers:
41547 @samp{y}, @samp{psr}, @samp{wim}, @samp{tbr}, @samp{pc}, @samp{npc},
41548 @samp{fsr}, and @samp{csr} for sparc32
41550 @samp{pc}, @samp{npc}, @samp{state}, @samp{fsr}, @samp{fprs}, and @samp{y}
41554 @node TIC6x Features
41555 @subsection TMS320C6x Features
41556 @cindex target descriptions, TIC6x features
41557 @cindex target descriptions, TMS320C6x features
41558 The @samp{org.gnu.gdb.tic6x.core} feature is required for TMS320C6x
41559 targets. It should contain registers @samp{A0} through @samp{A15},
41560 registers @samp{B0} through @samp{B15}, @samp{CSR} and @samp{PC}.
41562 The @samp{org.gnu.gdb.tic6x.gp} feature is optional. It should
41563 contain registers @samp{A16} through @samp{A31} and @samp{B16}
41564 through @samp{B31}.
41566 The @samp{org.gnu.gdb.tic6x.c6xp} feature is optional. It should
41567 contain registers @samp{TSR}, @samp{ILC} and @samp{RILC}.
41569 @node Operating System Information
41570 @appendix Operating System Information
41571 @cindex operating system information
41577 Users of @value{GDBN} often wish to obtain information about the state of
41578 the operating system running on the target---for example the list of
41579 processes, or the list of open files. This section describes the
41580 mechanism that makes it possible. This mechanism is similar to the
41581 target features mechanism (@pxref{Target Descriptions}), but focuses
41582 on a different aspect of target.
41584 Operating system information is retrived from the target via the
41585 remote protocol, using @samp{qXfer} requests (@pxref{qXfer osdata
41586 read}). The object name in the request should be @samp{osdata}, and
41587 the @var{annex} identifies the data to be fetched.
41590 @appendixsection Process list
41591 @cindex operating system information, process list
41593 When requesting the process list, the @var{annex} field in the
41594 @samp{qXfer} request should be @samp{processes}. The returned data is
41595 an XML document. The formal syntax of this document is defined in
41596 @file{gdb/features/osdata.dtd}.
41598 An example document is:
41601 <?xml version="1.0"?>
41602 <!DOCTYPE target SYSTEM "osdata.dtd">
41603 <osdata type="processes">
41605 <column name="pid">1</column>
41606 <column name="user">root</column>
41607 <column name="command">/sbin/init</column>
41608 <column name="cores">1,2,3</column>
41613 Each item should include a column whose name is @samp{pid}. The value
41614 of that column should identify the process on the target. The
41615 @samp{user} and @samp{command} columns are optional, and will be
41616 displayed by @value{GDBN}. The @samp{cores} column, if present,
41617 should contain a comma-separated list of cores that this process
41618 is running on. Target may provide additional columns,
41619 which @value{GDBN} currently ignores.
41621 @node Trace File Format
41622 @appendix Trace File Format
41623 @cindex trace file format
41625 The trace file comes in three parts: a header, a textual description
41626 section, and a trace frame section with binary data.
41628 The header has the form @code{\x7fTRACE0\n}. The first byte is
41629 @code{0x7f} so as to indicate that the file contains binary data,
41630 while the @code{0} is a version number that may have different values
41633 The description section consists of multiple lines of @sc{ascii} text
41634 separated by newline characters (@code{0xa}). The lines may include a
41635 variety of optional descriptive or context-setting information, such
41636 as tracepoint definitions or register set size. @value{GDBN} will
41637 ignore any line that it does not recognize. An empty line marks the end
41642 Specifies the size of a register block in bytes. This is equal to the
41643 size of a @code{g} packet payload in the remote protocol. @var{size}
41644 is an ascii decimal number. There should be only one such line in
41645 a single trace file.
41647 @item status @var{status}
41648 Trace status. @var{status} has the same format as a @code{qTStatus}
41649 remote packet reply. There should be only one such line in a single trace
41652 @item tp @var{payload}
41653 Tracepoint definition. The @var{payload} has the same format as
41654 @code{qTfP}/@code{qTsP} remote packet reply payload. A single tracepoint
41655 may take multiple lines of definition, corresponding to the multiple
41658 @item tsv @var{payload}
41659 Trace state variable definition. The @var{payload} has the same format as
41660 @code{qTfV}/@code{qTsV} remote packet reply payload. A single variable
41661 may take multiple lines of definition, corresponding to the multiple
41664 @item tdesc @var{payload}
41665 Target description in XML format. The @var{payload} is a single line of
41666 the XML file. All such lines should be concatenated together to get
41667 the original XML file. This file is in the same format as @code{qXfer}
41668 @code{features} payload, and corresponds to the main @code{target.xml}
41669 file. Includes are not allowed.
41673 The trace frame section consists of a number of consecutive frames.
41674 Each frame begins with a two-byte tracepoint number, followed by a
41675 four-byte size giving the amount of data in the frame. The data in
41676 the frame consists of a number of blocks, each introduced by a
41677 character indicating its type (at least register, memory, and trace
41678 state variable). The data in this section is raw binary, not a
41679 hexadecimal or other encoding; its endianness matches the target's
41682 @c FIXME bi-arch may require endianness/arch info in description section
41685 @item R @var{bytes}
41686 Register block. The number and ordering of bytes matches that of a
41687 @code{g} packet in the remote protocol. Note that these are the
41688 actual bytes, in target order, not a hexadecimal encoding.
41690 @item M @var{address} @var{length} @var{bytes}...
41691 Memory block. This is a contiguous block of memory, at the 8-byte
41692 address @var{address}, with a 2-byte length @var{length}, followed by
41693 @var{length} bytes.
41695 @item V @var{number} @var{value}
41696 Trace state variable block. This records the 8-byte signed value
41697 @var{value} of trace state variable numbered @var{number}.
41701 Future enhancements of the trace file format may include additional types
41704 @node Index Section Format
41705 @appendix @code{.gdb_index} section format
41706 @cindex .gdb_index section format
41707 @cindex index section format
41709 This section documents the index section that is created by @code{save
41710 gdb-index} (@pxref{Index Files}). The index section is
41711 DWARF-specific; some knowledge of DWARF is assumed in this
41714 The mapped index file format is designed to be directly
41715 @code{mmap}able on any architecture. In most cases, a datum is
41716 represented using a little-endian 32-bit integer value, called an
41717 @code{offset_type}. Big endian machines must byte-swap the values
41718 before using them. Exceptions to this rule are noted. The data is
41719 laid out such that alignment is always respected.
41721 A mapped index consists of several areas, laid out in order.
41725 The file header. This is a sequence of values, of @code{offset_type}
41726 unless otherwise noted:
41730 The version number, currently 8. Versions 1, 2 and 3 are obsolete.
41731 Version 4 uses a different hashing function from versions 5 and 6.
41732 Version 6 includes symbols for inlined functions, whereas versions 4
41733 and 5 do not. Version 7 adds attributes to the CU indices in the
41734 symbol table. Version 8 specifies that symbols from DWARF type units
41735 (@samp{DW_TAG_type_unit}) refer to the type unit's symbol table and not the
41736 compilation unit (@samp{DW_TAG_comp_unit}) using the type.
41738 @value{GDBN} will only read version 4, 5, or 6 indices
41739 by specifying @code{set use-deprecated-index-sections on}.
41740 GDB has a workaround for potentially broken version 7 indices so it is
41741 currently not flagged as deprecated.
41744 The offset, from the start of the file, of the CU list.
41747 The offset, from the start of the file, of the types CU list. Note
41748 that this area can be empty, in which case this offset will be equal
41749 to the next offset.
41752 The offset, from the start of the file, of the address area.
41755 The offset, from the start of the file, of the symbol table.
41758 The offset, from the start of the file, of the constant pool.
41762 The CU list. This is a sequence of pairs of 64-bit little-endian
41763 values, sorted by the CU offset. The first element in each pair is
41764 the offset of a CU in the @code{.debug_info} section. The second
41765 element in each pair is the length of that CU. References to a CU
41766 elsewhere in the map are done using a CU index, which is just the
41767 0-based index into this table. Note that if there are type CUs, then
41768 conceptually CUs and type CUs form a single list for the purposes of
41772 The types CU list. This is a sequence of triplets of 64-bit
41773 little-endian values. In a triplet, the first value is the CU offset,
41774 the second value is the type offset in the CU, and the third value is
41775 the type signature. The types CU list is not sorted.
41778 The address area. The address area consists of a sequence of address
41779 entries. Each address entry has three elements:
41783 The low address. This is a 64-bit little-endian value.
41786 The high address. This is a 64-bit little-endian value. Like
41787 @code{DW_AT_high_pc}, the value is one byte beyond the end.
41790 The CU index. This is an @code{offset_type} value.
41794 The symbol table. This is an open-addressed hash table. The size of
41795 the hash table is always a power of 2.
41797 Each slot in the hash table consists of a pair of @code{offset_type}
41798 values. The first value is the offset of the symbol's name in the
41799 constant pool. The second value is the offset of the CU vector in the
41802 If both values are 0, then this slot in the hash table is empty. This
41803 is ok because while 0 is a valid constant pool index, it cannot be a
41804 valid index for both a string and a CU vector.
41806 The hash value for a table entry is computed by applying an
41807 iterative hash function to the symbol's name. Starting with an
41808 initial value of @code{r = 0}, each (unsigned) character @samp{c} in
41809 the string is incorporated into the hash using the formula depending on the
41814 The formula is @code{r = r * 67 + c - 113}.
41816 @item Versions 5 to 7
41817 The formula is @code{r = r * 67 + tolower (c) - 113}.
41820 The terminating @samp{\0} is not incorporated into the hash.
41822 The step size used in the hash table is computed via
41823 @code{((hash * 17) & (size - 1)) | 1}, where @samp{hash} is the hash
41824 value, and @samp{size} is the size of the hash table. The step size
41825 is used to find the next candidate slot when handling a hash
41828 The names of C@t{++} symbols in the hash table are canonicalized. We
41829 don't currently have a simple description of the canonicalization
41830 algorithm; if you intend to create new index sections, you must read
41834 The constant pool. This is simply a bunch of bytes. It is organized
41835 so that alignment is correct: CU vectors are stored first, followed by
41838 A CU vector in the constant pool is a sequence of @code{offset_type}
41839 values. The first value is the number of CU indices in the vector.
41840 Each subsequent value is the index and symbol attributes of a CU in
41841 the CU list. This element in the hash table is used to indicate which
41842 CUs define the symbol and how the symbol is used.
41843 See below for the format of each CU index+attributes entry.
41845 A string in the constant pool is zero-terminated.
41848 Attributes were added to CU index values in @code{.gdb_index} version 7.
41849 If a symbol has multiple uses within a CU then there is one
41850 CU index+attributes value for each use.
41852 The format of each CU index+attributes entry is as follows
41858 This is the index of the CU in the CU list.
41860 These bits are reserved for future purposes and must be zero.
41862 The kind of the symbol in the CU.
41866 This value is reserved and should not be used.
41867 By reserving zero the full @code{offset_type} value is backwards compatible
41868 with previous versions of the index.
41870 The symbol is a type.
41872 The symbol is a variable or an enum value.
41874 The symbol is a function.
41876 Any other kind of symbol.
41878 These values are reserved.
41882 This bit is zero if the value is global and one if it is static.
41884 The determination of whether a symbol is global or static is complicated.
41885 The authorative reference is the file @file{dwarf2read.c} in
41886 @value{GDBN} sources.
41890 This pseudo-code describes the computation of a symbol's kind and
41891 global/static attributes in the index.
41894 is_external = get_attribute (die, DW_AT_external);
41895 language = get_attribute (cu_die, DW_AT_language);
41898 case DW_TAG_typedef:
41899 case DW_TAG_base_type:
41900 case DW_TAG_subrange_type:
41904 case DW_TAG_enumerator:
41906 is_static = language != CPLUS;
41908 case DW_TAG_subprogram:
41910 is_static = ! (is_external || language == ADA);
41912 case DW_TAG_constant:
41914 is_static = ! is_external;
41916 case DW_TAG_variable:
41918 is_static = ! is_external;
41920 case DW_TAG_namespace:
41924 case DW_TAG_class_type:
41925 case DW_TAG_interface_type:
41926 case DW_TAG_structure_type:
41927 case DW_TAG_union_type:
41928 case DW_TAG_enumeration_type:
41930 is_static = language != CPLUS;
41938 @appendix Manual pages
41942 * gdb man:: The GNU Debugger man page
41943 * gdbserver man:: Remote Server for the GNU Debugger man page
41944 * gcore man:: Generate a core file of a running program
41945 * gdbinit man:: gdbinit scripts
41951 @c man title gdb The GNU Debugger
41953 @c man begin SYNOPSIS gdb
41954 gdb [@option{-help}] [@option{-nh}] [@option{-nx}] [@option{-q}]
41955 [@option{-batch}] [@option{-cd=}@var{dir}] [@option{-f}]
41956 [@option{-b}@w{ }@var{bps}]
41957 [@option{-tty=}@var{dev}] [@option{-s} @var{symfile}]
41958 [@option{-e}@w{ }@var{prog}] [@option{-se}@w{ }@var{prog}]
41959 [@option{-c}@w{ }@var{core}] [@option{-p}@w{ }@var{procID}]
41960 [@option{-x}@w{ }@var{cmds}] [@option{-d}@w{ }@var{dir}]
41961 [@var{prog}|@var{prog} @var{procID}|@var{prog} @var{core}]
41964 @c man begin DESCRIPTION gdb
41965 The purpose of a debugger such as @value{GDBN} is to allow you to see what is
41966 going on ``inside'' another program while it executes -- or what another
41967 program was doing at the moment it crashed.
41969 @value{GDBN} can do four main kinds of things (plus other things in support of
41970 these) to help you catch bugs in the act:
41974 Start your program, specifying anything that might affect its behavior.
41977 Make your program stop on specified conditions.
41980 Examine what has happened, when your program has stopped.
41983 Change things in your program, so you can experiment with correcting the
41984 effects of one bug and go on to learn about another.
41987 You can use @value{GDBN} to debug programs written in C, C@t{++}, Fortran and
41990 @value{GDBN} is invoked with the shell command @code{gdb}. Once started, it reads
41991 commands from the terminal until you tell it to exit with the @value{GDBN}
41992 command @code{quit}. You can get online help from @value{GDBN} itself
41993 by using the command @code{help}.
41995 You can run @code{gdb} with no arguments or options; but the most
41996 usual way to start @value{GDBN} is with one argument or two, specifying an
41997 executable program as the argument:
42003 You can also start with both an executable program and a core file specified:
42009 You can, instead, specify a process ID as a second argument, if you want
42010 to debug a running process:
42018 would attach @value{GDBN} to process @code{1234} (unless you also have a file
42019 named @file{1234}; @value{GDBN} does check for a core file first).
42020 With option @option{-p} you can omit the @var{program} filename.
42022 Here are some of the most frequently needed @value{GDBN} commands:
42024 @c pod2man highlights the right hand side of the @item lines.
42026 @item break [@var{file}:]@var{function}
42027 Set a breakpoint at @var{function} (in @var{file}).
42029 @item run [@var{arglist}]
42030 Start your program (with @var{arglist}, if specified).
42033 Backtrace: display the program stack.
42035 @item print @var{expr}
42036 Display the value of an expression.
42039 Continue running your program (after stopping, e.g. at a breakpoint).
42042 Execute next program line (after stopping); step @emph{over} any
42043 function calls in the line.
42045 @item edit [@var{file}:]@var{function}
42046 look at the program line where it is presently stopped.
42048 @item list [@var{file}:]@var{function}
42049 type the text of the program in the vicinity of where it is presently stopped.
42052 Execute next program line (after stopping); step @emph{into} any
42053 function calls in the line.
42055 @item help [@var{name}]
42056 Show information about @value{GDBN} command @var{name}, or general information
42057 about using @value{GDBN}.
42060 Exit from @value{GDBN}.
42064 For full details on @value{GDBN},
42065 see @cite{Using GDB: A Guide to the GNU Source-Level Debugger},
42066 by Richard M. Stallman and Roland H. Pesch. The same text is available online
42067 as the @code{gdb} entry in the @code{info} program.
42071 @c man begin OPTIONS gdb
42072 Any arguments other than options specify an executable
42073 file and core file (or process ID); that is, the first argument
42074 encountered with no
42075 associated option flag is equivalent to a @option{-se} option, and the second,
42076 if any, is equivalent to a @option{-c} option if it's the name of a file.
42078 both long and short forms; both are shown here. The long forms are also
42079 recognized if you truncate them, so long as enough of the option is
42080 present to be unambiguous. (If you prefer, you can flag option
42081 arguments with @option{+} rather than @option{-}, though we illustrate the
42082 more usual convention.)
42084 All the options and command line arguments you give are processed
42085 in sequential order. The order makes a difference when the @option{-x}
42091 List all options, with brief explanations.
42093 @item -symbols=@var{file}
42094 @itemx -s @var{file}
42095 Read symbol table from file @var{file}.
42098 Enable writing into executable and core files.
42100 @item -exec=@var{file}
42101 @itemx -e @var{file}
42102 Use file @var{file} as the executable file to execute when
42103 appropriate, and for examining pure data in conjunction with a core
42106 @item -se=@var{file}
42107 Read symbol table from file @var{file} and use it as the executable
42110 @item -core=@var{file}
42111 @itemx -c @var{file}
42112 Use file @var{file} as a core dump to examine.
42114 @item -command=@var{file}
42115 @itemx -x @var{file}
42116 Execute @value{GDBN} commands from file @var{file}.
42118 @item -ex @var{command}
42119 Execute given @value{GDBN} @var{command}.
42121 @item -directory=@var{directory}
42122 @itemx -d @var{directory}
42123 Add @var{directory} to the path to search for source files.
42126 Do not execute commands from @file{~/.gdbinit}.
42130 Do not execute commands from any @file{.gdbinit} initialization files.
42134 ``Quiet''. Do not print the introductory and copyright messages. These
42135 messages are also suppressed in batch mode.
42138 Run in batch mode. Exit with status @code{0} after processing all the command
42139 files specified with @option{-x} (and @file{.gdbinit}, if not inhibited).
42140 Exit with nonzero status if an error occurs in executing the @value{GDBN}
42141 commands in the command files.
42143 Batch mode may be useful for running @value{GDBN} as a filter, for example to
42144 download and run a program on another computer; in order to make this
42145 more useful, the message
42148 Program exited normally.
42152 (which is ordinarily issued whenever a program running under @value{GDBN} control
42153 terminates) is not issued when running in batch mode.
42155 @item -cd=@var{directory}
42156 Run @value{GDBN} using @var{directory} as its working directory,
42157 instead of the current directory.
42161 Emacs sets this option when it runs @value{GDBN} as a subprocess. It tells
42162 @value{GDBN} to output the full file name and line number in a standard,
42163 recognizable fashion each time a stack frame is displayed (which
42164 includes each time the program stops). This recognizable format looks
42165 like two @samp{\032} characters, followed by the file name, line number
42166 and character position separated by colons, and a newline. The
42167 Emacs-to-@value{GDBN} interface program uses the two @samp{\032}
42168 characters as a signal to display the source code for the frame.
42171 Set the line speed (baud rate or bits per second) of any serial
42172 interface used by @value{GDBN} for remote debugging.
42174 @item -tty=@var{device}
42175 Run using @var{device} for your program's standard input and output.
42179 @c man begin SEEALSO gdb
42181 The full documentation for @value{GDBN} is maintained as a Texinfo manual.
42182 If the @code{info} and @code{gdb} programs and @value{GDBN}'s Texinfo
42183 documentation are properly installed at your site, the command
42190 should give you access to the complete manual.
42192 @cite{Using GDB: A Guide to the GNU Source-Level Debugger},
42193 Richard M. Stallman and Roland H. Pesch, July 1991.
42197 @node gdbserver man
42198 @heading gdbserver man
42200 @c man title gdbserver Remote Server for the GNU Debugger
42202 @c man begin SYNOPSIS gdbserver
42203 gdbserver @var{comm} @var{prog} [@var{args}@dots{}]
42205 gdbserver --attach @var{comm} @var{pid}
42207 gdbserver --multi @var{comm}
42211 @c man begin DESCRIPTION gdbserver
42212 @command{gdbserver} is a program that allows you to run @value{GDBN} on a different machine
42213 than the one which is running the program being debugged.
42216 @subheading Usage (server (target) side)
42219 Usage (server (target) side):
42222 First, you need to have a copy of the program you want to debug put onto
42223 the target system. The program can be stripped to save space if needed, as
42224 @command{gdbserver} doesn't care about symbols. All symbol handling is taken care of by
42225 the @value{GDBN} running on the host system.
42227 To use the server, you log on to the target system, and run the @command{gdbserver}
42228 program. You must tell it (a) how to communicate with @value{GDBN}, (b) the name of
42229 your program, and (c) its arguments. The general syntax is:
42232 target> gdbserver @var{comm} @var{program} [@var{args} ...]
42235 For example, using a serial port, you might say:
42239 @c @file would wrap it as F</dev/com1>.
42240 target> gdbserver /dev/com1 emacs foo.txt
42243 target> gdbserver @file{/dev/com1} emacs foo.txt
42247 This tells @command{gdbserver} to debug emacs with an argument of foo.txt, and
42248 to communicate with @value{GDBN} via @file{/dev/com1}. @command{gdbserver} now
42249 waits patiently for the host @value{GDBN} to communicate with it.
42251 To use a TCP connection, you could say:
42254 target> gdbserver host:2345 emacs foo.txt
42257 This says pretty much the same thing as the last example, except that we are
42258 going to communicate with the @code{host} @value{GDBN} via TCP. The @code{host:2345} argument means
42259 that we are expecting to see a TCP connection from @code{host} to local TCP port
42260 2345. (Currently, the @code{host} part is ignored.) You can choose any number you
42261 want for the port number as long as it does not conflict with any existing TCP
42262 ports on the target system. This same port number must be used in the host
42263 @value{GDBN}s @code{target remote} command, which will be described shortly. Note that if
42264 you chose a port number that conflicts with another service, @command{gdbserver} will
42265 print an error message and exit.
42267 @command{gdbserver} can also attach to running programs.
42268 This is accomplished via the @option{--attach} argument. The syntax is:
42271 target> gdbserver --attach @var{comm} @var{pid}
42274 @var{pid} is the process ID of a currently running process. It isn't
42275 necessary to point @command{gdbserver} at a binary for the running process.
42277 To start @code{gdbserver} without supplying an initial command to run
42278 or process ID to attach, use the @option{--multi} command line option.
42279 In such case you should connect using @kbd{target extended-remote} to start
42280 the program you want to debug.
42283 target> gdbserver --multi @var{comm}
42287 @subheading Usage (host side)
42293 You need an unstripped copy of the target program on your host system, since
42294 @value{GDBN} needs to examine it's symbol tables and such. Start up @value{GDBN} as you normally
42295 would, with the target program as the first argument. (You may need to use the
42296 @option{--baud} option if the serial line is running at anything except 9600 baud.)
42297 That is @code{gdb TARGET-PROG}, or @code{gdb --baud BAUD TARGET-PROG}. After that, the only
42298 new command you need to know about is @code{target remote}
42299 (or @code{target extended-remote}). Its argument is either
42300 a device name (usually a serial device, like @file{/dev/ttyb}), or a @code{HOST:PORT}
42301 descriptor. For example:
42305 @c @file would wrap it as F</dev/ttyb>.
42306 (gdb) target remote /dev/ttyb
42309 (gdb) target remote @file{/dev/ttyb}
42314 communicates with the server via serial line @file{/dev/ttyb}, and:
42317 (gdb) target remote the-target:2345
42321 communicates via a TCP connection to port 2345 on host `the-target', where
42322 you previously started up @command{gdbserver} with the same port number. Note that for
42323 TCP connections, you must start up @command{gdbserver} prior to using the `target remote'
42324 command, otherwise you may get an error that looks something like
42325 `Connection refused'.
42327 @command{gdbserver} can also debug multiple inferiors at once,
42330 the @value{GDBN} manual in node @code{Inferiors and Programs}
42331 -- shell command @code{info -f gdb -n 'Inferiors and Programs'}.
42334 @ref{Inferiors and Programs}.
42336 In such case use the @code{extended-remote} @value{GDBN} command variant:
42339 (gdb) target extended-remote the-target:2345
42342 The @command{gdbserver} option @option{--multi} may or may not be used in such
42346 @c man begin OPTIONS gdbserver
42347 There are three different modes for invoking @command{gdbserver}:
42352 Debug a specific program specified by its program name:
42355 gdbserver @var{comm} @var{prog} [@var{args}@dots{}]
42358 The @var{comm} parameter specifies how should the server communicate
42359 with @value{GDBN}; it is either a device name (to use a serial line),
42360 a TCP port number (@code{:1234}), or @code{-} or @code{stdio} to use
42361 stdin/stdout of @code{gdbserver}. Specify the name of the program to
42362 debug in @var{prog}. Any remaining arguments will be passed to the
42363 program verbatim. When the program exits, @value{GDBN} will close the
42364 connection, and @code{gdbserver} will exit.
42367 Debug a specific program by specifying the process ID of a running
42371 gdbserver --attach @var{comm} @var{pid}
42374 The @var{comm} parameter is as described above. Supply the process ID
42375 of a running program in @var{pid}; @value{GDBN} will do everything
42376 else. Like with the previous mode, when the process @var{pid} exits,
42377 @value{GDBN} will close the connection, and @code{gdbserver} will exit.
42380 Multi-process mode -- debug more than one program/process:
42383 gdbserver --multi @var{comm}
42386 In this mode, @value{GDBN} can instruct @command{gdbserver} which
42387 command(s) to run. Unlike the other 2 modes, @value{GDBN} will not
42388 close the connection when a process being debugged exits, so you can
42389 debug several processes in the same session.
42392 In each of the modes you may specify these options:
42397 List all options, with brief explanations.
42400 This option causes @command{gdbserver} to print its version number and exit.
42403 @command{gdbserver} will attach to a running program. The syntax is:
42406 target> gdbserver --attach @var{comm} @var{pid}
42409 @var{pid} is the process ID of a currently running process. It isn't
42410 necessary to point @command{gdbserver} at a binary for the running process.
42413 To start @code{gdbserver} without supplying an initial command to run
42414 or process ID to attach, use this command line option.
42415 Then you can connect using @kbd{target extended-remote} and start
42416 the program you want to debug. The syntax is:
42419 target> gdbserver --multi @var{comm}
42423 Instruct @code{gdbserver} to display extra status information about the debugging
42425 This option is intended for @code{gdbserver} development and for bug reports to
42428 @item --remote-debug
42429 Instruct @code{gdbserver} to display remote protocol debug output.
42430 This option is intended for @code{gdbserver} development and for bug reports to
42433 @item --debug-format=option1@r{[},option2,...@r{]}
42434 Instruct @code{gdbserver} to include extra information in each line
42435 of debugging output.
42436 @xref{Other Command-Line Arguments for gdbserver}.
42439 Specify a wrapper to launch programs
42440 for debugging. The option should be followed by the name of the
42441 wrapper, then any command-line arguments to pass to the wrapper, then
42442 @kbd{--} indicating the end of the wrapper arguments.
42445 By default, @command{gdbserver} keeps the listening TCP port open, so that
42446 additional connections are possible. However, if you start @code{gdbserver}
42447 with the @option{--once} option, it will stop listening for any further
42448 connection attempts after connecting to the first @value{GDBN} session.
42450 @c --disable-packet is not documented for users.
42452 @c --disable-randomization and --no-disable-randomization are superseded by
42453 @c QDisableRandomization.
42458 @c man begin SEEALSO gdbserver
42460 The full documentation for @value{GDBN} is maintained as a Texinfo manual.
42461 If the @code{info} and @code{gdb} programs and @value{GDBN}'s Texinfo
42462 documentation are properly installed at your site, the command
42468 should give you access to the complete manual.
42470 @cite{Using GDB: A Guide to the GNU Source-Level Debugger},
42471 Richard M. Stallman and Roland H. Pesch, July 1991.
42478 @c man title gcore Generate a core file of a running program
42481 @c man begin SYNOPSIS gcore
42482 gcore [-o @var{filename}] @var{pid}
42486 @c man begin DESCRIPTION gcore
42487 Generate a core dump of a running program with process ID @var{pid}.
42488 Produced file is equivalent to a kernel produced core file as if the process
42489 crashed (and if @kbd{ulimit -c} were used to set up an appropriate core dump
42490 limit). Unlike after a crash, after @command{gcore} the program remains
42491 running without any change.
42494 @c man begin OPTIONS gcore
42496 @item -o @var{filename}
42497 The optional argument
42498 @var{filename} specifies the file name where to put the core dump.
42499 If not specified, the file name defaults to @file{core.@var{pid}},
42500 where @var{pid} is the running program process ID.
42504 @c man begin SEEALSO gcore
42506 The full documentation for @value{GDBN} is maintained as a Texinfo manual.
42507 If the @code{info} and @code{gdb} programs and @value{GDBN}'s Texinfo
42508 documentation are properly installed at your site, the command
42515 should give you access to the complete manual.
42517 @cite{Using GDB: A Guide to the GNU Source-Level Debugger},
42518 Richard M. Stallman and Roland H. Pesch, July 1991.
42525 @c man title gdbinit GDB initialization scripts
42528 @c man begin SYNOPSIS gdbinit
42529 @ifset SYSTEM_GDBINIT
42530 @value{SYSTEM_GDBINIT}
42539 @c man begin DESCRIPTION gdbinit
42540 These files contain @value{GDBN} commands to automatically execute during
42541 @value{GDBN} startup. The lines of contents are canned sequences of commands,
42544 the @value{GDBN} manual in node @code{Sequences}
42545 -- shell command @code{info -f gdb -n Sequences}.
42551 Please read more in
42553 the @value{GDBN} manual in node @code{Startup}
42554 -- shell command @code{info -f gdb -n Startup}.
42561 @ifset SYSTEM_GDBINIT
42562 @item @value{SYSTEM_GDBINIT}
42564 @ifclear SYSTEM_GDBINIT
42565 @item (not enabled with @code{--with-system-gdbinit} during compilation)
42567 System-wide initialization file. It is executed unless user specified
42568 @value{GDBN} option @code{-nx} or @code{-n}.
42571 the @value{GDBN} manual in node @code{System-wide configuration}
42572 -- shell command @code{info -f gdb -n 'System-wide configuration'}.
42575 @ref{System-wide configuration}.
42579 User initialization file. It is executed unless user specified
42580 @value{GDBN} options @code{-nx}, @code{-n} or @code{-nh}.
42583 Initialization file for current directory. It may need to be enabled with
42584 @value{GDBN} security command @code{set auto-load local-gdbinit}.
42587 the @value{GDBN} manual in node @code{Init File in the Current Directory}
42588 -- shell command @code{info -f gdb -n 'Init File in the Current Directory'}.
42591 @ref{Init File in the Current Directory}.
42596 @c man begin SEEALSO gdbinit
42598 gdb(1), @code{info -f gdb -n Startup}
42600 The full documentation for @value{GDBN} is maintained as a Texinfo manual.
42601 If the @code{info} and @code{gdb} programs and @value{GDBN}'s Texinfo
42602 documentation are properly installed at your site, the command
42608 should give you access to the complete manual.
42610 @cite{Using GDB: A Guide to the GNU Source-Level Debugger},
42611 Richard M. Stallman and Roland H. Pesch, July 1991.
42617 @node GNU Free Documentation License
42618 @appendix GNU Free Documentation License
42621 @node Concept Index
42622 @unnumbered Concept Index
42626 @node Command and Variable Index
42627 @unnumbered Command, Variable, and Function Index
42632 % I think something like @@colophon should be in texinfo. In the
42634 \long\def\colophon{\hbox to0pt{}\vfill
42635 \centerline{The body of this manual is set in}
42636 \centerline{\fontname\tenrm,}
42637 \centerline{with headings in {\bf\fontname\tenbf}}
42638 \centerline{and examples in {\tt\fontname\tentt}.}
42639 \centerline{{\it\fontname\tenit\/},}
42640 \centerline{{\bf\fontname\tenbf}, and}
42641 \centerline{{\sl\fontname\tensl\/}}
42642 \centerline{are used for emphasis.}\vfill}
42644 % Blame: doc@@cygnus.com, 1991.