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 @item set startup-with-shell
2157 @itemx set startup-with-shell on
2158 @itemx set startup-with-shell off
2159 @itemx show set startup-with-shell
2160 On Unix systems, by default, if a shell is available on your target,
2161 @value{GDBN}) uses it to start your program. Arguments of the
2162 @code{run} command are passed to the shell, which does variable
2163 substitution, expands wildcard characters and performs redirection of
2164 I/O. In some circumstances, it may be useful to disable such use of a
2165 shell, for example, when debugging the shell itself or diagnosing
2166 startup failures such as:
2170 Starting program: ./a.out
2171 During startup program terminated with signal SIGSEGV, Segmentation fault.
2175 which indicates the shell or the wrapper specified with
2176 @samp{exec-wrapper} crashed, not your program. Most often, this is
2177 caused by something odd in your shell's non-interactive mode
2178 initialization file---such as @file{.cshrc} for C-shell,
2179 $@file{.zshenv} for the Z shell, or the file specified in the
2180 @samp{BASH_ENV} environment variable for BASH.
2182 @anchor{set auto-connect-native-target}
2183 @kindex set auto-connect-native-target
2184 @item set auto-connect-native-target
2185 @itemx set auto-connect-native-target on
2186 @itemx set auto-connect-native-target off
2187 @itemx show auto-connect-native-target
2189 By default, if not connected to any target yet (e.g., with
2190 @code{target remote}), the @code{run} command starts your program as a
2191 native process under @value{GDBN}, on your local machine. If you're
2192 sure you don't want to debug programs on your local machine, you can
2193 tell @value{GDBN} to not connect to the native target automatically
2194 with the @code{set auto-connect-native-target off} command.
2196 If @code{on}, which is the default, and if @value{GDBN} is not
2197 connected to a target already, the @code{run} command automaticaly
2198 connects to the native target, if one is available.
2200 If @code{off}, and if @value{GDBN} is not connected to a target
2201 already, the @code{run} command fails with an error:
2205 Don't know how to run. Try "help target".
2208 If @value{GDBN} is already connected to a target, @value{GDBN} always
2209 uses it with the @code{run} command.
2211 In any case, you can explicitly connect to the native target with the
2212 @code{target native} command. For example,
2215 (@value{GDBP}) set auto-connect-native-target off
2217 Don't know how to run. Try "help target".
2218 (@value{GDBP}) target native
2220 Starting program: ./a.out
2221 [Inferior 1 (process 10421) exited normally]
2224 In case you connected explicitly to the @code{native} target,
2225 @value{GDBN} remains connected even if all inferiors exit, ready for
2226 the next @code{run} command. Use the @code{disconnect} command to
2229 Examples of other commands that likewise respect the
2230 @code{auto-connect-native-target} setting: @code{attach}, @code{info
2231 proc}, @code{info os}.
2233 @kindex set disable-randomization
2234 @item set disable-randomization
2235 @itemx set disable-randomization on
2236 This option (enabled by default in @value{GDBN}) will turn off the native
2237 randomization of the virtual address space of the started program. This option
2238 is useful for multiple debugging sessions to make the execution better
2239 reproducible and memory addresses reusable across debugging sessions.
2241 This feature is implemented only on certain targets, including @sc{gnu}/Linux.
2242 On @sc{gnu}/Linux you can get the same behavior using
2245 (@value{GDBP}) set exec-wrapper setarch `uname -m` -R
2248 @item set disable-randomization off
2249 Leave the behavior of the started executable unchanged. Some bugs rear their
2250 ugly heads only when the program is loaded at certain addresses. If your bug
2251 disappears when you run the program under @value{GDBN}, that might be because
2252 @value{GDBN} by default disables the address randomization on platforms, such
2253 as @sc{gnu}/Linux, which do that for stand-alone programs. Use @kbd{set
2254 disable-randomization off} to try to reproduce such elusive bugs.
2256 On targets where it is available, virtual address space randomization
2257 protects the programs against certain kinds of security attacks. In these
2258 cases the attacker needs to know the exact location of a concrete executable
2259 code. Randomizing its location makes it impossible to inject jumps misusing
2260 a code at its expected addresses.
2262 Prelinking shared libraries provides a startup performance advantage but it
2263 makes addresses in these libraries predictable for privileged processes by
2264 having just unprivileged access at the target system. Reading the shared
2265 library binary gives enough information for assembling the malicious code
2266 misusing it. Still even a prelinked shared library can get loaded at a new
2267 random address just requiring the regular relocation process during the
2268 startup. Shared libraries not already prelinked are always loaded at
2269 a randomly chosen address.
2271 Position independent executables (PIE) contain position independent code
2272 similar to the shared libraries and therefore such executables get loaded at
2273 a randomly chosen address upon startup. PIE executables always load even
2274 already prelinked shared libraries at a random address. You can build such
2275 executable using @command{gcc -fPIE -pie}.
2277 Heap (malloc storage), stack and custom mmap areas are always placed randomly
2278 (as long as the randomization is enabled).
2280 @item show disable-randomization
2281 Show the current setting of the explicit disable of the native randomization of
2282 the virtual address space of the started program.
2287 @section Your Program's Arguments
2289 @cindex arguments (to your program)
2290 The arguments to your program can be specified by the arguments of the
2292 They are passed to a shell, which expands wildcard characters and
2293 performs redirection of I/O, and thence to your program. Your
2294 @code{SHELL} environment variable (if it exists) specifies what shell
2295 @value{GDBN} uses. If you do not define @code{SHELL}, @value{GDBN} uses
2296 the default shell (@file{/bin/sh} on Unix).
2298 On non-Unix systems, the program is usually invoked directly by
2299 @value{GDBN}, which emulates I/O redirection via the appropriate system
2300 calls, and the wildcard characters are expanded by the startup code of
2301 the program, not by the shell.
2303 @code{run} with no arguments uses the same arguments used by the previous
2304 @code{run}, or those set by the @code{set args} command.
2309 Specify the arguments to be used the next time your program is run. If
2310 @code{set args} has no arguments, @code{run} executes your program
2311 with no arguments. Once you have run your program with arguments,
2312 using @code{set args} before the next @code{run} is the only way to run
2313 it again without arguments.
2317 Show the arguments to give your program when it is started.
2321 @section Your Program's Environment
2323 @cindex environment (of your program)
2324 The @dfn{environment} consists of a set of environment variables and
2325 their values. Environment variables conventionally record such things as
2326 your user name, your home directory, your terminal type, and your search
2327 path for programs to run. Usually you set up environment variables with
2328 the shell and they are inherited by all the other programs you run. When
2329 debugging, it can be useful to try running your program with a modified
2330 environment without having to start @value{GDBN} over again.
2334 @item path @var{directory}
2335 Add @var{directory} to the front of the @code{PATH} environment variable
2336 (the search path for executables) that will be passed to your program.
2337 The value of @code{PATH} used by @value{GDBN} does not change.
2338 You may specify several directory names, separated by whitespace or by a
2339 system-dependent separator character (@samp{:} on Unix, @samp{;} on
2340 MS-DOS and MS-Windows). If @var{directory} is already in the path, it
2341 is moved to the front, so it is searched sooner.
2343 You can use the string @samp{$cwd} to refer to whatever is the current
2344 working directory at the time @value{GDBN} searches the path. If you
2345 use @samp{.} instead, it refers to the directory where you executed the
2346 @code{path} command. @value{GDBN} replaces @samp{.} in the
2347 @var{directory} argument (with the current path) before adding
2348 @var{directory} to the search path.
2349 @c 'path' is explicitly nonrepeatable, but RMS points out it is silly to
2350 @c document that, since repeating it would be a no-op.
2354 Display the list of search paths for executables (the @code{PATH}
2355 environment variable).
2357 @kindex show environment
2358 @item show environment @r{[}@var{varname}@r{]}
2359 Print the value of environment variable @var{varname} to be given to
2360 your program when it starts. If you do not supply @var{varname},
2361 print the names and values of all environment variables to be given to
2362 your program. You can abbreviate @code{environment} as @code{env}.
2364 @kindex set environment
2365 @item set environment @var{varname} @r{[}=@var{value}@r{]}
2366 Set environment variable @var{varname} to @var{value}. The value
2367 changes for your program (and the shell @value{GDBN} uses to launch
2368 it), not for @value{GDBN} itself. The @var{value} may be any string; the
2369 values of environment variables are just strings, and any
2370 interpretation is supplied by your program itself. The @var{value}
2371 parameter is optional; if it is eliminated, the variable is set to a
2373 @c "any string" here does not include leading, trailing
2374 @c blanks. Gnu asks: does anyone care?
2376 For example, this command:
2383 tells the debugged program, when subsequently run, that its user is named
2384 @samp{foo}. (The spaces around @samp{=} are used for clarity here; they
2385 are not actually required.)
2387 Note that on Unix systems, @value{GDBN} runs your program via a shell,
2388 which also inherits the environment set with @code{set environment}.
2389 If necessary, you can avoid that by using the @samp{env} program as a
2390 wrapper instead of using @code{set environment}. @xref{set
2391 exec-wrapper}, for an example doing just that.
2393 @kindex unset environment
2394 @item unset environment @var{varname}
2395 Remove variable @var{varname} from the environment to be passed to your
2396 program. This is different from @samp{set env @var{varname} =};
2397 @code{unset environment} removes the variable from the environment,
2398 rather than assigning it an empty value.
2401 @emph{Warning:} On Unix systems, @value{GDBN} runs your program using
2402 the shell indicated by your @code{SHELL} environment variable if it
2403 exists (or @code{/bin/sh} if not). If your @code{SHELL} variable
2404 names a shell that runs an initialization file when started
2405 non-interactively---such as @file{.cshrc} for C-shell, $@file{.zshenv}
2406 for the Z shell, or the file specified in the @samp{BASH_ENV}
2407 environment variable for BASH---any variables you set in that file
2408 affect your program. You may wish to move setting of environment
2409 variables to files that are only run when you sign on, such as
2410 @file{.login} or @file{.profile}.
2412 @node Working Directory
2413 @section Your Program's Working Directory
2415 @cindex working directory (of your program)
2416 Each time you start your program with @code{run}, it inherits its
2417 working directory from the current working directory of @value{GDBN}.
2418 The @value{GDBN} working directory is initially whatever it inherited
2419 from its parent process (typically the shell), but you can specify a new
2420 working directory in @value{GDBN} with the @code{cd} command.
2422 The @value{GDBN} working directory also serves as a default for the commands
2423 that specify files for @value{GDBN} to operate on. @xref{Files, ,Commands to
2428 @cindex change working directory
2429 @item cd @r{[}@var{directory}@r{]}
2430 Set the @value{GDBN} working directory to @var{directory}. If not
2431 given, @var{directory} uses @file{'~'}.
2435 Print the @value{GDBN} working directory.
2438 It is generally impossible to find the current working directory of
2439 the process being debugged (since a program can change its directory
2440 during its run). If you work on a system where @value{GDBN} is
2441 configured with the @file{/proc} support, you can use the @code{info
2442 proc} command (@pxref{SVR4 Process Information}) to find out the
2443 current working directory of the debuggee.
2446 @section Your Program's Input and Output
2451 By default, the program you run under @value{GDBN} does input and output to
2452 the same terminal that @value{GDBN} uses. @value{GDBN} switches the terminal
2453 to its own terminal modes to interact with you, but it records the terminal
2454 modes your program was using and switches back to them when you continue
2455 running your program.
2458 @kindex info terminal
2460 Displays information recorded by @value{GDBN} about the terminal modes your
2464 You can redirect your program's input and/or output using shell
2465 redirection with the @code{run} command. For example,
2472 starts your program, diverting its output to the file @file{outfile}.
2475 @cindex controlling terminal
2476 Another way to specify where your program should do input and output is
2477 with the @code{tty} command. This command accepts a file name as
2478 argument, and causes this file to be the default for future @code{run}
2479 commands. It also resets the controlling terminal for the child
2480 process, for future @code{run} commands. For example,
2487 directs that processes started with subsequent @code{run} commands
2488 default to do input and output on the terminal @file{/dev/ttyb} and have
2489 that as their controlling terminal.
2491 An explicit redirection in @code{run} overrides the @code{tty} command's
2492 effect on the input/output device, but not its effect on the controlling
2495 When you use the @code{tty} command or redirect input in the @code{run}
2496 command, only the input @emph{for your program} is affected. The input
2497 for @value{GDBN} still comes from your terminal. @code{tty} is an alias
2498 for @code{set inferior-tty}.
2500 @cindex inferior tty
2501 @cindex set inferior controlling terminal
2502 You can use the @code{show inferior-tty} command to tell @value{GDBN} to
2503 display the name of the terminal that will be used for future runs of your
2507 @item set inferior-tty [ @var{tty} ]
2508 @kindex set inferior-tty
2509 Set the tty for the program being debugged to @var{tty}. Omitting @var{tty}
2510 restores the default behavior, which is to use the same terminal as
2513 @item show inferior-tty
2514 @kindex show inferior-tty
2515 Show the current tty for the program being debugged.
2519 @section Debugging an Already-running Process
2524 @item attach @var{process-id}
2525 This command attaches to a running process---one that was started
2526 outside @value{GDBN}. (@code{info files} shows your active
2527 targets.) The command takes as argument a process ID. The usual way to
2528 find out the @var{process-id} of a Unix process is with the @code{ps} utility,
2529 or with the @samp{jobs -l} shell command.
2531 @code{attach} does not repeat if you press @key{RET} a second time after
2532 executing the command.
2535 To use @code{attach}, your program must be running in an environment
2536 which supports processes; for example, @code{attach} does not work for
2537 programs on bare-board targets that lack an operating system. You must
2538 also have permission to send the process a signal.
2540 When you use @code{attach}, the debugger finds the program running in
2541 the process first by looking in the current working directory, then (if
2542 the program is not found) by using the source file search path
2543 (@pxref{Source Path, ,Specifying Source Directories}). You can also use
2544 the @code{file} command to load the program. @xref{Files, ,Commands to
2547 The first thing @value{GDBN} does after arranging to debug the specified
2548 process is to stop it. You can examine and modify an attached process
2549 with all the @value{GDBN} commands that are ordinarily available when
2550 you start processes with @code{run}. You can insert breakpoints; you
2551 can step and continue; you can modify storage. If you would rather the
2552 process continue running, you may use the @code{continue} command after
2553 attaching @value{GDBN} to the process.
2558 When you have finished debugging the attached process, you can use the
2559 @code{detach} command to release it from @value{GDBN} control. Detaching
2560 the process continues its execution. After the @code{detach} command,
2561 that process and @value{GDBN} become completely independent once more, and you
2562 are ready to @code{attach} another process or start one with @code{run}.
2563 @code{detach} does not repeat if you press @key{RET} again after
2564 executing the command.
2567 If you exit @value{GDBN} while you have an attached process, you detach
2568 that process. If you use the @code{run} command, you kill that process.
2569 By default, @value{GDBN} asks for confirmation if you try to do either of these
2570 things; you can control whether or not you need to confirm by using the
2571 @code{set confirm} command (@pxref{Messages/Warnings, ,Optional Warnings and
2575 @section Killing the Child Process
2580 Kill the child process in which your program is running under @value{GDBN}.
2583 This command is useful if you wish to debug a core dump instead of a
2584 running process. @value{GDBN} ignores any core dump file while your program
2587 On some operating systems, a program cannot be executed outside @value{GDBN}
2588 while you have breakpoints set on it inside @value{GDBN}. You can use the
2589 @code{kill} command in this situation to permit running your program
2590 outside the debugger.
2592 The @code{kill} command is also useful if you wish to recompile and
2593 relink your program, since on many systems it is impossible to modify an
2594 executable file while it is running in a process. In this case, when you
2595 next type @code{run}, @value{GDBN} notices that the file has changed, and
2596 reads the symbol table again (while trying to preserve your current
2597 breakpoint settings).
2599 @node Inferiors and Programs
2600 @section Debugging Multiple Inferiors and Programs
2602 @value{GDBN} lets you run and debug multiple programs in a single
2603 session. In addition, @value{GDBN} on some systems may let you run
2604 several programs simultaneously (otherwise you have to exit from one
2605 before starting another). In the most general case, you can have
2606 multiple threads of execution in each of multiple processes, launched
2607 from multiple executables.
2610 @value{GDBN} represents the state of each program execution with an
2611 object called an @dfn{inferior}. An inferior typically corresponds to
2612 a process, but is more general and applies also to targets that do not
2613 have processes. Inferiors may be created before a process runs, and
2614 may be retained after a process exits. Inferiors have unique
2615 identifiers that are different from process ids. Usually each
2616 inferior will also have its own distinct address space, although some
2617 embedded targets may have several inferiors running in different parts
2618 of a single address space. Each inferior may in turn have multiple
2619 threads running in it.
2621 To find out what inferiors exist at any moment, use @w{@code{info
2625 @kindex info inferiors
2626 @item info inferiors
2627 Print a list of all inferiors currently being managed by @value{GDBN}.
2629 @value{GDBN} displays for each inferior (in this order):
2633 the inferior number assigned by @value{GDBN}
2636 the target system's inferior identifier
2639 the name of the executable the inferior is running.
2644 An asterisk @samp{*} preceding the @value{GDBN} inferior number
2645 indicates the current inferior.
2649 @c end table here to get a little more width for example
2652 (@value{GDBP}) info inferiors
2653 Num Description Executable
2654 2 process 2307 hello
2655 * 1 process 3401 goodbye
2658 To switch focus between inferiors, use the @code{inferior} command:
2661 @kindex inferior @var{infno}
2662 @item inferior @var{infno}
2663 Make inferior number @var{infno} the current inferior. The argument
2664 @var{infno} is the inferior number assigned by @value{GDBN}, as shown
2665 in the first field of the @samp{info inferiors} display.
2668 @vindex $_inferior@r{, convenience variable}
2669 The debugger convenience variable @samp{$_inferior} contains the
2670 number of the current inferior. You may find this useful in writing
2671 breakpoint conditional expressions, command scripts, and so forth.
2672 @xref{Convenience Vars,, Convenience Variables}, for general
2673 information on convenience variables.
2675 You can get multiple executables into a debugging session via the
2676 @code{add-inferior} and @w{@code{clone-inferior}} commands. On some
2677 systems @value{GDBN} can add inferiors to the debug session
2678 automatically by following calls to @code{fork} and @code{exec}. To
2679 remove inferiors from the debugging session use the
2680 @w{@code{remove-inferiors}} command.
2683 @kindex add-inferior
2684 @item add-inferior [ -copies @var{n} ] [ -exec @var{executable} ]
2685 Adds @var{n} inferiors to be run using @var{executable} as the
2686 executable; @var{n} defaults to 1. If no executable is specified,
2687 the inferiors begins empty, with no program. You can still assign or
2688 change the program assigned to the inferior at any time by using the
2689 @code{file} command with the executable name as its argument.
2691 @kindex clone-inferior
2692 @item clone-inferior [ -copies @var{n} ] [ @var{infno} ]
2693 Adds @var{n} inferiors ready to execute the same program as inferior
2694 @var{infno}; @var{n} defaults to 1, and @var{infno} defaults to the
2695 number of the current inferior. This is a convenient command when you
2696 want to run another instance of the inferior you are debugging.
2699 (@value{GDBP}) info inferiors
2700 Num Description Executable
2701 * 1 process 29964 helloworld
2702 (@value{GDBP}) clone-inferior
2705 (@value{GDBP}) info inferiors
2706 Num Description Executable
2708 * 1 process 29964 helloworld
2711 You can now simply switch focus to inferior 2 and run it.
2713 @kindex remove-inferiors
2714 @item remove-inferiors @var{infno}@dots{}
2715 Removes the inferior or inferiors @var{infno}@dots{}. It is not
2716 possible to remove an inferior that is running with this command. For
2717 those, use the @code{kill} or @code{detach} command first.
2721 To quit debugging one of the running inferiors that is not the current
2722 inferior, you can either detach from it by using the @w{@code{detach
2723 inferior}} command (allowing it to run independently), or kill it
2724 using the @w{@code{kill inferiors}} command:
2727 @kindex detach inferiors @var{infno}@dots{}
2728 @item detach inferior @var{infno}@dots{}
2729 Detach from the inferior or inferiors identified by @value{GDBN}
2730 inferior number(s) @var{infno}@dots{}. Note that the inferior's entry
2731 still stays on the list of inferiors shown by @code{info inferiors},
2732 but its Description will show @samp{<null>}.
2734 @kindex kill inferiors @var{infno}@dots{}
2735 @item kill inferiors @var{infno}@dots{}
2736 Kill the inferior or inferiors identified by @value{GDBN} inferior
2737 number(s) @var{infno}@dots{}. Note that the inferior's entry still
2738 stays on the list of inferiors shown by @code{info inferiors}, but its
2739 Description will show @samp{<null>}.
2742 After the successful completion of a command such as @code{detach},
2743 @code{detach inferiors}, @code{kill} or @code{kill inferiors}, or after
2744 a normal process exit, the inferior is still valid and listed with
2745 @code{info inferiors}, ready to be restarted.
2748 To be notified when inferiors are started or exit under @value{GDBN}'s
2749 control use @w{@code{set print inferior-events}}:
2752 @kindex set print inferior-events
2753 @cindex print messages on inferior start and exit
2754 @item set print inferior-events
2755 @itemx set print inferior-events on
2756 @itemx set print inferior-events off
2757 The @code{set print inferior-events} command allows you to enable or
2758 disable printing of messages when @value{GDBN} notices that new
2759 inferiors have started or that inferiors have exited or have been
2760 detached. By default, these messages will not be printed.
2762 @kindex show print inferior-events
2763 @item show print inferior-events
2764 Show whether messages will be printed when @value{GDBN} detects that
2765 inferiors have started, exited or have been detached.
2768 Many commands will work the same with multiple programs as with a
2769 single program: e.g., @code{print myglobal} will simply display the
2770 value of @code{myglobal} in the current inferior.
2773 Occasionaly, when debugging @value{GDBN} itself, it may be useful to
2774 get more info about the relationship of inferiors, programs, address
2775 spaces in a debug session. You can do that with the @w{@code{maint
2776 info program-spaces}} command.
2779 @kindex maint info program-spaces
2780 @item maint info program-spaces
2781 Print a list of all program spaces currently being managed by
2784 @value{GDBN} displays for each program space (in this order):
2788 the program space number assigned by @value{GDBN}
2791 the name of the executable loaded into the program space, with e.g.,
2792 the @code{file} command.
2797 An asterisk @samp{*} preceding the @value{GDBN} program space number
2798 indicates the current program space.
2800 In addition, below each program space line, @value{GDBN} prints extra
2801 information that isn't suitable to display in tabular form. For
2802 example, the list of inferiors bound to the program space.
2805 (@value{GDBP}) maint info program-spaces
2809 Bound inferiors: ID 1 (process 21561)
2812 Here we can see that no inferior is running the program @code{hello},
2813 while @code{process 21561} is running the program @code{goodbye}. On
2814 some targets, it is possible that multiple inferiors are bound to the
2815 same program space. The most common example is that of debugging both
2816 the parent and child processes of a @code{vfork} call. For example,
2819 (@value{GDBP}) maint info program-spaces
2822 Bound inferiors: ID 2 (process 18050), ID 1 (process 18045)
2825 Here, both inferior 2 and inferior 1 are running in the same program
2826 space as a result of inferior 1 having executed a @code{vfork} call.
2830 @section Debugging Programs with Multiple Threads
2832 @cindex threads of execution
2833 @cindex multiple threads
2834 @cindex switching threads
2835 In some operating systems, such as GNU/Linux and Solaris, a single program
2836 may have more than one @dfn{thread} of execution. The precise semantics
2837 of threads differ from one operating system to another, but in general
2838 the threads of a single program are akin to multiple processes---except
2839 that they share one address space (that is, they can all examine and
2840 modify the same variables). On the other hand, each thread has its own
2841 registers and execution stack, and perhaps private memory.
2843 @value{GDBN} provides these facilities for debugging multi-thread
2847 @item automatic notification of new threads
2848 @item @samp{thread @var{thread-id}}, a command to switch among threads
2849 @item @samp{info threads}, a command to inquire about existing threads
2850 @item @samp{thread apply [@var{thread-id-list}] [@var{all}] @var{args}},
2851 a command to apply a command to a list of threads
2852 @item thread-specific breakpoints
2853 @item @samp{set print thread-events}, which controls printing of
2854 messages on thread start and exit.
2855 @item @samp{set libthread-db-search-path @var{path}}, which lets
2856 the user specify which @code{libthread_db} to use if the default choice
2857 isn't compatible with the program.
2860 @cindex focus of debugging
2861 @cindex current thread
2862 The @value{GDBN} thread debugging facility allows you to observe all
2863 threads while your program runs---but whenever @value{GDBN} takes
2864 control, one thread in particular is always the focus of debugging.
2865 This thread is called the @dfn{current thread}. Debugging commands show
2866 program information from the perspective of the current thread.
2868 @cindex @code{New} @var{systag} message
2869 @cindex thread identifier (system)
2870 @c FIXME-implementors!! It would be more helpful if the [New...] message
2871 @c included GDB's numeric thread handle, so you could just go to that
2872 @c thread without first checking `info threads'.
2873 Whenever @value{GDBN} detects a new thread in your program, it displays
2874 the target system's identification for the thread with a message in the
2875 form @samp{[New @var{systag}]}, where @var{systag} is a thread identifier
2876 whose form varies depending on the particular system. For example, on
2877 @sc{gnu}/Linux, you might see
2880 [New Thread 0x41e02940 (LWP 25582)]
2884 when @value{GDBN} notices a new thread. In contrast, on other systems,
2885 the @var{systag} is simply something like @samp{process 368}, with no
2888 @c FIXME!! (1) Does the [New...] message appear even for the very first
2889 @c thread of a program, or does it only appear for the
2890 @c second---i.e.@: when it becomes obvious we have a multithread
2892 @c (2) *Is* there necessarily a first thread always? Or do some
2893 @c multithread systems permit starting a program with multiple
2894 @c threads ab initio?
2896 @anchor{thread numbers}
2897 @cindex thread number, per inferior
2898 @cindex thread identifier (GDB)
2899 For debugging purposes, @value{GDBN} associates its own thread number
2900 ---always a single integer---with each thread of an inferior. This
2901 number is unique between all threads of an inferior, but not unique
2902 between threads of different inferiors.
2904 @cindex qualified thread ID
2905 You can refer to a given thread in an inferior using the qualified
2906 @var{inferior-num}.@var{thread-num} syntax, also known as
2907 @dfn{qualified thread ID}, with @var{inferior-num} being the inferior
2908 number and @var{thread-num} being the thread number of the given
2909 inferior. For example, thread @code{2.3} refers to thread number 3 of
2910 inferior 2. If you omit @var{inferior-num} (e.g., @code{thread 3}),
2911 then @value{GDBN} infers you're referring to a thread of the current
2914 Until you create a second inferior, @value{GDBN} does not show the
2915 @var{inferior-num} part of thread IDs, even though you can always use
2916 the full @var{inferior-num}.@var{thread-num} form to refer to threads
2917 of inferior 1, the initial inferior.
2919 @anchor{thread ID lists}
2920 @cindex thread ID lists
2921 Some commands accept a space-separated @dfn{thread ID list} as
2922 argument. A list element can be:
2926 A thread ID as shown in the first field of the @samp{info threads}
2927 display, with or without an inferior qualifier. E.g., @samp{2.1} or
2931 A range of thread numbers, again with or without an inferior
2932 qualifier, as in @var{inf}.@var{thr1}-@var{thr2} or
2933 @var{thr1}-@var{thr2}. E.g., @samp{1.2-4} or @samp{2-4}.
2936 All threads of an inferior, specified with a star wildcard, with or
2937 without an inferior qualifier, as in @var{inf}.@code{*} (e.g.,
2938 @samp{1.*}) or @code{*}. The former refers to all threads of the
2939 given inferior, and the latter form without an inferior qualifier
2940 refers to all threads of the current inferior.
2944 For example, if the current inferior is 1, and inferior 7 has one
2945 thread with ID 7.1, the thread list @samp{1 2-3 4.5 6.7-9 7.*}
2946 includes threads 1 to 3 of inferior 1, thread 5 of inferior 4, threads
2947 7 to 9 of inferior 6 and all threads of inferior 7. That is, in
2948 expanded qualified form, the same as @samp{1.1 1.2 1.3 4.5 6.7 6.8 6.9
2952 @anchor{global thread numbers}
2953 @cindex global thread number
2954 @cindex global thread identifier (GDB)
2955 In addition to a @emph{per-inferior} number, each thread is also
2956 assigned a unique @emph{global} number, also known as @dfn{global
2957 thread ID}, a single integer. Unlike the thread number component of
2958 the thread ID, no two threads have the same global ID, even when
2959 you're debugging multiple inferiors.
2961 From @value{GDBN}'s perspective, a process always has at least one
2962 thread. In other words, @value{GDBN} assigns a thread number to the
2963 program's ``main thread'' even if the program is not multi-threaded.
2965 @vindex $_thread@r{, convenience variable}
2966 @vindex $_gthread@r{, convenience variable}
2967 The debugger convenience variables @samp{$_thread} and
2968 @samp{$_gthread} contain, respectively, the per-inferior thread number
2969 and the global thread number of the current thread. You may find this
2970 useful in writing breakpoint conditional expressions, command scripts,
2971 and so forth. @xref{Convenience Vars,, Convenience Variables}, for
2972 general information on convenience variables.
2974 If @value{GDBN} detects the program is multi-threaded, it augments the
2975 usual message about stopping at a breakpoint with the ID and name of
2976 the thread that hit the breakpoint.
2979 Thread 2 "client" hit Breakpoint 1, send_message () at client.c:68
2982 Likewise when the program receives a signal:
2985 Thread 1 "main" received signal SIGINT, Interrupt.
2989 @kindex info threads
2990 @item info threads @r{[}@var{thread-id-list}@r{]}
2992 Display information about one or more threads. With no arguments
2993 displays information about all threads. You can specify the list of
2994 threads that you want to display using the thread ID list syntax
2995 (@pxref{thread ID lists}).
2997 @value{GDBN} displays for each thread (in this order):
3001 the per-inferior thread number assigned by @value{GDBN}
3004 the global thread number assigned by @value{GDBN}, if the @samp{-gid}
3005 option was specified
3008 the target system's thread identifier (@var{systag})
3011 the thread's name, if one is known. A thread can either be named by
3012 the user (see @code{thread name}, below), or, in some cases, by the
3016 the current stack frame summary for that thread
3020 An asterisk @samp{*} to the left of the @value{GDBN} thread number
3021 indicates the current thread.
3025 @c end table here to get a little more width for example
3028 (@value{GDBP}) info threads
3030 * 1 process 35 thread 13 main (argc=1, argv=0x7ffffff8)
3031 2 process 35 thread 23 0x34e5 in sigpause ()
3032 3 process 35 thread 27 0x34e5 in sigpause ()
3036 If you're debugging multiple inferiors, @value{GDBN} displays thread
3037 IDs using the qualified @var{inferior-num}.@var{thread-num} format.
3038 Otherwise, only @var{thread-num} is shown.
3040 If you specify the @samp{-gid} option, @value{GDBN} displays a column
3041 indicating each thread's global thread ID:
3044 (@value{GDBP}) info threads
3045 Id GId Target Id Frame
3046 1.1 1 process 35 thread 13 main (argc=1, argv=0x7ffffff8)
3047 1.2 3 process 35 thread 23 0x34e5 in sigpause ()
3048 1.3 4 process 35 thread 27 0x34e5 in sigpause ()
3049 * 2.1 2 process 65 thread 1 main (argc=1, argv=0x7ffffff8)
3052 On Solaris, you can display more information about user threads with a
3053 Solaris-specific command:
3056 @item maint info sol-threads
3057 @kindex maint info sol-threads
3058 @cindex thread info (Solaris)
3059 Display info on Solaris user threads.
3063 @kindex thread @var{thread-id}
3064 @item thread @var{thread-id}
3065 Make thread ID @var{thread-id} the current thread. The command
3066 argument @var{thread-id} is the @value{GDBN} thread ID, as shown in
3067 the first field of the @samp{info threads} display, with or without an
3068 inferior qualifier (e.g., @samp{2.1} or @samp{1}).
3070 @value{GDBN} responds by displaying the system identifier of the
3071 thread you selected, and its current stack frame summary:
3074 (@value{GDBP}) thread 2
3075 [Switching to thread 2 (Thread 0xb7fdab70 (LWP 12747))]
3076 #0 some_function (ignore=0x0) at example.c:8
3077 8 printf ("hello\n");
3081 As with the @samp{[New @dots{}]} message, the form of the text after
3082 @samp{Switching to} depends on your system's conventions for identifying
3085 @kindex thread apply
3086 @cindex apply command to several threads
3087 @item thread apply [@var{thread-id-list} | all [-ascending]] @var{command}
3088 The @code{thread apply} command allows you to apply the named
3089 @var{command} to one or more threads. Specify the threads that you
3090 want affected using the thread ID list syntax (@pxref{thread ID
3091 lists}), or specify @code{all} to apply to all threads. To apply a
3092 command to all threads in descending order, type @kbd{thread apply all
3093 @var{command}}. To apply a command to all threads in ascending order,
3094 type @kbd{thread apply all -ascending @var{command}}.
3098 @cindex name a thread
3099 @item thread name [@var{name}]
3100 This command assigns a name to the current thread. If no argument is
3101 given, any existing user-specified name is removed. The thread name
3102 appears in the @samp{info threads} display.
3104 On some systems, such as @sc{gnu}/Linux, @value{GDBN} is able to
3105 determine the name of the thread as given by the OS. On these
3106 systems, a name specified with @samp{thread name} will override the
3107 system-give name, and removing the user-specified name will cause
3108 @value{GDBN} to once again display the system-specified name.
3111 @cindex search for a thread
3112 @item thread find [@var{regexp}]
3113 Search for and display thread ids whose name or @var{systag}
3114 matches the supplied regular expression.
3116 As well as being the complement to the @samp{thread name} command,
3117 this command also allows you to identify a thread by its target
3118 @var{systag}. For instance, on @sc{gnu}/Linux, the target @var{systag}
3122 (@value{GDBN}) thread find 26688
3123 Thread 4 has target id 'Thread 0x41e02940 (LWP 26688)'
3124 (@value{GDBN}) info thread 4
3126 4 Thread 0x41e02940 (LWP 26688) 0x00000031ca6cd372 in select ()
3129 @kindex set print thread-events
3130 @cindex print messages on thread start and exit
3131 @item set print thread-events
3132 @itemx set print thread-events on
3133 @itemx set print thread-events off
3134 The @code{set print thread-events} command allows you to enable or
3135 disable printing of messages when @value{GDBN} notices that new threads have
3136 started or that threads have exited. By default, these messages will
3137 be printed if detection of these events is supported by the target.
3138 Note that these messages cannot be disabled on all targets.
3140 @kindex show print thread-events
3141 @item show print thread-events
3142 Show whether messages will be printed when @value{GDBN} detects that threads
3143 have started and exited.
3146 @xref{Thread Stops,,Stopping and Starting Multi-thread Programs}, for
3147 more information about how @value{GDBN} behaves when you stop and start
3148 programs with multiple threads.
3150 @xref{Set Watchpoints,,Setting Watchpoints}, for information about
3151 watchpoints in programs with multiple threads.
3153 @anchor{set libthread-db-search-path}
3155 @kindex set libthread-db-search-path
3156 @cindex search path for @code{libthread_db}
3157 @item set libthread-db-search-path @r{[}@var{path}@r{]}
3158 If this variable is set, @var{path} is a colon-separated list of
3159 directories @value{GDBN} will use to search for @code{libthread_db}.
3160 If you omit @var{path}, @samp{libthread-db-search-path} will be reset to
3161 its default value (@code{$sdir:$pdir} on @sc{gnu}/Linux and Solaris systems).
3162 Internally, the default value comes from the @code{LIBTHREAD_DB_SEARCH_PATH}
3165 On @sc{gnu}/Linux and Solaris systems, @value{GDBN} uses a ``helper''
3166 @code{libthread_db} library to obtain information about threads in the
3167 inferior process. @value{GDBN} will use @samp{libthread-db-search-path}
3168 to find @code{libthread_db}. @value{GDBN} also consults first if inferior
3169 specific thread debugging library loading is enabled
3170 by @samp{set auto-load libthread-db} (@pxref{libthread_db.so.1 file}).
3172 A special entry @samp{$sdir} for @samp{libthread-db-search-path}
3173 refers to the default system directories that are
3174 normally searched for loading shared libraries. The @samp{$sdir} entry
3175 is the only kind not needing to be enabled by @samp{set auto-load libthread-db}
3176 (@pxref{libthread_db.so.1 file}).
3178 A special entry @samp{$pdir} for @samp{libthread-db-search-path}
3179 refers to the directory from which @code{libpthread}
3180 was loaded in the inferior process.
3182 For any @code{libthread_db} library @value{GDBN} finds in above directories,
3183 @value{GDBN} attempts to initialize it with the current inferior process.
3184 If this initialization fails (which could happen because of a version
3185 mismatch between @code{libthread_db} and @code{libpthread}), @value{GDBN}
3186 will unload @code{libthread_db}, and continue with the next directory.
3187 If none of @code{libthread_db} libraries initialize successfully,
3188 @value{GDBN} will issue a warning and thread debugging will be disabled.
3190 Setting @code{libthread-db-search-path} is currently implemented
3191 only on some platforms.
3193 @kindex show libthread-db-search-path
3194 @item show libthread-db-search-path
3195 Display current libthread_db search path.
3197 @kindex set debug libthread-db
3198 @kindex show debug libthread-db
3199 @cindex debugging @code{libthread_db}
3200 @item set debug libthread-db
3201 @itemx show debug libthread-db
3202 Turns on or off display of @code{libthread_db}-related events.
3203 Use @code{1} to enable, @code{0} to disable.
3207 @section Debugging Forks
3209 @cindex fork, debugging programs which call
3210 @cindex multiple processes
3211 @cindex processes, multiple
3212 On most systems, @value{GDBN} has no special support for debugging
3213 programs which create additional processes using the @code{fork}
3214 function. When a program forks, @value{GDBN} will continue to debug the
3215 parent process and the child process will run unimpeded. If you have
3216 set a breakpoint in any code which the child then executes, the child
3217 will get a @code{SIGTRAP} signal which (unless it catches the signal)
3218 will cause it to terminate.
3220 However, if you want to debug the child process there is a workaround
3221 which isn't too painful. Put a call to @code{sleep} in the code which
3222 the child process executes after the fork. It may be useful to sleep
3223 only if a certain environment variable is set, or a certain file exists,
3224 so that the delay need not occur when you don't want to run @value{GDBN}
3225 on the child. While the child is sleeping, use the @code{ps} program to
3226 get its process ID. Then tell @value{GDBN} (a new invocation of
3227 @value{GDBN} if you are also debugging the parent process) to attach to
3228 the child process (@pxref{Attach}). From that point on you can debug
3229 the child process just like any other process which you attached to.
3231 On some systems, @value{GDBN} provides support for debugging programs
3232 that create additional processes using the @code{fork} or @code{vfork}
3233 functions. On @sc{gnu}/Linux platforms, this feature is supported
3234 with kernel version 2.5.46 and later.
3236 The fork debugging commands are supported in native mode and when
3237 connected to @code{gdbserver} in either @code{target remote} mode or
3238 @code{target extended-remote} mode.
3240 By default, when a program forks, @value{GDBN} will continue to debug
3241 the parent process and the child process will run unimpeded.
3243 If you want to follow the child process instead of the parent process,
3244 use the command @w{@code{set follow-fork-mode}}.
3247 @kindex set follow-fork-mode
3248 @item set follow-fork-mode @var{mode}
3249 Set the debugger response to a program call of @code{fork} or
3250 @code{vfork}. A call to @code{fork} or @code{vfork} creates a new
3251 process. The @var{mode} argument can be:
3255 The original process is debugged after a fork. The child process runs
3256 unimpeded. This is the default.
3259 The new process is debugged after a fork. The parent process runs
3264 @kindex show follow-fork-mode
3265 @item show follow-fork-mode
3266 Display the current debugger response to a @code{fork} or @code{vfork} call.
3269 @cindex debugging multiple processes
3270 On Linux, if you want to debug both the parent and child processes, use the
3271 command @w{@code{set detach-on-fork}}.
3274 @kindex set detach-on-fork
3275 @item set detach-on-fork @var{mode}
3276 Tells gdb whether to detach one of the processes after a fork, or
3277 retain debugger control over them both.
3281 The child process (or parent process, depending on the value of
3282 @code{follow-fork-mode}) will be detached and allowed to run
3283 independently. This is the default.
3286 Both processes will be held under the control of @value{GDBN}.
3287 One process (child or parent, depending on the value of
3288 @code{follow-fork-mode}) is debugged as usual, while the other
3293 @kindex show detach-on-fork
3294 @item show detach-on-fork
3295 Show whether detach-on-fork mode is on/off.
3298 If you choose to set @samp{detach-on-fork} mode off, then @value{GDBN}
3299 will retain control of all forked processes (including nested forks).
3300 You can list the forked processes under the control of @value{GDBN} by
3301 using the @w{@code{info inferiors}} command, and switch from one fork
3302 to another by using the @code{inferior} command (@pxref{Inferiors and
3303 Programs, ,Debugging Multiple Inferiors and Programs}).
3305 To quit debugging one of the forked processes, you can either detach
3306 from it by using the @w{@code{detach inferiors}} command (allowing it
3307 to run independently), or kill it using the @w{@code{kill inferiors}}
3308 command. @xref{Inferiors and Programs, ,Debugging Multiple Inferiors
3311 If you ask to debug a child process and a @code{vfork} is followed by an
3312 @code{exec}, @value{GDBN} executes the new target up to the first
3313 breakpoint in the new target. If you have a breakpoint set on
3314 @code{main} in your original program, the breakpoint will also be set on
3315 the child process's @code{main}.
3317 On some systems, when a child process is spawned by @code{vfork}, you
3318 cannot debug the child or parent until an @code{exec} call completes.
3320 If you issue a @code{run} command to @value{GDBN} after an @code{exec}
3321 call executes, the new target restarts. To restart the parent
3322 process, use the @code{file} command with the parent executable name
3323 as its argument. By default, after an @code{exec} call executes,
3324 @value{GDBN} discards the symbols of the previous executable image.
3325 You can change this behaviour with the @w{@code{set follow-exec-mode}}
3329 @kindex set follow-exec-mode
3330 @item set follow-exec-mode @var{mode}
3332 Set debugger response to a program call of @code{exec}. An
3333 @code{exec} call replaces the program image of a process.
3335 @code{follow-exec-mode} can be:
3339 @value{GDBN} creates a new inferior and rebinds the process to this
3340 new inferior. The program the process was running before the
3341 @code{exec} call can be restarted afterwards by restarting the
3347 (@value{GDBP}) info inferiors
3349 Id Description Executable
3352 process 12020 is executing new program: prog2
3353 Program exited normally.
3354 (@value{GDBP}) info inferiors
3355 Id Description Executable
3361 @value{GDBN} keeps the process bound to the same inferior. The new
3362 executable image replaces the previous executable loaded in the
3363 inferior. Restarting the inferior after the @code{exec} call, with
3364 e.g., the @code{run} command, restarts the executable the process was
3365 running after the @code{exec} call. This is the default mode.
3370 (@value{GDBP}) info inferiors
3371 Id Description Executable
3374 process 12020 is executing new program: prog2
3375 Program exited normally.
3376 (@value{GDBP}) info inferiors
3377 Id Description Executable
3384 @code{follow-exec-mode} is supported in native mode and
3385 @code{target extended-remote} mode.
3387 You can use the @code{catch} command to make @value{GDBN} stop whenever
3388 a @code{fork}, @code{vfork}, or @code{exec} call is made. @xref{Set
3389 Catchpoints, ,Setting Catchpoints}.
3391 @node Checkpoint/Restart
3392 @section Setting a @emph{Bookmark} to Return to Later
3397 @cindex snapshot of a process
3398 @cindex rewind program state
3400 On certain operating systems@footnote{Currently, only
3401 @sc{gnu}/Linux.}, @value{GDBN} is able to save a @dfn{snapshot} of a
3402 program's state, called a @dfn{checkpoint}, and come back to it
3405 Returning to a checkpoint effectively undoes everything that has
3406 happened in the program since the @code{checkpoint} was saved. This
3407 includes changes in memory, registers, and even (within some limits)
3408 system state. Effectively, it is like going back in time to the
3409 moment when the checkpoint was saved.
3411 Thus, if you're stepping thru a program and you think you're
3412 getting close to the point where things go wrong, you can save
3413 a checkpoint. Then, if you accidentally go too far and miss
3414 the critical statement, instead of having to restart your program
3415 from the beginning, you can just go back to the checkpoint and
3416 start again from there.
3418 This can be especially useful if it takes a lot of time or
3419 steps to reach the point where you think the bug occurs.
3421 To use the @code{checkpoint}/@code{restart} method of debugging:
3426 Save a snapshot of the debugged program's current execution state.
3427 The @code{checkpoint} command takes no arguments, but each checkpoint
3428 is assigned a small integer id, similar to a breakpoint id.
3430 @kindex info checkpoints
3431 @item info checkpoints
3432 List the checkpoints that have been saved in the current debugging
3433 session. For each checkpoint, the following information will be
3440 @item Source line, or label
3443 @kindex restart @var{checkpoint-id}
3444 @item restart @var{checkpoint-id}
3445 Restore the program state that was saved as checkpoint number
3446 @var{checkpoint-id}. All program variables, registers, stack frames
3447 etc.@: will be returned to the values that they had when the checkpoint
3448 was saved. In essence, gdb will ``wind back the clock'' to the point
3449 in time when the checkpoint was saved.
3451 Note that breakpoints, @value{GDBN} variables, command history etc.
3452 are not affected by restoring a checkpoint. In general, a checkpoint
3453 only restores things that reside in the program being debugged, not in
3456 @kindex delete checkpoint @var{checkpoint-id}
3457 @item delete checkpoint @var{checkpoint-id}
3458 Delete the previously-saved checkpoint identified by @var{checkpoint-id}.
3462 Returning to a previously saved checkpoint will restore the user state
3463 of the program being debugged, plus a significant subset of the system
3464 (OS) state, including file pointers. It won't ``un-write'' data from
3465 a file, but it will rewind the file pointer to the previous location,
3466 so that the previously written data can be overwritten. For files
3467 opened in read mode, the pointer will also be restored so that the
3468 previously read data can be read again.
3470 Of course, characters that have been sent to a printer (or other
3471 external device) cannot be ``snatched back'', and characters received
3472 from eg.@: a serial device can be removed from internal program buffers,
3473 but they cannot be ``pushed back'' into the serial pipeline, ready to
3474 be received again. Similarly, the actual contents of files that have
3475 been changed cannot be restored (at this time).
3477 However, within those constraints, you actually can ``rewind'' your
3478 program to a previously saved point in time, and begin debugging it
3479 again --- and you can change the course of events so as to debug a
3480 different execution path this time.
3482 @cindex checkpoints and process id
3483 Finally, there is one bit of internal program state that will be
3484 different when you return to a checkpoint --- the program's process
3485 id. Each checkpoint will have a unique process id (or @var{pid}),
3486 and each will be different from the program's original @var{pid}.
3487 If your program has saved a local copy of its process id, this could
3488 potentially pose a problem.
3490 @subsection A Non-obvious Benefit of Using Checkpoints
3492 On some systems such as @sc{gnu}/Linux, address space randomization
3493 is performed on new processes for security reasons. This makes it
3494 difficult or impossible to set a breakpoint, or watchpoint, on an
3495 absolute address if you have to restart the program, since the
3496 absolute location of a symbol will change from one execution to the
3499 A checkpoint, however, is an @emph{identical} copy of a process.
3500 Therefore if you create a checkpoint at (eg.@:) the start of main,
3501 and simply return to that checkpoint instead of restarting the
3502 process, you can avoid the effects of address randomization and
3503 your symbols will all stay in the same place.
3506 @chapter Stopping and Continuing
3508 The principal purposes of using a debugger are so that you can stop your
3509 program before it terminates; or so that, if your program runs into
3510 trouble, you can investigate and find out why.
3512 Inside @value{GDBN}, your program may stop for any of several reasons,
3513 such as a signal, a breakpoint, or reaching a new line after a
3514 @value{GDBN} command such as @code{step}. You may then examine and
3515 change variables, set new breakpoints or remove old ones, and then
3516 continue execution. Usually, the messages shown by @value{GDBN} provide
3517 ample explanation of the status of your program---but you can also
3518 explicitly request this information at any time.
3521 @kindex info program
3523 Display information about the status of your program: whether it is
3524 running or not, what process it is, and why it stopped.
3528 * Breakpoints:: Breakpoints, watchpoints, and catchpoints
3529 * Continuing and Stepping:: Resuming execution
3530 * Skipping Over Functions and Files::
3531 Skipping over functions and files
3533 * Thread Stops:: Stopping and starting multi-thread programs
3537 @section Breakpoints, Watchpoints, and Catchpoints
3540 A @dfn{breakpoint} makes your program stop whenever a certain point in
3541 the program is reached. For each breakpoint, you can add conditions to
3542 control in finer detail whether your program stops. You can set
3543 breakpoints with the @code{break} command and its variants (@pxref{Set
3544 Breaks, ,Setting Breakpoints}), to specify the place where your program
3545 should stop by line number, function name or exact address in the
3548 On some systems, you can set breakpoints in shared libraries before
3549 the executable is run.
3552 @cindex data breakpoints
3553 @cindex memory tracing
3554 @cindex breakpoint on memory address
3555 @cindex breakpoint on variable modification
3556 A @dfn{watchpoint} is a special breakpoint that stops your program
3557 when the value of an expression changes. The expression may be a value
3558 of a variable, or it could involve values of one or more variables
3559 combined by operators, such as @samp{a + b}. This is sometimes called
3560 @dfn{data breakpoints}. You must use a different command to set
3561 watchpoints (@pxref{Set Watchpoints, ,Setting Watchpoints}), but aside
3562 from that, you can manage a watchpoint like any other breakpoint: you
3563 enable, disable, and delete both breakpoints and watchpoints using the
3566 You can arrange to have values from your program displayed automatically
3567 whenever @value{GDBN} stops at a breakpoint. @xref{Auto Display,,
3571 @cindex breakpoint on events
3572 A @dfn{catchpoint} is another special breakpoint that stops your program
3573 when a certain kind of event occurs, such as the throwing of a C@t{++}
3574 exception or the loading of a library. As with watchpoints, you use a
3575 different command to set a catchpoint (@pxref{Set Catchpoints, ,Setting
3576 Catchpoints}), but aside from that, you can manage a catchpoint like any
3577 other breakpoint. (To stop when your program receives a signal, use the
3578 @code{handle} command; see @ref{Signals, ,Signals}.)
3580 @cindex breakpoint numbers
3581 @cindex numbers for breakpoints
3582 @value{GDBN} assigns a number to each breakpoint, watchpoint, or
3583 catchpoint when you create it; these numbers are successive integers
3584 starting with one. In many of the commands for controlling various
3585 features of breakpoints you use the breakpoint number to say which
3586 breakpoint you want to change. Each breakpoint may be @dfn{enabled} or
3587 @dfn{disabled}; if disabled, it has no effect on your program until you
3590 @cindex breakpoint ranges
3591 @cindex breakpoint lists
3592 @cindex ranges of breakpoints
3593 @cindex lists of breakpoints
3594 Some @value{GDBN} commands accept a space-separated list of breakpoints
3595 on which to operate. A list element can be either a single breakpoint number,
3596 like @samp{5}, or a range of such numbers, like @samp{5-7}.
3597 When a breakpoint list is given to a command, all breakpoints in that list
3601 * Set Breaks:: Setting breakpoints
3602 * Set Watchpoints:: Setting watchpoints
3603 * Set Catchpoints:: Setting catchpoints
3604 * Delete Breaks:: Deleting breakpoints
3605 * Disabling:: Disabling breakpoints
3606 * Conditions:: Break conditions
3607 * Break Commands:: Breakpoint command lists
3608 * Dynamic Printf:: Dynamic printf
3609 * Save Breakpoints:: How to save breakpoints in a file
3610 * Static Probe Points:: Listing static probe points
3611 * Error in Breakpoints:: ``Cannot insert breakpoints''
3612 * Breakpoint-related Warnings:: ``Breakpoint address adjusted...''
3616 @subsection Setting Breakpoints
3618 @c FIXME LMB what does GDB do if no code on line of breakpt?
3619 @c consider in particular declaration with/without initialization.
3621 @c FIXME 2 is there stuff on this already? break at fun start, already init?
3624 @kindex b @r{(@code{break})}
3625 @vindex $bpnum@r{, convenience variable}
3626 @cindex latest breakpoint
3627 Breakpoints are set with the @code{break} command (abbreviated
3628 @code{b}). The debugger convenience variable @samp{$bpnum} records the
3629 number of the breakpoint you've set most recently; see @ref{Convenience
3630 Vars,, Convenience Variables}, for a discussion of what you can do with
3631 convenience variables.
3634 @item break @var{location}
3635 Set a breakpoint at the given @var{location}, which can specify a
3636 function name, a line number, or an address of an instruction.
3637 (@xref{Specify Location}, for a list of all the possible ways to
3638 specify a @var{location}.) The breakpoint will stop your program just
3639 before it executes any of the code in the specified @var{location}.
3641 When using source languages that permit overloading of symbols, such as
3642 C@t{++}, a function name may refer to more than one possible place to break.
3643 @xref{Ambiguous Expressions,,Ambiguous Expressions}, for a discussion of
3646 It is also possible to insert a breakpoint that will stop the program
3647 only if a specific thread (@pxref{Thread-Specific Breakpoints})
3648 or a specific task (@pxref{Ada Tasks}) hits that breakpoint.
3651 When called without any arguments, @code{break} sets a breakpoint at
3652 the next instruction to be executed in the selected stack frame
3653 (@pxref{Stack, ,Examining the Stack}). In any selected frame but the
3654 innermost, this makes your program stop as soon as control
3655 returns to that frame. This is similar to the effect of a
3656 @code{finish} command in the frame inside the selected frame---except
3657 that @code{finish} does not leave an active breakpoint. If you use
3658 @code{break} without an argument in the innermost frame, @value{GDBN} stops
3659 the next time it reaches the current location; this may be useful
3662 @value{GDBN} normally ignores breakpoints when it resumes execution, until at
3663 least one instruction has been executed. If it did not do this, you
3664 would be unable to proceed past a breakpoint without first disabling the
3665 breakpoint. This rule applies whether or not the breakpoint already
3666 existed when your program stopped.
3668 @item break @dots{} if @var{cond}
3669 Set a breakpoint with condition @var{cond}; evaluate the expression
3670 @var{cond} each time the breakpoint is reached, and stop only if the
3671 value is nonzero---that is, if @var{cond} evaluates as true.
3672 @samp{@dots{}} stands for one of the possible arguments described
3673 above (or no argument) specifying where to break. @xref{Conditions,
3674 ,Break Conditions}, for more information on breakpoint conditions.
3677 @item tbreak @var{args}
3678 Set a breakpoint enabled only for one stop. The @var{args} are the
3679 same as for the @code{break} command, and the breakpoint is set in the same
3680 way, but the breakpoint is automatically deleted after the first time your
3681 program stops there. @xref{Disabling, ,Disabling Breakpoints}.
3684 @cindex hardware breakpoints
3685 @item hbreak @var{args}
3686 Set a hardware-assisted breakpoint. The @var{args} are the same as for the
3687 @code{break} command and the breakpoint is set in the same way, but the
3688 breakpoint requires hardware support and some target hardware may not
3689 have this support. The main purpose of this is EPROM/ROM code
3690 debugging, so you can set a breakpoint at an instruction without
3691 changing the instruction. This can be used with the new trap-generation
3692 provided by SPARClite DSU and most x86-based targets. These targets
3693 will generate traps when a program accesses some data or instruction
3694 address that is assigned to the debug registers. However the hardware
3695 breakpoint registers can take a limited number of breakpoints. For
3696 example, on the DSU, only two data breakpoints can be set at a time, and
3697 @value{GDBN} will reject this command if more than two are used. Delete
3698 or disable unused hardware breakpoints before setting new ones
3699 (@pxref{Disabling, ,Disabling Breakpoints}).
3700 @xref{Conditions, ,Break Conditions}.
3701 For remote targets, you can restrict the number of hardware
3702 breakpoints @value{GDBN} will use, see @ref{set remote
3703 hardware-breakpoint-limit}.
3706 @item thbreak @var{args}
3707 Set a hardware-assisted breakpoint enabled only for one stop. The @var{args}
3708 are the same as for the @code{hbreak} command and the breakpoint is set in
3709 the same way. However, like the @code{tbreak} command,
3710 the breakpoint is automatically deleted after the
3711 first time your program stops there. Also, like the @code{hbreak}
3712 command, the breakpoint requires hardware support and some target hardware
3713 may not have this support. @xref{Disabling, ,Disabling Breakpoints}.
3714 See also @ref{Conditions, ,Break Conditions}.
3717 @cindex regular expression
3718 @cindex breakpoints at functions matching a regexp
3719 @cindex set breakpoints in many functions
3720 @item rbreak @var{regex}
3721 Set breakpoints on all functions matching the regular expression
3722 @var{regex}. This command sets an unconditional breakpoint on all
3723 matches, printing a list of all breakpoints it set. Once these
3724 breakpoints are set, they are treated just like the breakpoints set with
3725 the @code{break} command. You can delete them, disable them, or make
3726 them conditional the same way as any other breakpoint.
3728 The syntax of the regular expression is the standard one used with tools
3729 like @file{grep}. Note that this is different from the syntax used by
3730 shells, so for instance @code{foo*} matches all functions that include
3731 an @code{fo} followed by zero or more @code{o}s. There is an implicit
3732 @code{.*} leading and trailing the regular expression you supply, so to
3733 match only functions that begin with @code{foo}, use @code{^foo}.
3735 @cindex non-member C@t{++} functions, set breakpoint in
3736 When debugging C@t{++} programs, @code{rbreak} is useful for setting
3737 breakpoints on overloaded functions that are not members of any special
3740 @cindex set breakpoints on all functions
3741 The @code{rbreak} command can be used to set breakpoints in
3742 @strong{all} the functions in a program, like this:
3745 (@value{GDBP}) rbreak .
3748 @item rbreak @var{file}:@var{regex}
3749 If @code{rbreak} is called with a filename qualification, it limits
3750 the search for functions matching the given regular expression to the
3751 specified @var{file}. This can be used, for example, to set breakpoints on
3752 every function in a given file:
3755 (@value{GDBP}) rbreak file.c:.
3758 The colon separating the filename qualifier from the regex may
3759 optionally be surrounded by spaces.
3761 @kindex info breakpoints
3762 @cindex @code{$_} and @code{info breakpoints}
3763 @item info breakpoints @r{[}@var{list}@dots{}@r{]}
3764 @itemx info break @r{[}@var{list}@dots{}@r{]}
3765 Print a table of all breakpoints, watchpoints, and catchpoints set and
3766 not deleted. Optional argument @var{n} means print information only
3767 about the specified breakpoint(s) (or watchpoint(s) or catchpoint(s)).
3768 For each breakpoint, following columns are printed:
3771 @item Breakpoint Numbers
3773 Breakpoint, watchpoint, or catchpoint.
3775 Whether the breakpoint is marked to be disabled or deleted when hit.
3776 @item Enabled or Disabled
3777 Enabled breakpoints are marked with @samp{y}. @samp{n} marks breakpoints
3778 that are not enabled.
3780 Where the breakpoint is in your program, as a memory address. For a
3781 pending breakpoint whose address is not yet known, this field will
3782 contain @samp{<PENDING>}. Such breakpoint won't fire until a shared
3783 library that has the symbol or line referred by breakpoint is loaded.
3784 See below for details. A breakpoint with several locations will
3785 have @samp{<MULTIPLE>} in this field---see below for details.
3787 Where the breakpoint is in the source for your program, as a file and
3788 line number. For a pending breakpoint, the original string passed to
3789 the breakpoint command will be listed as it cannot be resolved until
3790 the appropriate shared library is loaded in the future.
3794 If a breakpoint is conditional, there are two evaluation modes: ``host'' and
3795 ``target''. If mode is ``host'', breakpoint condition evaluation is done by
3796 @value{GDBN} on the host's side. If it is ``target'', then the condition
3797 is evaluated by the target. The @code{info break} command shows
3798 the condition on the line following the affected breakpoint, together with
3799 its condition evaluation mode in between parentheses.
3801 Breakpoint commands, if any, are listed after that. A pending breakpoint is
3802 allowed to have a condition specified for it. The condition is not parsed for
3803 validity until a shared library is loaded that allows the pending
3804 breakpoint to resolve to a valid location.
3807 @code{info break} with a breakpoint
3808 number @var{n} as argument lists only that breakpoint. The
3809 convenience variable @code{$_} and the default examining-address for
3810 the @code{x} command are set to the address of the last breakpoint
3811 listed (@pxref{Memory, ,Examining Memory}).
3814 @code{info break} displays a count of the number of times the breakpoint
3815 has been hit. This is especially useful in conjunction with the
3816 @code{ignore} command. You can ignore a large number of breakpoint
3817 hits, look at the breakpoint info to see how many times the breakpoint
3818 was hit, and then run again, ignoring one less than that number. This
3819 will get you quickly to the last hit of that breakpoint.
3822 For a breakpoints with an enable count (xref) greater than 1,
3823 @code{info break} also displays that count.
3827 @value{GDBN} allows you to set any number of breakpoints at the same place in
3828 your program. There is nothing silly or meaningless about this. When
3829 the breakpoints are conditional, this is even useful
3830 (@pxref{Conditions, ,Break Conditions}).
3832 @cindex multiple locations, breakpoints
3833 @cindex breakpoints, multiple locations
3834 It is possible that a breakpoint corresponds to several locations
3835 in your program. Examples of this situation are:
3839 Multiple functions in the program may have the same name.
3842 For a C@t{++} constructor, the @value{NGCC} compiler generates several
3843 instances of the function body, used in different cases.
3846 For a C@t{++} template function, a given line in the function can
3847 correspond to any number of instantiations.
3850 For an inlined function, a given source line can correspond to
3851 several places where that function is inlined.
3854 In all those cases, @value{GDBN} will insert a breakpoint at all
3855 the relevant locations.
3857 A breakpoint with multiple locations is displayed in the breakpoint
3858 table using several rows---one header row, followed by one row for
3859 each breakpoint location. The header row has @samp{<MULTIPLE>} in the
3860 address column. The rows for individual locations contain the actual
3861 addresses for locations, and show the functions to which those
3862 locations belong. The number column for a location is of the form
3863 @var{breakpoint-number}.@var{location-number}.
3868 Num Type Disp Enb Address What
3869 1 breakpoint keep y <MULTIPLE>
3871 breakpoint already hit 1 time
3872 1.1 y 0x080486a2 in void foo<int>() at t.cc:8
3873 1.2 y 0x080486ca in void foo<double>() at t.cc:8
3876 Each location can be individually enabled or disabled by passing
3877 @var{breakpoint-number}.@var{location-number} as argument to the
3878 @code{enable} and @code{disable} commands. Note that you cannot
3879 delete the individual locations from the list, you can only delete the
3880 entire list of locations that belong to their parent breakpoint (with
3881 the @kbd{delete @var{num}} command, where @var{num} is the number of
3882 the parent breakpoint, 1 in the above example). Disabling or enabling
3883 the parent breakpoint (@pxref{Disabling}) affects all of the locations
3884 that belong to that breakpoint.
3886 @cindex pending breakpoints
3887 It's quite common to have a breakpoint inside a shared library.
3888 Shared libraries can be loaded and unloaded explicitly,
3889 and possibly repeatedly, as the program is executed. To support
3890 this use case, @value{GDBN} updates breakpoint locations whenever
3891 any shared library is loaded or unloaded. Typically, you would
3892 set a breakpoint in a shared library at the beginning of your
3893 debugging session, when the library is not loaded, and when the
3894 symbols from the library are not available. When you try to set
3895 breakpoint, @value{GDBN} will ask you if you want to set
3896 a so called @dfn{pending breakpoint}---breakpoint whose address
3897 is not yet resolved.
3899 After the program is run, whenever a new shared library is loaded,
3900 @value{GDBN} reevaluates all the breakpoints. When a newly loaded
3901 shared library contains the symbol or line referred to by some
3902 pending breakpoint, that breakpoint is resolved and becomes an
3903 ordinary breakpoint. When a library is unloaded, all breakpoints
3904 that refer to its symbols or source lines become pending again.
3906 This logic works for breakpoints with multiple locations, too. For
3907 example, if you have a breakpoint in a C@t{++} template function, and
3908 a newly loaded shared library has an instantiation of that template,
3909 a new location is added to the list of locations for the breakpoint.
3911 Except for having unresolved address, pending breakpoints do not
3912 differ from regular breakpoints. You can set conditions or commands,
3913 enable and disable them and perform other breakpoint operations.
3915 @value{GDBN} provides some additional commands for controlling what
3916 happens when the @samp{break} command cannot resolve breakpoint
3917 address specification to an address:
3919 @kindex set breakpoint pending
3920 @kindex show breakpoint pending
3922 @item set breakpoint pending auto
3923 This is the default behavior. When @value{GDBN} cannot find the breakpoint
3924 location, it queries you whether a pending breakpoint should be created.
3926 @item set breakpoint pending on
3927 This indicates that an unrecognized breakpoint location should automatically
3928 result in a pending breakpoint being created.
3930 @item set breakpoint pending off
3931 This indicates that pending breakpoints are not to be created. Any
3932 unrecognized breakpoint location results in an error. This setting does
3933 not affect any pending breakpoints previously created.
3935 @item show breakpoint pending
3936 Show the current behavior setting for creating pending breakpoints.
3939 The settings above only affect the @code{break} command and its
3940 variants. Once breakpoint is set, it will be automatically updated
3941 as shared libraries are loaded and unloaded.
3943 @cindex automatic hardware breakpoints
3944 For some targets, @value{GDBN} can automatically decide if hardware or
3945 software breakpoints should be used, depending on whether the
3946 breakpoint address is read-only or read-write. This applies to
3947 breakpoints set with the @code{break} command as well as to internal
3948 breakpoints set by commands like @code{next} and @code{finish}. For
3949 breakpoints set with @code{hbreak}, @value{GDBN} will always use hardware
3952 You can control this automatic behaviour with the following commands:
3954 @kindex set breakpoint auto-hw
3955 @kindex show breakpoint auto-hw
3957 @item set breakpoint auto-hw on
3958 This is the default behavior. When @value{GDBN} sets a breakpoint, it
3959 will try to use the target memory map to decide if software or hardware
3960 breakpoint must be used.
3962 @item set breakpoint auto-hw off
3963 This indicates @value{GDBN} should not automatically select breakpoint
3964 type. If the target provides a memory map, @value{GDBN} will warn when
3965 trying to set software breakpoint at a read-only address.
3968 @value{GDBN} normally implements breakpoints by replacing the program code
3969 at the breakpoint address with a special instruction, which, when
3970 executed, given control to the debugger. By default, the program
3971 code is so modified only when the program is resumed. As soon as
3972 the program stops, @value{GDBN} restores the original instructions. This
3973 behaviour guards against leaving breakpoints inserted in the
3974 target should gdb abrubptly disconnect. However, with slow remote
3975 targets, inserting and removing breakpoint can reduce the performance.
3976 This behavior can be controlled with the following commands::
3978 @kindex set breakpoint always-inserted
3979 @kindex show breakpoint always-inserted
3981 @item set breakpoint always-inserted off
3982 All breakpoints, including newly added by the user, are inserted in
3983 the target only when the target is resumed. All breakpoints are
3984 removed from the target when it stops. This is the default mode.
3986 @item set breakpoint always-inserted on
3987 Causes all breakpoints to be inserted in the target at all times. If
3988 the user adds a new breakpoint, or changes an existing breakpoint, the
3989 breakpoints in the target are updated immediately. A breakpoint is
3990 removed from the target only when breakpoint itself is deleted.
3993 @value{GDBN} handles conditional breakpoints by evaluating these conditions
3994 when a breakpoint breaks. If the condition is true, then the process being
3995 debugged stops, otherwise the process is resumed.
3997 If the target supports evaluating conditions on its end, @value{GDBN} may
3998 download the breakpoint, together with its conditions, to it.
4000 This feature can be controlled via the following commands:
4002 @kindex set breakpoint condition-evaluation
4003 @kindex show breakpoint condition-evaluation
4005 @item set breakpoint condition-evaluation host
4006 This option commands @value{GDBN} to evaluate the breakpoint
4007 conditions on the host's side. Unconditional breakpoints are sent to
4008 the target which in turn receives the triggers and reports them back to GDB
4009 for condition evaluation. This is the standard evaluation mode.
4011 @item set breakpoint condition-evaluation target
4012 This option commands @value{GDBN} to download breakpoint conditions
4013 to the target at the moment of their insertion. The target
4014 is responsible for evaluating the conditional expression and reporting
4015 breakpoint stop events back to @value{GDBN} whenever the condition
4016 is true. Due to limitations of target-side evaluation, some conditions
4017 cannot be evaluated there, e.g., conditions that depend on local data
4018 that is only known to the host. Examples include
4019 conditional expressions involving convenience variables, complex types
4020 that cannot be handled by the agent expression parser and expressions
4021 that are too long to be sent over to the target, specially when the
4022 target is a remote system. In these cases, the conditions will be
4023 evaluated by @value{GDBN}.
4025 @item set breakpoint condition-evaluation auto
4026 This is the default mode. If the target supports evaluating breakpoint
4027 conditions on its end, @value{GDBN} will download breakpoint conditions to
4028 the target (limitations mentioned previously apply). If the target does
4029 not support breakpoint condition evaluation, then @value{GDBN} will fallback
4030 to evaluating all these conditions on the host's side.
4034 @cindex negative breakpoint numbers
4035 @cindex internal @value{GDBN} breakpoints
4036 @value{GDBN} itself sometimes sets breakpoints in your program for
4037 special purposes, such as proper handling of @code{longjmp} (in C
4038 programs). These internal breakpoints are assigned negative numbers,
4039 starting with @code{-1}; @samp{info breakpoints} does not display them.
4040 You can see these breakpoints with the @value{GDBN} maintenance command
4041 @samp{maint info breakpoints} (@pxref{maint info breakpoints}).
4044 @node Set Watchpoints
4045 @subsection Setting Watchpoints
4047 @cindex setting watchpoints
4048 You can use a watchpoint to stop execution whenever the value of an
4049 expression changes, without having to predict a particular place where
4050 this may happen. (This is sometimes called a @dfn{data breakpoint}.)
4051 The expression may be as simple as the value of a single variable, or
4052 as complex as many variables combined by operators. Examples include:
4056 A reference to the value of a single variable.
4059 An address cast to an appropriate data type. For example,
4060 @samp{*(int *)0x12345678} will watch a 4-byte region at the specified
4061 address (assuming an @code{int} occupies 4 bytes).
4064 An arbitrarily complex expression, such as @samp{a*b + c/d}. The
4065 expression can use any operators valid in the program's native
4066 language (@pxref{Languages}).
4069 You can set a watchpoint on an expression even if the expression can
4070 not be evaluated yet. For instance, you can set a watchpoint on
4071 @samp{*global_ptr} before @samp{global_ptr} is initialized.
4072 @value{GDBN} will stop when your program sets @samp{global_ptr} and
4073 the expression produces a valid value. If the expression becomes
4074 valid in some other way than changing a variable (e.g.@: if the memory
4075 pointed to by @samp{*global_ptr} becomes readable as the result of a
4076 @code{malloc} call), @value{GDBN} may not stop until the next time
4077 the expression changes.
4079 @cindex software watchpoints
4080 @cindex hardware watchpoints
4081 Depending on your system, watchpoints may be implemented in software or
4082 hardware. @value{GDBN} does software watchpointing by single-stepping your
4083 program and testing the variable's value each time, which is hundreds of
4084 times slower than normal execution. (But this may still be worth it, to
4085 catch errors where you have no clue what part of your program is the
4088 On some systems, such as most PowerPC or x86-based targets,
4089 @value{GDBN} includes support for hardware watchpoints, which do not
4090 slow down the running of your program.
4094 @item watch @r{[}-l@r{|}-location@r{]} @var{expr} @r{[}thread @var{thread-id}@r{]} @r{[}mask @var{maskvalue}@r{]}
4095 Set a watchpoint for an expression. @value{GDBN} will break when the
4096 expression @var{expr} is written into by the program and its value
4097 changes. The simplest (and the most popular) use of this command is
4098 to watch the value of a single variable:
4101 (@value{GDBP}) watch foo
4104 If the command includes a @code{@r{[}thread @var{thread-id}@r{]}}
4105 argument, @value{GDBN} breaks only when the thread identified by
4106 @var{thread-id} changes the value of @var{expr}. If any other threads
4107 change the value of @var{expr}, @value{GDBN} will not break. Note
4108 that watchpoints restricted to a single thread in this way only work
4109 with Hardware Watchpoints.
4111 Ordinarily a watchpoint respects the scope of variables in @var{expr}
4112 (see below). The @code{-location} argument tells @value{GDBN} to
4113 instead watch the memory referred to by @var{expr}. In this case,
4114 @value{GDBN} will evaluate @var{expr}, take the address of the result,
4115 and watch the memory at that address. The type of the result is used
4116 to determine the size of the watched memory. If the expression's
4117 result does not have an address, then @value{GDBN} will print an
4120 The @code{@r{[}mask @var{maskvalue}@r{]}} argument allows creation
4121 of masked watchpoints, if the current architecture supports this
4122 feature (e.g., PowerPC Embedded architecture, see @ref{PowerPC
4123 Embedded}.) A @dfn{masked watchpoint} specifies a mask in addition
4124 to an address to watch. The mask specifies that some bits of an address
4125 (the bits which are reset in the mask) should be ignored when matching
4126 the address accessed by the inferior against the watchpoint address.
4127 Thus, a masked watchpoint watches many addresses simultaneously---those
4128 addresses whose unmasked bits are identical to the unmasked bits in the
4129 watchpoint address. The @code{mask} argument implies @code{-location}.
4133 (@value{GDBP}) watch foo mask 0xffff00ff
4134 (@value{GDBP}) watch *0xdeadbeef mask 0xffffff00
4138 @item rwatch @r{[}-l@r{|}-location@r{]} @var{expr} @r{[}thread @var{thread-id}@r{]} @r{[}mask @var{maskvalue}@r{]}
4139 Set a watchpoint that will break when the value of @var{expr} is read
4143 @item awatch @r{[}-l@r{|}-location@r{]} @var{expr} @r{[}thread @var{thread-id}@r{]} @r{[}mask @var{maskvalue}@r{]}
4144 Set a watchpoint that will break when @var{expr} is either read from
4145 or written into by the program.
4147 @kindex info watchpoints @r{[}@var{list}@dots{}@r{]}
4148 @item info watchpoints @r{[}@var{list}@dots{}@r{]}
4149 This command prints a list of watchpoints, using the same format as
4150 @code{info break} (@pxref{Set Breaks}).
4153 If you watch for a change in a numerically entered address you need to
4154 dereference it, as the address itself is just a constant number which will
4155 never change. @value{GDBN} refuses to create a watchpoint that watches
4156 a never-changing value:
4159 (@value{GDBP}) watch 0x600850
4160 Cannot watch constant value 0x600850.
4161 (@value{GDBP}) watch *(int *) 0x600850
4162 Watchpoint 1: *(int *) 6293584
4165 @value{GDBN} sets a @dfn{hardware watchpoint} if possible. Hardware
4166 watchpoints execute very quickly, and the debugger reports a change in
4167 value at the exact instruction where the change occurs. If @value{GDBN}
4168 cannot set a hardware watchpoint, it sets a software watchpoint, which
4169 executes more slowly and reports the change in value at the next
4170 @emph{statement}, not the instruction, after the change occurs.
4172 @cindex use only software watchpoints
4173 You can force @value{GDBN} to use only software watchpoints with the
4174 @kbd{set can-use-hw-watchpoints 0} command. With this variable set to
4175 zero, @value{GDBN} will never try to use hardware watchpoints, even if
4176 the underlying system supports them. (Note that hardware-assisted
4177 watchpoints that were set @emph{before} setting
4178 @code{can-use-hw-watchpoints} to zero will still use the hardware
4179 mechanism of watching expression values.)
4182 @item set can-use-hw-watchpoints
4183 @kindex set can-use-hw-watchpoints
4184 Set whether or not to use hardware watchpoints.
4186 @item show can-use-hw-watchpoints
4187 @kindex show can-use-hw-watchpoints
4188 Show the current mode of using hardware watchpoints.
4191 For remote targets, you can restrict the number of hardware
4192 watchpoints @value{GDBN} will use, see @ref{set remote
4193 hardware-breakpoint-limit}.
4195 When you issue the @code{watch} command, @value{GDBN} reports
4198 Hardware watchpoint @var{num}: @var{expr}
4202 if it was able to set a hardware watchpoint.
4204 Currently, the @code{awatch} and @code{rwatch} commands can only set
4205 hardware watchpoints, because accesses to data that don't change the
4206 value of the watched expression cannot be detected without examining
4207 every instruction as it is being executed, and @value{GDBN} does not do
4208 that currently. If @value{GDBN} finds that it is unable to set a
4209 hardware breakpoint with the @code{awatch} or @code{rwatch} command, it
4210 will print a message like this:
4213 Expression cannot be implemented with read/access watchpoint.
4216 Sometimes, @value{GDBN} cannot set a hardware watchpoint because the
4217 data type of the watched expression is wider than what a hardware
4218 watchpoint on the target machine can handle. For example, some systems
4219 can only watch regions that are up to 4 bytes wide; on such systems you
4220 cannot set hardware watchpoints for an expression that yields a
4221 double-precision floating-point number (which is typically 8 bytes
4222 wide). As a work-around, it might be possible to break the large region
4223 into a series of smaller ones and watch them with separate watchpoints.
4225 If you set too many hardware watchpoints, @value{GDBN} might be unable
4226 to insert all of them when you resume the execution of your program.
4227 Since the precise number of active watchpoints is unknown until such
4228 time as the program is about to be resumed, @value{GDBN} might not be
4229 able to warn you about this when you set the watchpoints, and the
4230 warning will be printed only when the program is resumed:
4233 Hardware watchpoint @var{num}: Could not insert watchpoint
4237 If this happens, delete or disable some of the watchpoints.
4239 Watching complex expressions that reference many variables can also
4240 exhaust the resources available for hardware-assisted watchpoints.
4241 That's because @value{GDBN} needs to watch every variable in the
4242 expression with separately allocated resources.
4244 If you call a function interactively using @code{print} or @code{call},
4245 any watchpoints you have set will be inactive until @value{GDBN} reaches another
4246 kind of breakpoint or the call completes.
4248 @value{GDBN} automatically deletes watchpoints that watch local
4249 (automatic) variables, or expressions that involve such variables, when
4250 they go out of scope, that is, when the execution leaves the block in
4251 which these variables were defined. In particular, when the program
4252 being debugged terminates, @emph{all} local variables go out of scope,
4253 and so only watchpoints that watch global variables remain set. If you
4254 rerun the program, you will need to set all such watchpoints again. One
4255 way of doing that would be to set a code breakpoint at the entry to the
4256 @code{main} function and when it breaks, set all the watchpoints.
4258 @cindex watchpoints and threads
4259 @cindex threads and watchpoints
4260 In multi-threaded programs, watchpoints will detect changes to the
4261 watched expression from every thread.
4264 @emph{Warning:} In multi-threaded programs, software watchpoints
4265 have only limited usefulness. If @value{GDBN} creates a software
4266 watchpoint, it can only watch the value of an expression @emph{in a
4267 single thread}. If you are confident that the expression can only
4268 change due to the current thread's activity (and if you are also
4269 confident that no other thread can become current), then you can use
4270 software watchpoints as usual. However, @value{GDBN} may not notice
4271 when a non-current thread's activity changes the expression. (Hardware
4272 watchpoints, in contrast, watch an expression in all threads.)
4275 @xref{set remote hardware-watchpoint-limit}.
4277 @node Set Catchpoints
4278 @subsection Setting Catchpoints
4279 @cindex catchpoints, setting
4280 @cindex exception handlers
4281 @cindex event handling
4283 You can use @dfn{catchpoints} to cause the debugger to stop for certain
4284 kinds of program events, such as C@t{++} exceptions or the loading of a
4285 shared library. Use the @code{catch} command to set a catchpoint.
4289 @item catch @var{event}
4290 Stop when @var{event} occurs. The @var{event} can be any of the following:
4293 @item throw @r{[}@var{regexp}@r{]}
4294 @itemx rethrow @r{[}@var{regexp}@r{]}
4295 @itemx catch @r{[}@var{regexp}@r{]}
4297 @kindex catch rethrow
4299 @cindex stop on C@t{++} exceptions
4300 The throwing, re-throwing, or catching of a C@t{++} exception.
4302 If @var{regexp} is given, then only exceptions whose type matches the
4303 regular expression will be caught.
4305 @vindex $_exception@r{, convenience variable}
4306 The convenience variable @code{$_exception} is available at an
4307 exception-related catchpoint, on some systems. This holds the
4308 exception being thrown.
4310 There are currently some limitations to C@t{++} exception handling in
4315 The support for these commands is system-dependent. Currently, only
4316 systems using the @samp{gnu-v3} C@t{++} ABI (@pxref{ABI}) are
4320 The regular expression feature and the @code{$_exception} convenience
4321 variable rely on the presence of some SDT probes in @code{libstdc++}.
4322 If these probes are not present, then these features cannot be used.
4323 These probes were first available in the GCC 4.8 release, but whether
4324 or not they are available in your GCC also depends on how it was
4328 The @code{$_exception} convenience variable is only valid at the
4329 instruction at which an exception-related catchpoint is set.
4332 When an exception-related catchpoint is hit, @value{GDBN} stops at a
4333 location in the system library which implements runtime exception
4334 support for C@t{++}, usually @code{libstdc++}. You can use @code{up}
4335 (@pxref{Selection}) to get to your code.
4338 If you call a function interactively, @value{GDBN} normally returns
4339 control to you when the function has finished executing. If the call
4340 raises an exception, however, the call may bypass the mechanism that
4341 returns control to you and cause your program either to abort or to
4342 simply continue running until it hits a breakpoint, catches a signal
4343 that @value{GDBN} is listening for, or exits. This is the case even if
4344 you set a catchpoint for the exception; catchpoints on exceptions are
4345 disabled within interactive calls. @xref{Calling}, for information on
4346 controlling this with @code{set unwind-on-terminating-exception}.
4349 You cannot raise an exception interactively.
4352 You cannot install an exception handler interactively.
4356 @kindex catch exception
4357 @cindex Ada exception catching
4358 @cindex catch Ada exceptions
4359 An Ada exception being raised. If an exception name is specified
4360 at the end of the command (eg @code{catch exception Program_Error}),
4361 the debugger will stop only when this specific exception is raised.
4362 Otherwise, the debugger stops execution when any Ada exception is raised.
4364 When inserting an exception catchpoint on a user-defined exception whose
4365 name is identical to one of the exceptions defined by the language, the
4366 fully qualified name must be used as the exception name. Otherwise,
4367 @value{GDBN} will assume that it should stop on the pre-defined exception
4368 rather than the user-defined one. For instance, assuming an exception
4369 called @code{Constraint_Error} is defined in package @code{Pck}, then
4370 the command to use to catch such exceptions is @kbd{catch exception
4371 Pck.Constraint_Error}.
4373 @item exception unhandled
4374 @kindex catch exception unhandled
4375 An exception that was raised but is not handled by the program.
4378 @kindex catch assert
4379 A failed Ada assertion.
4383 @cindex break on fork/exec
4384 A call to @code{exec}.
4387 @itemx syscall @r{[}@var{name} @r{|} @var{number} @r{|} @r{group:}@var{groupname} @r{|} @r{g:}@var{groupname}@r{]} @dots{}
4388 @kindex catch syscall
4389 @cindex break on a system call.
4390 A call to or return from a system call, a.k.a.@: @dfn{syscall}. A
4391 syscall is a mechanism for application programs to request a service
4392 from the operating system (OS) or one of the OS system services.
4393 @value{GDBN} can catch some or all of the syscalls issued by the
4394 debuggee, and show the related information for each syscall. If no
4395 argument is specified, calls to and returns from all system calls
4398 @var{name} can be any system call name that is valid for the
4399 underlying OS. Just what syscalls are valid depends on the OS. On
4400 GNU and Unix systems, you can find the full list of valid syscall
4401 names on @file{/usr/include/asm/unistd.h}.
4403 @c For MS-Windows, the syscall names and the corresponding numbers
4404 @c can be found, e.g., on this URL:
4405 @c http://www.metasploit.com/users/opcode/syscalls.html
4406 @c but we don't support Windows syscalls yet.
4408 Normally, @value{GDBN} knows in advance which syscalls are valid for
4409 each OS, so you can use the @value{GDBN} command-line completion
4410 facilities (@pxref{Completion,, command completion}) to list the
4413 You may also specify the system call numerically. A syscall's
4414 number is the value passed to the OS's syscall dispatcher to
4415 identify the requested service. When you specify the syscall by its
4416 name, @value{GDBN} uses its database of syscalls to convert the name
4417 into the corresponding numeric code, but using the number directly
4418 may be useful if @value{GDBN}'s database does not have the complete
4419 list of syscalls on your system (e.g., because @value{GDBN} lags
4420 behind the OS upgrades).
4422 You may specify a group of related syscalls to be caught at once using
4423 the @code{group:} syntax (@code{g:} is a shorter equivalent). For
4424 instance, on some platforms @value{GDBN} allows you to catch all
4425 network related syscalls, by passing the argument @code{group:network}
4426 to @code{catch syscall}. Note that not all syscall groups are
4427 available in every system. You can use the command completion
4428 facilities (@pxref{Completion,, command completion}) to list the
4429 syscall groups available on your environment.
4431 The example below illustrates how this command works if you don't provide
4435 (@value{GDBP}) catch syscall
4436 Catchpoint 1 (syscall)
4438 Starting program: /tmp/catch-syscall
4440 Catchpoint 1 (call to syscall 'close'), \
4441 0xffffe424 in __kernel_vsyscall ()
4445 Catchpoint 1 (returned from syscall 'close'), \
4446 0xffffe424 in __kernel_vsyscall ()
4450 Here is an example of catching a system call by name:
4453 (@value{GDBP}) catch syscall chroot
4454 Catchpoint 1 (syscall 'chroot' [61])
4456 Starting program: /tmp/catch-syscall
4458 Catchpoint 1 (call to syscall 'chroot'), \
4459 0xffffe424 in __kernel_vsyscall ()
4463 Catchpoint 1 (returned from syscall 'chroot'), \
4464 0xffffe424 in __kernel_vsyscall ()
4468 An example of specifying a system call numerically. In the case
4469 below, the syscall number has a corresponding entry in the XML
4470 file, so @value{GDBN} finds its name and prints it:
4473 (@value{GDBP}) catch syscall 252
4474 Catchpoint 1 (syscall(s) 'exit_group')
4476 Starting program: /tmp/catch-syscall
4478 Catchpoint 1 (call to syscall 'exit_group'), \
4479 0xffffe424 in __kernel_vsyscall ()
4483 Program exited normally.
4487 Here is an example of catching a syscall group:
4490 (@value{GDBP}) catch syscall group:process
4491 Catchpoint 1 (syscalls 'exit' [1] 'fork' [2] 'waitpid' [7]
4492 'execve' [11] 'wait4' [114] 'clone' [120] 'vfork' [190]
4493 'exit_group' [252] 'waitid' [284] 'unshare' [310])
4495 Starting program: /tmp/catch-syscall
4497 Catchpoint 1 (call to syscall fork), 0x00007ffff7df4e27 in open64 ()
4498 from /lib64/ld-linux-x86-64.so.2
4504 However, there can be situations when there is no corresponding name
4505 in XML file for that syscall number. In this case, @value{GDBN} prints
4506 a warning message saying that it was not able to find the syscall name,
4507 but the catchpoint will be set anyway. See the example below:
4510 (@value{GDBP}) catch syscall 764
4511 warning: The number '764' does not represent a known syscall.
4512 Catchpoint 2 (syscall 764)
4516 If you configure @value{GDBN} using the @samp{--without-expat} option,
4517 it will not be able to display syscall names. Also, if your
4518 architecture does not have an XML file describing its system calls,
4519 you will not be able to see the syscall names. It is important to
4520 notice that these two features are used for accessing the syscall
4521 name database. In either case, you will see a warning like this:
4524 (@value{GDBP}) catch syscall
4525 warning: Could not open "syscalls/i386-linux.xml"
4526 warning: Could not load the syscall XML file 'syscalls/i386-linux.xml'.
4527 GDB will not be able to display syscall names.
4528 Catchpoint 1 (syscall)
4532 Of course, the file name will change depending on your architecture and system.
4534 Still using the example above, you can also try to catch a syscall by its
4535 number. In this case, you would see something like:
4538 (@value{GDBP}) catch syscall 252
4539 Catchpoint 1 (syscall(s) 252)
4542 Again, in this case @value{GDBN} would not be able to display syscall's names.
4546 A call to @code{fork}.
4550 A call to @code{vfork}.
4552 @item load @r{[}regexp@r{]}
4553 @itemx unload @r{[}regexp@r{]}
4555 @kindex catch unload
4556 The loading or unloading of a shared library. If @var{regexp} is
4557 given, then the catchpoint will stop only if the regular expression
4558 matches one of the affected libraries.
4560 @item signal @r{[}@var{signal}@dots{} @r{|} @samp{all}@r{]}
4561 @kindex catch signal
4562 The delivery of a signal.
4564 With no arguments, this catchpoint will catch any signal that is not
4565 used internally by @value{GDBN}, specifically, all signals except
4566 @samp{SIGTRAP} and @samp{SIGINT}.
4568 With the argument @samp{all}, all signals, including those used by
4569 @value{GDBN}, will be caught. This argument cannot be used with other
4572 Otherwise, the arguments are a list of signal names as given to
4573 @code{handle} (@pxref{Signals}). Only signals specified in this list
4576 One reason that @code{catch signal} can be more useful than
4577 @code{handle} is that you can attach commands and conditions to the
4580 When a signal is caught by a catchpoint, the signal's @code{stop} and
4581 @code{print} settings, as specified by @code{handle}, are ignored.
4582 However, whether the signal is still delivered to the inferior depends
4583 on the @code{pass} setting; this can be changed in the catchpoint's
4588 @item tcatch @var{event}
4590 Set a catchpoint that is enabled only for one stop. The catchpoint is
4591 automatically deleted after the first time the event is caught.
4595 Use the @code{info break} command to list the current catchpoints.
4599 @subsection Deleting Breakpoints
4601 @cindex clearing breakpoints, watchpoints, catchpoints
4602 @cindex deleting breakpoints, watchpoints, catchpoints
4603 It is often necessary to eliminate a breakpoint, watchpoint, or
4604 catchpoint once it has done its job and you no longer want your program
4605 to stop there. This is called @dfn{deleting} the breakpoint. A
4606 breakpoint that has been deleted no longer exists; it is forgotten.
4608 With the @code{clear} command you can delete breakpoints according to
4609 where they are in your program. With the @code{delete} command you can
4610 delete individual breakpoints, watchpoints, or catchpoints by specifying
4611 their breakpoint numbers.
4613 It is not necessary to delete a breakpoint to proceed past it. @value{GDBN}
4614 automatically ignores breakpoints on the first instruction to be executed
4615 when you continue execution without changing the execution address.
4620 Delete any breakpoints at the next instruction to be executed in the
4621 selected stack frame (@pxref{Selection, ,Selecting a Frame}). When
4622 the innermost frame is selected, this is a good way to delete a
4623 breakpoint where your program just stopped.
4625 @item clear @var{location}
4626 Delete any breakpoints set at the specified @var{location}.
4627 @xref{Specify Location}, for the various forms of @var{location}; the
4628 most useful ones are listed below:
4631 @item clear @var{function}
4632 @itemx clear @var{filename}:@var{function}
4633 Delete any breakpoints set at entry to the named @var{function}.
4635 @item clear @var{linenum}
4636 @itemx clear @var{filename}:@var{linenum}
4637 Delete any breakpoints set at or within the code of the specified
4638 @var{linenum} of the specified @var{filename}.
4641 @cindex delete breakpoints
4643 @kindex d @r{(@code{delete})}
4644 @item delete @r{[}breakpoints@r{]} @r{[}@var{list}@dots{}@r{]}
4645 Delete the breakpoints, watchpoints, or catchpoints of the breakpoint
4646 list specified as argument. If no argument is specified, delete all
4647 breakpoints (@value{GDBN} asks confirmation, unless you have @code{set
4648 confirm off}). You can abbreviate this command as @code{d}.
4652 @subsection Disabling Breakpoints
4654 @cindex enable/disable a breakpoint
4655 Rather than deleting a breakpoint, watchpoint, or catchpoint, you might
4656 prefer to @dfn{disable} it. This makes the breakpoint inoperative as if
4657 it had been deleted, but remembers the information on the breakpoint so
4658 that you can @dfn{enable} it again later.
4660 You disable and enable breakpoints, watchpoints, and catchpoints with
4661 the @code{enable} and @code{disable} commands, optionally specifying
4662 one or more breakpoint numbers as arguments. Use @code{info break} to
4663 print a list of all breakpoints, watchpoints, and catchpoints if you
4664 do not know which numbers to use.
4666 Disabling and enabling a breakpoint that has multiple locations
4667 affects all of its locations.
4669 A breakpoint, watchpoint, or catchpoint can have any of several
4670 different states of enablement:
4674 Enabled. The breakpoint stops your program. A breakpoint set
4675 with the @code{break} command starts out in this state.
4677 Disabled. The breakpoint has no effect on your program.
4679 Enabled once. The breakpoint stops your program, but then becomes
4682 Enabled for a count. The breakpoint stops your program for the next
4683 N times, then becomes disabled.
4685 Enabled for deletion. The breakpoint stops your program, but
4686 immediately after it does so it is deleted permanently. A breakpoint
4687 set with the @code{tbreak} command starts out in this state.
4690 You can use the following commands to enable or disable breakpoints,
4691 watchpoints, and catchpoints:
4695 @kindex dis @r{(@code{disable})}
4696 @item disable @r{[}breakpoints@r{]} @r{[}@var{list}@dots{}@r{]}
4697 Disable the specified breakpoints---or all breakpoints, if none are
4698 listed. A disabled breakpoint has no effect but is not forgotten. All
4699 options such as ignore-counts, conditions and commands are remembered in
4700 case the breakpoint is enabled again later. You may abbreviate
4701 @code{disable} as @code{dis}.
4704 @item enable @r{[}breakpoints@r{]} @r{[}@var{list}@dots{}@r{]}
4705 Enable the specified breakpoints (or all defined breakpoints). They
4706 become effective once again in stopping your program.
4708 @item enable @r{[}breakpoints@r{]} once @var{list}@dots{}
4709 Enable the specified breakpoints temporarily. @value{GDBN} disables any
4710 of these breakpoints immediately after stopping your program.
4712 @item enable @r{[}breakpoints@r{]} count @var{count} @var{list}@dots{}
4713 Enable the specified breakpoints temporarily. @value{GDBN} records
4714 @var{count} with each of the specified breakpoints, and decrements a
4715 breakpoint's count when it is hit. When any count reaches 0,
4716 @value{GDBN} disables that breakpoint. If a breakpoint has an ignore
4717 count (@pxref{Conditions, ,Break Conditions}), that will be
4718 decremented to 0 before @var{count} is affected.
4720 @item enable @r{[}breakpoints@r{]} delete @var{list}@dots{}
4721 Enable the specified breakpoints to work once, then die. @value{GDBN}
4722 deletes any of these breakpoints as soon as your program stops there.
4723 Breakpoints set by the @code{tbreak} command start out in this state.
4726 @c FIXME: I think the following ``Except for [...] @code{tbreak}'' is
4727 @c confusing: tbreak is also initially enabled.
4728 Except for a breakpoint set with @code{tbreak} (@pxref{Set Breaks,
4729 ,Setting Breakpoints}), breakpoints that you set are initially enabled;
4730 subsequently, they become disabled or enabled only when you use one of
4731 the commands above. (The command @code{until} can set and delete a
4732 breakpoint of its own, but it does not change the state of your other
4733 breakpoints; see @ref{Continuing and Stepping, ,Continuing and
4737 @subsection Break Conditions
4738 @cindex conditional breakpoints
4739 @cindex breakpoint conditions
4741 @c FIXME what is scope of break condition expr? Context where wanted?
4742 @c in particular for a watchpoint?
4743 The simplest sort of breakpoint breaks every time your program reaches a
4744 specified place. You can also specify a @dfn{condition} for a
4745 breakpoint. A condition is just a Boolean expression in your
4746 programming language (@pxref{Expressions, ,Expressions}). A breakpoint with
4747 a condition evaluates the expression each time your program reaches it,
4748 and your program stops only if the condition is @emph{true}.
4750 This is the converse of using assertions for program validation; in that
4751 situation, you want to stop when the assertion is violated---that is,
4752 when the condition is false. In C, if you want to test an assertion expressed
4753 by the condition @var{assert}, you should set the condition
4754 @samp{! @var{assert}} on the appropriate breakpoint.
4756 Conditions are also accepted for watchpoints; you may not need them,
4757 since a watchpoint is inspecting the value of an expression anyhow---but
4758 it might be simpler, say, to just set a watchpoint on a variable name,
4759 and specify a condition that tests whether the new value is an interesting
4762 Break conditions can have side effects, and may even call functions in
4763 your program. This can be useful, for example, to activate functions
4764 that log program progress, or to use your own print functions to
4765 format special data structures. The effects are completely predictable
4766 unless there is another enabled breakpoint at the same address. (In
4767 that case, @value{GDBN} might see the other breakpoint first and stop your
4768 program without checking the condition of this one.) Note that
4769 breakpoint commands are usually more convenient and flexible than break
4771 purpose of performing side effects when a breakpoint is reached
4772 (@pxref{Break Commands, ,Breakpoint Command Lists}).
4774 Breakpoint conditions can also be evaluated on the target's side if
4775 the target supports it. Instead of evaluating the conditions locally,
4776 @value{GDBN} encodes the expression into an agent expression
4777 (@pxref{Agent Expressions}) suitable for execution on the target,
4778 independently of @value{GDBN}. Global variables become raw memory
4779 locations, locals become stack accesses, and so forth.
4781 In this case, @value{GDBN} will only be notified of a breakpoint trigger
4782 when its condition evaluates to true. This mechanism may provide faster
4783 response times depending on the performance characteristics of the target
4784 since it does not need to keep @value{GDBN} informed about
4785 every breakpoint trigger, even those with false conditions.
4787 Break conditions can be specified when a breakpoint is set, by using
4788 @samp{if} in the arguments to the @code{break} command. @xref{Set
4789 Breaks, ,Setting Breakpoints}. They can also be changed at any time
4790 with the @code{condition} command.
4792 You can also use the @code{if} keyword with the @code{watch} command.
4793 The @code{catch} command does not recognize the @code{if} keyword;
4794 @code{condition} is the only way to impose a further condition on a
4799 @item condition @var{bnum} @var{expression}
4800 Specify @var{expression} as the break condition for breakpoint,
4801 watchpoint, or catchpoint number @var{bnum}. After you set a condition,
4802 breakpoint @var{bnum} stops your program only if the value of
4803 @var{expression} is true (nonzero, in C). When you use
4804 @code{condition}, @value{GDBN} checks @var{expression} immediately for
4805 syntactic correctness, and to determine whether symbols in it have
4806 referents in the context of your breakpoint. If @var{expression} uses
4807 symbols not referenced in the context of the breakpoint, @value{GDBN}
4808 prints an error message:
4811 No symbol "foo" in current context.
4816 not actually evaluate @var{expression} at the time the @code{condition}
4817 command (or a command that sets a breakpoint with a condition, like
4818 @code{break if @dots{}}) is given, however. @xref{Expressions, ,Expressions}.
4820 @item condition @var{bnum}
4821 Remove the condition from breakpoint number @var{bnum}. It becomes
4822 an ordinary unconditional breakpoint.
4825 @cindex ignore count (of breakpoint)
4826 A special case of a breakpoint condition is to stop only when the
4827 breakpoint has been reached a certain number of times. This is so
4828 useful that there is a special way to do it, using the @dfn{ignore
4829 count} of the breakpoint. Every breakpoint has an ignore count, which
4830 is an integer. Most of the time, the ignore count is zero, and
4831 therefore has no effect. But if your program reaches a breakpoint whose
4832 ignore count is positive, then instead of stopping, it just decrements
4833 the ignore count by one and continues. As a result, if the ignore count
4834 value is @var{n}, the breakpoint does not stop the next @var{n} times
4835 your program reaches it.
4839 @item ignore @var{bnum} @var{count}
4840 Set the ignore count of breakpoint number @var{bnum} to @var{count}.
4841 The next @var{count} times the breakpoint is reached, your program's
4842 execution does not stop; other than to decrement the ignore count, @value{GDBN}
4845 To make the breakpoint stop the next time it is reached, specify
4848 When you use @code{continue} to resume execution of your program from a
4849 breakpoint, you can specify an ignore count directly as an argument to
4850 @code{continue}, rather than using @code{ignore}. @xref{Continuing and
4851 Stepping,,Continuing and Stepping}.
4853 If a breakpoint has a positive ignore count and a condition, the
4854 condition is not checked. Once the ignore count reaches zero,
4855 @value{GDBN} resumes checking the condition.
4857 You could achieve the effect of the ignore count with a condition such
4858 as @w{@samp{$foo-- <= 0}} using a debugger convenience variable that
4859 is decremented each time. @xref{Convenience Vars, ,Convenience
4863 Ignore counts apply to breakpoints, watchpoints, and catchpoints.
4866 @node Break Commands
4867 @subsection Breakpoint Command Lists
4869 @cindex breakpoint commands
4870 You can give any breakpoint (or watchpoint or catchpoint) a series of
4871 commands to execute when your program stops due to that breakpoint. For
4872 example, you might want to print the values of certain expressions, or
4873 enable other breakpoints.
4877 @kindex end@r{ (breakpoint commands)}
4878 @item commands @r{[}@var{list}@dots{}@r{]}
4879 @itemx @dots{} @var{command-list} @dots{}
4881 Specify a list of commands for the given breakpoints. The commands
4882 themselves appear on the following lines. Type a line containing just
4883 @code{end} to terminate the commands.
4885 To remove all commands from a breakpoint, type @code{commands} and
4886 follow it immediately with @code{end}; that is, give no commands.
4888 With no argument, @code{commands} refers to the last breakpoint,
4889 watchpoint, or catchpoint set (not to the breakpoint most recently
4890 encountered). If the most recent breakpoints were set with a single
4891 command, then the @code{commands} will apply to all the breakpoints
4892 set by that command. This applies to breakpoints set by
4893 @code{rbreak}, and also applies when a single @code{break} command
4894 creates multiple breakpoints (@pxref{Ambiguous Expressions,,Ambiguous
4898 Pressing @key{RET} as a means of repeating the last @value{GDBN} command is
4899 disabled within a @var{command-list}.
4901 You can use breakpoint commands to start your program up again. Simply
4902 use the @code{continue} command, or @code{step}, or any other command
4903 that resumes execution.
4905 Any other commands in the command list, after a command that resumes
4906 execution, are ignored. This is because any time you resume execution
4907 (even with a simple @code{next} or @code{step}), you may encounter
4908 another breakpoint---which could have its own command list, leading to
4909 ambiguities about which list to execute.
4912 If the first command you specify in a command list is @code{silent}, the
4913 usual message about stopping at a breakpoint is not printed. This may
4914 be desirable for breakpoints that are to print a specific message and
4915 then continue. If none of the remaining commands print anything, you
4916 see no sign that the breakpoint was reached. @code{silent} is
4917 meaningful only at the beginning of a breakpoint command list.
4919 The commands @code{echo}, @code{output}, and @code{printf} allow you to
4920 print precisely controlled output, and are often useful in silent
4921 breakpoints. @xref{Output, ,Commands for Controlled Output}.
4923 For example, here is how you could use breakpoint commands to print the
4924 value of @code{x} at entry to @code{foo} whenever @code{x} is positive.
4930 printf "x is %d\n",x
4935 One application for breakpoint commands is to compensate for one bug so
4936 you can test for another. Put a breakpoint just after the erroneous line
4937 of code, give it a condition to detect the case in which something
4938 erroneous has been done, and give it commands to assign correct values
4939 to any variables that need them. End with the @code{continue} command
4940 so that your program does not stop, and start with the @code{silent}
4941 command so that no output is produced. Here is an example:
4952 @node Dynamic Printf
4953 @subsection Dynamic Printf
4955 @cindex dynamic printf
4957 The dynamic printf command @code{dprintf} combines a breakpoint with
4958 formatted printing of your program's data to give you the effect of
4959 inserting @code{printf} calls into your program on-the-fly, without
4960 having to recompile it.
4962 In its most basic form, the output goes to the GDB console. However,
4963 you can set the variable @code{dprintf-style} for alternate handling.
4964 For instance, you can ask to format the output by calling your
4965 program's @code{printf} function. This has the advantage that the
4966 characters go to the program's output device, so they can recorded in
4967 redirects to files and so forth.
4969 If you are doing remote debugging with a stub or agent, you can also
4970 ask to have the printf handled by the remote agent. In addition to
4971 ensuring that the output goes to the remote program's device along
4972 with any other output the program might produce, you can also ask that
4973 the dprintf remain active even after disconnecting from the remote
4974 target. Using the stub/agent is also more efficient, as it can do
4975 everything without needing to communicate with @value{GDBN}.
4979 @item dprintf @var{location},@var{template},@var{expression}[,@var{expression}@dots{}]
4980 Whenever execution reaches @var{location}, print the values of one or
4981 more @var{expressions} under the control of the string @var{template}.
4982 To print several values, separate them with commas.
4984 @item set dprintf-style @var{style}
4985 Set the dprintf output to be handled in one of several different
4986 styles enumerated below. A change of style affects all existing
4987 dynamic printfs immediately. (If you need individual control over the
4988 print commands, simply define normal breakpoints with
4989 explicitly-supplied command lists.)
4993 @kindex dprintf-style gdb
4994 Handle the output using the @value{GDBN} @code{printf} command.
4997 @kindex dprintf-style call
4998 Handle the output by calling a function in your program (normally
5002 @kindex dprintf-style agent
5003 Have the remote debugging agent (such as @code{gdbserver}) handle
5004 the output itself. This style is only available for agents that
5005 support running commands on the target.
5008 @item set dprintf-function @var{function}
5009 Set the function to call if the dprintf style is @code{call}. By
5010 default its value is @code{printf}. You may set it to any expression.
5011 that @value{GDBN} can evaluate to a function, as per the @code{call}
5014 @item set dprintf-channel @var{channel}
5015 Set a ``channel'' for dprintf. If set to a non-empty value,
5016 @value{GDBN} will evaluate it as an expression and pass the result as
5017 a first argument to the @code{dprintf-function}, in the manner of
5018 @code{fprintf} and similar functions. Otherwise, the dprintf format
5019 string will be the first argument, in the manner of @code{printf}.
5021 As an example, if you wanted @code{dprintf} output to go to a logfile
5022 that is a standard I/O stream assigned to the variable @code{mylog},
5023 you could do the following:
5026 (gdb) set dprintf-style call
5027 (gdb) set dprintf-function fprintf
5028 (gdb) set dprintf-channel mylog
5029 (gdb) dprintf 25,"at line 25, glob=%d\n",glob
5030 Dprintf 1 at 0x123456: file main.c, line 25.
5032 1 dprintf keep y 0x00123456 in main at main.c:25
5033 call (void) fprintf (mylog,"at line 25, glob=%d\n",glob)
5038 Note that the @code{info break} displays the dynamic printf commands
5039 as normal breakpoint commands; you can thus easily see the effect of
5040 the variable settings.
5042 @item set disconnected-dprintf on
5043 @itemx set disconnected-dprintf off
5044 @kindex set disconnected-dprintf
5045 Choose whether @code{dprintf} commands should continue to run if
5046 @value{GDBN} has disconnected from the target. This only applies
5047 if the @code{dprintf-style} is @code{agent}.
5049 @item show disconnected-dprintf off
5050 @kindex show disconnected-dprintf
5051 Show the current choice for disconnected @code{dprintf}.
5055 @value{GDBN} does not check the validity of function and channel,
5056 relying on you to supply values that are meaningful for the contexts
5057 in which they are being used. For instance, the function and channel
5058 may be the values of local variables, but if that is the case, then
5059 all enabled dynamic prints must be at locations within the scope of
5060 those locals. If evaluation fails, @value{GDBN} will report an error.
5062 @node Save Breakpoints
5063 @subsection How to save breakpoints to a file
5065 To save breakpoint definitions to a file use the @w{@code{save
5066 breakpoints}} command.
5069 @kindex save breakpoints
5070 @cindex save breakpoints to a file for future sessions
5071 @item save breakpoints [@var{filename}]
5072 This command saves all current breakpoint definitions together with
5073 their commands and ignore counts, into a file @file{@var{filename}}
5074 suitable for use in a later debugging session. This includes all
5075 types of breakpoints (breakpoints, watchpoints, catchpoints,
5076 tracepoints). To read the saved breakpoint definitions, use the
5077 @code{source} command (@pxref{Command Files}). Note that watchpoints
5078 with expressions involving local variables may fail to be recreated
5079 because it may not be possible to access the context where the
5080 watchpoint is valid anymore. Because the saved breakpoint definitions
5081 are simply a sequence of @value{GDBN} commands that recreate the
5082 breakpoints, you can edit the file in your favorite editing program,
5083 and remove the breakpoint definitions you're not interested in, or
5084 that can no longer be recreated.
5087 @node Static Probe Points
5088 @subsection Static Probe Points
5090 @cindex static probe point, SystemTap
5091 @cindex static probe point, DTrace
5092 @value{GDBN} supports @dfn{SDT} probes in the code. @acronym{SDT} stands
5093 for Statically Defined Tracing, and the probes are designed to have a tiny
5094 runtime code and data footprint, and no dynamic relocations.
5096 Currently, the following types of probes are supported on
5097 ELF-compatible systems:
5101 @item @code{SystemTap} (@uref{http://sourceware.org/systemtap/})
5102 @acronym{SDT} probes@footnote{See
5103 @uref{http://sourceware.org/systemtap/wiki/AddingUserSpaceProbingToApps}
5104 for more information on how to add @code{SystemTap} @acronym{SDT}
5105 probes in your applications.}. @code{SystemTap} probes are usable
5106 from assembly, C and C@t{++} languages@footnote{See
5107 @uref{http://sourceware.org/systemtap/wiki/UserSpaceProbeImplementation}
5108 for a good reference on how the @acronym{SDT} probes are implemented.}.
5110 @item @code{DTrace} (@uref{http://oss.oracle.com/projects/DTrace})
5111 @acronym{USDT} probes. @code{DTrace} probes are usable from C and
5115 @cindex semaphores on static probe points
5116 Some @code{SystemTap} probes have an associated semaphore variable;
5117 for instance, this happens automatically if you defined your probe
5118 using a DTrace-style @file{.d} file. If your probe has a semaphore,
5119 @value{GDBN} will automatically enable it when you specify a
5120 breakpoint using the @samp{-probe-stap} notation. But, if you put a
5121 breakpoint at a probe's location by some other method (e.g.,
5122 @code{break file:line}), then @value{GDBN} will not automatically set
5123 the semaphore. @code{DTrace} probes do not support semaphores.
5125 You can examine the available static static probes using @code{info
5126 probes}, with optional arguments:
5130 @item info probes @r{[}@var{type}@r{]} @r{[}@var{provider} @r{[}@var{name} @r{[}@var{objfile}@r{]}@r{]}@r{]}
5131 If given, @var{type} is either @code{stap} for listing
5132 @code{SystemTap} probes or @code{dtrace} for listing @code{DTrace}
5133 probes. If omitted all probes are listed regardless of their types.
5135 If given, @var{provider} is a regular expression used to match against provider
5136 names when selecting which probes to list. If omitted, probes by all
5137 probes from all providers are listed.
5139 If given, @var{name} is a regular expression to match against probe names
5140 when selecting which probes to list. If omitted, probe names are not
5141 considered when deciding whether to display them.
5143 If given, @var{objfile} is a regular expression used to select which
5144 object files (executable or shared libraries) to examine. If not
5145 given, all object files are considered.
5147 @item info probes all
5148 List the available static probes, from all types.
5151 @cindex enabling and disabling probes
5152 Some probe points can be enabled and/or disabled. The effect of
5153 enabling or disabling a probe depends on the type of probe being
5154 handled. Some @code{DTrace} probes can be enabled or
5155 disabled, but @code{SystemTap} probes cannot be disabled.
5157 You can enable (or disable) one or more probes using the following
5158 commands, with optional arguments:
5161 @kindex enable probes
5162 @item enable probes @r{[}@var{provider} @r{[}@var{name} @r{[}@var{objfile}@r{]}@r{]}@r{]}
5163 If given, @var{provider} is a regular expression used to match against
5164 provider names when selecting which probes to enable. If omitted,
5165 all probes from all providers are enabled.
5167 If given, @var{name} is a regular expression to match against probe
5168 names when selecting which probes to enable. If omitted, probe names
5169 are not considered when deciding whether to enable them.
5171 If given, @var{objfile} is a regular expression used to select which
5172 object files (executable or shared libraries) to examine. If not
5173 given, all object files are considered.
5175 @kindex disable probes
5176 @item disable probes @r{[}@var{provider} @r{[}@var{name} @r{[}@var{objfile}@r{]}@r{]}@r{]}
5177 See the @code{enable probes} command above for a description of the
5178 optional arguments accepted by this command.
5181 @vindex $_probe_arg@r{, convenience variable}
5182 A probe may specify up to twelve arguments. These are available at the
5183 point at which the probe is defined---that is, when the current PC is
5184 at the probe's location. The arguments are available using the
5185 convenience variables (@pxref{Convenience Vars})
5186 @code{$_probe_arg0}@dots{}@code{$_probe_arg11}. In @code{SystemTap}
5187 probes each probe argument is an integer of the appropriate size;
5188 types are not preserved. In @code{DTrace} probes types are preserved
5189 provided that they are recognized as such by @value{GDBN}; otherwise
5190 the value of the probe argument will be a long integer. The
5191 convenience variable @code{$_probe_argc} holds the number of arguments
5192 at the current probe point.
5194 These variables are always available, but attempts to access them at
5195 any location other than a probe point will cause @value{GDBN} to give
5199 @c @ifclear BARETARGET
5200 @node Error in Breakpoints
5201 @subsection ``Cannot insert breakpoints''
5203 If you request too many active hardware-assisted breakpoints and
5204 watchpoints, you will see this error message:
5206 @c FIXME: the precise wording of this message may change; the relevant
5207 @c source change is not committed yet (Sep 3, 1999).
5209 Stopped; cannot insert breakpoints.
5210 You may have requested too many hardware breakpoints and watchpoints.
5214 This message is printed when you attempt to resume the program, since
5215 only then @value{GDBN} knows exactly how many hardware breakpoints and
5216 watchpoints it needs to insert.
5218 When this message is printed, you need to disable or remove some of the
5219 hardware-assisted breakpoints and watchpoints, and then continue.
5221 @node Breakpoint-related Warnings
5222 @subsection ``Breakpoint address adjusted...''
5223 @cindex breakpoint address adjusted
5225 Some processor architectures place constraints on the addresses at
5226 which breakpoints may be placed. For architectures thus constrained,
5227 @value{GDBN} will attempt to adjust the breakpoint's address to comply
5228 with the constraints dictated by the architecture.
5230 One example of such an architecture is the Fujitsu FR-V. The FR-V is
5231 a VLIW architecture in which a number of RISC-like instructions may be
5232 bundled together for parallel execution. The FR-V architecture
5233 constrains the location of a breakpoint instruction within such a
5234 bundle to the instruction with the lowest address. @value{GDBN}
5235 honors this constraint by adjusting a breakpoint's address to the
5236 first in the bundle.
5238 It is not uncommon for optimized code to have bundles which contain
5239 instructions from different source statements, thus it may happen that
5240 a breakpoint's address will be adjusted from one source statement to
5241 another. Since this adjustment may significantly alter @value{GDBN}'s
5242 breakpoint related behavior from what the user expects, a warning is
5243 printed when the breakpoint is first set and also when the breakpoint
5246 A warning like the one below is printed when setting a breakpoint
5247 that's been subject to address adjustment:
5250 warning: Breakpoint address adjusted from 0x00010414 to 0x00010410.
5253 Such warnings are printed both for user settable and @value{GDBN}'s
5254 internal breakpoints. If you see one of these warnings, you should
5255 verify that a breakpoint set at the adjusted address will have the
5256 desired affect. If not, the breakpoint in question may be removed and
5257 other breakpoints may be set which will have the desired behavior.
5258 E.g., it may be sufficient to place the breakpoint at a later
5259 instruction. A conditional breakpoint may also be useful in some
5260 cases to prevent the breakpoint from triggering too often.
5262 @value{GDBN} will also issue a warning when stopping at one of these
5263 adjusted breakpoints:
5266 warning: Breakpoint 1 address previously adjusted from 0x00010414
5270 When this warning is encountered, it may be too late to take remedial
5271 action except in cases where the breakpoint is hit earlier or more
5272 frequently than expected.
5274 @node Continuing and Stepping
5275 @section Continuing and Stepping
5279 @cindex resuming execution
5280 @dfn{Continuing} means resuming program execution until your program
5281 completes normally. In contrast, @dfn{stepping} means executing just
5282 one more ``step'' of your program, where ``step'' may mean either one
5283 line of source code, or one machine instruction (depending on what
5284 particular command you use). Either when continuing or when stepping,
5285 your program may stop even sooner, due to a breakpoint or a signal. (If
5286 it stops due to a signal, you may want to use @code{handle}, or use
5287 @samp{signal 0} to resume execution (@pxref{Signals, ,Signals}),
5288 or you may step into the signal's handler (@pxref{stepping and signal
5293 @kindex c @r{(@code{continue})}
5294 @kindex fg @r{(resume foreground execution)}
5295 @item continue @r{[}@var{ignore-count}@r{]}
5296 @itemx c @r{[}@var{ignore-count}@r{]}
5297 @itemx fg @r{[}@var{ignore-count}@r{]}
5298 Resume program execution, at the address where your program last stopped;
5299 any breakpoints set at that address are bypassed. The optional argument
5300 @var{ignore-count} allows you to specify a further number of times to
5301 ignore a breakpoint at this location; its effect is like that of
5302 @code{ignore} (@pxref{Conditions, ,Break Conditions}).
5304 The argument @var{ignore-count} is meaningful only when your program
5305 stopped due to a breakpoint. At other times, the argument to
5306 @code{continue} is ignored.
5308 The synonyms @code{c} and @code{fg} (for @dfn{foreground}, as the
5309 debugged program is deemed to be the foreground program) are provided
5310 purely for convenience, and have exactly the same behavior as
5314 To resume execution at a different place, you can use @code{return}
5315 (@pxref{Returning, ,Returning from a Function}) to go back to the
5316 calling function; or @code{jump} (@pxref{Jumping, ,Continuing at a
5317 Different Address}) to go to an arbitrary location in your program.
5319 A typical technique for using stepping is to set a breakpoint
5320 (@pxref{Breakpoints, ,Breakpoints; Watchpoints; and Catchpoints}) at the
5321 beginning of the function or the section of your program where a problem
5322 is believed to lie, run your program until it stops at that breakpoint,
5323 and then step through the suspect area, examining the variables that are
5324 interesting, until you see the problem happen.
5328 @kindex s @r{(@code{step})}
5330 Continue running your program until control reaches a different source
5331 line, then stop it and return control to @value{GDBN}. This command is
5332 abbreviated @code{s}.
5335 @c "without debugging information" is imprecise; actually "without line
5336 @c numbers in the debugging information". (gcc -g1 has debugging info but
5337 @c not line numbers). But it seems complex to try to make that
5338 @c distinction here.
5339 @emph{Warning:} If you use the @code{step} command while control is
5340 within a function that was compiled without debugging information,
5341 execution proceeds until control reaches a function that does have
5342 debugging information. Likewise, it will not step into a function which
5343 is compiled without debugging information. To step through functions
5344 without debugging information, use the @code{stepi} command, described
5348 The @code{step} command only stops at the first instruction of a source
5349 line. This prevents the multiple stops that could otherwise occur in
5350 @code{switch} statements, @code{for} loops, etc. @code{step} continues
5351 to stop if a function that has debugging information is called within
5352 the line. In other words, @code{step} @emph{steps inside} any functions
5353 called within the line.
5355 Also, the @code{step} command only enters a function if there is line
5356 number information for the function. Otherwise it acts like the
5357 @code{next} command. This avoids problems when using @code{cc -gl}
5358 on @acronym{MIPS} machines. Previously, @code{step} entered subroutines if there
5359 was any debugging information about the routine.
5361 @item step @var{count}
5362 Continue running as in @code{step}, but do so @var{count} times. If a
5363 breakpoint is reached, or a signal not related to stepping occurs before
5364 @var{count} steps, stepping stops right away.
5367 @kindex n @r{(@code{next})}
5368 @item next @r{[}@var{count}@r{]}
5369 Continue to the next source line in the current (innermost) stack frame.
5370 This is similar to @code{step}, but function calls that appear within
5371 the line of code are executed without stopping. Execution stops when
5372 control reaches a different line of code at the original stack level
5373 that was executing when you gave the @code{next} command. This command
5374 is abbreviated @code{n}.
5376 An argument @var{count} is a repeat count, as for @code{step}.
5379 @c FIX ME!! Do we delete this, or is there a way it fits in with
5380 @c the following paragraph? --- Vctoria
5382 @c @code{next} within a function that lacks debugging information acts like
5383 @c @code{step}, but any function calls appearing within the code of the
5384 @c function are executed without stopping.
5386 The @code{next} command only stops at the first instruction of a
5387 source line. This prevents multiple stops that could otherwise occur in
5388 @code{switch} statements, @code{for} loops, etc.
5390 @kindex set step-mode
5392 @cindex functions without line info, and stepping
5393 @cindex stepping into functions with no line info
5394 @itemx set step-mode on
5395 The @code{set step-mode on} command causes the @code{step} command to
5396 stop at the first instruction of a function which contains no debug line
5397 information rather than stepping over it.
5399 This is useful in cases where you may be interested in inspecting the
5400 machine instructions of a function which has no symbolic info and do not
5401 want @value{GDBN} to automatically skip over this function.
5403 @item set step-mode off
5404 Causes the @code{step} command to step over any functions which contains no
5405 debug information. This is the default.
5407 @item show step-mode
5408 Show whether @value{GDBN} will stop in or step over functions without
5409 source line debug information.
5412 @kindex fin @r{(@code{finish})}
5414 Continue running until just after function in the selected stack frame
5415 returns. Print the returned value (if any). This command can be
5416 abbreviated as @code{fin}.
5418 Contrast this with the @code{return} command (@pxref{Returning,
5419 ,Returning from a Function}).
5422 @kindex u @r{(@code{until})}
5423 @cindex run until specified location
5426 Continue running until a source line past the current line, in the
5427 current stack frame, is reached. This command is used to avoid single
5428 stepping through a loop more than once. It is like the @code{next}
5429 command, except that when @code{until} encounters a jump, it
5430 automatically continues execution until the program counter is greater
5431 than the address of the jump.
5433 This means that when you reach the end of a loop after single stepping
5434 though it, @code{until} makes your program continue execution until it
5435 exits the loop. In contrast, a @code{next} command at the end of a loop
5436 simply steps back to the beginning of the loop, which forces you to step
5437 through the next iteration.
5439 @code{until} always stops your program if it attempts to exit the current
5442 @code{until} may produce somewhat counterintuitive results if the order
5443 of machine code does not match the order of the source lines. For
5444 example, in the following excerpt from a debugging session, the @code{f}
5445 (@code{frame}) command shows that execution is stopped at line
5446 @code{206}; yet when we use @code{until}, we get to line @code{195}:
5450 #0 main (argc=4, argv=0xf7fffae8) at m4.c:206
5452 (@value{GDBP}) until
5453 195 for ( ; argc > 0; NEXTARG) @{
5456 This happened because, for execution efficiency, the compiler had
5457 generated code for the loop closure test at the end, rather than the
5458 start, of the loop---even though the test in a C @code{for}-loop is
5459 written before the body of the loop. The @code{until} command appeared
5460 to step back to the beginning of the loop when it advanced to this
5461 expression; however, it has not really gone to an earlier
5462 statement---not in terms of the actual machine code.
5464 @code{until} with no argument works by means of single
5465 instruction stepping, and hence is slower than @code{until} with an
5468 @item until @var{location}
5469 @itemx u @var{location}
5470 Continue running your program until either the specified @var{location} is
5471 reached, or the current stack frame returns. The location is any of
5472 the forms described in @ref{Specify Location}.
5473 This form of the command uses temporary breakpoints, and
5474 hence is quicker than @code{until} without an argument. The specified
5475 location is actually reached only if it is in the current frame. This
5476 implies that @code{until} can be used to skip over recursive function
5477 invocations. For instance in the code below, if the current location is
5478 line @code{96}, issuing @code{until 99} will execute the program up to
5479 line @code{99} in the same invocation of factorial, i.e., after the inner
5480 invocations have returned.
5483 94 int factorial (int value)
5485 96 if (value > 1) @{
5486 97 value *= factorial (value - 1);
5493 @kindex advance @var{location}
5494 @item advance @var{location}
5495 Continue running the program up to the given @var{location}. An argument is
5496 required, which should be of one of the forms described in
5497 @ref{Specify Location}.
5498 Execution will also stop upon exit from the current stack
5499 frame. This command is similar to @code{until}, but @code{advance} will
5500 not skip over recursive function calls, and the target location doesn't
5501 have to be in the same frame as the current one.
5505 @kindex si @r{(@code{stepi})}
5507 @itemx stepi @var{arg}
5509 Execute one machine instruction, then stop and return to the debugger.
5511 It is often useful to do @samp{display/i $pc} when stepping by machine
5512 instructions. This makes @value{GDBN} automatically display the next
5513 instruction to be executed, each time your program stops. @xref{Auto
5514 Display,, Automatic Display}.
5516 An argument is a repeat count, as in @code{step}.
5520 @kindex ni @r{(@code{nexti})}
5522 @itemx nexti @var{arg}
5524 Execute one machine instruction, but if it is a function call,
5525 proceed until the function returns.
5527 An argument is a repeat count, as in @code{next}.
5531 @anchor{range stepping}
5532 @cindex range stepping
5533 @cindex target-assisted range stepping
5534 By default, and if available, @value{GDBN} makes use of
5535 target-assisted @dfn{range stepping}. In other words, whenever you
5536 use a stepping command (e.g., @code{step}, @code{next}), @value{GDBN}
5537 tells the target to step the corresponding range of instruction
5538 addresses instead of issuing multiple single-steps. This speeds up
5539 line stepping, particularly for remote targets. Ideally, there should
5540 be no reason you would want to turn range stepping off. However, it's
5541 possible that a bug in the debug info, a bug in the remote stub (for
5542 remote targets), or even a bug in @value{GDBN} could make line
5543 stepping behave incorrectly when target-assisted range stepping is
5544 enabled. You can use the following command to turn off range stepping
5548 @kindex set range-stepping
5549 @kindex show range-stepping
5550 @item set range-stepping
5551 @itemx show range-stepping
5552 Control whether range stepping is enabled.
5554 If @code{on}, and the target supports it, @value{GDBN} tells the
5555 target to step a range of addresses itself, instead of issuing
5556 multiple single-steps. If @code{off}, @value{GDBN} always issues
5557 single-steps, even if range stepping is supported by the target. The
5558 default is @code{on}.
5562 @node Skipping Over Functions and Files
5563 @section Skipping Over Functions and Files
5564 @cindex skipping over functions and files
5566 The program you are debugging may contain some functions which are
5567 uninteresting to debug. The @code{skip} command lets you tell @value{GDBN} to
5568 skip a function, all functions in a file or a particular function in
5569 a particular file when stepping.
5571 For example, consider the following C function:
5582 Suppose you wish to step into the functions @code{foo} and @code{bar}, but you
5583 are not interested in stepping through @code{boring}. If you run @code{step}
5584 at line 103, you'll enter @code{boring()}, but if you run @code{next}, you'll
5585 step over both @code{foo} and @code{boring}!
5587 One solution is to @code{step} into @code{boring} and use the @code{finish}
5588 command to immediately exit it. But this can become tedious if @code{boring}
5589 is called from many places.
5591 A more flexible solution is to execute @kbd{skip boring}. This instructs
5592 @value{GDBN} never to step into @code{boring}. Now when you execute
5593 @code{step} at line 103, you'll step over @code{boring} and directly into
5596 Functions may be skipped by providing either a function name, linespec
5597 (@pxref{Specify Location}), regular expression that matches the function's
5598 name, file name or a @code{glob}-style pattern that matches the file name.
5600 On Posix systems the form of the regular expression is
5601 ``Extended Regular Expressions''. See for example @samp{man 7 regex}
5602 on @sc{gnu}/Linux systems. On non-Posix systems the form of the regular
5603 expression is whatever is provided by the @code{regcomp} function of
5604 the underlying system.
5605 See for example @samp{man 7 glob} on @sc{gnu}/Linux systems for a
5606 description of @code{glob}-style patterns.
5610 @item skip @r{[}@var{options}@r{]}
5611 The basic form of the @code{skip} command takes zero or more options
5612 that specify what to skip.
5613 The @var{options} argument is any useful combination of the following:
5616 @item -file @var{file}
5617 @itemx -fi @var{file}
5618 Functions in @var{file} will be skipped over when stepping.
5620 @item -gfile @var{file-glob-pattern}
5621 @itemx -gfi @var{file-glob-pattern}
5622 @cindex skipping over files via glob-style patterns
5623 Functions in files matching @var{file-glob-pattern} will be skipped
5627 (gdb) skip -gfi utils/*.c
5630 @item -function @var{linespec}
5631 @itemx -fu @var{linespec}
5632 Functions named by @var{linespec} or the function containing the line
5633 named by @var{linespec} will be skipped over when stepping.
5634 @xref{Specify Location}.
5636 @item -rfunction @var{regexp}
5637 @itemx -rfu @var{regexp}
5638 @cindex skipping over functions via regular expressions
5639 Functions whose name matches @var{regexp} will be skipped over when stepping.
5641 This form is useful for complex function names.
5642 For example, there is generally no need to step into C@t{++} @code{std::string}
5643 constructors or destructors. Plus with C@t{++} templates it can be hard to
5644 write out the full name of the function, and often it doesn't matter what
5645 the template arguments are. Specifying the function to be skipped as a
5646 regular expression makes this easier.
5649 (gdb) skip -rfu ^std::(allocator|basic_string)<.*>::~?\1 *\(
5652 If you want to skip every templated C@t{++} constructor and destructor
5653 in the @code{std} namespace you can do:
5656 (gdb) skip -rfu ^std::([a-zA-z0-9_]+)<.*>::~?\1 *\(
5660 If no options are specified, the function you're currently debugging
5663 @kindex skip function
5664 @item skip function @r{[}@var{linespec}@r{]}
5665 After running this command, the function named by @var{linespec} or the
5666 function containing the line named by @var{linespec} will be skipped over when
5667 stepping. @xref{Specify Location}.
5669 If you do not specify @var{linespec}, the function you're currently debugging
5672 (If you have a function called @code{file} that you want to skip, use
5673 @kbd{skip function file}.)
5676 @item skip file @r{[}@var{filename}@r{]}
5677 After running this command, any function whose source lives in @var{filename}
5678 will be skipped over when stepping.
5681 (gdb) skip file boring.c
5682 File boring.c will be skipped when stepping.
5685 If you do not specify @var{filename}, functions whose source lives in the file
5686 you're currently debugging will be skipped.
5689 Skips can be listed, deleted, disabled, and enabled, much like breakpoints.
5690 These are the commands for managing your list of skips:
5694 @item info skip @r{[}@var{range}@r{]}
5695 Print details about the specified skip(s). If @var{range} is not specified,
5696 print a table with details about all functions and files marked for skipping.
5697 @code{info skip} prints the following information about each skip:
5701 A number identifying this skip.
5702 @item Enabled or Disabled
5703 Enabled skips are marked with @samp{y}.
5704 Disabled skips are marked with @samp{n}.
5706 If the file name is a @samp{glob} pattern this is @samp{y}.
5707 Otherwise it is @samp{n}.
5709 The name or @samp{glob} pattern of the file to be skipped.
5710 If no file is specified this is @samp{<none>}.
5712 If the function name is a @samp{regular expression} this is @samp{y}.
5713 Otherwise it is @samp{n}.
5715 The name or regular expression of the function to skip.
5716 If no function is specified this is @samp{<none>}.
5720 @item skip delete @r{[}@var{range}@r{]}
5721 Delete the specified skip(s). If @var{range} is not specified, delete all
5725 @item skip enable @r{[}@var{range}@r{]}
5726 Enable the specified skip(s). If @var{range} is not specified, enable all
5729 @kindex skip disable
5730 @item skip disable @r{[}@var{range}@r{]}
5731 Disable the specified skip(s). If @var{range} is not specified, disable all
5740 A signal is an asynchronous event that can happen in a program. The
5741 operating system defines the possible kinds of signals, and gives each
5742 kind a name and a number. For example, in Unix @code{SIGINT} is the
5743 signal a program gets when you type an interrupt character (often @kbd{Ctrl-c});
5744 @code{SIGSEGV} is the signal a program gets from referencing a place in
5745 memory far away from all the areas in use; @code{SIGALRM} occurs when
5746 the alarm clock timer goes off (which happens only if your program has
5747 requested an alarm).
5749 @cindex fatal signals
5750 Some signals, including @code{SIGALRM}, are a normal part of the
5751 functioning of your program. Others, such as @code{SIGSEGV}, indicate
5752 errors; these signals are @dfn{fatal} (they kill your program immediately) if the
5753 program has not specified in advance some other way to handle the signal.
5754 @code{SIGINT} does not indicate an error in your program, but it is normally
5755 fatal so it can carry out the purpose of the interrupt: to kill the program.
5757 @value{GDBN} has the ability to detect any occurrence of a signal in your
5758 program. You can tell @value{GDBN} in advance what to do for each kind of
5761 @cindex handling signals
5762 Normally, @value{GDBN} is set up to let the non-erroneous signals like
5763 @code{SIGALRM} be silently passed to your program
5764 (so as not to interfere with their role in the program's functioning)
5765 but to stop your program immediately whenever an error signal happens.
5766 You can change these settings with the @code{handle} command.
5769 @kindex info signals
5773 Print a table of all the kinds of signals and how @value{GDBN} has been told to
5774 handle each one. You can use this to see the signal numbers of all
5775 the defined types of signals.
5777 @item info signals @var{sig}
5778 Similar, but print information only about the specified signal number.
5780 @code{info handle} is an alias for @code{info signals}.
5782 @item catch signal @r{[}@var{signal}@dots{} @r{|} @samp{all}@r{]}
5783 Set a catchpoint for the indicated signals. @xref{Set Catchpoints},
5784 for details about this command.
5787 @item handle @var{signal} @r{[}@var{keywords}@dots{}@r{]}
5788 Change the way @value{GDBN} handles signal @var{signal}. The @var{signal}
5789 can be the number of a signal or its name (with or without the
5790 @samp{SIG} at the beginning); a list of signal numbers of the form
5791 @samp{@var{low}-@var{high}}; or the word @samp{all}, meaning all the
5792 known signals. Optional arguments @var{keywords}, described below,
5793 say what change to make.
5797 The keywords allowed by the @code{handle} command can be abbreviated.
5798 Their full names are:
5802 @value{GDBN} should not stop your program when this signal happens. It may
5803 still print a message telling you that the signal has come in.
5806 @value{GDBN} should stop your program when this signal happens. This implies
5807 the @code{print} keyword as well.
5810 @value{GDBN} should print a message when this signal happens.
5813 @value{GDBN} should not mention the occurrence of the signal at all. This
5814 implies the @code{nostop} keyword as well.
5818 @value{GDBN} should allow your program to see this signal; your program
5819 can handle the signal, or else it may terminate if the signal is fatal
5820 and not handled. @code{pass} and @code{noignore} are synonyms.
5824 @value{GDBN} should not allow your program to see this signal.
5825 @code{nopass} and @code{ignore} are synonyms.
5829 When a signal stops your program, the signal is not visible to the
5831 continue. Your program sees the signal then, if @code{pass} is in
5832 effect for the signal in question @emph{at that time}. In other words,
5833 after @value{GDBN} reports a signal, you can use the @code{handle}
5834 command with @code{pass} or @code{nopass} to control whether your
5835 program sees that signal when you continue.
5837 The default is set to @code{nostop}, @code{noprint}, @code{pass} for
5838 non-erroneous signals such as @code{SIGALRM}, @code{SIGWINCH} and
5839 @code{SIGCHLD}, and to @code{stop}, @code{print}, @code{pass} for the
5842 You can also use the @code{signal} command to prevent your program from
5843 seeing a signal, or cause it to see a signal it normally would not see,
5844 or to give it any signal at any time. For example, if your program stopped
5845 due to some sort of memory reference error, you might store correct
5846 values into the erroneous variables and continue, hoping to see more
5847 execution; but your program would probably terminate immediately as
5848 a result of the fatal signal once it saw the signal. To prevent this,
5849 you can continue with @samp{signal 0}. @xref{Signaling, ,Giving your
5852 @cindex stepping and signal handlers
5853 @anchor{stepping and signal handlers}
5855 @value{GDBN} optimizes for stepping the mainline code. If a signal
5856 that has @code{handle nostop} and @code{handle pass} set arrives while
5857 a stepping command (e.g., @code{stepi}, @code{step}, @code{next}) is
5858 in progress, @value{GDBN} lets the signal handler run and then resumes
5859 stepping the mainline code once the signal handler returns. In other
5860 words, @value{GDBN} steps over the signal handler. This prevents
5861 signals that you've specified as not interesting (with @code{handle
5862 nostop}) from changing the focus of debugging unexpectedly. Note that
5863 the signal handler itself may still hit a breakpoint, stop for another
5864 signal that has @code{handle stop} in effect, or for any other event
5865 that normally results in stopping the stepping command sooner. Also
5866 note that @value{GDBN} still informs you that the program received a
5867 signal if @code{handle print} is set.
5869 @anchor{stepping into signal handlers}
5871 If you set @code{handle pass} for a signal, and your program sets up a
5872 handler for it, then issuing a stepping command, such as @code{step}
5873 or @code{stepi}, when your program is stopped due to the signal will
5874 step @emph{into} the signal handler (if the target supports that).
5876 Likewise, if you use the @code{queue-signal} command to queue a signal
5877 to be delivered to the current thread when execution of the thread
5878 resumes (@pxref{Signaling, ,Giving your Program a Signal}), then a
5879 stepping command will step into the signal handler.
5881 Here's an example, using @code{stepi} to step to the first instruction
5882 of @code{SIGUSR1}'s handler:
5885 (@value{GDBP}) handle SIGUSR1
5886 Signal Stop Print Pass to program Description
5887 SIGUSR1 Yes Yes Yes User defined signal 1
5891 Program received signal SIGUSR1, User defined signal 1.
5892 main () sigusr1.c:28
5895 sigusr1_handler () at sigusr1.c:9
5899 The same, but using @code{queue-signal} instead of waiting for the
5900 program to receive the signal first:
5905 (@value{GDBP}) queue-signal SIGUSR1
5907 sigusr1_handler () at sigusr1.c:9
5912 @cindex extra signal information
5913 @anchor{extra signal information}
5915 On some targets, @value{GDBN} can inspect extra signal information
5916 associated with the intercepted signal, before it is actually
5917 delivered to the program being debugged. This information is exported
5918 by the convenience variable @code{$_siginfo}, and consists of data
5919 that is passed by the kernel to the signal handler at the time of the
5920 receipt of a signal. The data type of the information itself is
5921 target dependent. You can see the data type using the @code{ptype
5922 $_siginfo} command. On Unix systems, it typically corresponds to the
5923 standard @code{siginfo_t} type, as defined in the @file{signal.h}
5926 Here's an example, on a @sc{gnu}/Linux system, printing the stray
5927 referenced address that raised a segmentation fault.
5931 (@value{GDBP}) continue
5932 Program received signal SIGSEGV, Segmentation fault.
5933 0x0000000000400766 in main ()
5935 (@value{GDBP}) ptype $_siginfo
5942 struct @{...@} _kill;
5943 struct @{...@} _timer;
5945 struct @{...@} _sigchld;
5946 struct @{...@} _sigfault;
5947 struct @{...@} _sigpoll;
5950 (@value{GDBP}) ptype $_siginfo._sifields._sigfault
5954 (@value{GDBP}) p $_siginfo._sifields._sigfault.si_addr
5955 $1 = (void *) 0x7ffff7ff7000
5959 Depending on target support, @code{$_siginfo} may also be writable.
5961 @cindex Intel MPX boundary violations
5962 @cindex boundary violations, Intel MPX
5963 On some targets, a @code{SIGSEGV} can be caused by a boundary
5964 violation, i.e., accessing an address outside of the allowed range.
5965 In those cases @value{GDBN} may displays additional information,
5966 depending on how @value{GDBN} has been told to handle the signal.
5967 With @code{handle stop SIGSEGV}, @value{GDBN} displays the violation
5968 kind: "Upper" or "Lower", the memory address accessed and the
5969 bounds, while with @code{handle nostop SIGSEGV} no additional
5970 information is displayed.
5972 The usual output of a segfault is:
5974 Program received signal SIGSEGV, Segmentation fault
5975 0x0000000000400d7c in upper () at i386-mpx-sigsegv.c:68
5976 68 value = *(p + len);
5979 While a bound violation is presented as:
5981 Program received signal SIGSEGV, Segmentation fault
5982 Upper bound violation while accessing address 0x7fffffffc3b3
5983 Bounds: [lower = 0x7fffffffc390, upper = 0x7fffffffc3a3]
5984 0x0000000000400d7c in upper () at i386-mpx-sigsegv.c:68
5985 68 value = *(p + len);
5989 @section Stopping and Starting Multi-thread Programs
5991 @cindex stopped threads
5992 @cindex threads, stopped
5994 @cindex continuing threads
5995 @cindex threads, continuing
5997 @value{GDBN} supports debugging programs with multiple threads
5998 (@pxref{Threads,, Debugging Programs with Multiple Threads}). There
5999 are two modes of controlling execution of your program within the
6000 debugger. In the default mode, referred to as @dfn{all-stop mode},
6001 when any thread in your program stops (for example, at a breakpoint
6002 or while being stepped), all other threads in the program are also stopped by
6003 @value{GDBN}. On some targets, @value{GDBN} also supports
6004 @dfn{non-stop mode}, in which other threads can continue to run freely while
6005 you examine the stopped thread in the debugger.
6008 * All-Stop Mode:: All threads stop when GDB takes control
6009 * Non-Stop Mode:: Other threads continue to execute
6010 * Background Execution:: Running your program asynchronously
6011 * Thread-Specific Breakpoints:: Controlling breakpoints
6012 * Interrupted System Calls:: GDB may interfere with system calls
6013 * Observer Mode:: GDB does not alter program behavior
6017 @subsection All-Stop Mode
6019 @cindex all-stop mode
6021 In all-stop mode, whenever your program stops under @value{GDBN} for any reason,
6022 @emph{all} threads of execution stop, not just the current thread. This
6023 allows you to examine the overall state of the program, including
6024 switching between threads, without worrying that things may change
6027 Conversely, whenever you restart the program, @emph{all} threads start
6028 executing. @emph{This is true even when single-stepping} with commands
6029 like @code{step} or @code{next}.
6031 In particular, @value{GDBN} cannot single-step all threads in lockstep.
6032 Since thread scheduling is up to your debugging target's operating
6033 system (not controlled by @value{GDBN}), other threads may
6034 execute more than one statement while the current thread completes a
6035 single step. Moreover, in general other threads stop in the middle of a
6036 statement, rather than at a clean statement boundary, when the program
6039 You might even find your program stopped in another thread after
6040 continuing or even single-stepping. This happens whenever some other
6041 thread runs into a breakpoint, a signal, or an exception before the
6042 first thread completes whatever you requested.
6044 @cindex automatic thread selection
6045 @cindex switching threads automatically
6046 @cindex threads, automatic switching
6047 Whenever @value{GDBN} stops your program, due to a breakpoint or a
6048 signal, it automatically selects the thread where that breakpoint or
6049 signal happened. @value{GDBN} alerts you to the context switch with a
6050 message such as @samp{[Switching to Thread @var{n}]} to identify the
6053 On some OSes, you can modify @value{GDBN}'s default behavior by
6054 locking the OS scheduler to allow only a single thread to run.
6057 @item set scheduler-locking @var{mode}
6058 @cindex scheduler locking mode
6059 @cindex lock scheduler
6060 Set the scheduler locking mode. It applies to normal execution,
6061 record mode, and replay mode. If it is @code{off}, then there is no
6062 locking and any thread may run at any time. If @code{on}, then only
6063 the current thread may run when the inferior is resumed. The
6064 @code{step} mode optimizes for single-stepping; it prevents other
6065 threads from preempting the current thread while you are stepping, so
6066 that the focus of debugging does not change unexpectedly. Other
6067 threads never get a chance to run when you step, and they are
6068 completely free to run when you use commands like @samp{continue},
6069 @samp{until}, or @samp{finish}. However, unless another thread hits a
6070 breakpoint during its timeslice, @value{GDBN} does not change the
6071 current thread away from the thread that you are debugging. The
6072 @code{replay} mode behaves like @code{off} in record mode and like
6073 @code{on} in replay mode.
6075 @item show scheduler-locking
6076 Display the current scheduler locking mode.
6079 @cindex resume threads of multiple processes simultaneously
6080 By default, when you issue one of the execution commands such as
6081 @code{continue}, @code{next} or @code{step}, @value{GDBN} allows only
6082 threads of the current inferior to run. For example, if @value{GDBN}
6083 is attached to two inferiors, each with two threads, the
6084 @code{continue} command resumes only the two threads of the current
6085 inferior. This is useful, for example, when you debug a program that
6086 forks and you want to hold the parent stopped (so that, for instance,
6087 it doesn't run to exit), while you debug the child. In other
6088 situations, you may not be interested in inspecting the current state
6089 of any of the processes @value{GDBN} is attached to, and you may want
6090 to resume them all until some breakpoint is hit. In the latter case,
6091 you can instruct @value{GDBN} to allow all threads of all the
6092 inferiors to run with the @w{@code{set schedule-multiple}} command.
6095 @kindex set schedule-multiple
6096 @item set schedule-multiple
6097 Set the mode for allowing threads of multiple processes to be resumed
6098 when an execution command is issued. When @code{on}, all threads of
6099 all processes are allowed to run. When @code{off}, only the threads
6100 of the current process are resumed. The default is @code{off}. The
6101 @code{scheduler-locking} mode takes precedence when set to @code{on},
6102 or while you are stepping and set to @code{step}.
6104 @item show schedule-multiple
6105 Display the current mode for resuming the execution of threads of
6110 @subsection Non-Stop Mode
6112 @cindex non-stop mode
6114 @c This section is really only a place-holder, and needs to be expanded
6115 @c with more details.
6117 For some multi-threaded targets, @value{GDBN} supports an optional
6118 mode of operation in which you can examine stopped program threads in
6119 the debugger while other threads continue to execute freely. This
6120 minimizes intrusion when debugging live systems, such as programs
6121 where some threads have real-time constraints or must continue to
6122 respond to external events. This is referred to as @dfn{non-stop} mode.
6124 In non-stop mode, when a thread stops to report a debugging event,
6125 @emph{only} that thread is stopped; @value{GDBN} does not stop other
6126 threads as well, in contrast to the all-stop mode behavior. Additionally,
6127 execution commands such as @code{continue} and @code{step} apply by default
6128 only to the current thread in non-stop mode, rather than all threads as
6129 in all-stop mode. This allows you to control threads explicitly in
6130 ways that are not possible in all-stop mode --- for example, stepping
6131 one thread while allowing others to run freely, stepping
6132 one thread while holding all others stopped, or stepping several threads
6133 independently and simultaneously.
6135 To enter non-stop mode, use this sequence of commands before you run
6136 or attach to your program:
6139 # If using the CLI, pagination breaks non-stop.
6142 # Finally, turn it on!
6146 You can use these commands to manipulate the non-stop mode setting:
6149 @kindex set non-stop
6150 @item set non-stop on
6151 Enable selection of non-stop mode.
6152 @item set non-stop off
6153 Disable selection of non-stop mode.
6154 @kindex show non-stop
6156 Show the current non-stop enablement setting.
6159 Note these commands only reflect whether non-stop mode is enabled,
6160 not whether the currently-executing program is being run in non-stop mode.
6161 In particular, the @code{set non-stop} preference is only consulted when
6162 @value{GDBN} starts or connects to the target program, and it is generally
6163 not possible to switch modes once debugging has started. Furthermore,
6164 since not all targets support non-stop mode, even when you have enabled
6165 non-stop mode, @value{GDBN} may still fall back to all-stop operation by
6168 In non-stop mode, all execution commands apply only to the current thread
6169 by default. That is, @code{continue} only continues one thread.
6170 To continue all threads, issue @code{continue -a} or @code{c -a}.
6172 You can use @value{GDBN}'s background execution commands
6173 (@pxref{Background Execution}) to run some threads in the background
6174 while you continue to examine or step others from @value{GDBN}.
6175 The MI execution commands (@pxref{GDB/MI Program Execution}) are
6176 always executed asynchronously in non-stop mode.
6178 Suspending execution is done with the @code{interrupt} command when
6179 running in the background, or @kbd{Ctrl-c} during foreground execution.
6180 In all-stop mode, this stops the whole process;
6181 but in non-stop mode the interrupt applies only to the current thread.
6182 To stop the whole program, use @code{interrupt -a}.
6184 Other execution commands do not currently support the @code{-a} option.
6186 In non-stop mode, when a thread stops, @value{GDBN} doesn't automatically make
6187 that thread current, as it does in all-stop mode. This is because the
6188 thread stop notifications are asynchronous with respect to @value{GDBN}'s
6189 command interpreter, and it would be confusing if @value{GDBN} unexpectedly
6190 changed to a different thread just as you entered a command to operate on the
6191 previously current thread.
6193 @node Background Execution
6194 @subsection Background Execution
6196 @cindex foreground execution
6197 @cindex background execution
6198 @cindex asynchronous execution
6199 @cindex execution, foreground, background and asynchronous
6201 @value{GDBN}'s execution commands have two variants: the normal
6202 foreground (synchronous) behavior, and a background
6203 (asynchronous) behavior. In foreground execution, @value{GDBN} waits for
6204 the program to report that some thread has stopped before prompting for
6205 another command. In background execution, @value{GDBN} immediately gives
6206 a command prompt so that you can issue other commands while your program runs.
6208 If the target doesn't support async mode, @value{GDBN} issues an error
6209 message if you attempt to use the background execution commands.
6211 To specify background execution, add a @code{&} to the command. For example,
6212 the background form of the @code{continue} command is @code{continue&}, or
6213 just @code{c&}. The execution commands that accept background execution
6219 @xref{Starting, , Starting your Program}.
6223 @xref{Attach, , Debugging an Already-running Process}.
6227 @xref{Continuing and Stepping, step}.
6231 @xref{Continuing and Stepping, stepi}.
6235 @xref{Continuing and Stepping, next}.
6239 @xref{Continuing and Stepping, nexti}.
6243 @xref{Continuing and Stepping, continue}.
6247 @xref{Continuing and Stepping, finish}.
6251 @xref{Continuing and Stepping, until}.
6255 Background execution is especially useful in conjunction with non-stop
6256 mode for debugging programs with multiple threads; see @ref{Non-Stop Mode}.
6257 However, you can also use these commands in the normal all-stop mode with
6258 the restriction that you cannot issue another execution command until the
6259 previous one finishes. Examples of commands that are valid in all-stop
6260 mode while the program is running include @code{help} and @code{info break}.
6262 You can interrupt your program while it is running in the background by
6263 using the @code{interrupt} command.
6270 Suspend execution of the running program. In all-stop mode,
6271 @code{interrupt} stops the whole process, but in non-stop mode, it stops
6272 only the current thread. To stop the whole program in non-stop mode,
6273 use @code{interrupt -a}.
6276 @node Thread-Specific Breakpoints
6277 @subsection Thread-Specific Breakpoints
6279 When your program has multiple threads (@pxref{Threads,, Debugging
6280 Programs with Multiple Threads}), you can choose whether to set
6281 breakpoints on all threads, or on a particular thread.
6284 @cindex breakpoints and threads
6285 @cindex thread breakpoints
6286 @kindex break @dots{} thread @var{thread-id}
6287 @item break @var{location} thread @var{thread-id}
6288 @itemx break @var{location} thread @var{thread-id} if @dots{}
6289 @var{location} specifies source lines; there are several ways of
6290 writing them (@pxref{Specify Location}), but the effect is always to
6291 specify some source line.
6293 Use the qualifier @samp{thread @var{thread-id}} with a breakpoint command
6294 to specify that you only want @value{GDBN} to stop the program when a
6295 particular thread reaches this breakpoint. The @var{thread-id} specifier
6296 is one of the thread identifiers assigned by @value{GDBN}, shown
6297 in the first column of the @samp{info threads} display.
6299 If you do not specify @samp{thread @var{thread-id}} when you set a
6300 breakpoint, the breakpoint applies to @emph{all} threads of your
6303 You can use the @code{thread} qualifier on conditional breakpoints as
6304 well; in this case, place @samp{thread @var{thread-id}} before or
6305 after the breakpoint condition, like this:
6308 (@value{GDBP}) break frik.c:13 thread 28 if bartab > lim
6313 Thread-specific breakpoints are automatically deleted when
6314 @value{GDBN} detects the corresponding thread is no longer in the
6315 thread list. For example:
6319 Thread-specific breakpoint 3 deleted - thread 28 no longer in the thread list.
6322 There are several ways for a thread to disappear, such as a regular
6323 thread exit, but also when you detach from the process with the
6324 @code{detach} command (@pxref{Attach, ,Debugging an Already-running
6325 Process}), or if @value{GDBN} loses the remote connection
6326 (@pxref{Remote Debugging}), etc. Note that with some targets,
6327 @value{GDBN} is only able to detect a thread has exited when the user
6328 explictly asks for the thread list with the @code{info threads}
6331 @node Interrupted System Calls
6332 @subsection Interrupted System Calls
6334 @cindex thread breakpoints and system calls
6335 @cindex system calls and thread breakpoints
6336 @cindex premature return from system calls
6337 There is an unfortunate side effect when using @value{GDBN} to debug
6338 multi-threaded programs. If one thread stops for a
6339 breakpoint, or for some other reason, and another thread is blocked in a
6340 system call, then the system call may return prematurely. This is a
6341 consequence of the interaction between multiple threads and the signals
6342 that @value{GDBN} uses to implement breakpoints and other events that
6345 To handle this problem, your program should check the return value of
6346 each system call and react appropriately. This is good programming
6349 For example, do not write code like this:
6355 The call to @code{sleep} will return early if a different thread stops
6356 at a breakpoint or for some other reason.
6358 Instead, write this:
6363 unslept = sleep (unslept);
6366 A system call is allowed to return early, so the system is still
6367 conforming to its specification. But @value{GDBN} does cause your
6368 multi-threaded program to behave differently than it would without
6371 Also, @value{GDBN} uses internal breakpoints in the thread library to
6372 monitor certain events such as thread creation and thread destruction.
6373 When such an event happens, a system call in another thread may return
6374 prematurely, even though your program does not appear to stop.
6377 @subsection Observer Mode
6379 If you want to build on non-stop mode and observe program behavior
6380 without any chance of disruption by @value{GDBN}, you can set
6381 variables to disable all of the debugger's attempts to modify state,
6382 whether by writing memory, inserting breakpoints, etc. These operate
6383 at a low level, intercepting operations from all commands.
6385 When all of these are set to @code{off}, then @value{GDBN} is said to
6386 be @dfn{observer mode}. As a convenience, the variable
6387 @code{observer} can be set to disable these, plus enable non-stop
6390 Note that @value{GDBN} will not prevent you from making nonsensical
6391 combinations of these settings. For instance, if you have enabled
6392 @code{may-insert-breakpoints} but disabled @code{may-write-memory},
6393 then breakpoints that work by writing trap instructions into the code
6394 stream will still not be able to be placed.
6399 @item set observer on
6400 @itemx set observer off
6401 When set to @code{on}, this disables all the permission variables
6402 below (except for @code{insert-fast-tracepoints}), plus enables
6403 non-stop debugging. Setting this to @code{off} switches back to
6404 normal debugging, though remaining in non-stop mode.
6407 Show whether observer mode is on or off.
6409 @kindex may-write-registers
6410 @item set may-write-registers on
6411 @itemx set may-write-registers off
6412 This controls whether @value{GDBN} will attempt to alter the values of
6413 registers, such as with assignment expressions in @code{print}, or the
6414 @code{jump} command. It defaults to @code{on}.
6416 @item show may-write-registers
6417 Show the current permission to write registers.
6419 @kindex may-write-memory
6420 @item set may-write-memory on
6421 @itemx set may-write-memory off
6422 This controls whether @value{GDBN} will attempt to alter the contents
6423 of memory, such as with assignment expressions in @code{print}. It
6424 defaults to @code{on}.
6426 @item show may-write-memory
6427 Show the current permission to write memory.
6429 @kindex may-insert-breakpoints
6430 @item set may-insert-breakpoints on
6431 @itemx set may-insert-breakpoints off
6432 This controls whether @value{GDBN} will attempt to insert breakpoints.
6433 This affects all breakpoints, including internal breakpoints defined
6434 by @value{GDBN}. It defaults to @code{on}.
6436 @item show may-insert-breakpoints
6437 Show the current permission to insert breakpoints.
6439 @kindex may-insert-tracepoints
6440 @item set may-insert-tracepoints on
6441 @itemx set may-insert-tracepoints off
6442 This controls whether @value{GDBN} will attempt to insert (regular)
6443 tracepoints at the beginning of a tracing experiment. It affects only
6444 non-fast tracepoints, fast tracepoints being under the control of
6445 @code{may-insert-fast-tracepoints}. It defaults to @code{on}.
6447 @item show may-insert-tracepoints
6448 Show the current permission to insert tracepoints.
6450 @kindex may-insert-fast-tracepoints
6451 @item set may-insert-fast-tracepoints on
6452 @itemx set may-insert-fast-tracepoints off
6453 This controls whether @value{GDBN} will attempt to insert fast
6454 tracepoints at the beginning of a tracing experiment. It affects only
6455 fast tracepoints, regular (non-fast) tracepoints being under the
6456 control of @code{may-insert-tracepoints}. It defaults to @code{on}.
6458 @item show may-insert-fast-tracepoints
6459 Show the current permission to insert fast tracepoints.
6461 @kindex may-interrupt
6462 @item set may-interrupt on
6463 @itemx set may-interrupt off
6464 This controls whether @value{GDBN} will attempt to interrupt or stop
6465 program execution. When this variable is @code{off}, the
6466 @code{interrupt} command will have no effect, nor will
6467 @kbd{Ctrl-c}. It defaults to @code{on}.
6469 @item show may-interrupt
6470 Show the current permission to interrupt or stop the program.
6474 @node Reverse Execution
6475 @chapter Running programs backward
6476 @cindex reverse execution
6477 @cindex running programs backward
6479 When you are debugging a program, it is not unusual to realize that
6480 you have gone too far, and some event of interest has already happened.
6481 If the target environment supports it, @value{GDBN} can allow you to
6482 ``rewind'' the program by running it backward.
6484 A target environment that supports reverse execution should be able
6485 to ``undo'' the changes in machine state that have taken place as the
6486 program was executing normally. Variables, registers etc.@: should
6487 revert to their previous values. Obviously this requires a great
6488 deal of sophistication on the part of the target environment; not
6489 all target environments can support reverse execution.
6491 When a program is executed in reverse, the instructions that
6492 have most recently been executed are ``un-executed'', in reverse
6493 order. The program counter runs backward, following the previous
6494 thread of execution in reverse. As each instruction is ``un-executed'',
6495 the values of memory and/or registers that were changed by that
6496 instruction are reverted to their previous states. After executing
6497 a piece of source code in reverse, all side effects of that code
6498 should be ``undone'', and all variables should be returned to their
6499 prior values@footnote{
6500 Note that some side effects are easier to undo than others. For instance,
6501 memory and registers are relatively easy, but device I/O is hard. Some
6502 targets may be able undo things like device I/O, and some may not.
6504 The contract between @value{GDBN} and the reverse executing target
6505 requires only that the target do something reasonable when
6506 @value{GDBN} tells it to execute backwards, and then report the
6507 results back to @value{GDBN}. Whatever the target reports back to
6508 @value{GDBN}, @value{GDBN} will report back to the user. @value{GDBN}
6509 assumes that the memory and registers that the target reports are in a
6510 consistant state, but @value{GDBN} accepts whatever it is given.
6513 If you are debugging in a target environment that supports
6514 reverse execution, @value{GDBN} provides the following commands.
6517 @kindex reverse-continue
6518 @kindex rc @r{(@code{reverse-continue})}
6519 @item reverse-continue @r{[}@var{ignore-count}@r{]}
6520 @itemx rc @r{[}@var{ignore-count}@r{]}
6521 Beginning at the point where your program last stopped, start executing
6522 in reverse. Reverse execution will stop for breakpoints and synchronous
6523 exceptions (signals), just like normal execution. Behavior of
6524 asynchronous signals depends on the target environment.
6526 @kindex reverse-step
6527 @kindex rs @r{(@code{step})}
6528 @item reverse-step @r{[}@var{count}@r{]}
6529 Run the program backward until control reaches the start of a
6530 different source line; then stop it, and return control to @value{GDBN}.
6532 Like the @code{step} command, @code{reverse-step} will only stop
6533 at the beginning of a source line. It ``un-executes'' the previously
6534 executed source line. If the previous source line included calls to
6535 debuggable functions, @code{reverse-step} will step (backward) into
6536 the called function, stopping at the beginning of the @emph{last}
6537 statement in the called function (typically a return statement).
6539 Also, as with the @code{step} command, if non-debuggable functions are
6540 called, @code{reverse-step} will run thru them backward without stopping.
6542 @kindex reverse-stepi
6543 @kindex rsi @r{(@code{reverse-stepi})}
6544 @item reverse-stepi @r{[}@var{count}@r{]}
6545 Reverse-execute one machine instruction. Note that the instruction
6546 to be reverse-executed is @emph{not} the one pointed to by the program
6547 counter, but the instruction executed prior to that one. For instance,
6548 if the last instruction was a jump, @code{reverse-stepi} will take you
6549 back from the destination of the jump to the jump instruction itself.
6551 @kindex reverse-next
6552 @kindex rn @r{(@code{reverse-next})}
6553 @item reverse-next @r{[}@var{count}@r{]}
6554 Run backward to the beginning of the previous line executed in
6555 the current (innermost) stack frame. If the line contains function
6556 calls, they will be ``un-executed'' without stopping. Starting from
6557 the first line of a function, @code{reverse-next} will take you back
6558 to the caller of that function, @emph{before} the function was called,
6559 just as the normal @code{next} command would take you from the last
6560 line of a function back to its return to its caller
6561 @footnote{Unless the code is too heavily optimized.}.
6563 @kindex reverse-nexti
6564 @kindex rni @r{(@code{reverse-nexti})}
6565 @item reverse-nexti @r{[}@var{count}@r{]}
6566 Like @code{nexti}, @code{reverse-nexti} executes a single instruction
6567 in reverse, except that called functions are ``un-executed'' atomically.
6568 That is, if the previously executed instruction was a return from
6569 another function, @code{reverse-nexti} will continue to execute
6570 in reverse until the call to that function (from the current stack
6573 @kindex reverse-finish
6574 @item reverse-finish
6575 Just as the @code{finish} command takes you to the point where the
6576 current function returns, @code{reverse-finish} takes you to the point
6577 where it was called. Instead of ending up at the end of the current
6578 function invocation, you end up at the beginning.
6580 @kindex set exec-direction
6581 @item set exec-direction
6582 Set the direction of target execution.
6583 @item set exec-direction reverse
6584 @cindex execute forward or backward in time
6585 @value{GDBN} will perform all execution commands in reverse, until the
6586 exec-direction mode is changed to ``forward''. Affected commands include
6587 @code{step, stepi, next, nexti, continue, and finish}. The @code{return}
6588 command cannot be used in reverse mode.
6589 @item set exec-direction forward
6590 @value{GDBN} will perform all execution commands in the normal fashion.
6591 This is the default.
6595 @node Process Record and Replay
6596 @chapter Recording Inferior's Execution and Replaying It
6597 @cindex process record and replay
6598 @cindex recording inferior's execution and replaying it
6600 On some platforms, @value{GDBN} provides a special @dfn{process record
6601 and replay} target that can record a log of the process execution, and
6602 replay it later with both forward and reverse execution commands.
6605 When this target is in use, if the execution log includes the record
6606 for the next instruction, @value{GDBN} will debug in @dfn{replay
6607 mode}. In the replay mode, the inferior does not really execute code
6608 instructions. Instead, all the events that normally happen during
6609 code execution are taken from the execution log. While code is not
6610 really executed in replay mode, the values of registers (including the
6611 program counter register) and the memory of the inferior are still
6612 changed as they normally would. Their contents are taken from the
6616 If the record for the next instruction is not in the execution log,
6617 @value{GDBN} will debug in @dfn{record mode}. In this mode, the
6618 inferior executes normally, and @value{GDBN} records the execution log
6621 The process record and replay target supports reverse execution
6622 (@pxref{Reverse Execution}), even if the platform on which the
6623 inferior runs does not. However, the reverse execution is limited in
6624 this case by the range of the instructions recorded in the execution
6625 log. In other words, reverse execution on platforms that don't
6626 support it directly can only be done in the replay mode.
6628 When debugging in the reverse direction, @value{GDBN} will work in
6629 replay mode as long as the execution log includes the record for the
6630 previous instruction; otherwise, it will work in record mode, if the
6631 platform supports reverse execution, or stop if not.
6633 For architecture environments that support process record and replay,
6634 @value{GDBN} provides the following commands:
6637 @kindex target record
6638 @kindex target record-full
6639 @kindex target record-btrace
6642 @kindex record btrace
6643 @kindex record btrace bts
6644 @kindex record btrace pt
6650 @kindex rec btrace bts
6651 @kindex rec btrace pt
6654 @item record @var{method}
6655 This command starts the process record and replay target. The
6656 recording method can be specified as parameter. Without a parameter
6657 the command uses the @code{full} recording method. The following
6658 recording methods are available:
6662 Full record/replay recording using @value{GDBN}'s software record and
6663 replay implementation. This method allows replaying and reverse
6666 @item btrace @var{format}
6667 Hardware-supported instruction recording. This method does not record
6668 data. Further, the data is collected in a ring buffer so old data will
6669 be overwritten when the buffer is full. It allows limited reverse
6670 execution. Variables and registers are not available during reverse
6671 execution. In remote debugging, recording continues on disconnect.
6672 Recorded data can be inspected after reconnecting. The recording may
6673 be stopped using @code{record stop}.
6675 The recording format can be specified as parameter. Without a parameter
6676 the command chooses the recording format. The following recording
6677 formats are available:
6681 @cindex branch trace store
6682 Use the @dfn{Branch Trace Store} (@acronym{BTS}) recording format. In
6683 this format, the processor stores a from/to record for each executed
6684 branch in the btrace ring buffer.
6687 @cindex Intel Processor Trace
6688 Use the @dfn{Intel Processor Trace} recording format. In this
6689 format, the processor stores the execution trace in a compressed form
6690 that is afterwards decoded by @value{GDBN}.
6692 The trace can be recorded with very low overhead. The compressed
6693 trace format also allows small trace buffers to already contain a big
6694 number of instructions compared to @acronym{BTS}.
6696 Decoding the recorded execution trace, on the other hand, is more
6697 expensive than decoding @acronym{BTS} trace. This is mostly due to the
6698 increased number of instructions to process. You should increase the
6699 buffer-size with care.
6702 Not all recording formats may be available on all processors.
6705 The process record and replay target can only debug a process that is
6706 already running. Therefore, you need first to start the process with
6707 the @kbd{run} or @kbd{start} commands, and then start the recording
6708 with the @kbd{record @var{method}} command.
6710 @cindex displaced stepping, and process record and replay
6711 Displaced stepping (@pxref{Maintenance Commands,, displaced stepping})
6712 will be automatically disabled when process record and replay target
6713 is started. That's because the process record and replay target
6714 doesn't support displaced stepping.
6716 @cindex non-stop mode, and process record and replay
6717 @cindex asynchronous execution, and process record and replay
6718 If the inferior is in the non-stop mode (@pxref{Non-Stop Mode}) or in
6719 the asynchronous execution mode (@pxref{Background Execution}), not
6720 all recording methods are available. The @code{full} recording method
6721 does not support these two modes.
6726 Stop the process record and replay target. When process record and
6727 replay target stops, the entire execution log will be deleted and the
6728 inferior will either be terminated, or will remain in its final state.
6730 When you stop the process record and replay target in record mode (at
6731 the end of the execution log), the inferior will be stopped at the
6732 next instruction that would have been recorded. In other words, if
6733 you record for a while and then stop recording, the inferior process
6734 will be left in the same state as if the recording never happened.
6736 On the other hand, if the process record and replay target is stopped
6737 while in replay mode (that is, not at the end of the execution log,
6738 but at some earlier point), the inferior process will become ``live''
6739 at that earlier state, and it will then be possible to continue the
6740 usual ``live'' debugging of the process from that state.
6742 When the inferior process exits, or @value{GDBN} detaches from it,
6743 process record and replay target will automatically stop itself.
6747 Go to a specific location in the execution log. There are several
6748 ways to specify the location to go to:
6751 @item record goto begin
6752 @itemx record goto start
6753 Go to the beginning of the execution log.
6755 @item record goto end
6756 Go to the end of the execution log.
6758 @item record goto @var{n}
6759 Go to instruction number @var{n} in the execution log.
6763 @item record save @var{filename}
6764 Save the execution log to a file @file{@var{filename}}.
6765 Default filename is @file{gdb_record.@var{process_id}}, where
6766 @var{process_id} is the process ID of the inferior.
6768 This command may not be available for all recording methods.
6770 @kindex record restore
6771 @item record restore @var{filename}
6772 Restore the execution log from a file @file{@var{filename}}.
6773 File must have been created with @code{record save}.
6775 @kindex set record full
6776 @item set record full insn-number-max @var{limit}
6777 @itemx set record full insn-number-max unlimited
6778 Set the limit of instructions to be recorded for the @code{full}
6779 recording method. Default value is 200000.
6781 If @var{limit} is a positive number, then @value{GDBN} will start
6782 deleting instructions from the log once the number of the record
6783 instructions becomes greater than @var{limit}. For every new recorded
6784 instruction, @value{GDBN} will delete the earliest recorded
6785 instruction to keep the number of recorded instructions at the limit.
6786 (Since deleting recorded instructions loses information, @value{GDBN}
6787 lets you control what happens when the limit is reached, by means of
6788 the @code{stop-at-limit} option, described below.)
6790 If @var{limit} is @code{unlimited} or zero, @value{GDBN} will never
6791 delete recorded instructions from the execution log. The number of
6792 recorded instructions is limited only by the available memory.
6794 @kindex show record full
6795 @item show record full insn-number-max
6796 Show the limit of instructions to be recorded with the @code{full}
6799 @item set record full stop-at-limit
6800 Control the behavior of the @code{full} recording method when the
6801 number of recorded instructions reaches the limit. If ON (the
6802 default), @value{GDBN} will stop when the limit is reached for the
6803 first time and ask you whether you want to stop the inferior or
6804 continue running it and recording the execution log. If you decide
6805 to continue recording, each new recorded instruction will cause the
6806 oldest one to be deleted.
6808 If this option is OFF, @value{GDBN} will automatically delete the
6809 oldest record to make room for each new one, without asking.
6811 @item show record full stop-at-limit
6812 Show the current setting of @code{stop-at-limit}.
6814 @item set record full memory-query
6815 Control the behavior when @value{GDBN} is unable to record memory
6816 changes caused by an instruction for the @code{full} recording method.
6817 If ON, @value{GDBN} will query whether to stop the inferior in that
6820 If this option is OFF (the default), @value{GDBN} will automatically
6821 ignore the effect of such instructions on memory. Later, when
6822 @value{GDBN} replays this execution log, it will mark the log of this
6823 instruction as not accessible, and it will not affect the replay
6826 @item show record full memory-query
6827 Show the current setting of @code{memory-query}.
6829 @kindex set record btrace
6830 The @code{btrace} record target does not trace data. As a
6831 convenience, when replaying, @value{GDBN} reads read-only memory off
6832 the live program directly, assuming that the addresses of the
6833 read-only areas don't change. This for example makes it possible to
6834 disassemble code while replaying, but not to print variables.
6835 In some cases, being able to inspect variables might be useful.
6836 You can use the following command for that:
6838 @item set record btrace replay-memory-access
6839 Control the behavior of the @code{btrace} recording method when
6840 accessing memory during replay. If @code{read-only} (the default),
6841 @value{GDBN} will only allow accesses to read-only memory.
6842 If @code{read-write}, @value{GDBN} will allow accesses to read-only
6843 and to read-write memory. Beware that the accessed memory corresponds
6844 to the live target and not necessarily to the current replay
6847 @kindex show record btrace
6848 @item show record btrace replay-memory-access
6849 Show the current setting of @code{replay-memory-access}.
6851 @kindex set record btrace bts
6852 @item set record btrace bts buffer-size @var{size}
6853 @itemx set record btrace bts buffer-size unlimited
6854 Set the requested ring buffer size for branch tracing in @acronym{BTS}
6855 format. Default is 64KB.
6857 If @var{size} is a positive number, then @value{GDBN} will try to
6858 allocate a buffer of at least @var{size} bytes for each new thread
6859 that uses the btrace recording method and the @acronym{BTS} format.
6860 The actually obtained buffer size may differ from the requested
6861 @var{size}. Use the @code{info record} command to see the actual
6862 buffer size for each thread that uses the btrace recording method and
6863 the @acronym{BTS} format.
6865 If @var{limit} is @code{unlimited} or zero, @value{GDBN} will try to
6866 allocate a buffer of 4MB.
6868 Bigger buffers mean longer traces. On the other hand, @value{GDBN} will
6869 also need longer to process the branch trace data before it can be used.
6871 @item show record btrace bts buffer-size @var{size}
6872 Show the current setting of the requested ring buffer size for branch
6873 tracing in @acronym{BTS} format.
6875 @kindex set record btrace pt
6876 @item set record btrace pt buffer-size @var{size}
6877 @itemx set record btrace pt buffer-size unlimited
6878 Set the requested ring buffer size for branch tracing in Intel
6879 Processor Trace format. Default is 16KB.
6881 If @var{size} is a positive number, then @value{GDBN} will try to
6882 allocate a buffer of at least @var{size} bytes for each new thread
6883 that uses the btrace recording method and the Intel Processor Trace
6884 format. The actually obtained buffer size may differ from the
6885 requested @var{size}. Use the @code{info record} command to see the
6886 actual buffer size for each thread.
6888 If @var{limit} is @code{unlimited} or zero, @value{GDBN} will try to
6889 allocate a buffer of 4MB.
6891 Bigger buffers mean longer traces. On the other hand, @value{GDBN} will
6892 also need longer to process the branch trace data before it can be used.
6894 @item show record btrace pt buffer-size @var{size}
6895 Show the current setting of the requested ring buffer size for branch
6896 tracing in Intel Processor Trace format.
6900 Show various statistics about the recording depending on the recording
6905 For the @code{full} recording method, it shows the state of process
6906 record and its in-memory execution log buffer, including:
6910 Whether in record mode or replay mode.
6912 Lowest recorded instruction number (counting from when the current execution log started recording instructions).
6914 Highest recorded instruction number.
6916 Current instruction about to be replayed (if in replay mode).
6918 Number of instructions contained in the execution log.
6920 Maximum number of instructions that may be contained in the execution log.
6924 For the @code{btrace} recording method, it shows:
6930 Number of instructions that have been recorded.
6932 Number of blocks of sequential control-flow formed by the recorded
6935 Whether in record mode or replay mode.
6938 For the @code{bts} recording format, it also shows:
6941 Size of the perf ring buffer.
6944 For the @code{pt} recording format, it also shows:
6947 Size of the perf ring buffer.
6951 @kindex record delete
6954 When record target runs in replay mode (``in the past''), delete the
6955 subsequent execution log and begin to record a new execution log starting
6956 from the current address. This means you will abandon the previously
6957 recorded ``future'' and begin recording a new ``future''.
6959 @kindex record instruction-history
6960 @kindex rec instruction-history
6961 @item record instruction-history
6962 Disassembles instructions from the recorded execution log. By
6963 default, ten instructions are disassembled. This can be changed using
6964 the @code{set record instruction-history-size} command. Instructions
6965 are printed in execution order.
6967 It can also print mixed source+disassembly if you specify the the
6968 @code{/m} or @code{/s} modifier, and print the raw instructions in hex
6969 as well as in symbolic form by specifying the @code{/r} modifier.
6971 The current position marker is printed for the instruction at the
6972 current program counter value. This instruction can appear multiple
6973 times in the trace and the current position marker will be printed
6974 every time. To omit the current position marker, specify the
6977 To better align the printed instructions when the trace contains
6978 instructions from more than one function, the function name may be
6979 omitted by specifying the @code{/f} modifier.
6981 Speculatively executed instructions are prefixed with @samp{?}. This
6982 feature is not available for all recording formats.
6984 There are several ways to specify what part of the execution log to
6988 @item record instruction-history @var{insn}
6989 Disassembles ten instructions starting from instruction number
6992 @item record instruction-history @var{insn}, +/-@var{n}
6993 Disassembles @var{n} instructions around instruction number
6994 @var{insn}. If @var{n} is preceded with @code{+}, disassembles
6995 @var{n} instructions after instruction number @var{insn}. If
6996 @var{n} is preceded with @code{-}, disassembles @var{n}
6997 instructions before instruction number @var{insn}.
6999 @item record instruction-history
7000 Disassembles ten more instructions after the last disassembly.
7002 @item record instruction-history -
7003 Disassembles ten more instructions before the last disassembly.
7005 @item record instruction-history @var{begin}, @var{end}
7006 Disassembles instructions beginning with instruction number
7007 @var{begin} until instruction number @var{end}. The instruction
7008 number @var{end} is included.
7011 This command may not be available for all recording methods.
7014 @item set record instruction-history-size @var{size}
7015 @itemx set record instruction-history-size unlimited
7016 Define how many instructions to disassemble in the @code{record
7017 instruction-history} command. The default value is 10.
7018 A @var{size} of @code{unlimited} means unlimited instructions.
7021 @item show record instruction-history-size
7022 Show how many instructions to disassemble in the @code{record
7023 instruction-history} command.
7025 @kindex record function-call-history
7026 @kindex rec function-call-history
7027 @item record function-call-history
7028 Prints the execution history at function granularity. It prints one
7029 line for each sequence of instructions that belong to the same
7030 function giving the name of that function, the source lines
7031 for this instruction sequence (if the @code{/l} modifier is
7032 specified), and the instructions numbers that form the sequence (if
7033 the @code{/i} modifier is specified). The function names are indented
7034 to reflect the call stack depth if the @code{/c} modifier is
7035 specified. The @code{/l}, @code{/i}, and @code{/c} modifiers can be
7039 (@value{GDBP}) @b{list 1, 10}
7050 (@value{GDBP}) @b{record function-call-history /ilc}
7051 1 bar inst 1,4 at foo.c:6,8
7052 2 foo inst 5,10 at foo.c:2,3
7053 3 bar inst 11,13 at foo.c:9,10
7056 By default, ten lines are printed. This can be changed using the
7057 @code{set record function-call-history-size} command. Functions are
7058 printed in execution order. There are several ways to specify what
7062 @item record function-call-history @var{func}
7063 Prints ten functions starting from function number @var{func}.
7065 @item record function-call-history @var{func}, +/-@var{n}
7066 Prints @var{n} functions around function number @var{func}. If
7067 @var{n} is preceded with @code{+}, prints @var{n} functions after
7068 function number @var{func}. If @var{n} is preceded with @code{-},
7069 prints @var{n} functions before function number @var{func}.
7071 @item record function-call-history
7072 Prints ten more functions after the last ten-line print.
7074 @item record function-call-history -
7075 Prints ten more functions before the last ten-line print.
7077 @item record function-call-history @var{begin}, @var{end}
7078 Prints functions beginning with function number @var{begin} until
7079 function number @var{end}. The function number @var{end} is included.
7082 This command may not be available for all recording methods.
7084 @item set record function-call-history-size @var{size}
7085 @itemx set record function-call-history-size unlimited
7086 Define how many lines to print in the
7087 @code{record function-call-history} command. The default value is 10.
7088 A size of @code{unlimited} means unlimited lines.
7090 @item show record function-call-history-size
7091 Show how many lines to print in the
7092 @code{record function-call-history} command.
7097 @chapter Examining the Stack
7099 When your program has stopped, the first thing you need to know is where it
7100 stopped and how it got there.
7103 Each time your program performs a function call, information about the call
7105 That information includes the location of the call in your program,
7106 the arguments of the call,
7107 and the local variables of the function being called.
7108 The information is saved in a block of data called a @dfn{stack frame}.
7109 The stack frames are allocated in a region of memory called the @dfn{call
7112 When your program stops, the @value{GDBN} commands for examining the
7113 stack allow you to see all of this information.
7115 @cindex selected frame
7116 One of the stack frames is @dfn{selected} by @value{GDBN} and many
7117 @value{GDBN} commands refer implicitly to the selected frame. In
7118 particular, whenever you ask @value{GDBN} for the value of a variable in
7119 your program, the value is found in the selected frame. There are
7120 special @value{GDBN} commands to select whichever frame you are
7121 interested in. @xref{Selection, ,Selecting a Frame}.
7123 When your program stops, @value{GDBN} automatically selects the
7124 currently executing frame and describes it briefly, similar to the
7125 @code{frame} command (@pxref{Frame Info, ,Information about a Frame}).
7128 * Frames:: Stack frames
7129 * Backtrace:: Backtraces
7130 * Selection:: Selecting a frame
7131 * Frame Info:: Information on a frame
7132 * Frame Filter Management:: Managing frame filters
7137 @section Stack Frames
7139 @cindex frame, definition
7141 The call stack is divided up into contiguous pieces called @dfn{stack
7142 frames}, or @dfn{frames} for short; each frame is the data associated
7143 with one call to one function. The frame contains the arguments given
7144 to the function, the function's local variables, and the address at
7145 which the function is executing.
7147 @cindex initial frame
7148 @cindex outermost frame
7149 @cindex innermost frame
7150 When your program is started, the stack has only one frame, that of the
7151 function @code{main}. This is called the @dfn{initial} frame or the
7152 @dfn{outermost} frame. Each time a function is called, a new frame is
7153 made. Each time a function returns, the frame for that function invocation
7154 is eliminated. If a function is recursive, there can be many frames for
7155 the same function. The frame for the function in which execution is
7156 actually occurring is called the @dfn{innermost} frame. This is the most
7157 recently created of all the stack frames that still exist.
7159 @cindex frame pointer
7160 Inside your program, stack frames are identified by their addresses. A
7161 stack frame consists of many bytes, each of which has its own address; each
7162 kind of computer has a convention for choosing one byte whose
7163 address serves as the address of the frame. Usually this address is kept
7164 in a register called the @dfn{frame pointer register}
7165 (@pxref{Registers, $fp}) while execution is going on in that frame.
7167 @cindex frame number
7168 @value{GDBN} assigns numbers to all existing stack frames, starting with
7169 zero for the innermost frame, one for the frame that called it,
7170 and so on upward. These numbers do not really exist in your program;
7171 they are assigned by @value{GDBN} to give you a way of designating stack
7172 frames in @value{GDBN} commands.
7174 @c The -fomit-frame-pointer below perennially causes hbox overflow
7175 @c underflow problems.
7176 @cindex frameless execution
7177 Some compilers provide a way to compile functions so that they operate
7178 without stack frames. (For example, the @value{NGCC} option
7180 @samp{-fomit-frame-pointer}
7182 generates functions without a frame.)
7183 This is occasionally done with heavily used library functions to save
7184 the frame setup time. @value{GDBN} has limited facilities for dealing
7185 with these function invocations. If the innermost function invocation
7186 has no stack frame, @value{GDBN} nevertheless regards it as though
7187 it had a separate frame, which is numbered zero as usual, allowing
7188 correct tracing of the function call chain. However, @value{GDBN} has
7189 no provision for frameless functions elsewhere in the stack.
7195 @cindex call stack traces
7196 A backtrace is a summary of how your program got where it is. It shows one
7197 line per frame, for many frames, starting with the currently executing
7198 frame (frame zero), followed by its caller (frame one), and on up the
7201 @anchor{backtrace-command}
7204 @kindex bt @r{(@code{backtrace})}
7207 Print a backtrace of the entire stack: one line per frame for all
7208 frames in the stack.
7210 You can stop the backtrace at any time by typing the system interrupt
7211 character, normally @kbd{Ctrl-c}.
7213 @item backtrace @var{n}
7215 Similar, but print only the innermost @var{n} frames.
7217 @item backtrace -@var{n}
7219 Similar, but print only the outermost @var{n} frames.
7221 @item backtrace full
7223 @itemx bt full @var{n}
7224 @itemx bt full -@var{n}
7225 Print the values of the local variables also. As described above,
7226 @var{n} specifies the number of frames to print.
7228 @item backtrace no-filters
7229 @itemx bt no-filters
7230 @itemx bt no-filters @var{n}
7231 @itemx bt no-filters -@var{n}
7232 @itemx bt no-filters full
7233 @itemx bt no-filters full @var{n}
7234 @itemx bt no-filters full -@var{n}
7235 Do not run Python frame filters on this backtrace. @xref{Frame
7236 Filter API}, for more information. Additionally use @ref{disable
7237 frame-filter all} to turn off all frame filters. This is only
7238 relevant when @value{GDBN} has been configured with @code{Python}
7244 The names @code{where} and @code{info stack} (abbreviated @code{info s})
7245 are additional aliases for @code{backtrace}.
7247 @cindex multiple threads, backtrace
7248 In a multi-threaded program, @value{GDBN} by default shows the
7249 backtrace only for the current thread. To display the backtrace for
7250 several or all of the threads, use the command @code{thread apply}
7251 (@pxref{Threads, thread apply}). For example, if you type @kbd{thread
7252 apply all backtrace}, @value{GDBN} will display the backtrace for all
7253 the threads; this is handy when you debug a core dump of a
7254 multi-threaded program.
7256 Each line in the backtrace shows the frame number and the function name.
7257 The program counter value is also shown---unless you use @code{set
7258 print address off}. The backtrace also shows the source file name and
7259 line number, as well as the arguments to the function. The program
7260 counter value is omitted if it is at the beginning of the code for that
7263 Here is an example of a backtrace. It was made with the command
7264 @samp{bt 3}, so it shows the innermost three frames.
7268 #0 m4_traceon (obs=0x24eb0, argc=1, argv=0x2b8c8)
7270 #1 0x6e38 in expand_macro (sym=0x2b600, data=...) at macro.c:242
7271 #2 0x6840 in expand_token (obs=0x0, t=177664, td=0xf7fffb08)
7273 (More stack frames follow...)
7278 The display for frame zero does not begin with a program counter
7279 value, indicating that your program has stopped at the beginning of the
7280 code for line @code{993} of @code{builtin.c}.
7283 The value of parameter @code{data} in frame 1 has been replaced by
7284 @code{@dots{}}. By default, @value{GDBN} prints the value of a parameter
7285 only if it is a scalar (integer, pointer, enumeration, etc). See command
7286 @kbd{set print frame-arguments} in @ref{Print Settings} for more details
7287 on how to configure the way function parameter values are printed.
7289 @cindex optimized out, in backtrace
7290 @cindex function call arguments, optimized out
7291 If your program was compiled with optimizations, some compilers will
7292 optimize away arguments passed to functions if those arguments are
7293 never used after the call. Such optimizations generate code that
7294 passes arguments through registers, but doesn't store those arguments
7295 in the stack frame. @value{GDBN} has no way of displaying such
7296 arguments in stack frames other than the innermost one. Here's what
7297 such a backtrace might look like:
7301 #0 m4_traceon (obs=0x24eb0, argc=1, argv=0x2b8c8)
7303 #1 0x6e38 in expand_macro (sym=<optimized out>) at macro.c:242
7304 #2 0x6840 in expand_token (obs=0x0, t=<optimized out>, td=0xf7fffb08)
7306 (More stack frames follow...)
7311 The values of arguments that were not saved in their stack frames are
7312 shown as @samp{<optimized out>}.
7314 If you need to display the values of such optimized-out arguments,
7315 either deduce that from other variables whose values depend on the one
7316 you are interested in, or recompile without optimizations.
7318 @cindex backtrace beyond @code{main} function
7319 @cindex program entry point
7320 @cindex startup code, and backtrace
7321 Most programs have a standard user entry point---a place where system
7322 libraries and startup code transition into user code. For C this is
7323 @code{main}@footnote{
7324 Note that embedded programs (the so-called ``free-standing''
7325 environment) are not required to have a @code{main} function as the
7326 entry point. They could even have multiple entry points.}.
7327 When @value{GDBN} finds the entry function in a backtrace
7328 it will terminate the backtrace, to avoid tracing into highly
7329 system-specific (and generally uninteresting) code.
7331 If you need to examine the startup code, or limit the number of levels
7332 in a backtrace, you can change this behavior:
7335 @item set backtrace past-main
7336 @itemx set backtrace past-main on
7337 @kindex set backtrace
7338 Backtraces will continue past the user entry point.
7340 @item set backtrace past-main off
7341 Backtraces will stop when they encounter the user entry point. This is the
7344 @item show backtrace past-main
7345 @kindex show backtrace
7346 Display the current user entry point backtrace policy.
7348 @item set backtrace past-entry
7349 @itemx set backtrace past-entry on
7350 Backtraces will continue past the internal entry point of an application.
7351 This entry point is encoded by the linker when the application is built,
7352 and is likely before the user entry point @code{main} (or equivalent) is called.
7354 @item set backtrace past-entry off
7355 Backtraces will stop when they encounter the internal entry point of an
7356 application. This is the default.
7358 @item show backtrace past-entry
7359 Display the current internal entry point backtrace policy.
7361 @item set backtrace limit @var{n}
7362 @itemx set backtrace limit 0
7363 @itemx set backtrace limit unlimited
7364 @cindex backtrace limit
7365 Limit the backtrace to @var{n} levels. A value of @code{unlimited}
7366 or zero means unlimited levels.
7368 @item show backtrace limit
7369 Display the current limit on backtrace levels.
7372 You can control how file names are displayed.
7375 @item set filename-display
7376 @itemx set filename-display relative
7377 @cindex filename-display
7378 Display file names relative to the compilation directory. This is the default.
7380 @item set filename-display basename
7381 Display only basename of a filename.
7383 @item set filename-display absolute
7384 Display an absolute filename.
7386 @item show filename-display
7387 Show the current way to display filenames.
7391 @section Selecting a Frame
7393 Most commands for examining the stack and other data in your program work on
7394 whichever stack frame is selected at the moment. Here are the commands for
7395 selecting a stack frame; all of them finish by printing a brief description
7396 of the stack frame just selected.
7399 @kindex frame@r{, selecting}
7400 @kindex f @r{(@code{frame})}
7403 Select frame number @var{n}. Recall that frame zero is the innermost
7404 (currently executing) frame, frame one is the frame that called the
7405 innermost one, and so on. The highest-numbered frame is the one for
7408 @item frame @var{stack-addr} [ @var{pc-addr} ]
7409 @itemx f @var{stack-addr} [ @var{pc-addr} ]
7410 Select the frame at address @var{stack-addr}. This is useful mainly if the
7411 chaining of stack frames has been damaged by a bug, making it
7412 impossible for @value{GDBN} to assign numbers properly to all frames. In
7413 addition, this can be useful when your program has multiple stacks and
7414 switches between them. The optional @var{pc-addr} can also be given to
7415 specify the value of PC for the stack frame.
7419 Move @var{n} frames up the stack; @var{n} defaults to 1. For positive
7420 numbers @var{n}, this advances toward the outermost frame, to higher
7421 frame numbers, to frames that have existed longer.
7424 @kindex do @r{(@code{down})}
7426 Move @var{n} frames down the stack; @var{n} defaults to 1. For
7427 positive numbers @var{n}, this advances toward the innermost frame, to
7428 lower frame numbers, to frames that were created more recently.
7429 You may abbreviate @code{down} as @code{do}.
7432 All of these commands end by printing two lines of output describing the
7433 frame. The first line shows the frame number, the function name, the
7434 arguments, and the source file and line number of execution in that
7435 frame. The second line shows the text of that source line.
7443 #1 0x22f0 in main (argc=1, argv=0xf7fffbf4, env=0xf7fffbfc)
7445 10 read_input_file (argv[i]);
7449 After such a printout, the @code{list} command with no arguments
7450 prints ten lines centered on the point of execution in the frame.
7451 You can also edit the program at the point of execution with your favorite
7452 editing program by typing @code{edit}.
7453 @xref{List, ,Printing Source Lines},
7457 @kindex select-frame
7459 The @code{select-frame} command is a variant of @code{frame} that does
7460 not display the new frame after selecting it. This command is
7461 intended primarily for use in @value{GDBN} command scripts, where the
7462 output might be unnecessary and distracting.
7464 @kindex down-silently
7466 @item up-silently @var{n}
7467 @itemx down-silently @var{n}
7468 These two commands are variants of @code{up} and @code{down},
7469 respectively; they differ in that they do their work silently, without
7470 causing display of the new frame. They are intended primarily for use
7471 in @value{GDBN} command scripts, where the output might be unnecessary and
7476 @section Information About a Frame
7478 There are several other commands to print information about the selected
7484 When used without any argument, this command does not change which
7485 frame is selected, but prints a brief description of the currently
7486 selected stack frame. It can be abbreviated @code{f}. With an
7487 argument, this command is used to select a stack frame.
7488 @xref{Selection, ,Selecting a Frame}.
7491 @kindex info f @r{(@code{info frame})}
7494 This command prints a verbose description of the selected stack frame,
7499 the address of the frame
7501 the address of the next frame down (called by this frame)
7503 the address of the next frame up (caller of this frame)
7505 the language in which the source code corresponding to this frame is written
7507 the address of the frame's arguments
7509 the address of the frame's local variables
7511 the program counter saved in it (the address of execution in the caller frame)
7513 which registers were saved in the frame
7516 @noindent The verbose description is useful when
7517 something has gone wrong that has made the stack format fail to fit
7518 the usual conventions.
7520 @item info frame @var{addr}
7521 @itemx info f @var{addr}
7522 Print a verbose description of the frame at address @var{addr}, without
7523 selecting that frame. The selected frame remains unchanged by this
7524 command. This requires the same kind of address (more than one for some
7525 architectures) that you specify in the @code{frame} command.
7526 @xref{Selection, ,Selecting a Frame}.
7530 Print the arguments of the selected frame, each on a separate line.
7534 Print the local variables of the selected frame, each on a separate
7535 line. These are all variables (declared either static or automatic)
7536 accessible at the point of execution of the selected frame.
7540 @node Frame Filter Management
7541 @section Management of Frame Filters.
7542 @cindex managing frame filters
7544 Frame filters are Python based utilities to manage and decorate the
7545 output of frames. @xref{Frame Filter API}, for further information.
7547 Managing frame filters is performed by several commands available
7548 within @value{GDBN}, detailed here.
7551 @kindex info frame-filter
7552 @item info frame-filter
7553 Print a list of installed frame filters from all dictionaries, showing
7554 their name, priority and enabled status.
7556 @kindex disable frame-filter
7557 @anchor{disable frame-filter all}
7558 @item disable frame-filter @var{filter-dictionary} @var{filter-name}
7559 Disable a frame filter in the dictionary matching
7560 @var{filter-dictionary} and @var{filter-name}. The
7561 @var{filter-dictionary} may be @code{all}, @code{global},
7562 @code{progspace}, or the name of the object file where the frame filter
7563 dictionary resides. When @code{all} is specified, all frame filters
7564 across all dictionaries are disabled. The @var{filter-name} is the name
7565 of the frame filter and is used when @code{all} is not the option for
7566 @var{filter-dictionary}. A disabled frame-filter is not deleted, it
7567 may be enabled again later.
7569 @kindex enable frame-filter
7570 @item enable frame-filter @var{filter-dictionary} @var{filter-name}
7571 Enable a frame filter in the dictionary matching
7572 @var{filter-dictionary} and @var{filter-name}. The
7573 @var{filter-dictionary} may be @code{all}, @code{global},
7574 @code{progspace} or the name of the object file where the frame filter
7575 dictionary resides. When @code{all} is specified, all frame filters across
7576 all dictionaries are enabled. The @var{filter-name} is the name of the frame
7577 filter and is used when @code{all} is not the option for
7578 @var{filter-dictionary}.
7583 (gdb) info frame-filter
7585 global frame-filters:
7586 Priority Enabled Name
7587 1000 No PrimaryFunctionFilter
7590 progspace /build/test frame-filters:
7591 Priority Enabled Name
7592 100 Yes ProgspaceFilter
7594 objfile /build/test frame-filters:
7595 Priority Enabled Name
7596 999 Yes BuildProgra Filter
7598 (gdb) disable frame-filter /build/test BuildProgramFilter
7599 (gdb) info frame-filter
7601 global frame-filters:
7602 Priority Enabled Name
7603 1000 No PrimaryFunctionFilter
7606 progspace /build/test frame-filters:
7607 Priority Enabled Name
7608 100 Yes ProgspaceFilter
7610 objfile /build/test frame-filters:
7611 Priority Enabled Name
7612 999 No BuildProgramFilter
7614 (gdb) enable frame-filter global PrimaryFunctionFilter
7615 (gdb) info frame-filter
7617 global frame-filters:
7618 Priority Enabled Name
7619 1000 Yes PrimaryFunctionFilter
7622 progspace /build/test frame-filters:
7623 Priority Enabled Name
7624 100 Yes ProgspaceFilter
7626 objfile /build/test frame-filters:
7627 Priority Enabled Name
7628 999 No BuildProgramFilter
7631 @kindex set frame-filter priority
7632 @item set frame-filter priority @var{filter-dictionary} @var{filter-name} @var{priority}
7633 Set the @var{priority} of a frame filter in the dictionary matching
7634 @var{filter-dictionary}, and the frame filter name matching
7635 @var{filter-name}. The @var{filter-dictionary} may be @code{global},
7636 @code{progspace} or the name of the object file where the frame filter
7637 dictionary resides. The @var{priority} is an integer.
7639 @kindex show frame-filter priority
7640 @item show frame-filter priority @var{filter-dictionary} @var{filter-name}
7641 Show the @var{priority} of a frame filter in the dictionary matching
7642 @var{filter-dictionary}, and the frame filter name matching
7643 @var{filter-name}. The @var{filter-dictionary} may be @code{global},
7644 @code{progspace} or the name of the object file where the frame filter
7650 (gdb) info frame-filter
7652 global frame-filters:
7653 Priority Enabled Name
7654 1000 Yes PrimaryFunctionFilter
7657 progspace /build/test frame-filters:
7658 Priority Enabled Name
7659 100 Yes ProgspaceFilter
7661 objfile /build/test frame-filters:
7662 Priority Enabled Name
7663 999 No BuildProgramFilter
7665 (gdb) set frame-filter priority global Reverse 50
7666 (gdb) info frame-filter
7668 global frame-filters:
7669 Priority Enabled Name
7670 1000 Yes PrimaryFunctionFilter
7673 progspace /build/test frame-filters:
7674 Priority Enabled Name
7675 100 Yes ProgspaceFilter
7677 objfile /build/test frame-filters:
7678 Priority Enabled Name
7679 999 No BuildProgramFilter
7684 @chapter Examining Source Files
7686 @value{GDBN} can print parts of your program's source, since the debugging
7687 information recorded in the program tells @value{GDBN} what source files were
7688 used to build it. When your program stops, @value{GDBN} spontaneously prints
7689 the line where it stopped. Likewise, when you select a stack frame
7690 (@pxref{Selection, ,Selecting a Frame}), @value{GDBN} prints the line where
7691 execution in that frame has stopped. You can print other portions of
7692 source files by explicit command.
7694 If you use @value{GDBN} through its @sc{gnu} Emacs interface, you may
7695 prefer to use Emacs facilities to view source; see @ref{Emacs, ,Using
7696 @value{GDBN} under @sc{gnu} Emacs}.
7699 * List:: Printing source lines
7700 * Specify Location:: How to specify code locations
7701 * Edit:: Editing source files
7702 * Search:: Searching source files
7703 * Source Path:: Specifying source directories
7704 * Machine Code:: Source and machine code
7708 @section Printing Source Lines
7711 @kindex l @r{(@code{list})}
7712 To print lines from a source file, use the @code{list} command
7713 (abbreviated @code{l}). By default, ten lines are printed.
7714 There are several ways to specify what part of the file you want to
7715 print; see @ref{Specify Location}, for the full list.
7717 Here are the forms of the @code{list} command most commonly used:
7720 @item list @var{linenum}
7721 Print lines centered around line number @var{linenum} in the
7722 current source file.
7724 @item list @var{function}
7725 Print lines centered around the beginning of function
7729 Print more lines. If the last lines printed were printed with a
7730 @code{list} command, this prints lines following the last lines
7731 printed; however, if the last line printed was a solitary line printed
7732 as part of displaying a stack frame (@pxref{Stack, ,Examining the
7733 Stack}), this prints lines centered around that line.
7736 Print lines just before the lines last printed.
7739 @cindex @code{list}, how many lines to display
7740 By default, @value{GDBN} prints ten source lines with any of these forms of
7741 the @code{list} command. You can change this using @code{set listsize}:
7744 @kindex set listsize
7745 @item set listsize @var{count}
7746 @itemx set listsize unlimited
7747 Make the @code{list} command display @var{count} source lines (unless
7748 the @code{list} argument explicitly specifies some other number).
7749 Setting @var{count} to @code{unlimited} or 0 means there's no limit.
7751 @kindex show listsize
7753 Display the number of lines that @code{list} prints.
7756 Repeating a @code{list} command with @key{RET} discards the argument,
7757 so it is equivalent to typing just @code{list}. This is more useful
7758 than listing the same lines again. An exception is made for an
7759 argument of @samp{-}; that argument is preserved in repetition so that
7760 each repetition moves up in the source file.
7762 In general, the @code{list} command expects you to supply zero, one or two
7763 @dfn{locations}. Locations specify source lines; there are several ways
7764 of writing them (@pxref{Specify Location}), but the effect is always
7765 to specify some source line.
7767 Here is a complete description of the possible arguments for @code{list}:
7770 @item list @var{location}
7771 Print lines centered around the line specified by @var{location}.
7773 @item list @var{first},@var{last}
7774 Print lines from @var{first} to @var{last}. Both arguments are
7775 locations. When a @code{list} command has two locations, and the
7776 source file of the second location is omitted, this refers to
7777 the same source file as the first location.
7779 @item list ,@var{last}
7780 Print lines ending with @var{last}.
7782 @item list @var{first},
7783 Print lines starting with @var{first}.
7786 Print lines just after the lines last printed.
7789 Print lines just before the lines last printed.
7792 As described in the preceding table.
7795 @node Specify Location
7796 @section Specifying a Location
7797 @cindex specifying location
7799 @cindex source location
7802 * Linespec Locations:: Linespec locations
7803 * Explicit Locations:: Explicit locations
7804 * Address Locations:: Address locations
7807 Several @value{GDBN} commands accept arguments that specify a location
7808 of your program's code. Since @value{GDBN} is a source-level
7809 debugger, a location usually specifies some line in the source code.
7810 Locations may be specified using three different formats:
7811 linespec locations, explicit locations, or address locations.
7813 @node Linespec Locations
7814 @subsection Linespec Locations
7815 @cindex linespec locations
7817 A @dfn{linespec} is a colon-separated list of source location parameters such
7818 as file name, function name, etc. Here are all the different ways of
7819 specifying a linespec:
7823 Specifies the line number @var{linenum} of the current source file.
7826 @itemx +@var{offset}
7827 Specifies the line @var{offset} lines before or after the @dfn{current
7828 line}. For the @code{list} command, the current line is the last one
7829 printed; for the breakpoint commands, this is the line at which
7830 execution stopped in the currently selected @dfn{stack frame}
7831 (@pxref{Frames, ,Frames}, for a description of stack frames.) When
7832 used as the second of the two linespecs in a @code{list} command,
7833 this specifies the line @var{offset} lines up or down from the first
7836 @item @var{filename}:@var{linenum}
7837 Specifies the line @var{linenum} in the source file @var{filename}.
7838 If @var{filename} is a relative file name, then it will match any
7839 source file name with the same trailing components. For example, if
7840 @var{filename} is @samp{gcc/expr.c}, then it will match source file
7841 name of @file{/build/trunk/gcc/expr.c}, but not
7842 @file{/build/trunk/libcpp/expr.c} or @file{/build/trunk/gcc/x-expr.c}.
7844 @item @var{function}
7845 Specifies the line that begins the body of the function @var{function}.
7846 For example, in C, this is the line with the open brace.
7848 @item @var{function}:@var{label}
7849 Specifies the line where @var{label} appears in @var{function}.
7851 @item @var{filename}:@var{function}
7852 Specifies the line that begins the body of the function @var{function}
7853 in the file @var{filename}. You only need the file name with a
7854 function name to avoid ambiguity when there are identically named
7855 functions in different source files.
7858 Specifies the line at which the label named @var{label} appears
7859 in the function corresponding to the currently selected stack frame.
7860 If there is no current selected stack frame (for instance, if the inferior
7861 is not running), then @value{GDBN} will not search for a label.
7863 @cindex breakpoint at static probe point
7864 @item -pstap|-probe-stap @r{[}@var{objfile}:@r{[}@var{provider}:@r{]}@r{]}@var{name}
7865 The @sc{gnu}/Linux tool @code{SystemTap} provides a way for
7866 applications to embed static probes. @xref{Static Probe Points}, for more
7867 information on finding and using static probes. This form of linespec
7868 specifies the location of such a static probe.
7870 If @var{objfile} is given, only probes coming from that shared library
7871 or executable matching @var{objfile} as a regular expression are considered.
7872 If @var{provider} is given, then only probes from that provider are considered.
7873 If several probes match the spec, @value{GDBN} will insert a breakpoint at
7874 each one of those probes.
7877 @node Explicit Locations
7878 @subsection Explicit Locations
7879 @cindex explicit locations
7881 @dfn{Explicit locations} allow the user to directly specify the source
7882 location's parameters using option-value pairs.
7884 Explicit locations are useful when several functions, labels, or
7885 file names have the same name (base name for files) in the program's
7886 sources. In these cases, explicit locations point to the source
7887 line you meant more accurately and unambiguously. Also, using
7888 explicit locations might be faster in large programs.
7890 For example, the linespec @samp{foo:bar} may refer to a function @code{bar}
7891 defined in the file named @file{foo} or the label @code{bar} in a function
7892 named @code{foo}. @value{GDBN} must search either the file system or
7893 the symbol table to know.
7895 The list of valid explicit location options is summarized in the
7899 @item -source @var{filename}
7900 The value specifies the source file name. To differentiate between
7901 files with the same base name, prepend as many directories as is necessary
7902 to uniquely identify the desired file, e.g., @file{foo/bar/baz.c}. Otherwise
7903 @value{GDBN} will use the first file it finds with the given base
7904 name. This option requires the use of either @code{-function} or @code{-line}.
7906 @item -function @var{function}
7907 The value specifies the name of a function. Operations
7908 on function locations unmodified by other options (such as @code{-label}
7909 or @code{-line}) refer to the line that begins the body of the function.
7910 In C, for example, this is the line with the open brace.
7912 @item -label @var{label}
7913 The value specifies the name of a label. When the function
7914 name is not specified, the label is searched in the function of the currently
7915 selected stack frame.
7917 @item -line @var{number}
7918 The value specifies a line offset for the location. The offset may either
7919 be absolute (@code{-line 3}) or relative (@code{-line +3}), depending on
7920 the command. When specified without any other options, the line offset is
7921 relative to the current line.
7924 Explicit location options may be abbreviated by omitting any non-unique
7925 trailing characters from the option name, e.g., @code{break -s main.c -li 3}.
7927 @node Address Locations
7928 @subsection Address Locations
7929 @cindex address locations
7931 @dfn{Address locations} indicate a specific program address. They have
7932 the generalized form *@var{address}.
7934 For line-oriented commands, such as @code{list} and @code{edit}, this
7935 specifies a source line that contains @var{address}. For @code{break} and
7936 other breakpoint-oriented commands, this can be used to set breakpoints in
7937 parts of your program which do not have debugging information or
7940 Here @var{address} may be any expression valid in the current working
7941 language (@pxref{Languages, working language}) that specifies a code
7942 address. In addition, as a convenience, @value{GDBN} extends the
7943 semantics of expressions used in locations to cover several situations
7944 that frequently occur during debugging. Here are the various forms
7948 @item @var{expression}
7949 Any expression valid in the current working language.
7951 @item @var{funcaddr}
7952 An address of a function or procedure derived from its name. In C,
7953 C@t{++}, Objective-C, Fortran, minimal, and assembly, this is
7954 simply the function's name @var{function} (and actually a special case
7955 of a valid expression). In Pascal and Modula-2, this is
7956 @code{&@var{function}}. In Ada, this is @code{@var{function}'Address}
7957 (although the Pascal form also works).
7959 This form specifies the address of the function's first instruction,
7960 before the stack frame and arguments have been set up.
7962 @item '@var{filename}':@var{funcaddr}
7963 Like @var{funcaddr} above, but also specifies the name of the source
7964 file explicitly. This is useful if the name of the function does not
7965 specify the function unambiguously, e.g., if there are several
7966 functions with identical names in different source files.
7970 @section Editing Source Files
7971 @cindex editing source files
7974 @kindex e @r{(@code{edit})}
7975 To edit the lines in a source file, use the @code{edit} command.
7976 The editing program of your choice
7977 is invoked with the current line set to
7978 the active line in the program.
7979 Alternatively, there are several ways to specify what part of the file you
7980 want to print if you want to see other parts of the program:
7983 @item edit @var{location}
7984 Edit the source file specified by @code{location}. Editing starts at
7985 that @var{location}, e.g., at the specified source line of the
7986 specified file. @xref{Specify Location}, for all the possible forms
7987 of the @var{location} argument; here are the forms of the @code{edit}
7988 command most commonly used:
7991 @item edit @var{number}
7992 Edit the current source file with @var{number} as the active line number.
7994 @item edit @var{function}
7995 Edit the file containing @var{function} at the beginning of its definition.
8000 @subsection Choosing your Editor
8001 You can customize @value{GDBN} to use any editor you want
8003 The only restriction is that your editor (say @code{ex}), recognizes the
8004 following command-line syntax:
8006 ex +@var{number} file
8008 The optional numeric value +@var{number} specifies the number of the line in
8009 the file where to start editing.}.
8010 By default, it is @file{@value{EDITOR}}, but you can change this
8011 by setting the environment variable @code{EDITOR} before using
8012 @value{GDBN}. For example, to configure @value{GDBN} to use the
8013 @code{vi} editor, you could use these commands with the @code{sh} shell:
8019 or in the @code{csh} shell,
8021 setenv EDITOR /usr/bin/vi
8026 @section Searching Source Files
8027 @cindex searching source files
8029 There are two commands for searching through the current source file for a
8034 @kindex forward-search
8035 @kindex fo @r{(@code{forward-search})}
8036 @item forward-search @var{regexp}
8037 @itemx search @var{regexp}
8038 The command @samp{forward-search @var{regexp}} checks each line,
8039 starting with the one following the last line listed, for a match for
8040 @var{regexp}. It lists the line that is found. You can use the
8041 synonym @samp{search @var{regexp}} or abbreviate the command name as
8044 @kindex reverse-search
8045 @item reverse-search @var{regexp}
8046 The command @samp{reverse-search @var{regexp}} checks each line, starting
8047 with the one before the last line listed and going backward, for a match
8048 for @var{regexp}. It lists the line that is found. You can abbreviate
8049 this command as @code{rev}.
8053 @section Specifying Source Directories
8056 @cindex directories for source files
8057 Executable programs sometimes do not record the directories of the source
8058 files from which they were compiled, just the names. Even when they do,
8059 the directories could be moved between the compilation and your debugging
8060 session. @value{GDBN} has a list of directories to search for source files;
8061 this is called the @dfn{source path}. Each time @value{GDBN} wants a source file,
8062 it tries all the directories in the list, in the order they are present
8063 in the list, until it finds a file with the desired name.
8065 For example, suppose an executable references the file
8066 @file{/usr/src/foo-1.0/lib/foo.c}, and our source path is
8067 @file{/mnt/cross}. The file is first looked up literally; if this
8068 fails, @file{/mnt/cross/usr/src/foo-1.0/lib/foo.c} is tried; if this
8069 fails, @file{/mnt/cross/foo.c} is opened; if this fails, an error
8070 message is printed. @value{GDBN} does not look up the parts of the
8071 source file name, such as @file{/mnt/cross/src/foo-1.0/lib/foo.c}.
8072 Likewise, the subdirectories of the source path are not searched: if
8073 the source path is @file{/mnt/cross}, and the binary refers to
8074 @file{foo.c}, @value{GDBN} would not find it under
8075 @file{/mnt/cross/usr/src/foo-1.0/lib}.
8077 Plain file names, relative file names with leading directories, file
8078 names containing dots, etc.@: are all treated as described above; for
8079 instance, if the source path is @file{/mnt/cross}, and the source file
8080 is recorded as @file{../lib/foo.c}, @value{GDBN} would first try
8081 @file{../lib/foo.c}, then @file{/mnt/cross/../lib/foo.c}, and after
8082 that---@file{/mnt/cross/foo.c}.
8084 Note that the executable search path is @emph{not} used to locate the
8087 Whenever you reset or rearrange the source path, @value{GDBN} clears out
8088 any information it has cached about where source files are found and where
8089 each line is in the file.
8093 When you start @value{GDBN}, its source path includes only @samp{cdir}
8094 and @samp{cwd}, in that order.
8095 To add other directories, use the @code{directory} command.
8097 The search path is used to find both program source files and @value{GDBN}
8098 script files (read using the @samp{-command} option and @samp{source} command).
8100 In addition to the source path, @value{GDBN} provides a set of commands
8101 that manage a list of source path substitution rules. A @dfn{substitution
8102 rule} specifies how to rewrite source directories stored in the program's
8103 debug information in case the sources were moved to a different
8104 directory between compilation and debugging. A rule is made of
8105 two strings, the first specifying what needs to be rewritten in
8106 the path, and the second specifying how it should be rewritten.
8107 In @ref{set substitute-path}, we name these two parts @var{from} and
8108 @var{to} respectively. @value{GDBN} does a simple string replacement
8109 of @var{from} with @var{to} at the start of the directory part of the
8110 source file name, and uses that result instead of the original file
8111 name to look up the sources.
8113 Using the previous example, suppose the @file{foo-1.0} tree has been
8114 moved from @file{/usr/src} to @file{/mnt/cross}, then you can tell
8115 @value{GDBN} to replace @file{/usr/src} in all source path names with
8116 @file{/mnt/cross}. The first lookup will then be
8117 @file{/mnt/cross/foo-1.0/lib/foo.c} in place of the original location
8118 of @file{/usr/src/foo-1.0/lib/foo.c}. To define a source path
8119 substitution rule, use the @code{set substitute-path} command
8120 (@pxref{set substitute-path}).
8122 To avoid unexpected substitution results, a rule is applied only if the
8123 @var{from} part of the directory name ends at a directory separator.
8124 For instance, a rule substituting @file{/usr/source} into
8125 @file{/mnt/cross} will be applied to @file{/usr/source/foo-1.0} but
8126 not to @file{/usr/sourceware/foo-2.0}. And because the substitution
8127 is applied only at the beginning of the directory name, this rule will
8128 not be applied to @file{/root/usr/source/baz.c} either.
8130 In many cases, you can achieve the same result using the @code{directory}
8131 command. However, @code{set substitute-path} can be more efficient in
8132 the case where the sources are organized in a complex tree with multiple
8133 subdirectories. With the @code{directory} command, you need to add each
8134 subdirectory of your project. If you moved the entire tree while
8135 preserving its internal organization, then @code{set substitute-path}
8136 allows you to direct the debugger to all the sources with one single
8139 @code{set substitute-path} is also more than just a shortcut command.
8140 The source path is only used if the file at the original location no
8141 longer exists. On the other hand, @code{set substitute-path} modifies
8142 the debugger behavior to look at the rewritten location instead. So, if
8143 for any reason a source file that is not relevant to your executable is
8144 located at the original location, a substitution rule is the only
8145 method available to point @value{GDBN} at the new location.
8147 @cindex @samp{--with-relocated-sources}
8148 @cindex default source path substitution
8149 You can configure a default source path substitution rule by
8150 configuring @value{GDBN} with the
8151 @samp{--with-relocated-sources=@var{dir}} option. The @var{dir}
8152 should be the name of a directory under @value{GDBN}'s configured
8153 prefix (set with @samp{--prefix} or @samp{--exec-prefix}), and
8154 directory names in debug information under @var{dir} will be adjusted
8155 automatically if the installed @value{GDBN} is moved to a new
8156 location. This is useful if @value{GDBN}, libraries or executables
8157 with debug information and corresponding source code are being moved
8161 @item directory @var{dirname} @dots{}
8162 @item dir @var{dirname} @dots{}
8163 Add directory @var{dirname} to the front of the source path. Several
8164 directory names may be given to this command, separated by @samp{:}
8165 (@samp{;} on MS-DOS and MS-Windows, where @samp{:} usually appears as
8166 part of absolute file names) or
8167 whitespace. You may specify a directory that is already in the source
8168 path; this moves it forward, so @value{GDBN} searches it sooner.
8172 @vindex $cdir@r{, convenience variable}
8173 @vindex $cwd@r{, convenience variable}
8174 @cindex compilation directory
8175 @cindex current directory
8176 @cindex working directory
8177 @cindex directory, current
8178 @cindex directory, compilation
8179 You can use the string @samp{$cdir} to refer to the compilation
8180 directory (if one is recorded), and @samp{$cwd} to refer to the current
8181 working directory. @samp{$cwd} is not the same as @samp{.}---the former
8182 tracks the current working directory as it changes during your @value{GDBN}
8183 session, while the latter is immediately expanded to the current
8184 directory at the time you add an entry to the source path.
8187 Reset the source path to its default value (@samp{$cdir:$cwd} on Unix systems). This requires confirmation.
8189 @c RET-repeat for @code{directory} is explicitly disabled, but since
8190 @c repeating it would be a no-op we do not say that. (thanks to RMS)
8192 @item set directories @var{path-list}
8193 @kindex set directories
8194 Set the source path to @var{path-list}.
8195 @samp{$cdir:$cwd} are added if missing.
8197 @item show directories
8198 @kindex show directories
8199 Print the source path: show which directories it contains.
8201 @anchor{set substitute-path}
8202 @item set substitute-path @var{from} @var{to}
8203 @kindex set substitute-path
8204 Define a source path substitution rule, and add it at the end of the
8205 current list of existing substitution rules. If a rule with the same
8206 @var{from} was already defined, then the old rule is also deleted.
8208 For example, if the file @file{/foo/bar/baz.c} was moved to
8209 @file{/mnt/cross/baz.c}, then the command
8212 (@value{GDBP}) set substitute-path /foo/bar /mnt/cross
8216 will tell @value{GDBN} to replace @samp{/foo/bar} with
8217 @samp{/mnt/cross}, which will allow @value{GDBN} to find the file
8218 @file{baz.c} even though it was moved.
8220 In the case when more than one substitution rule have been defined,
8221 the rules are evaluated one by one in the order where they have been
8222 defined. The first one matching, if any, is selected to perform
8225 For instance, if we had entered the following commands:
8228 (@value{GDBP}) set substitute-path /usr/src/include /mnt/include
8229 (@value{GDBP}) set substitute-path /usr/src /mnt/src
8233 @value{GDBN} would then rewrite @file{/usr/src/include/defs.h} into
8234 @file{/mnt/include/defs.h} by using the first rule. However, it would
8235 use the second rule to rewrite @file{/usr/src/lib/foo.c} into
8236 @file{/mnt/src/lib/foo.c}.
8239 @item unset substitute-path [path]
8240 @kindex unset substitute-path
8241 If a path is specified, search the current list of substitution rules
8242 for a rule that would rewrite that path. Delete that rule if found.
8243 A warning is emitted by the debugger if no rule could be found.
8245 If no path is specified, then all substitution rules are deleted.
8247 @item show substitute-path [path]
8248 @kindex show substitute-path
8249 If a path is specified, then print the source path substitution rule
8250 which would rewrite that path, if any.
8252 If no path is specified, then print all existing source path substitution
8257 If your source path is cluttered with directories that are no longer of
8258 interest, @value{GDBN} may sometimes cause confusion by finding the wrong
8259 versions of source. You can correct the situation as follows:
8263 Use @code{directory} with no argument to reset the source path to its default value.
8266 Use @code{directory} with suitable arguments to reinstall the
8267 directories you want in the source path. You can add all the
8268 directories in one command.
8272 @section Source and Machine Code
8273 @cindex source line and its code address
8275 You can use the command @code{info line} to map source lines to program
8276 addresses (and vice versa), and the command @code{disassemble} to display
8277 a range of addresses as machine instructions. You can use the command
8278 @code{set disassemble-next-line} to set whether to disassemble next
8279 source line when execution stops. When run under @sc{gnu} Emacs
8280 mode, the @code{info line} command causes the arrow to point to the
8281 line specified. Also, @code{info line} prints addresses in symbolic form as
8286 @item info line @var{location}
8287 Print the starting and ending addresses of the compiled code for
8288 source line @var{location}. You can specify source lines in any of
8289 the ways documented in @ref{Specify Location}.
8292 For example, we can use @code{info line} to discover the location of
8293 the object code for the first line of function
8294 @code{m4_changequote}:
8296 @c FIXME: I think this example should also show the addresses in
8297 @c symbolic form, as they usually would be displayed.
8299 (@value{GDBP}) info line m4_changequote
8300 Line 895 of "builtin.c" starts at pc 0x634c and ends at 0x6350.
8304 @cindex code address and its source line
8305 We can also inquire (using @code{*@var{addr}} as the form for
8306 @var{location}) what source line covers a particular address:
8308 (@value{GDBP}) info line *0x63ff
8309 Line 926 of "builtin.c" starts at pc 0x63e4 and ends at 0x6404.
8312 @cindex @code{$_} and @code{info line}
8313 @cindex @code{x} command, default address
8314 @kindex x@r{(examine), and} info line
8315 After @code{info line}, the default address for the @code{x} command
8316 is changed to the starting address of the line, so that @samp{x/i} is
8317 sufficient to begin examining the machine code (@pxref{Memory,
8318 ,Examining Memory}). Also, this address is saved as the value of the
8319 convenience variable @code{$_} (@pxref{Convenience Vars, ,Convenience
8324 @cindex assembly instructions
8325 @cindex instructions, assembly
8326 @cindex machine instructions
8327 @cindex listing machine instructions
8329 @itemx disassemble /m
8330 @itemx disassemble /s
8331 @itemx disassemble /r
8332 This specialized command dumps a range of memory as machine
8333 instructions. It can also print mixed source+disassembly by specifying
8334 the @code{/m} or @code{/s} modifier and print the raw instructions in hex
8335 as well as in symbolic form by specifying the @code{/r} modifier.
8336 The default memory range is the function surrounding the
8337 program counter of the selected frame. A single argument to this
8338 command is a program counter value; @value{GDBN} dumps the function
8339 surrounding this value. When two arguments are given, they should
8340 be separated by a comma, possibly surrounded by whitespace. The
8341 arguments specify a range of addresses to dump, in one of two forms:
8344 @item @var{start},@var{end}
8345 the addresses from @var{start} (inclusive) to @var{end} (exclusive)
8346 @item @var{start},+@var{length}
8347 the addresses from @var{start} (inclusive) to
8348 @code{@var{start}+@var{length}} (exclusive).
8352 When 2 arguments are specified, the name of the function is also
8353 printed (since there could be several functions in the given range).
8355 The argument(s) can be any expression yielding a numeric value, such as
8356 @samp{0x32c4}, @samp{&main+10} or @samp{$pc - 8}.
8358 If the range of memory being disassembled contains current program counter,
8359 the instruction at that location is shown with a @code{=>} marker.
8362 The following example shows the disassembly of a range of addresses of
8363 HP PA-RISC 2.0 code:
8366 (@value{GDBP}) disas 0x32c4, 0x32e4
8367 Dump of assembler code from 0x32c4 to 0x32e4:
8368 0x32c4 <main+204>: addil 0,dp
8369 0x32c8 <main+208>: ldw 0x22c(sr0,r1),r26
8370 0x32cc <main+212>: ldil 0x3000,r31
8371 0x32d0 <main+216>: ble 0x3f8(sr4,r31)
8372 0x32d4 <main+220>: ldo 0(r31),rp
8373 0x32d8 <main+224>: addil -0x800,dp
8374 0x32dc <main+228>: ldo 0x588(r1),r26
8375 0x32e0 <main+232>: ldil 0x3000,r31
8376 End of assembler dump.
8379 Here is an example showing mixed source+assembly for Intel x86
8380 with @code{/m} or @code{/s}, when the program is stopped just after
8381 function prologue in a non-optimized function with no inline code.
8384 (@value{GDBP}) disas /m main
8385 Dump of assembler code for function main:
8387 0x08048330 <+0>: push %ebp
8388 0x08048331 <+1>: mov %esp,%ebp
8389 0x08048333 <+3>: sub $0x8,%esp
8390 0x08048336 <+6>: and $0xfffffff0,%esp
8391 0x08048339 <+9>: sub $0x10,%esp
8393 6 printf ("Hello.\n");
8394 => 0x0804833c <+12>: movl $0x8048440,(%esp)
8395 0x08048343 <+19>: call 0x8048284 <puts@@plt>
8399 0x08048348 <+24>: mov $0x0,%eax
8400 0x0804834d <+29>: leave
8401 0x0804834e <+30>: ret
8403 End of assembler dump.
8406 The @code{/m} option is deprecated as its output is not useful when
8407 there is either inlined code or re-ordered code.
8408 The @code{/s} option is the preferred choice.
8409 Here is an example for AMD x86-64 showing the difference between
8410 @code{/m} output and @code{/s} output.
8411 This example has one inline function defined in a header file,
8412 and the code is compiled with @samp{-O2} optimization.
8413 Note how the @code{/m} output is missing the disassembly of
8414 several instructions that are present in the @code{/s} output.
8444 (@value{GDBP}) disas /m main
8445 Dump of assembler code for function main:
8449 0x0000000000400400 <+0>: mov 0x200c2e(%rip),%eax # 0x601034 <y>
8450 0x0000000000400417 <+23>: mov %eax,0x200c13(%rip) # 0x601030 <x>
8454 0x000000000040041d <+29>: xor %eax,%eax
8455 0x000000000040041f <+31>: retq
8456 0x0000000000400420 <+32>: add %eax,%eax
8457 0x0000000000400422 <+34>: jmp 0x400417 <main+23>
8459 End of assembler dump.
8460 (@value{GDBP}) disas /s main
8461 Dump of assembler code for function main:
8465 0x0000000000400400 <+0>: mov 0x200c2e(%rip),%eax # 0x601034 <y>
8469 0x0000000000400406 <+6>: test %eax,%eax
8470 0x0000000000400408 <+8>: js 0x400420 <main+32>
8475 0x000000000040040a <+10>: lea 0xa(%rax),%edx
8476 0x000000000040040d <+13>: test %eax,%eax
8477 0x000000000040040f <+15>: mov $0x1,%eax
8478 0x0000000000400414 <+20>: cmovne %edx,%eax
8482 0x0000000000400417 <+23>: mov %eax,0x200c13(%rip) # 0x601030 <x>
8486 0x000000000040041d <+29>: xor %eax,%eax
8487 0x000000000040041f <+31>: retq
8491 0x0000000000400420 <+32>: add %eax,%eax
8492 0x0000000000400422 <+34>: jmp 0x400417 <main+23>
8493 End of assembler dump.
8496 Here is another example showing raw instructions in hex for AMD x86-64,
8499 (gdb) disas /r 0x400281,+10
8500 Dump of assembler code from 0x400281 to 0x40028b:
8501 0x0000000000400281: 38 36 cmp %dh,(%rsi)
8502 0x0000000000400283: 2d 36 34 2e 73 sub $0x732e3436,%eax
8503 0x0000000000400288: 6f outsl %ds:(%rsi),(%dx)
8504 0x0000000000400289: 2e 32 00 xor %cs:(%rax),%al
8505 End of assembler dump.
8508 Addresses cannot be specified as a location (@pxref{Specify Location}).
8509 So, for example, if you want to disassemble function @code{bar}
8510 in file @file{foo.c}, you must type @samp{disassemble 'foo.c'::bar}
8511 and not @samp{disassemble foo.c:bar}.
8513 Some architectures have more than one commonly-used set of instruction
8514 mnemonics or other syntax.
8516 For programs that were dynamically linked and use shared libraries,
8517 instructions that call functions or branch to locations in the shared
8518 libraries might show a seemingly bogus location---it's actually a
8519 location of the relocation table. On some architectures, @value{GDBN}
8520 might be able to resolve these to actual function names.
8523 @kindex set disassembler-options
8524 @cindex disassembler options
8525 @item set disassembler-options @var{option1}[,@var{option2}@dots{}]
8526 This command controls the passing of target specific information to
8527 the disassembler. For a list of valid options, please refer to the
8528 @code{-M}/@code{--disassembler-options} section of the @samp{objdump}
8529 manual and/or the output of @kbd{objdump --help}
8530 (@pxref{objdump,,objdump,binutils.info,The GNU Binary Utilities}).
8531 The default value is the empty string.
8533 If it is necessary to specify more than one disassembler option, then
8534 multiple options can be placed together into a comma separated list.
8535 Currently this command is only supported on targets ARM, PowerPC
8538 @kindex show disassembler-options
8539 @item show disassembler-options
8540 Show the current setting of the disassembler options.
8544 @kindex set disassembly-flavor
8545 @cindex Intel disassembly flavor
8546 @cindex AT&T disassembly flavor
8547 @item set disassembly-flavor @var{instruction-set}
8548 Select the instruction set to use when disassembling the
8549 program via the @code{disassemble} or @code{x/i} commands.
8551 Currently this command is only defined for the Intel x86 family. You
8552 can set @var{instruction-set} to either @code{intel} or @code{att}.
8553 The default is @code{att}, the AT&T flavor used by default by Unix
8554 assemblers for x86-based targets.
8556 @kindex show disassembly-flavor
8557 @item show disassembly-flavor
8558 Show the current setting of the disassembly flavor.
8562 @kindex set disassemble-next-line
8563 @kindex show disassemble-next-line
8564 @item set disassemble-next-line
8565 @itemx show disassemble-next-line
8566 Control whether or not @value{GDBN} will disassemble the next source
8567 line or instruction when execution stops. If ON, @value{GDBN} will
8568 display disassembly of the next source line when execution of the
8569 program being debugged stops. This is @emph{in addition} to
8570 displaying the source line itself, which @value{GDBN} always does if
8571 possible. If the next source line cannot be displayed for some reason
8572 (e.g., if @value{GDBN} cannot find the source file, or there's no line
8573 info in the debug info), @value{GDBN} will display disassembly of the
8574 next @emph{instruction} instead of showing the next source line. If
8575 AUTO, @value{GDBN} will display disassembly of next instruction only
8576 if the source line cannot be displayed. This setting causes
8577 @value{GDBN} to display some feedback when you step through a function
8578 with no line info or whose source file is unavailable. The default is
8579 OFF, which means never display the disassembly of the next line or
8585 @chapter Examining Data
8587 @cindex printing data
8588 @cindex examining data
8591 The usual way to examine data in your program is with the @code{print}
8592 command (abbreviated @code{p}), or its synonym @code{inspect}. It
8593 evaluates and prints the value of an expression of the language your
8594 program is written in (@pxref{Languages, ,Using @value{GDBN} with
8595 Different Languages}). It may also print the expression using a
8596 Python-based pretty-printer (@pxref{Pretty Printing}).
8599 @item print @var{expr}
8600 @itemx print /@var{f} @var{expr}
8601 @var{expr} is an expression (in the source language). By default the
8602 value of @var{expr} is printed in a format appropriate to its data type;
8603 you can choose a different format by specifying @samp{/@var{f}}, where
8604 @var{f} is a letter specifying the format; see @ref{Output Formats,,Output
8608 @itemx print /@var{f}
8609 @cindex reprint the last value
8610 If you omit @var{expr}, @value{GDBN} displays the last value again (from the
8611 @dfn{value history}; @pxref{Value History, ,Value History}). This allows you to
8612 conveniently inspect the same value in an alternative format.
8615 A more low-level way of examining data is with the @code{x} command.
8616 It examines data in memory at a specified address and prints it in a
8617 specified format. @xref{Memory, ,Examining Memory}.
8619 If you are interested in information about types, or about how the
8620 fields of a struct or a class are declared, use the @code{ptype @var{exp}}
8621 command rather than @code{print}. @xref{Symbols, ,Examining the Symbol
8624 @cindex exploring hierarchical data structures
8626 Another way of examining values of expressions and type information is
8627 through the Python extension command @code{explore} (available only if
8628 the @value{GDBN} build is configured with @code{--with-python}). It
8629 offers an interactive way to start at the highest level (or, the most
8630 abstract level) of the data type of an expression (or, the data type
8631 itself) and explore all the way down to leaf scalar values/fields
8632 embedded in the higher level data types.
8635 @item explore @var{arg}
8636 @var{arg} is either an expression (in the source language), or a type
8637 visible in the current context of the program being debugged.
8640 The working of the @code{explore} command can be illustrated with an
8641 example. If a data type @code{struct ComplexStruct} is defined in your
8651 struct ComplexStruct
8653 struct SimpleStruct *ss_p;
8659 followed by variable declarations as
8662 struct SimpleStruct ss = @{ 10, 1.11 @};
8663 struct ComplexStruct cs = @{ &ss, @{ 0, 1, 2, 3, 4, 5, 6, 7, 8, 9 @} @};
8667 then, the value of the variable @code{cs} can be explored using the
8668 @code{explore} command as follows.
8672 The value of `cs' is a struct/class of type `struct ComplexStruct' with
8673 the following fields:
8675 ss_p = <Enter 0 to explore this field of type `struct SimpleStruct *'>
8676 arr = <Enter 1 to explore this field of type `int [10]'>
8678 Enter the field number of choice:
8682 Since the fields of @code{cs} are not scalar values, you are being
8683 prompted to chose the field you want to explore. Let's say you choose
8684 the field @code{ss_p} by entering @code{0}. Then, since this field is a
8685 pointer, you will be asked if it is pointing to a single value. From
8686 the declaration of @code{cs} above, it is indeed pointing to a single
8687 value, hence you enter @code{y}. If you enter @code{n}, then you will
8688 be asked if it were pointing to an array of values, in which case this
8689 field will be explored as if it were an array.
8692 `cs.ss_p' is a pointer to a value of type `struct SimpleStruct'
8693 Continue exploring it as a pointer to a single value [y/n]: y
8694 The value of `*(cs.ss_p)' is a struct/class of type `struct
8695 SimpleStruct' with the following fields:
8697 i = 10 .. (Value of type `int')
8698 d = 1.1100000000000001 .. (Value of type `double')
8700 Press enter to return to parent value:
8704 If the field @code{arr} of @code{cs} was chosen for exploration by
8705 entering @code{1} earlier, then since it is as array, you will be
8706 prompted to enter the index of the element in the array that you want
8710 `cs.arr' is an array of `int'.
8711 Enter the index of the element you want to explore in `cs.arr': 5
8713 `(cs.arr)[5]' is a scalar value of type `int'.
8717 Press enter to return to parent value:
8720 In general, at any stage of exploration, you can go deeper towards the
8721 leaf values by responding to the prompts appropriately, or hit the
8722 return key to return to the enclosing data structure (the @i{higher}
8723 level data structure).
8725 Similar to exploring values, you can use the @code{explore} command to
8726 explore types. Instead of specifying a value (which is typically a
8727 variable name or an expression valid in the current context of the
8728 program being debugged), you specify a type name. If you consider the
8729 same example as above, your can explore the type
8730 @code{struct ComplexStruct} by passing the argument
8731 @code{struct ComplexStruct} to the @code{explore} command.
8734 (gdb) explore struct ComplexStruct
8738 By responding to the prompts appropriately in the subsequent interactive
8739 session, you can explore the type @code{struct ComplexStruct} in a
8740 manner similar to how the value @code{cs} was explored in the above
8743 The @code{explore} command also has two sub-commands,
8744 @code{explore value} and @code{explore type}. The former sub-command is
8745 a way to explicitly specify that value exploration of the argument is
8746 being invoked, while the latter is a way to explicitly specify that type
8747 exploration of the argument is being invoked.
8750 @item explore value @var{expr}
8751 @cindex explore value
8752 This sub-command of @code{explore} explores the value of the
8753 expression @var{expr} (if @var{expr} is an expression valid in the
8754 current context of the program being debugged). The behavior of this
8755 command is identical to that of the behavior of the @code{explore}
8756 command being passed the argument @var{expr}.
8758 @item explore type @var{arg}
8759 @cindex explore type
8760 This sub-command of @code{explore} explores the type of @var{arg} (if
8761 @var{arg} is a type visible in the current context of program being
8762 debugged), or the type of the value/expression @var{arg} (if @var{arg}
8763 is an expression valid in the current context of the program being
8764 debugged). If @var{arg} is a type, then the behavior of this command is
8765 identical to that of the @code{explore} command being passed the
8766 argument @var{arg}. If @var{arg} is an expression, then the behavior of
8767 this command will be identical to that of the @code{explore} command
8768 being passed the type of @var{arg} as the argument.
8772 * Expressions:: Expressions
8773 * Ambiguous Expressions:: Ambiguous Expressions
8774 * Variables:: Program variables
8775 * Arrays:: Artificial arrays
8776 * Output Formats:: Output formats
8777 * Memory:: Examining memory
8778 * Auto Display:: Automatic display
8779 * Print Settings:: Print settings
8780 * Pretty Printing:: Python pretty printing
8781 * Value History:: Value history
8782 * Convenience Vars:: Convenience variables
8783 * Convenience Funs:: Convenience functions
8784 * Registers:: Registers
8785 * Floating Point Hardware:: Floating point hardware
8786 * Vector Unit:: Vector Unit
8787 * OS Information:: Auxiliary data provided by operating system
8788 * Memory Region Attributes:: Memory region attributes
8789 * Dump/Restore Files:: Copy between memory and a file
8790 * Core File Generation:: Cause a program dump its core
8791 * Character Sets:: Debugging programs that use a different
8792 character set than GDB does
8793 * Caching Target Data:: Data caching for targets
8794 * Searching Memory:: Searching memory for a sequence of bytes
8795 * Value Sizes:: Managing memory allocated for values
8799 @section Expressions
8802 @code{print} and many other @value{GDBN} commands accept an expression and
8803 compute its value. Any kind of constant, variable or operator defined
8804 by the programming language you are using is valid in an expression in
8805 @value{GDBN}. This includes conditional expressions, function calls,
8806 casts, and string constants. It also includes preprocessor macros, if
8807 you compiled your program to include this information; see
8810 @cindex arrays in expressions
8811 @value{GDBN} supports array constants in expressions input by
8812 the user. The syntax is @{@var{element}, @var{element}@dots{}@}. For example,
8813 you can use the command @code{print @{1, 2, 3@}} to create an array
8814 of three integers. If you pass an array to a function or assign it
8815 to a program variable, @value{GDBN} copies the array to memory that
8816 is @code{malloc}ed in the target program.
8818 Because C is so widespread, most of the expressions shown in examples in
8819 this manual are in C. @xref{Languages, , Using @value{GDBN} with Different
8820 Languages}, for information on how to use expressions in other
8823 In this section, we discuss operators that you can use in @value{GDBN}
8824 expressions regardless of your programming language.
8826 @cindex casts, in expressions
8827 Casts are supported in all languages, not just in C, because it is so
8828 useful to cast a number into a pointer in order to examine a structure
8829 at that address in memory.
8830 @c FIXME: casts supported---Mod2 true?
8832 @value{GDBN} supports these operators, in addition to those common
8833 to programming languages:
8837 @samp{@@} is a binary operator for treating parts of memory as arrays.
8838 @xref{Arrays, ,Artificial Arrays}, for more information.
8841 @samp{::} allows you to specify a variable in terms of the file or
8842 function where it is defined. @xref{Variables, ,Program Variables}.
8844 @cindex @{@var{type}@}
8845 @cindex type casting memory
8846 @cindex memory, viewing as typed object
8847 @cindex casts, to view memory
8848 @item @{@var{type}@} @var{addr}
8849 Refers to an object of type @var{type} stored at address @var{addr} in
8850 memory. The address @var{addr} may be any expression whose value is
8851 an integer or pointer (but parentheses are required around binary
8852 operators, just as in a cast). This construct is allowed regardless
8853 of what kind of data is normally supposed to reside at @var{addr}.
8856 @node Ambiguous Expressions
8857 @section Ambiguous Expressions
8858 @cindex ambiguous expressions
8860 Expressions can sometimes contain some ambiguous elements. For instance,
8861 some programming languages (notably Ada, C@t{++} and Objective-C) permit
8862 a single function name to be defined several times, for application in
8863 different contexts. This is called @dfn{overloading}. Another example
8864 involving Ada is generics. A @dfn{generic package} is similar to C@t{++}
8865 templates and is typically instantiated several times, resulting in
8866 the same function name being defined in different contexts.
8868 In some cases and depending on the language, it is possible to adjust
8869 the expression to remove the ambiguity. For instance in C@t{++}, you
8870 can specify the signature of the function you want to break on, as in
8871 @kbd{break @var{function}(@var{types})}. In Ada, using the fully
8872 qualified name of your function often makes the expression unambiguous
8875 When an ambiguity that needs to be resolved is detected, the debugger
8876 has the capability to display a menu of numbered choices for each
8877 possibility, and then waits for the selection with the prompt @samp{>}.
8878 The first option is always @samp{[0] cancel}, and typing @kbd{0 @key{RET}}
8879 aborts the current command. If the command in which the expression was
8880 used allows more than one choice to be selected, the next option in the
8881 menu is @samp{[1] all}, and typing @kbd{1 @key{RET}} selects all possible
8884 For example, the following session excerpt shows an attempt to set a
8885 breakpoint at the overloaded symbol @code{String::after}.
8886 We choose three particular definitions of that function name:
8888 @c FIXME! This is likely to change to show arg type lists, at least
8891 (@value{GDBP}) b String::after
8894 [2] file:String.cc; line number:867
8895 [3] file:String.cc; line number:860
8896 [4] file:String.cc; line number:875
8897 [5] file:String.cc; line number:853
8898 [6] file:String.cc; line number:846
8899 [7] file:String.cc; line number:735
8901 Breakpoint 1 at 0xb26c: file String.cc, line 867.
8902 Breakpoint 2 at 0xb344: file String.cc, line 875.
8903 Breakpoint 3 at 0xafcc: file String.cc, line 846.
8904 Multiple breakpoints were set.
8905 Use the "delete" command to delete unwanted
8912 @kindex set multiple-symbols
8913 @item set multiple-symbols @var{mode}
8914 @cindex multiple-symbols menu
8916 This option allows you to adjust the debugger behavior when an expression
8919 By default, @var{mode} is set to @code{all}. If the command with which
8920 the expression is used allows more than one choice, then @value{GDBN}
8921 automatically selects all possible choices. For instance, inserting
8922 a breakpoint on a function using an ambiguous name results in a breakpoint
8923 inserted on each possible match. However, if a unique choice must be made,
8924 then @value{GDBN} uses the menu to help you disambiguate the expression.
8925 For instance, printing the address of an overloaded function will result
8926 in the use of the menu.
8928 When @var{mode} is set to @code{ask}, the debugger always uses the menu
8929 when an ambiguity is detected.
8931 Finally, when @var{mode} is set to @code{cancel}, the debugger reports
8932 an error due to the ambiguity and the command is aborted.
8934 @kindex show multiple-symbols
8935 @item show multiple-symbols
8936 Show the current value of the @code{multiple-symbols} setting.
8940 @section Program Variables
8942 The most common kind of expression to use is the name of a variable
8945 Variables in expressions are understood in the selected stack frame
8946 (@pxref{Selection, ,Selecting a Frame}); they must be either:
8950 global (or file-static)
8957 visible according to the scope rules of the
8958 programming language from the point of execution in that frame
8961 @noindent This means that in the function
8976 you can examine and use the variable @code{a} whenever your program is
8977 executing within the function @code{foo}, but you can only use or
8978 examine the variable @code{b} while your program is executing inside
8979 the block where @code{b} is declared.
8981 @cindex variable name conflict
8982 There is an exception: you can refer to a variable or function whose
8983 scope is a single source file even if the current execution point is not
8984 in this file. But it is possible to have more than one such variable or
8985 function with the same name (in different source files). If that
8986 happens, referring to that name has unpredictable effects. If you wish,
8987 you can specify a static variable in a particular function or file by
8988 using the colon-colon (@code{::}) notation:
8990 @cindex colon-colon, context for variables/functions
8992 @c info cannot cope with a :: index entry, but why deprive hard copy readers?
8993 @cindex @code{::}, context for variables/functions
8996 @var{file}::@var{variable}
8997 @var{function}::@var{variable}
9001 Here @var{file} or @var{function} is the name of the context for the
9002 static @var{variable}. In the case of file names, you can use quotes to
9003 make sure @value{GDBN} parses the file name as a single word---for example,
9004 to print a global value of @code{x} defined in @file{f2.c}:
9007 (@value{GDBP}) p 'f2.c'::x
9010 The @code{::} notation is normally used for referring to
9011 static variables, since you typically disambiguate uses of local variables
9012 in functions by selecting the appropriate frame and using the
9013 simple name of the variable. However, you may also use this notation
9014 to refer to local variables in frames enclosing the selected frame:
9023 process (a); /* Stop here */
9034 For example, if there is a breakpoint at the commented line,
9035 here is what you might see
9036 when the program stops after executing the call @code{bar(0)}:
9041 (@value{GDBP}) p bar::a
9044 #2 0x080483d0 in foo (a=5) at foobar.c:12
9047 (@value{GDBP}) p bar::a
9051 @cindex C@t{++} scope resolution
9052 These uses of @samp{::} are very rarely in conflict with the very
9053 similar use of the same notation in C@t{++}. When they are in
9054 conflict, the C@t{++} meaning takes precedence; however, this can be
9055 overridden by quoting the file or function name with single quotes.
9057 For example, suppose the program is stopped in a method of a class
9058 that has a field named @code{includefile}, and there is also an
9059 include file named @file{includefile} that defines a variable,
9063 (@value{GDBP}) p includefile
9065 (@value{GDBP}) p includefile::some_global
9066 A syntax error in expression, near `'.
9067 (@value{GDBP}) p 'includefile'::some_global
9071 @cindex wrong values
9072 @cindex variable values, wrong
9073 @cindex function entry/exit, wrong values of variables
9074 @cindex optimized code, wrong values of variables
9076 @emph{Warning:} Occasionally, a local variable may appear to have the
9077 wrong value at certain points in a function---just after entry to a new
9078 scope, and just before exit.
9080 You may see this problem when you are stepping by machine instructions.
9081 This is because, on most machines, it takes more than one instruction to
9082 set up a stack frame (including local variable definitions); if you are
9083 stepping by machine instructions, variables may appear to have the wrong
9084 values until the stack frame is completely built. On exit, it usually
9085 also takes more than one machine instruction to destroy a stack frame;
9086 after you begin stepping through that group of instructions, local
9087 variable definitions may be gone.
9089 This may also happen when the compiler does significant optimizations.
9090 To be sure of always seeing accurate values, turn off all optimization
9093 @cindex ``No symbol "foo" in current context''
9094 Another possible effect of compiler optimizations is to optimize
9095 unused variables out of existence, or assign variables to registers (as
9096 opposed to memory addresses). Depending on the support for such cases
9097 offered by the debug info format used by the compiler, @value{GDBN}
9098 might not be able to display values for such local variables. If that
9099 happens, @value{GDBN} will print a message like this:
9102 No symbol "foo" in current context.
9105 To solve such problems, either recompile without optimizations, or use a
9106 different debug info format, if the compiler supports several such
9107 formats. @xref{Compilation}, for more information on choosing compiler
9108 options. @xref{C, ,C and C@t{++}}, for more information about debug
9109 info formats that are best suited to C@t{++} programs.
9111 If you ask to print an object whose contents are unknown to
9112 @value{GDBN}, e.g., because its data type is not completely specified
9113 by the debug information, @value{GDBN} will say @samp{<incomplete
9114 type>}. @xref{Symbols, incomplete type}, for more about this.
9116 If you append @kbd{@@entry} string to a function parameter name you get its
9117 value at the time the function got called. If the value is not available an
9118 error message is printed. Entry values are available only with some compilers.
9119 Entry values are normally also printed at the function parameter list according
9120 to @ref{set print entry-values}.
9123 Breakpoint 1, d (i=30) at gdb.base/entry-value.c:29
9129 (gdb) print i@@entry
9133 Strings are identified as arrays of @code{char} values without specified
9134 signedness. Arrays of either @code{signed char} or @code{unsigned char} get
9135 printed as arrays of 1 byte sized integers. @code{-fsigned-char} or
9136 @code{-funsigned-char} @value{NGCC} options have no effect as @value{GDBN}
9137 defines literal string type @code{"char"} as @code{char} without a sign.
9142 signed char var1[] = "A";
9145 You get during debugging
9150 $2 = @{65 'A', 0 '\0'@}
9154 @section Artificial Arrays
9156 @cindex artificial array
9158 @kindex @@@r{, referencing memory as an array}
9159 It is often useful to print out several successive objects of the
9160 same type in memory; a section of an array, or an array of
9161 dynamically determined size for which only a pointer exists in the
9164 You can do this by referring to a contiguous span of memory as an
9165 @dfn{artificial array}, using the binary operator @samp{@@}. The left
9166 operand of @samp{@@} should be the first element of the desired array
9167 and be an individual object. The right operand should be the desired length
9168 of the array. The result is an array value whose elements are all of
9169 the type of the left argument. The first element is actually the left
9170 argument; the second element comes from bytes of memory immediately
9171 following those that hold the first element, and so on. Here is an
9172 example. If a program says
9175 int *array = (int *) malloc (len * sizeof (int));
9179 you can print the contents of @code{array} with
9185 The left operand of @samp{@@} must reside in memory. Array values made
9186 with @samp{@@} in this way behave just like other arrays in terms of
9187 subscripting, and are coerced to pointers when used in expressions.
9188 Artificial arrays most often appear in expressions via the value history
9189 (@pxref{Value History, ,Value History}), after printing one out.
9191 Another way to create an artificial array is to use a cast.
9192 This re-interprets a value as if it were an array.
9193 The value need not be in memory:
9195 (@value{GDBP}) p/x (short[2])0x12345678
9196 $1 = @{0x1234, 0x5678@}
9199 As a convenience, if you leave the array length out (as in
9200 @samp{(@var{type}[])@var{value}}) @value{GDBN} calculates the size to fill
9201 the value (as @samp{sizeof(@var{value})/sizeof(@var{type})}:
9203 (@value{GDBP}) p/x (short[])0x12345678
9204 $2 = @{0x1234, 0x5678@}
9207 Sometimes the artificial array mechanism is not quite enough; in
9208 moderately complex data structures, the elements of interest may not
9209 actually be adjacent---for example, if you are interested in the values
9210 of pointers in an array. One useful work-around in this situation is
9211 to use a convenience variable (@pxref{Convenience Vars, ,Convenience
9212 Variables}) as a counter in an expression that prints the first
9213 interesting value, and then repeat that expression via @key{RET}. For
9214 instance, suppose you have an array @code{dtab} of pointers to
9215 structures, and you are interested in the values of a field @code{fv}
9216 in each structure. Here is an example of what you might type:
9226 @node Output Formats
9227 @section Output Formats
9229 @cindex formatted output
9230 @cindex output formats
9231 By default, @value{GDBN} prints a value according to its data type. Sometimes
9232 this is not what you want. For example, you might want to print a number
9233 in hex, or a pointer in decimal. Or you might want to view data in memory
9234 at a certain address as a character string or as an instruction. To do
9235 these things, specify an @dfn{output format} when you print a value.
9237 The simplest use of output formats is to say how to print a value
9238 already computed. This is done by starting the arguments of the
9239 @code{print} command with a slash and a format letter. The format
9240 letters supported are:
9244 Regard the bits of the value as an integer, and print the integer in
9248 Print as integer in signed decimal.
9251 Print as integer in unsigned decimal.
9254 Print as integer in octal.
9257 Print as integer in binary. The letter @samp{t} stands for ``two''.
9258 @footnote{@samp{b} cannot be used because these format letters are also
9259 used with the @code{x} command, where @samp{b} stands for ``byte'';
9260 see @ref{Memory,,Examining Memory}.}
9263 @cindex unknown address, locating
9264 @cindex locate address
9265 Print as an address, both absolute in hexadecimal and as an offset from
9266 the nearest preceding symbol. You can use this format used to discover
9267 where (in what function) an unknown address is located:
9270 (@value{GDBP}) p/a 0x54320
9271 $3 = 0x54320 <_initialize_vx+396>
9275 The command @code{info symbol 0x54320} yields similar results.
9276 @xref{Symbols, info symbol}.
9279 Regard as an integer and print it as a character constant. This
9280 prints both the numerical value and its character representation. The
9281 character representation is replaced with the octal escape @samp{\nnn}
9282 for characters outside the 7-bit @sc{ascii} range.
9284 Without this format, @value{GDBN} displays @code{char},
9285 @w{@code{unsigned char}}, and @w{@code{signed char}} data as character
9286 constants. Single-byte members of vectors are displayed as integer
9290 Regard the bits of the value as a floating point number and print
9291 using typical floating point syntax.
9294 @cindex printing strings
9295 @cindex printing byte arrays
9296 Regard as a string, if possible. With this format, pointers to single-byte
9297 data are displayed as null-terminated strings and arrays of single-byte data
9298 are displayed as fixed-length strings. Other values are displayed in their
9301 Without this format, @value{GDBN} displays pointers to and arrays of
9302 @code{char}, @w{@code{unsigned char}}, and @w{@code{signed char}} as
9303 strings. Single-byte members of a vector are displayed as an integer
9307 Like @samp{x} formatting, the value is treated as an integer and
9308 printed as hexadecimal, but leading zeros are printed to pad the value
9309 to the size of the integer type.
9312 @cindex raw printing
9313 Print using the @samp{raw} formatting. By default, @value{GDBN} will
9314 use a Python-based pretty-printer, if one is available (@pxref{Pretty
9315 Printing}). This typically results in a higher-level display of the
9316 value's contents. The @samp{r} format bypasses any Python
9317 pretty-printer which might exist.
9320 For example, to print the program counter in hex (@pxref{Registers}), type
9327 Note that no space is required before the slash; this is because command
9328 names in @value{GDBN} cannot contain a slash.
9330 To reprint the last value in the value history with a different format,
9331 you can use the @code{print} command with just a format and no
9332 expression. For example, @samp{p/x} reprints the last value in hex.
9335 @section Examining Memory
9337 You can use the command @code{x} (for ``examine'') to examine memory in
9338 any of several formats, independently of your program's data types.
9340 @cindex examining memory
9342 @kindex x @r{(examine memory)}
9343 @item x/@var{nfu} @var{addr}
9346 Use the @code{x} command to examine memory.
9349 @var{n}, @var{f}, and @var{u} are all optional parameters that specify how
9350 much memory to display and how to format it; @var{addr} is an
9351 expression giving the address where you want to start displaying memory.
9352 If you use defaults for @var{nfu}, you need not type the slash @samp{/}.
9353 Several commands set convenient defaults for @var{addr}.
9356 @item @var{n}, the repeat count
9357 The repeat count is a decimal integer; the default is 1. It specifies
9358 how much memory (counting by units @var{u}) to display. If a negative
9359 number is specified, memory is examined backward from @var{addr}.
9360 @c This really is **decimal**; unaffected by 'set radix' as of GDB
9363 @item @var{f}, the display format
9364 The display format is one of the formats used by @code{print}
9365 (@samp{x}, @samp{d}, @samp{u}, @samp{o}, @samp{t}, @samp{a}, @samp{c},
9366 @samp{f}, @samp{s}), and in addition @samp{i} (for machine instructions).
9367 The default is @samp{x} (hexadecimal) initially. The default changes
9368 each time you use either @code{x} or @code{print}.
9370 @item @var{u}, the unit size
9371 The unit size is any of
9377 Halfwords (two bytes).
9379 Words (four bytes). This is the initial default.
9381 Giant words (eight bytes).
9384 Each time you specify a unit size with @code{x}, that size becomes the
9385 default unit the next time you use @code{x}. For the @samp{i} format,
9386 the unit size is ignored and is normally not written. For the @samp{s} format,
9387 the unit size defaults to @samp{b}, unless it is explicitly given.
9388 Use @kbd{x /hs} to display 16-bit char strings and @kbd{x /ws} to display
9389 32-bit strings. The next use of @kbd{x /s} will again display 8-bit strings.
9390 Note that the results depend on the programming language of the
9391 current compilation unit. If the language is C, the @samp{s}
9392 modifier will use the UTF-16 encoding while @samp{w} will use
9393 UTF-32. The encoding is set by the programming language and cannot
9396 @item @var{addr}, starting display address
9397 @var{addr} is the address where you want @value{GDBN} to begin displaying
9398 memory. The expression need not have a pointer value (though it may);
9399 it is always interpreted as an integer address of a byte of memory.
9400 @xref{Expressions, ,Expressions}, for more information on expressions. The default for
9401 @var{addr} is usually just after the last address examined---but several
9402 other commands also set the default address: @code{info breakpoints} (to
9403 the address of the last breakpoint listed), @code{info line} (to the
9404 starting address of a line), and @code{print} (if you use it to display
9405 a value from memory).
9408 For example, @samp{x/3uh 0x54320} is a request to display three halfwords
9409 (@code{h}) of memory, formatted as unsigned decimal integers (@samp{u}),
9410 starting at address @code{0x54320}. @samp{x/4xw $sp} prints the four
9411 words (@samp{w}) of memory above the stack pointer (here, @samp{$sp};
9412 @pxref{Registers, ,Registers}) in hexadecimal (@samp{x}).
9414 You can also specify a negative repeat count to examine memory backward
9415 from the given address. For example, @samp{x/-3uh 0x54320} prints three
9416 halfwords (@code{h}) at @code{0x54314}, @code{0x54328}, and @code{0x5431c}.
9418 Since the letters indicating unit sizes are all distinct from the
9419 letters specifying output formats, you do not have to remember whether
9420 unit size or format comes first; either order works. The output
9421 specifications @samp{4xw} and @samp{4wx} mean exactly the same thing.
9422 (However, the count @var{n} must come first; @samp{wx4} does not work.)
9424 Even though the unit size @var{u} is ignored for the formats @samp{s}
9425 and @samp{i}, you might still want to use a count @var{n}; for example,
9426 @samp{3i} specifies that you want to see three machine instructions,
9427 including any operands. For convenience, especially when used with
9428 the @code{display} command, the @samp{i} format also prints branch delay
9429 slot instructions, if any, beyond the count specified, which immediately
9430 follow the last instruction that is within the count. The command
9431 @code{disassemble} gives an alternative way of inspecting machine
9432 instructions; see @ref{Machine Code,,Source and Machine Code}.
9434 If a negative repeat count is specified for the formats @samp{s} or @samp{i},
9435 the command displays null-terminated strings or instructions before the given
9436 address as many as the absolute value of the given number. For the @samp{i}
9437 format, we use line number information in the debug info to accurately locate
9438 instruction boundaries while disassembling backward. If line info is not
9439 available, the command stops examining memory with an error message.
9441 All the defaults for the arguments to @code{x} are designed to make it
9442 easy to continue scanning memory with minimal specifications each time
9443 you use @code{x}. For example, after you have inspected three machine
9444 instructions with @samp{x/3i @var{addr}}, you can inspect the next seven
9445 with just @samp{x/7}. If you use @key{RET} to repeat the @code{x} command,
9446 the repeat count @var{n} is used again; the other arguments default as
9447 for successive uses of @code{x}.
9449 When examining machine instructions, the instruction at current program
9450 counter is shown with a @code{=>} marker. For example:
9453 (@value{GDBP}) x/5i $pc-6
9454 0x804837f <main+11>: mov %esp,%ebp
9455 0x8048381 <main+13>: push %ecx
9456 0x8048382 <main+14>: sub $0x4,%esp
9457 => 0x8048385 <main+17>: movl $0x8048460,(%esp)
9458 0x804838c <main+24>: call 0x80482d4 <puts@@plt>
9461 @cindex @code{$_}, @code{$__}, and value history
9462 The addresses and contents printed by the @code{x} command are not saved
9463 in the value history because there is often too much of them and they
9464 would get in the way. Instead, @value{GDBN} makes these values available for
9465 subsequent use in expressions as values of the convenience variables
9466 @code{$_} and @code{$__}. After an @code{x} command, the last address
9467 examined is available for use in expressions in the convenience variable
9468 @code{$_}. The contents of that address, as examined, are available in
9469 the convenience variable @code{$__}.
9471 If the @code{x} command has a repeat count, the address and contents saved
9472 are from the last memory unit printed; this is not the same as the last
9473 address printed if several units were printed on the last line of output.
9475 @anchor{addressable memory unit}
9476 @cindex addressable memory unit
9477 Most targets have an addressable memory unit size of 8 bits. This means
9478 that to each memory address are associated 8 bits of data. Some
9479 targets, however, have other addressable memory unit sizes.
9480 Within @value{GDBN} and this document, the term
9481 @dfn{addressable memory unit} (or @dfn{memory unit} for short) is used
9482 when explicitly referring to a chunk of data of that size. The word
9483 @dfn{byte} is used to refer to a chunk of data of 8 bits, regardless of
9484 the addressable memory unit size of the target. For most systems,
9485 addressable memory unit is a synonym of byte.
9487 @cindex remote memory comparison
9488 @cindex target memory comparison
9489 @cindex verify remote memory image
9490 @cindex verify target memory image
9491 When you are debugging a program running on a remote target machine
9492 (@pxref{Remote Debugging}), you may wish to verify the program's image
9493 in the remote machine's memory against the executable file you
9494 downloaded to the target. Or, on any target, you may want to check
9495 whether the program has corrupted its own read-only sections. The
9496 @code{compare-sections} command is provided for such situations.
9499 @kindex compare-sections
9500 @item compare-sections @r{[}@var{section-name}@r{|}@code{-r}@r{]}
9501 Compare the data of a loadable section @var{section-name} in the
9502 executable file of the program being debugged with the same section in
9503 the target machine's memory, and report any mismatches. With no
9504 arguments, compares all loadable sections. With an argument of
9505 @code{-r}, compares all loadable read-only sections.
9507 Note: for remote targets, this command can be accelerated if the
9508 target supports computing the CRC checksum of a block of memory
9509 (@pxref{qCRC packet}).
9513 @section Automatic Display
9514 @cindex automatic display
9515 @cindex display of expressions
9517 If you find that you want to print the value of an expression frequently
9518 (to see how it changes), you might want to add it to the @dfn{automatic
9519 display list} so that @value{GDBN} prints its value each time your program stops.
9520 Each expression added to the list is given a number to identify it;
9521 to remove an expression from the list, you specify that number.
9522 The automatic display looks like this:
9526 3: bar[5] = (struct hack *) 0x3804
9530 This display shows item numbers, expressions and their current values. As with
9531 displays you request manually using @code{x} or @code{print}, you can
9532 specify the output format you prefer; in fact, @code{display} decides
9533 whether to use @code{print} or @code{x} depending your format
9534 specification---it uses @code{x} if you specify either the @samp{i}
9535 or @samp{s} format, or a unit size; otherwise it uses @code{print}.
9539 @item display @var{expr}
9540 Add the expression @var{expr} to the list of expressions to display
9541 each time your program stops. @xref{Expressions, ,Expressions}.
9543 @code{display} does not repeat if you press @key{RET} again after using it.
9545 @item display/@var{fmt} @var{expr}
9546 For @var{fmt} specifying only a display format and not a size or
9547 count, add the expression @var{expr} to the auto-display list but
9548 arrange to display it each time in the specified format @var{fmt}.
9549 @xref{Output Formats,,Output Formats}.
9551 @item display/@var{fmt} @var{addr}
9552 For @var{fmt} @samp{i} or @samp{s}, or including a unit-size or a
9553 number of units, add the expression @var{addr} as a memory address to
9554 be examined each time your program stops. Examining means in effect
9555 doing @samp{x/@var{fmt} @var{addr}}. @xref{Memory, ,Examining Memory}.
9558 For example, @samp{display/i $pc} can be helpful, to see the machine
9559 instruction about to be executed each time execution stops (@samp{$pc}
9560 is a common name for the program counter; @pxref{Registers, ,Registers}).
9563 @kindex delete display
9565 @item undisplay @var{dnums}@dots{}
9566 @itemx delete display @var{dnums}@dots{}
9567 Remove items from the list of expressions to display. Specify the
9568 numbers of the displays that you want affected with the command
9569 argument @var{dnums}. It can be a single display number, one of the
9570 numbers shown in the first field of the @samp{info display} display;
9571 or it could be a range of display numbers, as in @code{2-4}.
9573 @code{undisplay} does not repeat if you press @key{RET} after using it.
9574 (Otherwise you would just get the error @samp{No display number @dots{}}.)
9576 @kindex disable display
9577 @item disable display @var{dnums}@dots{}
9578 Disable the display of item numbers @var{dnums}. A disabled display
9579 item is not printed automatically, but is not forgotten. It may be
9580 enabled again later. Specify the numbers of the displays that you
9581 want affected with the command argument @var{dnums}. It can be a
9582 single display number, one of the numbers shown in the first field of
9583 the @samp{info display} display; or it could be a range of display
9584 numbers, as in @code{2-4}.
9586 @kindex enable display
9587 @item enable display @var{dnums}@dots{}
9588 Enable display of item numbers @var{dnums}. It becomes effective once
9589 again in auto display of its expression, until you specify otherwise.
9590 Specify the numbers of the displays that you want affected with the
9591 command argument @var{dnums}. It can be a single display number, one
9592 of the numbers shown in the first field of the @samp{info display}
9593 display; or it could be a range of display numbers, as in @code{2-4}.
9596 Display the current values of the expressions on the list, just as is
9597 done when your program stops.
9599 @kindex info display
9601 Print the list of expressions previously set up to display
9602 automatically, each one with its item number, but without showing the
9603 values. This includes disabled expressions, which are marked as such.
9604 It also includes expressions which would not be displayed right now
9605 because they refer to automatic variables not currently available.
9608 @cindex display disabled out of scope
9609 If a display expression refers to local variables, then it does not make
9610 sense outside the lexical context for which it was set up. Such an
9611 expression is disabled when execution enters a context where one of its
9612 variables is not defined. For example, if you give the command
9613 @code{display last_char} while inside a function with an argument
9614 @code{last_char}, @value{GDBN} displays this argument while your program
9615 continues to stop inside that function. When it stops elsewhere---where
9616 there is no variable @code{last_char}---the display is disabled
9617 automatically. The next time your program stops where @code{last_char}
9618 is meaningful, you can enable the display expression once again.
9620 @node Print Settings
9621 @section Print Settings
9623 @cindex format options
9624 @cindex print settings
9625 @value{GDBN} provides the following ways to control how arrays, structures,
9626 and symbols are printed.
9629 These settings are useful for debugging programs in any language:
9633 @item set print address
9634 @itemx set print address on
9635 @cindex print/don't print memory addresses
9636 @value{GDBN} prints memory addresses showing the location of stack
9637 traces, structure values, pointer values, breakpoints, and so forth,
9638 even when it also displays the contents of those addresses. The default
9639 is @code{on}. For example, this is what a stack frame display looks like with
9640 @code{set print address on}:
9645 #0 set_quotes (lq=0x34c78 "<<", rq=0x34c88 ">>")
9647 530 if (lquote != def_lquote)
9651 @item set print address off
9652 Do not print addresses when displaying their contents. For example,
9653 this is the same stack frame displayed with @code{set print address off}:
9657 (@value{GDBP}) set print addr off
9659 #0 set_quotes (lq="<<", rq=">>") at input.c:530
9660 530 if (lquote != def_lquote)
9664 You can use @samp{set print address off} to eliminate all machine
9665 dependent displays from the @value{GDBN} interface. For example, with
9666 @code{print address off}, you should get the same text for backtraces on
9667 all machines---whether or not they involve pointer arguments.
9670 @item show print address
9671 Show whether or not addresses are to be printed.
9674 When @value{GDBN} prints a symbolic address, it normally prints the
9675 closest earlier symbol plus an offset. If that symbol does not uniquely
9676 identify the address (for example, it is a name whose scope is a single
9677 source file), you may need to clarify. One way to do this is with
9678 @code{info line}, for example @samp{info line *0x4537}. Alternately,
9679 you can set @value{GDBN} to print the source file and line number when
9680 it prints a symbolic address:
9683 @item set print symbol-filename on
9684 @cindex source file and line of a symbol
9685 @cindex symbol, source file and line
9686 Tell @value{GDBN} to print the source file name and line number of a
9687 symbol in the symbolic form of an address.
9689 @item set print symbol-filename off
9690 Do not print source file name and line number of a symbol. This is the
9693 @item show print symbol-filename
9694 Show whether or not @value{GDBN} will print the source file name and
9695 line number of a symbol in the symbolic form of an address.
9698 Another situation where it is helpful to show symbol filenames and line
9699 numbers is when disassembling code; @value{GDBN} shows you the line
9700 number and source file that corresponds to each instruction.
9702 Also, you may wish to see the symbolic form only if the address being
9703 printed is reasonably close to the closest earlier symbol:
9706 @item set print max-symbolic-offset @var{max-offset}
9707 @itemx set print max-symbolic-offset unlimited
9708 @cindex maximum value for offset of closest symbol
9709 Tell @value{GDBN} to only display the symbolic form of an address if the
9710 offset between the closest earlier symbol and the address is less than
9711 @var{max-offset}. The default is @code{unlimited}, which tells @value{GDBN}
9712 to always print the symbolic form of an address if any symbol precedes
9713 it. Zero is equivalent to @code{unlimited}.
9715 @item show print max-symbolic-offset
9716 Ask how large the maximum offset is that @value{GDBN} prints in a
9720 @cindex wild pointer, interpreting
9721 @cindex pointer, finding referent
9722 If you have a pointer and you are not sure where it points, try
9723 @samp{set print symbol-filename on}. Then you can determine the name
9724 and source file location of the variable where it points, using
9725 @samp{p/a @var{pointer}}. This interprets the address in symbolic form.
9726 For example, here @value{GDBN} shows that a variable @code{ptt} points
9727 at another variable @code{t}, defined in @file{hi2.c}:
9730 (@value{GDBP}) set print symbol-filename on
9731 (@value{GDBP}) p/a ptt
9732 $4 = 0xe008 <t in hi2.c>
9736 @emph{Warning:} For pointers that point to a local variable, @samp{p/a}
9737 does not show the symbol name and filename of the referent, even with
9738 the appropriate @code{set print} options turned on.
9741 You can also enable @samp{/a}-like formatting all the time using
9742 @samp{set print symbol on}:
9745 @item set print symbol on
9746 Tell @value{GDBN} to print the symbol corresponding to an address, if
9749 @item set print symbol off
9750 Tell @value{GDBN} not to print the symbol corresponding to an
9751 address. In this mode, @value{GDBN} will still print the symbol
9752 corresponding to pointers to functions. This is the default.
9754 @item show print symbol
9755 Show whether @value{GDBN} will display the symbol corresponding to an
9759 Other settings control how different kinds of objects are printed:
9762 @item set print array
9763 @itemx set print array on
9764 @cindex pretty print arrays
9765 Pretty print arrays. This format is more convenient to read,
9766 but uses more space. The default is off.
9768 @item set print array off
9769 Return to compressed format for arrays.
9771 @item show print array
9772 Show whether compressed or pretty format is selected for displaying
9775 @cindex print array indexes
9776 @item set print array-indexes
9777 @itemx set print array-indexes on
9778 Print the index of each element when displaying arrays. May be more
9779 convenient to locate a given element in the array or quickly find the
9780 index of a given element in that printed array. The default is off.
9782 @item set print array-indexes off
9783 Stop printing element indexes when displaying arrays.
9785 @item show print array-indexes
9786 Show whether the index of each element is printed when displaying
9789 @item set print elements @var{number-of-elements}
9790 @itemx set print elements unlimited
9791 @cindex number of array elements to print
9792 @cindex limit on number of printed array elements
9793 Set a limit on how many elements of an array @value{GDBN} will print.
9794 If @value{GDBN} is printing a large array, it stops printing after it has
9795 printed the number of elements set by the @code{set print elements} command.
9796 This limit also applies to the display of strings.
9797 When @value{GDBN} starts, this limit is set to 200.
9798 Setting @var{number-of-elements} to @code{unlimited} or zero means
9799 that the number of elements to print is unlimited.
9801 @item show print elements
9802 Display the number of elements of a large array that @value{GDBN} will print.
9803 If the number is 0, then the printing is unlimited.
9805 @item set print frame-arguments @var{value}
9806 @kindex set print frame-arguments
9807 @cindex printing frame argument values
9808 @cindex print all frame argument values
9809 @cindex print frame argument values for scalars only
9810 @cindex do not print frame argument values
9811 This command allows to control how the values of arguments are printed
9812 when the debugger prints a frame (@pxref{Frames}). The possible
9817 The values of all arguments are printed.
9820 Print the value of an argument only if it is a scalar. The value of more
9821 complex arguments such as arrays, structures, unions, etc, is replaced
9822 by @code{@dots{}}. This is the default. Here is an example where
9823 only scalar arguments are shown:
9826 #1 0x08048361 in call_me (i=3, s=@dots{}, ss=0xbf8d508c, u=@dots{}, e=green)
9831 None of the argument values are printed. Instead, the value of each argument
9832 is replaced by @code{@dots{}}. In this case, the example above now becomes:
9835 #1 0x08048361 in call_me (i=@dots{}, s=@dots{}, ss=@dots{}, u=@dots{}, e=@dots{})
9840 By default, only scalar arguments are printed. This command can be used
9841 to configure the debugger to print the value of all arguments, regardless
9842 of their type. However, it is often advantageous to not print the value
9843 of more complex parameters. For instance, it reduces the amount of
9844 information printed in each frame, making the backtrace more readable.
9845 Also, it improves performance when displaying Ada frames, because
9846 the computation of large arguments can sometimes be CPU-intensive,
9847 especially in large applications. Setting @code{print frame-arguments}
9848 to @code{scalars} (the default) or @code{none} avoids this computation,
9849 thus speeding up the display of each Ada frame.
9851 @item show print frame-arguments
9852 Show how the value of arguments should be displayed when printing a frame.
9854 @item set print raw frame-arguments on
9855 Print frame arguments in raw, non pretty-printed, form.
9857 @item set print raw frame-arguments off
9858 Print frame arguments in pretty-printed form, if there is a pretty-printer
9859 for the value (@pxref{Pretty Printing}),
9860 otherwise print the value in raw form.
9861 This is the default.
9863 @item show print raw frame-arguments
9864 Show whether to print frame arguments in raw form.
9866 @anchor{set print entry-values}
9867 @item set print entry-values @var{value}
9868 @kindex set print entry-values
9869 Set printing of frame argument values at function entry. In some cases
9870 @value{GDBN} can determine the value of function argument which was passed by
9871 the function caller, even if the value was modified inside the called function
9872 and therefore is different. With optimized code, the current value could be
9873 unavailable, but the entry value may still be known.
9875 The default value is @code{default} (see below for its description). Older
9876 @value{GDBN} behaved as with the setting @code{no}. Compilers not supporting
9877 this feature will behave in the @code{default} setting the same way as with the
9880 This functionality is currently supported only by DWARF 2 debugging format and
9881 the compiler has to produce @samp{DW_TAG_call_site} tags. With
9882 @value{NGCC}, you need to specify @option{-O -g} during compilation, to get
9885 The @var{value} parameter can be one of the following:
9889 Print only actual parameter values, never print values from function entry
9893 #0 different (val=6)
9894 #0 lost (val=<optimized out>)
9896 #0 invalid (val=<optimized out>)
9900 Print only parameter values from function entry point. The actual parameter
9901 values are never printed.
9903 #0 equal (val@@entry=5)
9904 #0 different (val@@entry=5)
9905 #0 lost (val@@entry=5)
9906 #0 born (val@@entry=<optimized out>)
9907 #0 invalid (val@@entry=<optimized out>)
9911 Print only parameter values from function entry point. If value from function
9912 entry point is not known while the actual value is known, print the actual
9913 value for such parameter.
9915 #0 equal (val@@entry=5)
9916 #0 different (val@@entry=5)
9917 #0 lost (val@@entry=5)
9919 #0 invalid (val@@entry=<optimized out>)
9923 Print actual parameter values. If actual parameter value is not known while
9924 value from function entry point is known, print the entry point value for such
9928 #0 different (val=6)
9929 #0 lost (val@@entry=5)
9931 #0 invalid (val=<optimized out>)
9935 Always print both the actual parameter value and its value from function entry
9936 point, even if values of one or both are not available due to compiler
9939 #0 equal (val=5, val@@entry=5)
9940 #0 different (val=6, val@@entry=5)
9941 #0 lost (val=<optimized out>, val@@entry=5)
9942 #0 born (val=10, val@@entry=<optimized out>)
9943 #0 invalid (val=<optimized out>, val@@entry=<optimized out>)
9947 Print the actual parameter value if it is known and also its value from
9948 function entry point if it is known. If neither is known, print for the actual
9949 value @code{<optimized out>}. If not in MI mode (@pxref{GDB/MI}) and if both
9950 values are known and identical, print the shortened
9951 @code{param=param@@entry=VALUE} notation.
9953 #0 equal (val=val@@entry=5)
9954 #0 different (val=6, val@@entry=5)
9955 #0 lost (val@@entry=5)
9957 #0 invalid (val=<optimized out>)
9961 Always print the actual parameter value. Print also its value from function
9962 entry point, but only if it is known. If not in MI mode (@pxref{GDB/MI}) and
9963 if both values are known and identical, print the shortened
9964 @code{param=param@@entry=VALUE} notation.
9966 #0 equal (val=val@@entry=5)
9967 #0 different (val=6, val@@entry=5)
9968 #0 lost (val=<optimized out>, val@@entry=5)
9970 #0 invalid (val=<optimized out>)
9974 For analysis messages on possible failures of frame argument values at function
9975 entry resolution see @ref{set debug entry-values}.
9977 @item show print entry-values
9978 Show the method being used for printing of frame argument values at function
9981 @item set print repeats @var{number-of-repeats}
9982 @itemx set print repeats unlimited
9983 @cindex repeated array elements
9984 Set the threshold for suppressing display of repeated array
9985 elements. When the number of consecutive identical elements of an
9986 array exceeds the threshold, @value{GDBN} prints the string
9987 @code{"<repeats @var{n} times>"}, where @var{n} is the number of
9988 identical repetitions, instead of displaying the identical elements
9989 themselves. Setting the threshold to @code{unlimited} or zero will
9990 cause all elements to be individually printed. The default threshold
9993 @item show print repeats
9994 Display the current threshold for printing repeated identical
9997 @item set print null-stop
9998 @cindex @sc{null} elements in arrays
9999 Cause @value{GDBN} to stop printing the characters of an array when the first
10000 @sc{null} is encountered. This is useful when large arrays actually
10001 contain only short strings.
10002 The default is off.
10004 @item show print null-stop
10005 Show whether @value{GDBN} stops printing an array on the first
10006 @sc{null} character.
10008 @item set print pretty on
10009 @cindex print structures in indented form
10010 @cindex indentation in structure display
10011 Cause @value{GDBN} to print structures in an indented format with one member
10012 per line, like this:
10027 @item set print pretty off
10028 Cause @value{GDBN} to print structures in a compact format, like this:
10032 $1 = @{next = 0x0, flags = @{sweet = 1, sour = 1@}, \
10033 meat = 0x54 "Pork"@}
10038 This is the default format.
10040 @item show print pretty
10041 Show which format @value{GDBN} is using to print structures.
10043 @item set print sevenbit-strings on
10044 @cindex eight-bit characters in strings
10045 @cindex octal escapes in strings
10046 Print using only seven-bit characters; if this option is set,
10047 @value{GDBN} displays any eight-bit characters (in strings or
10048 character values) using the notation @code{\}@var{nnn}. This setting is
10049 best if you are working in English (@sc{ascii}) and you use the
10050 high-order bit of characters as a marker or ``meta'' bit.
10052 @item set print sevenbit-strings off
10053 Print full eight-bit characters. This allows the use of more
10054 international character sets, and is the default.
10056 @item show print sevenbit-strings
10057 Show whether or not @value{GDBN} is printing only seven-bit characters.
10059 @item set print union on
10060 @cindex unions in structures, printing
10061 Tell @value{GDBN} to print unions which are contained in structures
10062 and other unions. This is the default setting.
10064 @item set print union off
10065 Tell @value{GDBN} not to print unions which are contained in
10066 structures and other unions. @value{GDBN} will print @code{"@{...@}"}
10069 @item show print union
10070 Ask @value{GDBN} whether or not it will print unions which are contained in
10071 structures and other unions.
10073 For example, given the declarations
10076 typedef enum @{Tree, Bug@} Species;
10077 typedef enum @{Big_tree, Acorn, Seedling@} Tree_forms;
10078 typedef enum @{Caterpillar, Cocoon, Butterfly@}
10089 struct thing foo = @{Tree, @{Acorn@}@};
10093 with @code{set print union on} in effect @samp{p foo} would print
10096 $1 = @{it = Tree, form = @{tree = Acorn, bug = Cocoon@}@}
10100 and with @code{set print union off} in effect it would print
10103 $1 = @{it = Tree, form = @{...@}@}
10107 @code{set print union} affects programs written in C-like languages
10113 These settings are of interest when debugging C@t{++} programs:
10116 @cindex demangling C@t{++} names
10117 @item set print demangle
10118 @itemx set print demangle on
10119 Print C@t{++} names in their source form rather than in the encoded
10120 (``mangled'') form passed to the assembler and linker for type-safe
10121 linkage. The default is on.
10123 @item show print demangle
10124 Show whether C@t{++} names are printed in mangled or demangled form.
10126 @item set print asm-demangle
10127 @itemx set print asm-demangle on
10128 Print C@t{++} names in their source form rather than their mangled form, even
10129 in assembler code printouts such as instruction disassemblies.
10130 The default is off.
10132 @item show print asm-demangle
10133 Show whether C@t{++} names in assembly listings are printed in mangled
10136 @cindex C@t{++} symbol decoding style
10137 @cindex symbol decoding style, C@t{++}
10138 @kindex set demangle-style
10139 @item set demangle-style @var{style}
10140 Choose among several encoding schemes used by different compilers to
10141 represent C@t{++} names. The choices for @var{style} are currently:
10145 Allow @value{GDBN} to choose a decoding style by inspecting your program.
10146 This is the default.
10149 Decode based on the @sc{gnu} C@t{++} compiler (@code{g++}) encoding algorithm.
10152 Decode based on the HP ANSI C@t{++} (@code{aCC}) encoding algorithm.
10155 Decode based on the Lucid C@t{++} compiler (@code{lcc}) encoding algorithm.
10158 Decode using the algorithm in the @cite{C@t{++} Annotated Reference Manual}.
10159 @strong{Warning:} this setting alone is not sufficient to allow
10160 debugging @code{cfront}-generated executables. @value{GDBN} would
10161 require further enhancement to permit that.
10164 If you omit @var{style}, you will see a list of possible formats.
10166 @item show demangle-style
10167 Display the encoding style currently in use for decoding C@t{++} symbols.
10169 @item set print object
10170 @itemx set print object on
10171 @cindex derived type of an object, printing
10172 @cindex display derived types
10173 When displaying a pointer to an object, identify the @emph{actual}
10174 (derived) type of the object rather than the @emph{declared} type, using
10175 the virtual function table. Note that the virtual function table is
10176 required---this feature can only work for objects that have run-time
10177 type identification; a single virtual method in the object's declared
10178 type is sufficient. Note that this setting is also taken into account when
10179 working with variable objects via MI (@pxref{GDB/MI}).
10181 @item set print object off
10182 Display only the declared type of objects, without reference to the
10183 virtual function table. This is the default setting.
10185 @item show print object
10186 Show whether actual, or declared, object types are displayed.
10188 @item set print static-members
10189 @itemx set print static-members on
10190 @cindex static members of C@t{++} objects
10191 Print static members when displaying a C@t{++} object. The default is on.
10193 @item set print static-members off
10194 Do not print static members when displaying a C@t{++} object.
10196 @item show print static-members
10197 Show whether C@t{++} static members are printed or not.
10199 @item set print pascal_static-members
10200 @itemx set print pascal_static-members on
10201 @cindex static members of Pascal objects
10202 @cindex Pascal objects, static members display
10203 Print static members when displaying a Pascal object. The default is on.
10205 @item set print pascal_static-members off
10206 Do not print static members when displaying a Pascal object.
10208 @item show print pascal_static-members
10209 Show whether Pascal static members are printed or not.
10211 @c These don't work with HP ANSI C++ yet.
10212 @item set print vtbl
10213 @itemx set print vtbl on
10214 @cindex pretty print C@t{++} virtual function tables
10215 @cindex virtual functions (C@t{++}) display
10216 @cindex VTBL display
10217 Pretty print C@t{++} virtual function tables. The default is off.
10218 (The @code{vtbl} commands do not work on programs compiled with the HP
10219 ANSI C@t{++} compiler (@code{aCC}).)
10221 @item set print vtbl off
10222 Do not pretty print C@t{++} virtual function tables.
10224 @item show print vtbl
10225 Show whether C@t{++} virtual function tables are pretty printed, or not.
10228 @node Pretty Printing
10229 @section Pretty Printing
10231 @value{GDBN} provides a mechanism to allow pretty-printing of values using
10232 Python code. It greatly simplifies the display of complex objects. This
10233 mechanism works for both MI and the CLI.
10236 * Pretty-Printer Introduction:: Introduction to pretty-printers
10237 * Pretty-Printer Example:: An example pretty-printer
10238 * Pretty-Printer Commands:: Pretty-printer commands
10241 @node Pretty-Printer Introduction
10242 @subsection Pretty-Printer Introduction
10244 When @value{GDBN} prints a value, it first sees if there is a pretty-printer
10245 registered for the value. If there is then @value{GDBN} invokes the
10246 pretty-printer to print the value. Otherwise the value is printed normally.
10248 Pretty-printers are normally named. This makes them easy to manage.
10249 The @samp{info pretty-printer} command will list all the installed
10250 pretty-printers with their names.
10251 If a pretty-printer can handle multiple data types, then its
10252 @dfn{subprinters} are the printers for the individual data types.
10253 Each such subprinter has its own name.
10254 The format of the name is @var{printer-name};@var{subprinter-name}.
10256 Pretty-printers are installed by @dfn{registering} them with @value{GDBN}.
10257 Typically they are automatically loaded and registered when the corresponding
10258 debug information is loaded, thus making them available without having to
10259 do anything special.
10261 There are three places where a pretty-printer can be registered.
10265 Pretty-printers registered globally are available when debugging
10269 Pretty-printers registered with a program space are available only
10270 when debugging that program.
10271 @xref{Progspaces In Python}, for more details on program spaces in Python.
10274 Pretty-printers registered with an objfile are loaded and unloaded
10275 with the corresponding objfile (e.g., shared library).
10276 @xref{Objfiles In Python}, for more details on objfiles in Python.
10279 @xref{Selecting Pretty-Printers}, for further information on how
10280 pretty-printers are selected,
10282 @xref{Writing a Pretty-Printer}, for implementing pretty printers
10285 @node Pretty-Printer Example
10286 @subsection Pretty-Printer Example
10288 Here is how a C@t{++} @code{std::string} looks without a pretty-printer:
10291 (@value{GDBP}) print s
10293 static npos = 4294967295,
10295 <std::allocator<char>> = @{
10296 <__gnu_cxx::new_allocator<char>> = @{
10297 <No data fields>@}, <No data fields>
10299 members of std::basic_string<char, std::char_traits<char>,
10300 std::allocator<char> >::_Alloc_hider:
10301 _M_p = 0x804a014 "abcd"
10306 With a pretty-printer for @code{std::string} only the contents are printed:
10309 (@value{GDBP}) print s
10313 @node Pretty-Printer Commands
10314 @subsection Pretty-Printer Commands
10315 @cindex pretty-printer commands
10318 @kindex info pretty-printer
10319 @item info pretty-printer [@var{object-regexp} [@var{name-regexp}]]
10320 Print the list of installed pretty-printers.
10321 This includes disabled pretty-printers, which are marked as such.
10323 @var{object-regexp} is a regular expression matching the objects
10324 whose pretty-printers to list.
10325 Objects can be @code{global}, the program space's file
10326 (@pxref{Progspaces In Python}),
10327 and the object files within that program space (@pxref{Objfiles In Python}).
10328 @xref{Selecting Pretty-Printers}, for details on how @value{GDBN}
10329 looks up a printer from these three objects.
10331 @var{name-regexp} is a regular expression matching the name of the printers
10334 @kindex disable pretty-printer
10335 @item disable pretty-printer [@var{object-regexp} [@var{name-regexp}]]
10336 Disable pretty-printers matching @var{object-regexp} and @var{name-regexp}.
10337 A disabled pretty-printer is not forgotten, it may be enabled again later.
10339 @kindex enable pretty-printer
10340 @item enable pretty-printer [@var{object-regexp} [@var{name-regexp}]]
10341 Enable pretty-printers matching @var{object-regexp} and @var{name-regexp}.
10346 Suppose we have three pretty-printers installed: one from library1.so
10347 named @code{foo} that prints objects of type @code{foo}, and
10348 another from library2.so named @code{bar} that prints two types of objects,
10349 @code{bar1} and @code{bar2}.
10352 (gdb) info pretty-printer
10359 (gdb) info pretty-printer library2
10364 (gdb) disable pretty-printer library1
10366 2 of 3 printers enabled
10367 (gdb) info pretty-printer
10374 (gdb) disable pretty-printer library2 bar:bar1
10376 1 of 3 printers enabled
10377 (gdb) info pretty-printer library2
10384 (gdb) disable pretty-printer library2 bar
10386 0 of 3 printers enabled
10387 (gdb) info pretty-printer library2
10396 Note that for @code{bar} the entire printer can be disabled,
10397 as can each individual subprinter.
10399 @node Value History
10400 @section Value History
10402 @cindex value history
10403 @cindex history of values printed by @value{GDBN}
10404 Values printed by the @code{print} command are saved in the @value{GDBN}
10405 @dfn{value history}. This allows you to refer to them in other expressions.
10406 Values are kept until the symbol table is re-read or discarded
10407 (for example with the @code{file} or @code{symbol-file} commands).
10408 When the symbol table changes, the value history is discarded,
10409 since the values may contain pointers back to the types defined in the
10414 @cindex history number
10415 The values printed are given @dfn{history numbers} by which you can
10416 refer to them. These are successive integers starting with one.
10417 @code{print} shows you the history number assigned to a value by
10418 printing @samp{$@var{num} = } before the value; here @var{num} is the
10421 To refer to any previous value, use @samp{$} followed by the value's
10422 history number. The way @code{print} labels its output is designed to
10423 remind you of this. Just @code{$} refers to the most recent value in
10424 the history, and @code{$$} refers to the value before that.
10425 @code{$$@var{n}} refers to the @var{n}th value from the end; @code{$$2}
10426 is the value just prior to @code{$$}, @code{$$1} is equivalent to
10427 @code{$$}, and @code{$$0} is equivalent to @code{$}.
10429 For example, suppose you have just printed a pointer to a structure and
10430 want to see the contents of the structure. It suffices to type
10436 If you have a chain of structures where the component @code{next} points
10437 to the next one, you can print the contents of the next one with this:
10444 You can print successive links in the chain by repeating this
10445 command---which you can do by just typing @key{RET}.
10447 Note that the history records values, not expressions. If the value of
10448 @code{x} is 4 and you type these commands:
10456 then the value recorded in the value history by the @code{print} command
10457 remains 4 even though the value of @code{x} has changed.
10460 @kindex show values
10462 Print the last ten values in the value history, with their item numbers.
10463 This is like @samp{p@ $$9} repeated ten times, except that @code{show
10464 values} does not change the history.
10466 @item show values @var{n}
10467 Print ten history values centered on history item number @var{n}.
10469 @item show values +
10470 Print ten history values just after the values last printed. If no more
10471 values are available, @code{show values +} produces no display.
10474 Pressing @key{RET} to repeat @code{show values @var{n}} has exactly the
10475 same effect as @samp{show values +}.
10477 @node Convenience Vars
10478 @section Convenience Variables
10480 @cindex convenience variables
10481 @cindex user-defined variables
10482 @value{GDBN} provides @dfn{convenience variables} that you can use within
10483 @value{GDBN} to hold on to a value and refer to it later. These variables
10484 exist entirely within @value{GDBN}; they are not part of your program, and
10485 setting a convenience variable has no direct effect on further execution
10486 of your program. That is why you can use them freely.
10488 Convenience variables are prefixed with @samp{$}. Any name preceded by
10489 @samp{$} can be used for a convenience variable, unless it is one of
10490 the predefined machine-specific register names (@pxref{Registers, ,Registers}).
10491 (Value history references, in contrast, are @emph{numbers} preceded
10492 by @samp{$}. @xref{Value History, ,Value History}.)
10494 You can save a value in a convenience variable with an assignment
10495 expression, just as you would set a variable in your program.
10499 set $foo = *object_ptr
10503 would save in @code{$foo} the value contained in the object pointed to by
10506 Using a convenience variable for the first time creates it, but its
10507 value is @code{void} until you assign a new value. You can alter the
10508 value with another assignment at any time.
10510 Convenience variables have no fixed types. You can assign a convenience
10511 variable any type of value, including structures and arrays, even if
10512 that variable already has a value of a different type. The convenience
10513 variable, when used as an expression, has the type of its current value.
10516 @kindex show convenience
10517 @cindex show all user variables and functions
10518 @item show convenience
10519 Print a list of convenience variables used so far, and their values,
10520 as well as a list of the convenience functions.
10521 Abbreviated @code{show conv}.
10523 @kindex init-if-undefined
10524 @cindex convenience variables, initializing
10525 @item init-if-undefined $@var{variable} = @var{expression}
10526 Set a convenience variable if it has not already been set. This is useful
10527 for user-defined commands that keep some state. It is similar, in concept,
10528 to using local static variables with initializers in C (except that
10529 convenience variables are global). It can also be used to allow users to
10530 override default values used in a command script.
10532 If the variable is already defined then the expression is not evaluated so
10533 any side-effects do not occur.
10536 One of the ways to use a convenience variable is as a counter to be
10537 incremented or a pointer to be advanced. For example, to print
10538 a field from successive elements of an array of structures:
10542 print bar[$i++]->contents
10546 Repeat that command by typing @key{RET}.
10548 Some convenience variables are created automatically by @value{GDBN} and given
10549 values likely to be useful.
10552 @vindex $_@r{, convenience variable}
10554 The variable @code{$_} is automatically set by the @code{x} command to
10555 the last address examined (@pxref{Memory, ,Examining Memory}). Other
10556 commands which provide a default address for @code{x} to examine also
10557 set @code{$_} to that address; these commands include @code{info line}
10558 and @code{info breakpoint}. The type of @code{$_} is @code{void *}
10559 except when set by the @code{x} command, in which case it is a pointer
10560 to the type of @code{$__}.
10562 @vindex $__@r{, convenience variable}
10564 The variable @code{$__} is automatically set by the @code{x} command
10565 to the value found in the last address examined. Its type is chosen
10566 to match the format in which the data was printed.
10569 @vindex $_exitcode@r{, convenience variable}
10570 When the program being debugged terminates normally, @value{GDBN}
10571 automatically sets this variable to the exit code of the program, and
10572 resets @code{$_exitsignal} to @code{void}.
10575 @vindex $_exitsignal@r{, convenience variable}
10576 When the program being debugged dies due to an uncaught signal,
10577 @value{GDBN} automatically sets this variable to that signal's number,
10578 and resets @code{$_exitcode} to @code{void}.
10580 To distinguish between whether the program being debugged has exited
10581 (i.e., @code{$_exitcode} is not @code{void}) or signalled (i.e.,
10582 @code{$_exitsignal} is not @code{void}), the convenience function
10583 @code{$_isvoid} can be used (@pxref{Convenience Funs,, Convenience
10584 Functions}). For example, considering the following source code:
10587 #include <signal.h>
10590 main (int argc, char *argv[])
10597 A valid way of telling whether the program being debugged has exited
10598 or signalled would be:
10601 (@value{GDBP}) define has_exited_or_signalled
10602 Type commands for definition of ``has_exited_or_signalled''.
10603 End with a line saying just ``end''.
10604 >if $_isvoid ($_exitsignal)
10605 >echo The program has exited\n
10607 >echo The program has signalled\n
10613 Program terminated with signal SIGALRM, Alarm clock.
10614 The program no longer exists.
10615 (@value{GDBP}) has_exited_or_signalled
10616 The program has signalled
10619 As can be seen, @value{GDBN} correctly informs that the program being
10620 debugged has signalled, since it calls @code{raise} and raises a
10621 @code{SIGALRM} signal. If the program being debugged had not called
10622 @code{raise}, then @value{GDBN} would report a normal exit:
10625 (@value{GDBP}) has_exited_or_signalled
10626 The program has exited
10630 The variable @code{$_exception} is set to the exception object being
10631 thrown at an exception-related catchpoint. @xref{Set Catchpoints}.
10634 @itemx $_probe_arg0@dots{}$_probe_arg11
10635 Arguments to a static probe. @xref{Static Probe Points}.
10638 @vindex $_sdata@r{, inspect, convenience variable}
10639 The variable @code{$_sdata} contains extra collected static tracepoint
10640 data. @xref{Tracepoint Actions,,Tracepoint Action Lists}. Note that
10641 @code{$_sdata} could be empty, if not inspecting a trace buffer, or
10642 if extra static tracepoint data has not been collected.
10645 @vindex $_siginfo@r{, convenience variable}
10646 The variable @code{$_siginfo} contains extra signal information
10647 (@pxref{extra signal information}). Note that @code{$_siginfo}
10648 could be empty, if the application has not yet received any signals.
10649 For example, it will be empty before you execute the @code{run} command.
10652 @vindex $_tlb@r{, convenience variable}
10653 The variable @code{$_tlb} is automatically set when debugging
10654 applications running on MS-Windows in native mode or connected to
10655 gdbserver that supports the @code{qGetTIBAddr} request.
10656 @xref{General Query Packets}.
10657 This variable contains the address of the thread information block.
10660 The number of the current inferior. @xref{Inferiors and
10661 Programs, ,Debugging Multiple Inferiors and Programs}.
10664 The thread number of the current thread. @xref{thread numbers}.
10667 The global number of the current thread. @xref{global thread numbers}.
10671 @node Convenience Funs
10672 @section Convenience Functions
10674 @cindex convenience functions
10675 @value{GDBN} also supplies some @dfn{convenience functions}. These
10676 have a syntax similar to convenience variables. A convenience
10677 function can be used in an expression just like an ordinary function;
10678 however, a convenience function is implemented internally to
10681 These functions do not require @value{GDBN} to be configured with
10682 @code{Python} support, which means that they are always available.
10686 @item $_isvoid (@var{expr})
10687 @findex $_isvoid@r{, convenience function}
10688 Return one if the expression @var{expr} is @code{void}. Otherwise it
10691 A @code{void} expression is an expression where the type of the result
10692 is @code{void}. For example, you can examine a convenience variable
10693 (see @ref{Convenience Vars,, Convenience Variables}) to check whether
10697 (@value{GDBP}) print $_exitcode
10699 (@value{GDBP}) print $_isvoid ($_exitcode)
10702 Starting program: ./a.out
10703 [Inferior 1 (process 29572) exited normally]
10704 (@value{GDBP}) print $_exitcode
10706 (@value{GDBP}) print $_isvoid ($_exitcode)
10710 In the example above, we used @code{$_isvoid} to check whether
10711 @code{$_exitcode} is @code{void} before and after the execution of the
10712 program being debugged. Before the execution there is no exit code to
10713 be examined, therefore @code{$_exitcode} is @code{void}. After the
10714 execution the program being debugged returned zero, therefore
10715 @code{$_exitcode} is zero, which means that it is not @code{void}
10718 The @code{void} expression can also be a call of a function from the
10719 program being debugged. For example, given the following function:
10728 The result of calling it inside @value{GDBN} is @code{void}:
10731 (@value{GDBP}) print foo ()
10733 (@value{GDBP}) print $_isvoid (foo ())
10735 (@value{GDBP}) set $v = foo ()
10736 (@value{GDBP}) print $v
10738 (@value{GDBP}) print $_isvoid ($v)
10744 These functions require @value{GDBN} to be configured with
10745 @code{Python} support.
10749 @item $_memeq(@var{buf1}, @var{buf2}, @var{length})
10750 @findex $_memeq@r{, convenience function}
10751 Returns one if the @var{length} bytes at the addresses given by
10752 @var{buf1} and @var{buf2} are equal.
10753 Otherwise it returns zero.
10755 @item $_regex(@var{str}, @var{regex})
10756 @findex $_regex@r{, convenience function}
10757 Returns one if the string @var{str} matches the regular expression
10758 @var{regex}. Otherwise it returns zero.
10759 The syntax of the regular expression is that specified by @code{Python}'s
10760 regular expression support.
10762 @item $_streq(@var{str1}, @var{str2})
10763 @findex $_streq@r{, convenience function}
10764 Returns one if the strings @var{str1} and @var{str2} are equal.
10765 Otherwise it returns zero.
10767 @item $_strlen(@var{str})
10768 @findex $_strlen@r{, convenience function}
10769 Returns the length of string @var{str}.
10771 @item $_caller_is(@var{name}@r{[}, @var{number_of_frames}@r{]})
10772 @findex $_caller_is@r{, convenience function}
10773 Returns one if the calling function's name is equal to @var{name}.
10774 Otherwise it returns zero.
10776 If the optional argument @var{number_of_frames} is provided,
10777 it is the number of frames up in the stack to look.
10785 at testsuite/gdb.python/py-caller-is.c:21
10786 #1 0x00000000004005a0 in middle_func ()
10787 at testsuite/gdb.python/py-caller-is.c:27
10788 #2 0x00000000004005ab in top_func ()
10789 at testsuite/gdb.python/py-caller-is.c:33
10790 #3 0x00000000004005b6 in main ()
10791 at testsuite/gdb.python/py-caller-is.c:39
10792 (gdb) print $_caller_is ("middle_func")
10794 (gdb) print $_caller_is ("top_func", 2)
10798 @item $_caller_matches(@var{regexp}@r{[}, @var{number_of_frames}@r{]})
10799 @findex $_caller_matches@r{, convenience function}
10800 Returns one if the calling function's name matches the regular expression
10801 @var{regexp}. Otherwise it returns zero.
10803 If the optional argument @var{number_of_frames} is provided,
10804 it is the number of frames up in the stack to look.
10807 @item $_any_caller_is(@var{name}@r{[}, @var{number_of_frames}@r{]})
10808 @findex $_any_caller_is@r{, convenience function}
10809 Returns one if any calling function's name is equal to @var{name}.
10810 Otherwise it returns zero.
10812 If the optional argument @var{number_of_frames} is provided,
10813 it is the number of frames up in the stack to look.
10816 This function differs from @code{$_caller_is} in that this function
10817 checks all stack frames from the immediate caller to the frame specified
10818 by @var{number_of_frames}, whereas @code{$_caller_is} only checks the
10819 frame specified by @var{number_of_frames}.
10821 @item $_any_caller_matches(@var{regexp}@r{[}, @var{number_of_frames}@r{]})
10822 @findex $_any_caller_matches@r{, convenience function}
10823 Returns one if any calling function's name matches the regular expression
10824 @var{regexp}. Otherwise it returns zero.
10826 If the optional argument @var{number_of_frames} is provided,
10827 it is the number of frames up in the stack to look.
10830 This function differs from @code{$_caller_matches} in that this function
10831 checks all stack frames from the immediate caller to the frame specified
10832 by @var{number_of_frames}, whereas @code{$_caller_matches} only checks the
10833 frame specified by @var{number_of_frames}.
10835 @item $_as_string(@var{value})
10836 @findex $_as_string@r{, convenience function}
10837 Return the string representation of @var{value}.
10839 This function is useful to obtain the textual label (enumerator) of an
10840 enumeration value. For example, assuming the variable @var{node} is of
10841 an enumerated type:
10844 (gdb) printf "Visiting node of type %s\n", $_as_string(node)
10845 Visiting node of type NODE_INTEGER
10850 @value{GDBN} provides the ability to list and get help on
10851 convenience functions.
10854 @item help function
10855 @kindex help function
10856 @cindex show all convenience functions
10857 Print a list of all convenience functions.
10864 You can refer to machine register contents, in expressions, as variables
10865 with names starting with @samp{$}. The names of registers are different
10866 for each machine; use @code{info registers} to see the names used on
10870 @kindex info registers
10871 @item info registers
10872 Print the names and values of all registers except floating-point
10873 and vector registers (in the selected stack frame).
10875 @kindex info all-registers
10876 @cindex floating point registers
10877 @item info all-registers
10878 Print the names and values of all registers, including floating-point
10879 and vector registers (in the selected stack frame).
10881 @item info registers @var{regname} @dots{}
10882 Print the @dfn{relativized} value of each specified register @var{regname}.
10883 As discussed in detail below, register values are normally relative to
10884 the selected stack frame. The @var{regname} may be any register name valid on
10885 the machine you are using, with or without the initial @samp{$}.
10888 @anchor{standard registers}
10889 @cindex stack pointer register
10890 @cindex program counter register
10891 @cindex process status register
10892 @cindex frame pointer register
10893 @cindex standard registers
10894 @value{GDBN} has four ``standard'' register names that are available (in
10895 expressions) on most machines---whenever they do not conflict with an
10896 architecture's canonical mnemonics for registers. The register names
10897 @code{$pc} and @code{$sp} are used for the program counter register and
10898 the stack pointer. @code{$fp} is used for a register that contains a
10899 pointer to the current stack frame, and @code{$ps} is used for a
10900 register that contains the processor status. For example,
10901 you could print the program counter in hex with
10908 or print the instruction to be executed next with
10915 or add four to the stack pointer@footnote{This is a way of removing
10916 one word from the stack, on machines where stacks grow downward in
10917 memory (most machines, nowadays). This assumes that the innermost
10918 stack frame is selected; setting @code{$sp} is not allowed when other
10919 stack frames are selected. To pop entire frames off the stack,
10920 regardless of machine architecture, use @code{return};
10921 see @ref{Returning, ,Returning from a Function}.} with
10927 Whenever possible, these four standard register names are available on
10928 your machine even though the machine has different canonical mnemonics,
10929 so long as there is no conflict. The @code{info registers} command
10930 shows the canonical names. For example, on the SPARC, @code{info
10931 registers} displays the processor status register as @code{$psr} but you
10932 can also refer to it as @code{$ps}; and on x86-based machines @code{$ps}
10933 is an alias for the @sc{eflags} register.
10935 @value{GDBN} always considers the contents of an ordinary register as an
10936 integer when the register is examined in this way. Some machines have
10937 special registers which can hold nothing but floating point; these
10938 registers are considered to have floating point values. There is no way
10939 to refer to the contents of an ordinary register as floating point value
10940 (although you can @emph{print} it as a floating point value with
10941 @samp{print/f $@var{regname}}).
10943 Some registers have distinct ``raw'' and ``virtual'' data formats. This
10944 means that the data format in which the register contents are saved by
10945 the operating system is not the same one that your program normally
10946 sees. For example, the registers of the 68881 floating point
10947 coprocessor are always saved in ``extended'' (raw) format, but all C
10948 programs expect to work with ``double'' (virtual) format. In such
10949 cases, @value{GDBN} normally works with the virtual format only (the format
10950 that makes sense for your program), but the @code{info registers} command
10951 prints the data in both formats.
10953 @cindex SSE registers (x86)
10954 @cindex MMX registers (x86)
10955 Some machines have special registers whose contents can be interpreted
10956 in several different ways. For example, modern x86-based machines
10957 have SSE and MMX registers that can hold several values packed
10958 together in several different formats. @value{GDBN} refers to such
10959 registers in @code{struct} notation:
10962 (@value{GDBP}) print $xmm1
10964 v4_float = @{0, 3.43859137e-038, 1.54142831e-044, 1.821688e-044@},
10965 v2_double = @{9.92129282474342e-303, 2.7585945287983262e-313@},
10966 v16_int8 = "\000\000\000\000\3706;\001\v\000\000\000\r\000\000",
10967 v8_int16 = @{0, 0, 14072, 315, 11, 0, 13, 0@},
10968 v4_int32 = @{0, 20657912, 11, 13@},
10969 v2_int64 = @{88725056443645952, 55834574859@},
10970 uint128 = 0x0000000d0000000b013b36f800000000
10975 To set values of such registers, you need to tell @value{GDBN} which
10976 view of the register you wish to change, as if you were assigning
10977 value to a @code{struct} member:
10980 (@value{GDBP}) set $xmm1.uint128 = 0x000000000000000000000000FFFFFFFF
10983 Normally, register values are relative to the selected stack frame
10984 (@pxref{Selection, ,Selecting a Frame}). This means that you get the
10985 value that the register would contain if all stack frames farther in
10986 were exited and their saved registers restored. In order to see the
10987 true contents of hardware registers, you must select the innermost
10988 frame (with @samp{frame 0}).
10990 @cindex caller-saved registers
10991 @cindex call-clobbered registers
10992 @cindex volatile registers
10993 @cindex <not saved> values
10994 Usually ABIs reserve some registers as not needed to be saved by the
10995 callee (a.k.a.: ``caller-saved'', ``call-clobbered'' or ``volatile''
10996 registers). It may therefore not be possible for @value{GDBN} to know
10997 the value a register had before the call (in other words, in the outer
10998 frame), if the register value has since been changed by the callee.
10999 @value{GDBN} tries to deduce where the inner frame saved
11000 (``callee-saved'') registers, from the debug info, unwind info, or the
11001 machine code generated by your compiler. If some register is not
11002 saved, and @value{GDBN} knows the register is ``caller-saved'' (via
11003 its own knowledge of the ABI, or because the debug/unwind info
11004 explicitly says the register's value is undefined), @value{GDBN}
11005 displays @w{@samp{<not saved>}} as the register's value. With targets
11006 that @value{GDBN} has no knowledge of the register saving convention,
11007 if a register was not saved by the callee, then its value and location
11008 in the outer frame are assumed to be the same of the inner frame.
11009 This is usually harmless, because if the register is call-clobbered,
11010 the caller either does not care what is in the register after the
11011 call, or has code to restore the value that it does care about. Note,
11012 however, that if you change such a register in the outer frame, you
11013 may also be affecting the inner frame. Also, the more ``outer'' the
11014 frame is you're looking at, the more likely a call-clobbered
11015 register's value is to be wrong, in the sense that it doesn't actually
11016 represent the value the register had just before the call.
11018 @node Floating Point Hardware
11019 @section Floating Point Hardware
11020 @cindex floating point
11022 Depending on the configuration, @value{GDBN} may be able to give
11023 you more information about the status of the floating point hardware.
11028 Display hardware-dependent information about the floating
11029 point unit. The exact contents and layout vary depending on the
11030 floating point chip. Currently, @samp{info float} is supported on
11031 the ARM and x86 machines.
11035 @section Vector Unit
11036 @cindex vector unit
11038 Depending on the configuration, @value{GDBN} may be able to give you
11039 more information about the status of the vector unit.
11042 @kindex info vector
11044 Display information about the vector unit. The exact contents and
11045 layout vary depending on the hardware.
11048 @node OS Information
11049 @section Operating System Auxiliary Information
11050 @cindex OS information
11052 @value{GDBN} provides interfaces to useful OS facilities that can help
11053 you debug your program.
11055 @cindex auxiliary vector
11056 @cindex vector, auxiliary
11057 Some operating systems supply an @dfn{auxiliary vector} to programs at
11058 startup. This is akin to the arguments and environment that you
11059 specify for a program, but contains a system-dependent variety of
11060 binary values that tell system libraries important details about the
11061 hardware, operating system, and process. Each value's purpose is
11062 identified by an integer tag; the meanings are well-known but system-specific.
11063 Depending on the configuration and operating system facilities,
11064 @value{GDBN} may be able to show you this information. For remote
11065 targets, this functionality may further depend on the remote stub's
11066 support of the @samp{qXfer:auxv:read} packet, see
11067 @ref{qXfer auxiliary vector read}.
11072 Display the auxiliary vector of the inferior, which can be either a
11073 live process or a core dump file. @value{GDBN} prints each tag value
11074 numerically, and also shows names and text descriptions for recognized
11075 tags. Some values in the vector are numbers, some bit masks, and some
11076 pointers to strings or other data. @value{GDBN} displays each value in the
11077 most appropriate form for a recognized tag, and in hexadecimal for
11078 an unrecognized tag.
11081 On some targets, @value{GDBN} can access operating system-specific
11082 information and show it to you. The types of information available
11083 will differ depending on the type of operating system running on the
11084 target. The mechanism used to fetch the data is described in
11085 @ref{Operating System Information}. For remote targets, this
11086 functionality depends on the remote stub's support of the
11087 @samp{qXfer:osdata:read} packet, see @ref{qXfer osdata read}.
11091 @item info os @var{infotype}
11093 Display OS information of the requested type.
11095 On @sc{gnu}/Linux, the following values of @var{infotype} are valid:
11097 @anchor{linux info os infotypes}
11099 @kindex info os cpus
11101 Display the list of all CPUs/cores. For each CPU/core, @value{GDBN} prints
11102 the available fields from /proc/cpuinfo. For each supported architecture
11103 different fields are available. Two common entries are processor which gives
11104 CPU number and bogomips; a system constant that is calculated during
11105 kernel initialization.
11107 @kindex info os files
11109 Display the list of open file descriptors on the target. For each
11110 file descriptor, @value{GDBN} prints the identifier of the process
11111 owning the descriptor, the command of the owning process, the value
11112 of the descriptor, and the target of the descriptor.
11114 @kindex info os modules
11116 Display the list of all loaded kernel modules on the target. For each
11117 module, @value{GDBN} prints the module name, the size of the module in
11118 bytes, the number of times the module is used, the dependencies of the
11119 module, the status of the module, and the address of the loaded module
11122 @kindex info os msg
11124 Display the list of all System V message queues on the target. For each
11125 message queue, @value{GDBN} prints the message queue key, the message
11126 queue identifier, the access permissions, the current number of bytes
11127 on the queue, the current number of messages on the queue, the processes
11128 that last sent and received a message on the queue, the user and group
11129 of the owner and creator of the message queue, the times at which a
11130 message was last sent and received on the queue, and the time at which
11131 the message queue was last changed.
11133 @kindex info os processes
11135 Display the list of processes on the target. For each process,
11136 @value{GDBN} prints the process identifier, the name of the user, the
11137 command corresponding to the process, and the list of processor cores
11138 that the process is currently running on. (To understand what these
11139 properties mean, for this and the following info types, please consult
11140 the general @sc{gnu}/Linux documentation.)
11142 @kindex info os procgroups
11144 Display the list of process groups on the target. For each process,
11145 @value{GDBN} prints the identifier of the process group that it belongs
11146 to, the command corresponding to the process group leader, the process
11147 identifier, and the command line of the process. The list is sorted
11148 first by the process group identifier, then by the process identifier,
11149 so that processes belonging to the same process group are grouped together
11150 and the process group leader is listed first.
11152 @kindex info os semaphores
11154 Display the list of all System V semaphore sets on the target. For each
11155 semaphore set, @value{GDBN} prints the semaphore set key, the semaphore
11156 set identifier, the access permissions, the number of semaphores in the
11157 set, the user and group of the owner and creator of the semaphore set,
11158 and the times at which the semaphore set was operated upon and changed.
11160 @kindex info os shm
11162 Display the list of all System V shared-memory regions on the target.
11163 For each shared-memory region, @value{GDBN} prints the region key,
11164 the shared-memory identifier, the access permissions, the size of the
11165 region, the process that created the region, the process that last
11166 attached to or detached from the region, the current number of live
11167 attaches to the region, and the times at which the region was last
11168 attached to, detach from, and changed.
11170 @kindex info os sockets
11172 Display the list of Internet-domain sockets on the target. For each
11173 socket, @value{GDBN} prints the address and port of the local and
11174 remote endpoints, the current state of the connection, the creator of
11175 the socket, the IP address family of the socket, and the type of the
11178 @kindex info os threads
11180 Display the list of threads running on the target. For each thread,
11181 @value{GDBN} prints the identifier of the process that the thread
11182 belongs to, the command of the process, the thread identifier, and the
11183 processor core that it is currently running on. The main thread of a
11184 process is not listed.
11188 If @var{infotype} is omitted, then list the possible values for
11189 @var{infotype} and the kind of OS information available for each
11190 @var{infotype}. If the target does not return a list of possible
11191 types, this command will report an error.
11194 @node Memory Region Attributes
11195 @section Memory Region Attributes
11196 @cindex memory region attributes
11198 @dfn{Memory region attributes} allow you to describe special handling
11199 required by regions of your target's memory. @value{GDBN} uses
11200 attributes to determine whether to allow certain types of memory
11201 accesses; whether to use specific width accesses; and whether to cache
11202 target memory. By default the description of memory regions is
11203 fetched from the target (if the current target supports this), but the
11204 user can override the fetched regions.
11206 Defined memory regions can be individually enabled and disabled. When a
11207 memory region is disabled, @value{GDBN} uses the default attributes when
11208 accessing memory in that region. Similarly, if no memory regions have
11209 been defined, @value{GDBN} uses the default attributes when accessing
11212 When a memory region is defined, it is given a number to identify it;
11213 to enable, disable, or remove a memory region, you specify that number.
11217 @item mem @var{lower} @var{upper} @var{attributes}@dots{}
11218 Define a memory region bounded by @var{lower} and @var{upper} with
11219 attributes @var{attributes}@dots{}, and add it to the list of regions
11220 monitored by @value{GDBN}. Note that @var{upper} == 0 is a special
11221 case: it is treated as the target's maximum memory address.
11222 (0xffff on 16 bit targets, 0xffffffff on 32 bit targets, etc.)
11225 Discard any user changes to the memory regions and use target-supplied
11226 regions, if available, or no regions if the target does not support.
11229 @item delete mem @var{nums}@dots{}
11230 Remove memory regions @var{nums}@dots{} from the list of regions
11231 monitored by @value{GDBN}.
11233 @kindex disable mem
11234 @item disable mem @var{nums}@dots{}
11235 Disable monitoring of memory regions @var{nums}@dots{}.
11236 A disabled memory region is not forgotten.
11237 It may be enabled again later.
11240 @item enable mem @var{nums}@dots{}
11241 Enable monitoring of memory regions @var{nums}@dots{}.
11245 Print a table of all defined memory regions, with the following columns
11249 @item Memory Region Number
11250 @item Enabled or Disabled.
11251 Enabled memory regions are marked with @samp{y}.
11252 Disabled memory regions are marked with @samp{n}.
11255 The address defining the inclusive lower bound of the memory region.
11258 The address defining the exclusive upper bound of the memory region.
11261 The list of attributes set for this memory region.
11266 @subsection Attributes
11268 @subsubsection Memory Access Mode
11269 The access mode attributes set whether @value{GDBN} may make read or
11270 write accesses to a memory region.
11272 While these attributes prevent @value{GDBN} from performing invalid
11273 memory accesses, they do nothing to prevent the target system, I/O DMA,
11274 etc.@: from accessing memory.
11278 Memory is read only.
11280 Memory is write only.
11282 Memory is read/write. This is the default.
11285 @subsubsection Memory Access Size
11286 The access size attribute tells @value{GDBN} to use specific sized
11287 accesses in the memory region. Often memory mapped device registers
11288 require specific sized accesses. If no access size attribute is
11289 specified, @value{GDBN} may use accesses of any size.
11293 Use 8 bit memory accesses.
11295 Use 16 bit memory accesses.
11297 Use 32 bit memory accesses.
11299 Use 64 bit memory accesses.
11302 @c @subsubsection Hardware/Software Breakpoints
11303 @c The hardware/software breakpoint attributes set whether @value{GDBN}
11304 @c will use hardware or software breakpoints for the internal breakpoints
11305 @c used by the step, next, finish, until, etc. commands.
11309 @c Always use hardware breakpoints
11310 @c @item swbreak (default)
11313 @subsubsection Data Cache
11314 The data cache attributes set whether @value{GDBN} will cache target
11315 memory. While this generally improves performance by reducing debug
11316 protocol overhead, it can lead to incorrect results because @value{GDBN}
11317 does not know about volatile variables or memory mapped device
11322 Enable @value{GDBN} to cache target memory.
11324 Disable @value{GDBN} from caching target memory. This is the default.
11327 @subsection Memory Access Checking
11328 @value{GDBN} can be instructed to refuse accesses to memory that is
11329 not explicitly described. This can be useful if accessing such
11330 regions has undesired effects for a specific target, or to provide
11331 better error checking. The following commands control this behaviour.
11334 @kindex set mem inaccessible-by-default
11335 @item set mem inaccessible-by-default [on|off]
11336 If @code{on} is specified, make @value{GDBN} treat memory not
11337 explicitly described by the memory ranges as non-existent and refuse accesses
11338 to such memory. The checks are only performed if there's at least one
11339 memory range defined. If @code{off} is specified, make @value{GDBN}
11340 treat the memory not explicitly described by the memory ranges as RAM.
11341 The default value is @code{on}.
11342 @kindex show mem inaccessible-by-default
11343 @item show mem inaccessible-by-default
11344 Show the current handling of accesses to unknown memory.
11348 @c @subsubsection Memory Write Verification
11349 @c The memory write verification attributes set whether @value{GDBN}
11350 @c will re-reads data after each write to verify the write was successful.
11354 @c @item noverify (default)
11357 @node Dump/Restore Files
11358 @section Copy Between Memory and a File
11359 @cindex dump/restore files
11360 @cindex append data to a file
11361 @cindex dump data to a file
11362 @cindex restore data from a file
11364 You can use the commands @code{dump}, @code{append}, and
11365 @code{restore} to copy data between target memory and a file. The
11366 @code{dump} and @code{append} commands write data to a file, and the
11367 @code{restore} command reads data from a file back into the inferior's
11368 memory. Files may be in binary, Motorola S-record, Intel hex,
11369 Tektronix Hex, or Verilog Hex format; however, @value{GDBN} can only
11370 append to binary files, and cannot read from Verilog Hex files.
11375 @item dump @r{[}@var{format}@r{]} memory @var{filename} @var{start_addr} @var{end_addr}
11376 @itemx dump @r{[}@var{format}@r{]} value @var{filename} @var{expr}
11377 Dump the contents of memory from @var{start_addr} to @var{end_addr},
11378 or the value of @var{expr}, to @var{filename} in the given format.
11380 The @var{format} parameter may be any one of:
11387 Motorola S-record format.
11389 Tektronix Hex format.
11391 Verilog Hex format.
11394 @value{GDBN} uses the same definitions of these formats as the
11395 @sc{gnu} binary utilities, like @samp{objdump} and @samp{objcopy}. If
11396 @var{format} is omitted, @value{GDBN} dumps the data in raw binary
11400 @item append @r{[}binary@r{]} memory @var{filename} @var{start_addr} @var{end_addr}
11401 @itemx append @r{[}binary@r{]} value @var{filename} @var{expr}
11402 Append the contents of memory from @var{start_addr} to @var{end_addr},
11403 or the value of @var{expr}, to the file @var{filename}, in raw binary form.
11404 (@value{GDBN} can only append data to files in raw binary form.)
11407 @item restore @var{filename} @r{[}binary@r{]} @var{bias} @var{start} @var{end}
11408 Restore the contents of file @var{filename} into memory. The
11409 @code{restore} command can automatically recognize any known @sc{bfd}
11410 file format, except for raw binary. To restore a raw binary file you
11411 must specify the optional keyword @code{binary} after the filename.
11413 If @var{bias} is non-zero, its value will be added to the addresses
11414 contained in the file. Binary files always start at address zero, so
11415 they will be restored at address @var{bias}. Other bfd files have
11416 a built-in location; they will be restored at offset @var{bias}
11417 from that location.
11419 If @var{start} and/or @var{end} are non-zero, then only data between
11420 file offset @var{start} and file offset @var{end} will be restored.
11421 These offsets are relative to the addresses in the file, before
11422 the @var{bias} argument is applied.
11426 @node Core File Generation
11427 @section How to Produce a Core File from Your Program
11428 @cindex dump core from inferior
11430 A @dfn{core file} or @dfn{core dump} is a file that records the memory
11431 image of a running process and its process status (register values
11432 etc.). Its primary use is post-mortem debugging of a program that
11433 crashed while it ran outside a debugger. A program that crashes
11434 automatically produces a core file, unless this feature is disabled by
11435 the user. @xref{Files}, for information on invoking @value{GDBN} in
11436 the post-mortem debugging mode.
11438 Occasionally, you may wish to produce a core file of the program you
11439 are debugging in order to preserve a snapshot of its state.
11440 @value{GDBN} has a special command for that.
11444 @kindex generate-core-file
11445 @item generate-core-file [@var{file}]
11446 @itemx gcore [@var{file}]
11447 Produce a core dump of the inferior process. The optional argument
11448 @var{file} specifies the file name where to put the core dump. If not
11449 specified, the file name defaults to @file{core.@var{pid}}, where
11450 @var{pid} is the inferior process ID.
11452 Note that this command is implemented only for some systems (as of
11453 this writing, @sc{gnu}/Linux, FreeBSD, Solaris, and S390).
11455 On @sc{gnu}/Linux, this command can take into account the value of the
11456 file @file{/proc/@var{pid}/coredump_filter} when generating the core
11457 dump (@pxref{set use-coredump-filter}).
11459 @kindex set use-coredump-filter
11460 @anchor{set use-coredump-filter}
11461 @item set use-coredump-filter on
11462 @itemx set use-coredump-filter off
11463 Enable or disable the use of the file
11464 @file{/proc/@var{pid}/coredump_filter} when generating core dump
11465 files. This file is used by the Linux kernel to decide what types of
11466 memory mappings will be dumped or ignored when generating a core dump
11467 file. @var{pid} is the process ID of a currently running process.
11469 To make use of this feature, you have to write in the
11470 @file{/proc/@var{pid}/coredump_filter} file a value, in hexadecimal,
11471 which is a bit mask representing the memory mapping types. If a bit
11472 is set in the bit mask, then the memory mappings of the corresponding
11473 types will be dumped; otherwise, they will be ignored. This
11474 configuration is inherited by child processes. For more information
11475 about the bits that can be set in the
11476 @file{/proc/@var{pid}/coredump_filter} file, please refer to the
11477 manpage of @code{core(5)}.
11479 By default, this option is @code{on}. If this option is turned
11480 @code{off}, @value{GDBN} does not read the @file{coredump_filter} file
11481 and instead uses the same default value as the Linux kernel in order
11482 to decide which pages will be dumped in the core dump file. This
11483 value is currently @code{0x33}, which means that bits @code{0}
11484 (anonymous private mappings), @code{1} (anonymous shared mappings),
11485 @code{4} (ELF headers) and @code{5} (private huge pages) are active.
11486 This will cause these memory mappings to be dumped automatically.
11489 @node Character Sets
11490 @section Character Sets
11491 @cindex character sets
11493 @cindex translating between character sets
11494 @cindex host character set
11495 @cindex target character set
11497 If the program you are debugging uses a different character set to
11498 represent characters and strings than the one @value{GDBN} uses itself,
11499 @value{GDBN} can automatically translate between the character sets for
11500 you. The character set @value{GDBN} uses we call the @dfn{host
11501 character set}; the one the inferior program uses we call the
11502 @dfn{target character set}.
11504 For example, if you are running @value{GDBN} on a @sc{gnu}/Linux system, which
11505 uses the ISO Latin 1 character set, but you are using @value{GDBN}'s
11506 remote protocol (@pxref{Remote Debugging}) to debug a program
11507 running on an IBM mainframe, which uses the @sc{ebcdic} character set,
11508 then the host character set is Latin-1, and the target character set is
11509 @sc{ebcdic}. If you give @value{GDBN} the command @code{set
11510 target-charset EBCDIC-US}, then @value{GDBN} translates between
11511 @sc{ebcdic} and Latin 1 as you print character or string values, or use
11512 character and string literals in expressions.
11514 @value{GDBN} has no way to automatically recognize which character set
11515 the inferior program uses; you must tell it, using the @code{set
11516 target-charset} command, described below.
11518 Here are the commands for controlling @value{GDBN}'s character set
11522 @item set target-charset @var{charset}
11523 @kindex set target-charset
11524 Set the current target character set to @var{charset}. To display the
11525 list of supported target character sets, type
11526 @kbd{@w{set target-charset @key{TAB}@key{TAB}}}.
11528 @item set host-charset @var{charset}
11529 @kindex set host-charset
11530 Set the current host character set to @var{charset}.
11532 By default, @value{GDBN} uses a host character set appropriate to the
11533 system it is running on; you can override that default using the
11534 @code{set host-charset} command. On some systems, @value{GDBN} cannot
11535 automatically determine the appropriate host character set. In this
11536 case, @value{GDBN} uses @samp{UTF-8}.
11538 @value{GDBN} can only use certain character sets as its host character
11539 set. If you type @kbd{@w{set host-charset @key{TAB}@key{TAB}}},
11540 @value{GDBN} will list the host character sets it supports.
11542 @item set charset @var{charset}
11543 @kindex set charset
11544 Set the current host and target character sets to @var{charset}. As
11545 above, if you type @kbd{@w{set charset @key{TAB}@key{TAB}}},
11546 @value{GDBN} will list the names of the character sets that can be used
11547 for both host and target.
11550 @kindex show charset
11551 Show the names of the current host and target character sets.
11553 @item show host-charset
11554 @kindex show host-charset
11555 Show the name of the current host character set.
11557 @item show target-charset
11558 @kindex show target-charset
11559 Show the name of the current target character set.
11561 @item set target-wide-charset @var{charset}
11562 @kindex set target-wide-charset
11563 Set the current target's wide character set to @var{charset}. This is
11564 the character set used by the target's @code{wchar_t} type. To
11565 display the list of supported wide character sets, type
11566 @kbd{@w{set target-wide-charset @key{TAB}@key{TAB}}}.
11568 @item show target-wide-charset
11569 @kindex show target-wide-charset
11570 Show the name of the current target's wide character set.
11573 Here is an example of @value{GDBN}'s character set support in action.
11574 Assume that the following source code has been placed in the file
11575 @file{charset-test.c}:
11581 = @{72, 101, 108, 108, 111, 44, 32, 119,
11582 111, 114, 108, 100, 33, 10, 0@};
11583 char ibm1047_hello[]
11584 = @{200, 133, 147, 147, 150, 107, 64, 166,
11585 150, 153, 147, 132, 90, 37, 0@};
11589 printf ("Hello, world!\n");
11593 In this program, @code{ascii_hello} and @code{ibm1047_hello} are arrays
11594 containing the string @samp{Hello, world!} followed by a newline,
11595 encoded in the @sc{ascii} and @sc{ibm1047} character sets.
11597 We compile the program, and invoke the debugger on it:
11600 $ gcc -g charset-test.c -o charset-test
11601 $ gdb -nw charset-test
11602 GNU gdb 2001-12-19-cvs
11603 Copyright 2001 Free Software Foundation, Inc.
11608 We can use the @code{show charset} command to see what character sets
11609 @value{GDBN} is currently using to interpret and display characters and
11613 (@value{GDBP}) show charset
11614 The current host and target character set is `ISO-8859-1'.
11618 For the sake of printing this manual, let's use @sc{ascii} as our
11619 initial character set:
11621 (@value{GDBP}) set charset ASCII
11622 (@value{GDBP}) show charset
11623 The current host and target character set is `ASCII'.
11627 Let's assume that @sc{ascii} is indeed the correct character set for our
11628 host system --- in other words, let's assume that if @value{GDBN} prints
11629 characters using the @sc{ascii} character set, our terminal will display
11630 them properly. Since our current target character set is also
11631 @sc{ascii}, the contents of @code{ascii_hello} print legibly:
11634 (@value{GDBP}) print ascii_hello
11635 $1 = 0x401698 "Hello, world!\n"
11636 (@value{GDBP}) print ascii_hello[0]
11641 @value{GDBN} uses the target character set for character and string
11642 literals you use in expressions:
11645 (@value{GDBP}) print '+'
11650 The @sc{ascii} character set uses the number 43 to encode the @samp{+}
11653 @value{GDBN} relies on the user to tell it which character set the
11654 target program uses. If we print @code{ibm1047_hello} while our target
11655 character set is still @sc{ascii}, we get jibberish:
11658 (@value{GDBP}) print ibm1047_hello
11659 $4 = 0x4016a8 "\310\205\223\223\226k@@\246\226\231\223\204Z%"
11660 (@value{GDBP}) print ibm1047_hello[0]
11665 If we invoke the @code{set target-charset} followed by @key{TAB}@key{TAB},
11666 @value{GDBN} tells us the character sets it supports:
11669 (@value{GDBP}) set target-charset
11670 ASCII EBCDIC-US IBM1047 ISO-8859-1
11671 (@value{GDBP}) set target-charset
11674 We can select @sc{ibm1047} as our target character set, and examine the
11675 program's strings again. Now the @sc{ascii} string is wrong, but
11676 @value{GDBN} translates the contents of @code{ibm1047_hello} from the
11677 target character set, @sc{ibm1047}, to the host character set,
11678 @sc{ascii}, and they display correctly:
11681 (@value{GDBP}) set target-charset IBM1047
11682 (@value{GDBP}) show charset
11683 The current host character set is `ASCII'.
11684 The current target character set is `IBM1047'.
11685 (@value{GDBP}) print ascii_hello
11686 $6 = 0x401698 "\110\145%%?\054\040\167?\162%\144\041\012"
11687 (@value{GDBP}) print ascii_hello[0]
11689 (@value{GDBP}) print ibm1047_hello
11690 $8 = 0x4016a8 "Hello, world!\n"
11691 (@value{GDBP}) print ibm1047_hello[0]
11696 As above, @value{GDBN} uses the target character set for character and
11697 string literals you use in expressions:
11700 (@value{GDBP}) print '+'
11705 The @sc{ibm1047} character set uses the number 78 to encode the @samp{+}
11708 @node Caching Target Data
11709 @section Caching Data of Targets
11710 @cindex caching data of targets
11712 @value{GDBN} caches data exchanged between the debugger and a target.
11713 Each cache is associated with the address space of the inferior.
11714 @xref{Inferiors and Programs}, about inferior and address space.
11715 Such caching generally improves performance in remote debugging
11716 (@pxref{Remote Debugging}), because it reduces the overhead of the
11717 remote protocol by bundling memory reads and writes into large chunks.
11718 Unfortunately, simply caching everything would lead to incorrect results,
11719 since @value{GDBN} does not necessarily know anything about volatile
11720 values, memory-mapped I/O addresses, etc. Furthermore, in non-stop mode
11721 (@pxref{Non-Stop Mode}) memory can be changed @emph{while} a gdb command
11723 Therefore, by default, @value{GDBN} only caches data
11724 known to be on the stack@footnote{In non-stop mode, it is moderately
11725 rare for a running thread to modify the stack of a stopped thread
11726 in a way that would interfere with a backtrace, and caching of
11727 stack reads provides a significant speed up of remote backtraces.} or
11728 in the code segment.
11729 Other regions of memory can be explicitly marked as
11730 cacheable; @pxref{Memory Region Attributes}.
11733 @kindex set remotecache
11734 @item set remotecache on
11735 @itemx set remotecache off
11736 This option no longer does anything; it exists for compatibility
11739 @kindex show remotecache
11740 @item show remotecache
11741 Show the current state of the obsolete remotecache flag.
11743 @kindex set stack-cache
11744 @item set stack-cache on
11745 @itemx set stack-cache off
11746 Enable or disable caching of stack accesses. When @code{on}, use
11747 caching. By default, this option is @code{on}.
11749 @kindex show stack-cache
11750 @item show stack-cache
11751 Show the current state of data caching for memory accesses.
11753 @kindex set code-cache
11754 @item set code-cache on
11755 @itemx set code-cache off
11756 Enable or disable caching of code segment accesses. When @code{on},
11757 use caching. By default, this option is @code{on}. This improves
11758 performance of disassembly in remote debugging.
11760 @kindex show code-cache
11761 @item show code-cache
11762 Show the current state of target memory cache for code segment
11765 @kindex info dcache
11766 @item info dcache @r{[}line@r{]}
11767 Print the information about the performance of data cache of the
11768 current inferior's address space. The information displayed
11769 includes the dcache width and depth, and for each cache line, its
11770 number, address, and how many times it was referenced. This
11771 command is useful for debugging the data cache operation.
11773 If a line number is specified, the contents of that line will be
11776 @item set dcache size @var{size}
11777 @cindex dcache size
11778 @kindex set dcache size
11779 Set maximum number of entries in dcache (dcache depth above).
11781 @item set dcache line-size @var{line-size}
11782 @cindex dcache line-size
11783 @kindex set dcache line-size
11784 Set number of bytes each dcache entry caches (dcache width above).
11785 Must be a power of 2.
11787 @item show dcache size
11788 @kindex show dcache size
11789 Show maximum number of dcache entries. @xref{Caching Target Data, info dcache}.
11791 @item show dcache line-size
11792 @kindex show dcache line-size
11793 Show default size of dcache lines.
11797 @node Searching Memory
11798 @section Search Memory
11799 @cindex searching memory
11801 Memory can be searched for a particular sequence of bytes with the
11802 @code{find} command.
11806 @item find @r{[}/@var{sn}@r{]} @var{start_addr}, +@var{len}, @var{val1} @r{[}, @var{val2}, @dots{}@r{]}
11807 @itemx find @r{[}/@var{sn}@r{]} @var{start_addr}, @var{end_addr}, @var{val1} @r{[}, @var{val2}, @dots{}@r{]}
11808 Search memory for the sequence of bytes specified by @var{val1}, @var{val2},
11809 etc. The search begins at address @var{start_addr} and continues for either
11810 @var{len} bytes or through to @var{end_addr} inclusive.
11813 @var{s} and @var{n} are optional parameters.
11814 They may be specified in either order, apart or together.
11817 @item @var{s}, search query size
11818 The size of each search query value.
11824 halfwords (two bytes)
11828 giant words (eight bytes)
11831 All values are interpreted in the current language.
11832 This means, for example, that if the current source language is C/C@t{++}
11833 then searching for the string ``hello'' includes the trailing '\0'.
11835 If the value size is not specified, it is taken from the
11836 value's type in the current language.
11837 This is useful when one wants to specify the search
11838 pattern as a mixture of types.
11839 Note that this means, for example, that in the case of C-like languages
11840 a search for an untyped 0x42 will search for @samp{(int) 0x42}
11841 which is typically four bytes.
11843 @item @var{n}, maximum number of finds
11844 The maximum number of matches to print. The default is to print all finds.
11847 You can use strings as search values. Quote them with double-quotes
11849 The string value is copied into the search pattern byte by byte,
11850 regardless of the endianness of the target and the size specification.
11852 The address of each match found is printed as well as a count of the
11853 number of matches found.
11855 The address of the last value found is stored in convenience variable
11857 A count of the number of matches is stored in @samp{$numfound}.
11859 For example, if stopped at the @code{printf} in this function:
11865 static char hello[] = "hello-hello";
11866 static struct @{ char c; short s; int i; @}
11867 __attribute__ ((packed)) mixed
11868 = @{ 'c', 0x1234, 0x87654321 @};
11869 printf ("%s\n", hello);
11874 you get during debugging:
11877 (gdb) find &hello[0], +sizeof(hello), "hello"
11878 0x804956d <hello.1620+6>
11880 (gdb) find &hello[0], +sizeof(hello), 'h', 'e', 'l', 'l', 'o'
11881 0x8049567 <hello.1620>
11882 0x804956d <hello.1620+6>
11884 (gdb) find /b1 &hello[0], +sizeof(hello), 'h', 0x65, 'l'
11885 0x8049567 <hello.1620>
11887 (gdb) find &mixed, +sizeof(mixed), (char) 'c', (short) 0x1234, (int) 0x87654321
11888 0x8049560 <mixed.1625>
11890 (gdb) print $numfound
11893 $2 = (void *) 0x8049560
11897 @section Value Sizes
11899 Whenever @value{GDBN} prints a value memory will be allocated within
11900 @value{GDBN} to hold the contents of the value. It is possible in
11901 some languages with dynamic typing systems, that an invalid program
11902 may indicate a value that is incorrectly large, this in turn may cause
11903 @value{GDBN} to try and allocate an overly large ammount of memory.
11906 @kindex set max-value-size
11907 @item set max-value-size @var{bytes}
11908 @itemx set max-value-size unlimited
11909 Set the maximum size of memory that @value{GDBN} will allocate for the
11910 contents of a value to @var{bytes}, trying to display a value that
11911 requires more memory than that will result in an error.
11913 Setting this variable does not effect values that have already been
11914 allocated within @value{GDBN}, only future allocations.
11916 There's a minimum size that @code{max-value-size} can be set to in
11917 order that @value{GDBN} can still operate correctly, this minimum is
11918 currently 16 bytes.
11920 The limit applies to the results of some subexpressions as well as to
11921 complete expressions. For example, an expression denoting a simple
11922 integer component, such as @code{x.y.z}, may fail if the size of
11923 @var{x.y} is dynamic and exceeds @var{bytes}. On the other hand,
11924 @value{GDBN} is sometimes clever; the expression @code{A[i]}, where
11925 @var{A} is an array variable with non-constant size, will generally
11926 succeed regardless of the bounds on @var{A}, as long as the component
11927 size is less than @var{bytes}.
11929 The default value of @code{max-value-size} is currently 64k.
11931 @kindex show max-value-size
11932 @item show max-value-size
11933 Show the maximum size of memory, in bytes, that @value{GDBN} will
11934 allocate for the contents of a value.
11937 @node Optimized Code
11938 @chapter Debugging Optimized Code
11939 @cindex optimized code, debugging
11940 @cindex debugging optimized code
11942 Almost all compilers support optimization. With optimization
11943 disabled, the compiler generates assembly code that corresponds
11944 directly to your source code, in a simplistic way. As the compiler
11945 applies more powerful optimizations, the generated assembly code
11946 diverges from your original source code. With help from debugging
11947 information generated by the compiler, @value{GDBN} can map from
11948 the running program back to constructs from your original source.
11950 @value{GDBN} is more accurate with optimization disabled. If you
11951 can recompile without optimization, it is easier to follow the
11952 progress of your program during debugging. But, there are many cases
11953 where you may need to debug an optimized version.
11955 When you debug a program compiled with @samp{-g -O}, remember that the
11956 optimizer has rearranged your code; the debugger shows you what is
11957 really there. Do not be too surprised when the execution path does not
11958 exactly match your source file! An extreme example: if you define a
11959 variable, but never use it, @value{GDBN} never sees that
11960 variable---because the compiler optimizes it out of existence.
11962 Some things do not work as well with @samp{-g -O} as with just
11963 @samp{-g}, particularly on machines with instruction scheduling. If in
11964 doubt, recompile with @samp{-g} alone, and if this fixes the problem,
11965 please report it to us as a bug (including a test case!).
11966 @xref{Variables}, for more information about debugging optimized code.
11969 * Inline Functions:: How @value{GDBN} presents inlining
11970 * Tail Call Frames:: @value{GDBN} analysis of jumps to functions
11973 @node Inline Functions
11974 @section Inline Functions
11975 @cindex inline functions, debugging
11977 @dfn{Inlining} is an optimization that inserts a copy of the function
11978 body directly at each call site, instead of jumping to a shared
11979 routine. @value{GDBN} displays inlined functions just like
11980 non-inlined functions. They appear in backtraces. You can view their
11981 arguments and local variables, step into them with @code{step}, skip
11982 them with @code{next}, and escape from them with @code{finish}.
11983 You can check whether a function was inlined by using the
11984 @code{info frame} command.
11986 For @value{GDBN} to support inlined functions, the compiler must
11987 record information about inlining in the debug information ---
11988 @value{NGCC} using the @sc{dwarf 2} format does this, and several
11989 other compilers do also. @value{GDBN} only supports inlined functions
11990 when using @sc{dwarf 2}. Versions of @value{NGCC} before 4.1
11991 do not emit two required attributes (@samp{DW_AT_call_file} and
11992 @samp{DW_AT_call_line}); @value{GDBN} does not display inlined
11993 function calls with earlier versions of @value{NGCC}. It instead
11994 displays the arguments and local variables of inlined functions as
11995 local variables in the caller.
11997 The body of an inlined function is directly included at its call site;
11998 unlike a non-inlined function, there are no instructions devoted to
11999 the call. @value{GDBN} still pretends that the call site and the
12000 start of the inlined function are different instructions. Stepping to
12001 the call site shows the call site, and then stepping again shows
12002 the first line of the inlined function, even though no additional
12003 instructions are executed.
12005 This makes source-level debugging much clearer; you can see both the
12006 context of the call and then the effect of the call. Only stepping by
12007 a single instruction using @code{stepi} or @code{nexti} does not do
12008 this; single instruction steps always show the inlined body.
12010 There are some ways that @value{GDBN} does not pretend that inlined
12011 function calls are the same as normal calls:
12015 Setting breakpoints at the call site of an inlined function may not
12016 work, because the call site does not contain any code. @value{GDBN}
12017 may incorrectly move the breakpoint to the next line of the enclosing
12018 function, after the call. This limitation will be removed in a future
12019 version of @value{GDBN}; until then, set a breakpoint on an earlier line
12020 or inside the inlined function instead.
12023 @value{GDBN} cannot locate the return value of inlined calls after
12024 using the @code{finish} command. This is a limitation of compiler-generated
12025 debugging information; after @code{finish}, you can step to the next line
12026 and print a variable where your program stored the return value.
12030 @node Tail Call Frames
12031 @section Tail Call Frames
12032 @cindex tail call frames, debugging
12034 Function @code{B} can call function @code{C} in its very last statement. In
12035 unoptimized compilation the call of @code{C} is immediately followed by return
12036 instruction at the end of @code{B} code. Optimizing compiler may replace the
12037 call and return in function @code{B} into one jump to function @code{C}
12038 instead. Such use of a jump instruction is called @dfn{tail call}.
12040 During execution of function @code{C}, there will be no indication in the
12041 function call stack frames that it was tail-called from @code{B}. If function
12042 @code{A} regularly calls function @code{B} which tail-calls function @code{C},
12043 then @value{GDBN} will see @code{A} as the caller of @code{C}. However, in
12044 some cases @value{GDBN} can determine that @code{C} was tail-called from
12045 @code{B}, and it will then create fictitious call frame for that, with the
12046 return address set up as if @code{B} called @code{C} normally.
12048 This functionality is currently supported only by DWARF 2 debugging format and
12049 the compiler has to produce @samp{DW_TAG_call_site} tags. With
12050 @value{NGCC}, you need to specify @option{-O -g} during compilation, to get
12053 @kbd{info frame} command (@pxref{Frame Info}) will indicate the tail call frame
12054 kind by text @code{tail call frame} such as in this sample @value{GDBN} output:
12058 0x40066b <b(int, double)+11>: jmp 0x400640 <c(int, double)>
12060 Stack level 1, frame at 0x7fffffffda30:
12061 rip = 0x40066d in b (amd64-entry-value.cc:59); saved rip 0x4004c5
12062 tail call frame, caller of frame at 0x7fffffffda30
12063 source language c++.
12064 Arglist at unknown address.
12065 Locals at unknown address, Previous frame's sp is 0x7fffffffda30
12068 The detection of all the possible code path executions can find them ambiguous.
12069 There is no execution history stored (possible @ref{Reverse Execution} is never
12070 used for this purpose) and the last known caller could have reached the known
12071 callee by multiple different jump sequences. In such case @value{GDBN} still
12072 tries to show at least all the unambiguous top tail callers and all the
12073 unambiguous bottom tail calees, if any.
12076 @anchor{set debug entry-values}
12077 @item set debug entry-values
12078 @kindex set debug entry-values
12079 When set to on, enables printing of analysis messages for both frame argument
12080 values at function entry and tail calls. It will show all the possible valid
12081 tail calls code paths it has considered. It will also print the intersection
12082 of them with the final unambiguous (possibly partial or even empty) code path
12085 @item show debug entry-values
12086 @kindex show debug entry-values
12087 Show the current state of analysis messages printing for both frame argument
12088 values at function entry and tail calls.
12091 The analysis messages for tail calls can for example show why the virtual tail
12092 call frame for function @code{c} has not been recognized (due to the indirect
12093 reference by variable @code{x}):
12096 static void __attribute__((noinline, noclone)) c (void);
12097 void (*x) (void) = c;
12098 static void __attribute__((noinline, noclone)) a (void) @{ x++; @}
12099 static void __attribute__((noinline, noclone)) c (void) @{ a (); @}
12100 int main (void) @{ x (); return 0; @}
12102 Breakpoint 1, DW_OP_entry_value resolving cannot find
12103 DW_TAG_call_site 0x40039a in main
12105 3 static void __attribute__((noinline, noclone)) a (void) @{ x++; @}
12108 #1 0x000000000040039a in main () at t.c:5
12111 Another possibility is an ambiguous virtual tail call frames resolution:
12115 static void __attribute__((noinline, noclone)) f (void) @{ i++; @}
12116 static void __attribute__((noinline, noclone)) e (void) @{ f (); @}
12117 static void __attribute__((noinline, noclone)) d (void) @{ f (); @}
12118 static void __attribute__((noinline, noclone)) c (void) @{ d (); @}
12119 static void __attribute__((noinline, noclone)) b (void)
12120 @{ if (i) c (); else e (); @}
12121 static void __attribute__((noinline, noclone)) a (void) @{ b (); @}
12122 int main (void) @{ a (); return 0; @}
12124 tailcall: initial: 0x4004d2(a) 0x4004ce(b) 0x4004b2(c) 0x4004a2(d)
12125 tailcall: compare: 0x4004d2(a) 0x4004cc(b) 0x400492(e)
12126 tailcall: reduced: 0x4004d2(a) |
12129 #1 0x00000000004004d2 in a () at t.c:8
12130 #2 0x0000000000400395 in main () at t.c:9
12133 @set CALLSEQ1A @code{main@value{ARROW}a@value{ARROW}b@value{ARROW}c@value{ARROW}d@value{ARROW}f}
12134 @set CALLSEQ2A @code{main@value{ARROW}a@value{ARROW}b@value{ARROW}e@value{ARROW}f}
12136 @c Convert CALLSEQ#A to CALLSEQ#B depending on HAVE_MAKEINFO_CLICK.
12137 @ifset HAVE_MAKEINFO_CLICK
12138 @set ARROW @click{}
12139 @set CALLSEQ1B @clicksequence{@value{CALLSEQ1A}}
12140 @set CALLSEQ2B @clicksequence{@value{CALLSEQ2A}}
12142 @ifclear HAVE_MAKEINFO_CLICK
12144 @set CALLSEQ1B @value{CALLSEQ1A}
12145 @set CALLSEQ2B @value{CALLSEQ2A}
12148 Frames #0 and #2 are real, #1 is a virtual tail call frame.
12149 The code can have possible execution paths @value{CALLSEQ1B} or
12150 @value{CALLSEQ2B}, @value{GDBN} cannot find which one from the inferior state.
12152 @code{initial:} state shows some random possible calling sequence @value{GDBN}
12153 has found. It then finds another possible calling sequcen - that one is
12154 prefixed by @code{compare:}. The non-ambiguous intersection of these two is
12155 printed as the @code{reduced:} calling sequence. That one could have many
12156 futher @code{compare:} and @code{reduced:} statements as long as there remain
12157 any non-ambiguous sequence entries.
12159 For the frame of function @code{b} in both cases there are different possible
12160 @code{$pc} values (@code{0x4004cc} or @code{0x4004ce}), therefore this frame is
12161 also ambigous. The only non-ambiguous frame is the one for function @code{a},
12162 therefore this one is displayed to the user while the ambiguous frames are
12165 There can be also reasons why printing of frame argument values at function
12170 static void __attribute__((noinline, noclone)) c (int i) @{ v++; @}
12171 static void __attribute__((noinline, noclone)) a (int i);
12172 static void __attribute__((noinline, noclone)) b (int i) @{ a (i); @}
12173 static void __attribute__((noinline, noclone)) a (int i)
12174 @{ if (i) b (i - 1); else c (0); @}
12175 int main (void) @{ a (5); return 0; @}
12178 #0 c (i=i@@entry=0) at t.c:2
12179 #1 0x0000000000400428 in a (DW_OP_entry_value resolving has found
12180 function "a" at 0x400420 can call itself via tail calls
12181 i=<optimized out>) at t.c:6
12182 #2 0x000000000040036e in main () at t.c:7
12185 @value{GDBN} cannot find out from the inferior state if and how many times did
12186 function @code{a} call itself (via function @code{b}) as these calls would be
12187 tail calls. Such tail calls would modify thue @code{i} variable, therefore
12188 @value{GDBN} cannot be sure the value it knows would be right - @value{GDBN}
12189 prints @code{<optimized out>} instead.
12192 @chapter C Preprocessor Macros
12194 Some languages, such as C and C@t{++}, provide a way to define and invoke
12195 ``preprocessor macros'' which expand into strings of tokens.
12196 @value{GDBN} can evaluate expressions containing macro invocations, show
12197 the result of macro expansion, and show a macro's definition, including
12198 where it was defined.
12200 You may need to compile your program specially to provide @value{GDBN}
12201 with information about preprocessor macros. Most compilers do not
12202 include macros in their debugging information, even when you compile
12203 with the @option{-g} flag. @xref{Compilation}.
12205 A program may define a macro at one point, remove that definition later,
12206 and then provide a different definition after that. Thus, at different
12207 points in the program, a macro may have different definitions, or have
12208 no definition at all. If there is a current stack frame, @value{GDBN}
12209 uses the macros in scope at that frame's source code line. Otherwise,
12210 @value{GDBN} uses the macros in scope at the current listing location;
12213 Whenever @value{GDBN} evaluates an expression, it always expands any
12214 macro invocations present in the expression. @value{GDBN} also provides
12215 the following commands for working with macros explicitly.
12219 @kindex macro expand
12220 @cindex macro expansion, showing the results of preprocessor
12221 @cindex preprocessor macro expansion, showing the results of
12222 @cindex expanding preprocessor macros
12223 @item macro expand @var{expression}
12224 @itemx macro exp @var{expression}
12225 Show the results of expanding all preprocessor macro invocations in
12226 @var{expression}. Since @value{GDBN} simply expands macros, but does
12227 not parse the result, @var{expression} need not be a valid expression;
12228 it can be any string of tokens.
12231 @item macro expand-once @var{expression}
12232 @itemx macro exp1 @var{expression}
12233 @cindex expand macro once
12234 @i{(This command is not yet implemented.)} Show the results of
12235 expanding those preprocessor macro invocations that appear explicitly in
12236 @var{expression}. Macro invocations appearing in that expansion are
12237 left unchanged. This command allows you to see the effect of a
12238 particular macro more clearly, without being confused by further
12239 expansions. Since @value{GDBN} simply expands macros, but does not
12240 parse the result, @var{expression} need not be a valid expression; it
12241 can be any string of tokens.
12244 @cindex macro definition, showing
12245 @cindex definition of a macro, showing
12246 @cindex macros, from debug info
12247 @item info macro [-a|-all] [--] @var{macro}
12248 Show the current definition or all definitions of the named @var{macro},
12249 and describe the source location or compiler command-line where that
12250 definition was established. The optional double dash is to signify the end of
12251 argument processing and the beginning of @var{macro} for non C-like macros where
12252 the macro may begin with a hyphen.
12254 @kindex info macros
12255 @item info macros @var{location}
12256 Show all macro definitions that are in effect at the location specified
12257 by @var{location}, and describe the source location or compiler
12258 command-line where those definitions were established.
12260 @kindex macro define
12261 @cindex user-defined macros
12262 @cindex defining macros interactively
12263 @cindex macros, user-defined
12264 @item macro define @var{macro} @var{replacement-list}
12265 @itemx macro define @var{macro}(@var{arglist}) @var{replacement-list}
12266 Introduce a definition for a preprocessor macro named @var{macro},
12267 invocations of which are replaced by the tokens given in
12268 @var{replacement-list}. The first form of this command defines an
12269 ``object-like'' macro, which takes no arguments; the second form
12270 defines a ``function-like'' macro, which takes the arguments given in
12273 A definition introduced by this command is in scope in every
12274 expression evaluated in @value{GDBN}, until it is removed with the
12275 @code{macro undef} command, described below. The definition overrides
12276 all definitions for @var{macro} present in the program being debugged,
12277 as well as any previous user-supplied definition.
12279 @kindex macro undef
12280 @item macro undef @var{macro}
12281 Remove any user-supplied definition for the macro named @var{macro}.
12282 This command only affects definitions provided with the @code{macro
12283 define} command, described above; it cannot remove definitions present
12284 in the program being debugged.
12288 List all the macros defined using the @code{macro define} command.
12291 @cindex macros, example of debugging with
12292 Here is a transcript showing the above commands in action. First, we
12293 show our source files:
12298 #include "sample.h"
12301 #define ADD(x) (M + x)
12306 printf ("Hello, world!\n");
12308 printf ("We're so creative.\n");
12310 printf ("Goodbye, world!\n");
12317 Now, we compile the program using the @sc{gnu} C compiler,
12318 @value{NGCC}. We pass the @option{-gdwarf-2}@footnote{This is the
12319 minimum. Recent versions of @value{NGCC} support @option{-gdwarf-3}
12320 and @option{-gdwarf-4}; we recommend always choosing the most recent
12321 version of DWARF.} @emph{and} @option{-g3} flags to ensure the compiler
12322 includes information about preprocessor macros in the debugging
12326 $ gcc -gdwarf-2 -g3 sample.c -o sample
12330 Now, we start @value{GDBN} on our sample program:
12334 GNU gdb 2002-05-06-cvs
12335 Copyright 2002 Free Software Foundation, Inc.
12336 GDB is free software, @dots{}
12340 We can expand macros and examine their definitions, even when the
12341 program is not running. @value{GDBN} uses the current listing position
12342 to decide which macro definitions are in scope:
12345 (@value{GDBP}) list main
12348 5 #define ADD(x) (M + x)
12353 10 printf ("Hello, world!\n");
12355 12 printf ("We're so creative.\n");
12356 (@value{GDBP}) info macro ADD
12357 Defined at /home/jimb/gdb/macros/play/sample.c:5
12358 #define ADD(x) (M + x)
12359 (@value{GDBP}) info macro Q
12360 Defined at /home/jimb/gdb/macros/play/sample.h:1
12361 included at /home/jimb/gdb/macros/play/sample.c:2
12363 (@value{GDBP}) macro expand ADD(1)
12364 expands to: (42 + 1)
12365 (@value{GDBP}) macro expand-once ADD(1)
12366 expands to: once (M + 1)
12370 In the example above, note that @code{macro expand-once} expands only
12371 the macro invocation explicit in the original text --- the invocation of
12372 @code{ADD} --- but does not expand the invocation of the macro @code{M},
12373 which was introduced by @code{ADD}.
12375 Once the program is running, @value{GDBN} uses the macro definitions in
12376 force at the source line of the current stack frame:
12379 (@value{GDBP}) break main
12380 Breakpoint 1 at 0x8048370: file sample.c, line 10.
12382 Starting program: /home/jimb/gdb/macros/play/sample
12384 Breakpoint 1, main () at sample.c:10
12385 10 printf ("Hello, world!\n");
12389 At line 10, the definition of the macro @code{N} at line 9 is in force:
12392 (@value{GDBP}) info macro N
12393 Defined at /home/jimb/gdb/macros/play/sample.c:9
12395 (@value{GDBP}) macro expand N Q M
12396 expands to: 28 < 42
12397 (@value{GDBP}) print N Q M
12402 As we step over directives that remove @code{N}'s definition, and then
12403 give it a new definition, @value{GDBN} finds the definition (or lack
12404 thereof) in force at each point:
12407 (@value{GDBP}) next
12409 12 printf ("We're so creative.\n");
12410 (@value{GDBP}) info macro N
12411 The symbol `N' has no definition as a C/C++ preprocessor macro
12412 at /home/jimb/gdb/macros/play/sample.c:12
12413 (@value{GDBP}) next
12415 14 printf ("Goodbye, world!\n");
12416 (@value{GDBP}) info macro N
12417 Defined at /home/jimb/gdb/macros/play/sample.c:13
12419 (@value{GDBP}) macro expand N Q M
12420 expands to: 1729 < 42
12421 (@value{GDBP}) print N Q M
12426 In addition to source files, macros can be defined on the compilation command
12427 line using the @option{-D@var{name}=@var{value}} syntax. For macros defined in
12428 such a way, @value{GDBN} displays the location of their definition as line zero
12429 of the source file submitted to the compiler.
12432 (@value{GDBP}) info macro __STDC__
12433 Defined at /home/jimb/gdb/macros/play/sample.c:0
12440 @chapter Tracepoints
12441 @c This chapter is based on the documentation written by Michael
12442 @c Snyder, David Taylor, Jim Blandy, and Elena Zannoni.
12444 @cindex tracepoints
12445 In some applications, it is not feasible for the debugger to interrupt
12446 the program's execution long enough for the developer to learn
12447 anything helpful about its behavior. If the program's correctness
12448 depends on its real-time behavior, delays introduced by a debugger
12449 might cause the program to change its behavior drastically, or perhaps
12450 fail, even when the code itself is correct. It is useful to be able
12451 to observe the program's behavior without interrupting it.
12453 Using @value{GDBN}'s @code{trace} and @code{collect} commands, you can
12454 specify locations in the program, called @dfn{tracepoints}, and
12455 arbitrary expressions to evaluate when those tracepoints are reached.
12456 Later, using the @code{tfind} command, you can examine the values
12457 those expressions had when the program hit the tracepoints. The
12458 expressions may also denote objects in memory---structures or arrays,
12459 for example---whose values @value{GDBN} should record; while visiting
12460 a particular tracepoint, you may inspect those objects as if they were
12461 in memory at that moment. However, because @value{GDBN} records these
12462 values without interacting with you, it can do so quickly and
12463 unobtrusively, hopefully not disturbing the program's behavior.
12465 The tracepoint facility is currently available only for remote
12466 targets. @xref{Targets}. In addition, your remote target must know
12467 how to collect trace data. This functionality is implemented in the
12468 remote stub; however, none of the stubs distributed with @value{GDBN}
12469 support tracepoints as of this writing. The format of the remote
12470 packets used to implement tracepoints are described in @ref{Tracepoint
12473 It is also possible to get trace data from a file, in a manner reminiscent
12474 of corefiles; you specify the filename, and use @code{tfind} to search
12475 through the file. @xref{Trace Files}, for more details.
12477 This chapter describes the tracepoint commands and features.
12480 * Set Tracepoints::
12481 * Analyze Collected Data::
12482 * Tracepoint Variables::
12486 @node Set Tracepoints
12487 @section Commands to Set Tracepoints
12489 Before running such a @dfn{trace experiment}, an arbitrary number of
12490 tracepoints can be set. A tracepoint is actually a special type of
12491 breakpoint (@pxref{Set Breaks}), so you can manipulate it using
12492 standard breakpoint commands. For instance, as with breakpoints,
12493 tracepoint numbers are successive integers starting from one, and many
12494 of the commands associated with tracepoints take the tracepoint number
12495 as their argument, to identify which tracepoint to work on.
12497 For each tracepoint, you can specify, in advance, some arbitrary set
12498 of data that you want the target to collect in the trace buffer when
12499 it hits that tracepoint. The collected data can include registers,
12500 local variables, or global data. Later, you can use @value{GDBN}
12501 commands to examine the values these data had at the time the
12502 tracepoint was hit.
12504 Tracepoints do not support every breakpoint feature. Ignore counts on
12505 tracepoints have no effect, and tracepoints cannot run @value{GDBN}
12506 commands when they are hit. Tracepoints may not be thread-specific
12509 @cindex fast tracepoints
12510 Some targets may support @dfn{fast tracepoints}, which are inserted in
12511 a different way (such as with a jump instead of a trap), that is
12512 faster but possibly restricted in where they may be installed.
12514 @cindex static tracepoints
12515 @cindex markers, static tracepoints
12516 @cindex probing markers, static tracepoints
12517 Regular and fast tracepoints are dynamic tracing facilities, meaning
12518 that they can be used to insert tracepoints at (almost) any location
12519 in the target. Some targets may also support controlling @dfn{static
12520 tracepoints} from @value{GDBN}. With static tracing, a set of
12521 instrumentation points, also known as @dfn{markers}, are embedded in
12522 the target program, and can be activated or deactivated by name or
12523 address. These are usually placed at locations which facilitate
12524 investigating what the target is actually doing. @value{GDBN}'s
12525 support for static tracing includes being able to list instrumentation
12526 points, and attach them with @value{GDBN} defined high level
12527 tracepoints that expose the whole range of convenience of
12528 @value{GDBN}'s tracepoints support. Namely, support for collecting
12529 registers values and values of global or local (to the instrumentation
12530 point) variables; tracepoint conditions and trace state variables.
12531 The act of installing a @value{GDBN} static tracepoint on an
12532 instrumentation point, or marker, is referred to as @dfn{probing} a
12533 static tracepoint marker.
12535 @code{gdbserver} supports tracepoints on some target systems.
12536 @xref{Server,,Tracepoints support in @code{gdbserver}}.
12538 This section describes commands to set tracepoints and associated
12539 conditions and actions.
12542 * Create and Delete Tracepoints::
12543 * Enable and Disable Tracepoints::
12544 * Tracepoint Passcounts::
12545 * Tracepoint Conditions::
12546 * Trace State Variables::
12547 * Tracepoint Actions::
12548 * Listing Tracepoints::
12549 * Listing Static Tracepoint Markers::
12550 * Starting and Stopping Trace Experiments::
12551 * Tracepoint Restrictions::
12554 @node Create and Delete Tracepoints
12555 @subsection Create and Delete Tracepoints
12558 @cindex set tracepoint
12560 @item trace @var{location}
12561 The @code{trace} command is very similar to the @code{break} command.
12562 Its argument @var{location} can be any valid location.
12563 @xref{Specify Location}. The @code{trace} command defines a tracepoint,
12564 which is a point in the target program where the debugger will briefly stop,
12565 collect some data, and then allow the program to continue. Setting a tracepoint
12566 or changing its actions takes effect immediately if the remote stub
12567 supports the @samp{InstallInTrace} feature (@pxref{install tracepoint
12569 If remote stub doesn't support the @samp{InstallInTrace} feature, all
12570 these changes don't take effect until the next @code{tstart}
12571 command, and once a trace experiment is running, further changes will
12572 not have any effect until the next trace experiment starts. In addition,
12573 @value{GDBN} supports @dfn{pending tracepoints}---tracepoints whose
12574 address is not yet resolved. (This is similar to pending breakpoints.)
12575 Pending tracepoints are not downloaded to the target and not installed
12576 until they are resolved. The resolution of pending tracepoints requires
12577 @value{GDBN} support---when debugging with the remote target, and
12578 @value{GDBN} disconnects from the remote stub (@pxref{disconnected
12579 tracing}), pending tracepoints can not be resolved (and downloaded to
12580 the remote stub) while @value{GDBN} is disconnected.
12582 Here are some examples of using the @code{trace} command:
12585 (@value{GDBP}) @b{trace foo.c:121} // a source file and line number
12587 (@value{GDBP}) @b{trace +2} // 2 lines forward
12589 (@value{GDBP}) @b{trace my_function} // first source line of function
12591 (@value{GDBP}) @b{trace *my_function} // EXACT start address of function
12593 (@value{GDBP}) @b{trace *0x2117c4} // an address
12597 You can abbreviate @code{trace} as @code{tr}.
12599 @item trace @var{location} if @var{cond}
12600 Set a tracepoint with condition @var{cond}; evaluate the expression
12601 @var{cond} each time the tracepoint is reached, and collect data only
12602 if the value is nonzero---that is, if @var{cond} evaluates as true.
12603 @xref{Tracepoint Conditions, ,Tracepoint Conditions}, for more
12604 information on tracepoint conditions.
12606 @item ftrace @var{location} [ if @var{cond} ]
12607 @cindex set fast tracepoint
12608 @cindex fast tracepoints, setting
12610 The @code{ftrace} command sets a fast tracepoint. For targets that
12611 support them, fast tracepoints will use a more efficient but possibly
12612 less general technique to trigger data collection, such as a jump
12613 instruction instead of a trap, or some sort of hardware support. It
12614 may not be possible to create a fast tracepoint at the desired
12615 location, in which case the command will exit with an explanatory
12618 @value{GDBN} handles arguments to @code{ftrace} exactly as for
12621 On 32-bit x86-architecture systems, fast tracepoints normally need to
12622 be placed at an instruction that is 5 bytes or longer, but can be
12623 placed at 4-byte instructions if the low 64K of memory of the target
12624 program is available to install trampolines. Some Unix-type systems,
12625 such as @sc{gnu}/Linux, exclude low addresses from the program's
12626 address space; but for instance with the Linux kernel it is possible
12627 to let @value{GDBN} use this area by doing a @command{sysctl} command
12628 to set the @code{mmap_min_addr} kernel parameter, as in
12631 sudo sysctl -w vm.mmap_min_addr=32768
12635 which sets the low address to 32K, which leaves plenty of room for
12636 trampolines. The minimum address should be set to a page boundary.
12638 @item strace @var{location} [ if @var{cond} ]
12639 @cindex set static tracepoint
12640 @cindex static tracepoints, setting
12641 @cindex probe static tracepoint marker
12643 The @code{strace} command sets a static tracepoint. For targets that
12644 support it, setting a static tracepoint probes a static
12645 instrumentation point, or marker, found at @var{location}. It may not
12646 be possible to set a static tracepoint at the desired location, in
12647 which case the command will exit with an explanatory message.
12649 @value{GDBN} handles arguments to @code{strace} exactly as for
12650 @code{trace}, with the addition that the user can also specify
12651 @code{-m @var{marker}} as @var{location}. This probes the marker
12652 identified by the @var{marker} string identifier. This identifier
12653 depends on the static tracepoint backend library your program is
12654 using. You can find all the marker identifiers in the @samp{ID} field
12655 of the @code{info static-tracepoint-markers} command output.
12656 @xref{Listing Static Tracepoint Markers,,Listing Static Tracepoint
12657 Markers}. For example, in the following small program using the UST
12663 trace_mark(ust, bar33, "str %s", "FOOBAZ");
12668 the marker id is composed of joining the first two arguments to the
12669 @code{trace_mark} call with a slash, which translates to:
12672 (@value{GDBP}) info static-tracepoint-markers
12673 Cnt Enb ID Address What
12674 1 n ust/bar33 0x0000000000400ddc in main at stexample.c:22
12680 so you may probe the marker above with:
12683 (@value{GDBP}) strace -m ust/bar33
12686 Static tracepoints accept an extra collect action --- @code{collect
12687 $_sdata}. This collects arbitrary user data passed in the probe point
12688 call to the tracing library. In the UST example above, you'll see
12689 that the third argument to @code{trace_mark} is a printf-like format
12690 string. The user data is then the result of running that formating
12691 string against the following arguments. Note that @code{info
12692 static-tracepoint-markers} command output lists that format string in
12693 the @samp{Data:} field.
12695 You can inspect this data when analyzing the trace buffer, by printing
12696 the $_sdata variable like any other variable available to
12697 @value{GDBN}. @xref{Tracepoint Actions,,Tracepoint Action Lists}.
12700 @cindex last tracepoint number
12701 @cindex recent tracepoint number
12702 @cindex tracepoint number
12703 The convenience variable @code{$tpnum} records the tracepoint number
12704 of the most recently set tracepoint.
12706 @kindex delete tracepoint
12707 @cindex tracepoint deletion
12708 @item delete tracepoint @r{[}@var{num}@r{]}
12709 Permanently delete one or more tracepoints. With no argument, the
12710 default is to delete all tracepoints. Note that the regular
12711 @code{delete} command can remove tracepoints also.
12716 (@value{GDBP}) @b{delete trace 1 2 3} // remove three tracepoints
12718 (@value{GDBP}) @b{delete trace} // remove all tracepoints
12722 You can abbreviate this command as @code{del tr}.
12725 @node Enable and Disable Tracepoints
12726 @subsection Enable and Disable Tracepoints
12728 These commands are deprecated; they are equivalent to plain @code{disable} and @code{enable}.
12731 @kindex disable tracepoint
12732 @item disable tracepoint @r{[}@var{num}@r{]}
12733 Disable tracepoint @var{num}, or all tracepoints if no argument
12734 @var{num} is given. A disabled tracepoint will have no effect during
12735 a trace experiment, but it is not forgotten. You can re-enable
12736 a disabled tracepoint using the @code{enable tracepoint} command.
12737 If the command is issued during a trace experiment and the debug target
12738 has support for disabling tracepoints during a trace experiment, then the
12739 change will be effective immediately. Otherwise, it will be applied to the
12740 next trace experiment.
12742 @kindex enable tracepoint
12743 @item enable tracepoint @r{[}@var{num}@r{]}
12744 Enable tracepoint @var{num}, or all tracepoints. If this command is
12745 issued during a trace experiment and the debug target supports enabling
12746 tracepoints during a trace experiment, then the enabled tracepoints will
12747 become effective immediately. Otherwise, they will become effective the
12748 next time a trace experiment is run.
12751 @node Tracepoint Passcounts
12752 @subsection Tracepoint Passcounts
12756 @cindex tracepoint pass count
12757 @item passcount @r{[}@var{n} @r{[}@var{num}@r{]]}
12758 Set the @dfn{passcount} of a tracepoint. The passcount is a way to
12759 automatically stop a trace experiment. If a tracepoint's passcount is
12760 @var{n}, then the trace experiment will be automatically stopped on
12761 the @var{n}'th time that tracepoint is hit. If the tracepoint number
12762 @var{num} is not specified, the @code{passcount} command sets the
12763 passcount of the most recently defined tracepoint. If no passcount is
12764 given, the trace experiment will run until stopped explicitly by the
12770 (@value{GDBP}) @b{passcount 5 2} // Stop on the 5th execution of
12771 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// tracepoint 2}
12773 (@value{GDBP}) @b{passcount 12} // Stop on the 12th execution of the
12774 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// most recently defined tracepoint.}
12775 (@value{GDBP}) @b{trace foo}
12776 (@value{GDBP}) @b{pass 3}
12777 (@value{GDBP}) @b{trace bar}
12778 (@value{GDBP}) @b{pass 2}
12779 (@value{GDBP}) @b{trace baz}
12780 (@value{GDBP}) @b{pass 1} // Stop tracing when foo has been
12781 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// executed 3 times OR when bar has}
12782 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// been executed 2 times}
12783 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// OR when baz has been executed 1 time.}
12787 @node Tracepoint Conditions
12788 @subsection Tracepoint Conditions
12789 @cindex conditional tracepoints
12790 @cindex tracepoint conditions
12792 The simplest sort of tracepoint collects data every time your program
12793 reaches a specified place. You can also specify a @dfn{condition} for
12794 a tracepoint. A condition is just a Boolean expression in your
12795 programming language (@pxref{Expressions, ,Expressions}). A
12796 tracepoint with a condition evaluates the expression each time your
12797 program reaches it, and data collection happens only if the condition
12800 Tracepoint conditions can be specified when a tracepoint is set, by
12801 using @samp{if} in the arguments to the @code{trace} command.
12802 @xref{Create and Delete Tracepoints, ,Setting Tracepoints}. They can
12803 also be set or changed at any time with the @code{condition} command,
12804 just as with breakpoints.
12806 Unlike breakpoint conditions, @value{GDBN} does not actually evaluate
12807 the conditional expression itself. Instead, @value{GDBN} encodes the
12808 expression into an agent expression (@pxref{Agent Expressions})
12809 suitable for execution on the target, independently of @value{GDBN}.
12810 Global variables become raw memory locations, locals become stack
12811 accesses, and so forth.
12813 For instance, suppose you have a function that is usually called
12814 frequently, but should not be called after an error has occurred. You
12815 could use the following tracepoint command to collect data about calls
12816 of that function that happen while the error code is propagating
12817 through the program; an unconditional tracepoint could end up
12818 collecting thousands of useless trace frames that you would have to
12822 (@value{GDBP}) @kbd{trace normal_operation if errcode > 0}
12825 @node Trace State Variables
12826 @subsection Trace State Variables
12827 @cindex trace state variables
12829 A @dfn{trace state variable} is a special type of variable that is
12830 created and managed by target-side code. The syntax is the same as
12831 that for GDB's convenience variables (a string prefixed with ``$''),
12832 but they are stored on the target. They must be created explicitly,
12833 using a @code{tvariable} command. They are always 64-bit signed
12836 Trace state variables are remembered by @value{GDBN}, and downloaded
12837 to the target along with tracepoint information when the trace
12838 experiment starts. There are no intrinsic limits on the number of
12839 trace state variables, beyond memory limitations of the target.
12841 @cindex convenience variables, and trace state variables
12842 Although trace state variables are managed by the target, you can use
12843 them in print commands and expressions as if they were convenience
12844 variables; @value{GDBN} will get the current value from the target
12845 while the trace experiment is running. Trace state variables share
12846 the same namespace as other ``$'' variables, which means that you
12847 cannot have trace state variables with names like @code{$23} or
12848 @code{$pc}, nor can you have a trace state variable and a convenience
12849 variable with the same name.
12853 @item tvariable $@var{name} [ = @var{expression} ]
12855 The @code{tvariable} command creates a new trace state variable named
12856 @code{$@var{name}}, and optionally gives it an initial value of
12857 @var{expression}. The @var{expression} is evaluated when this command is
12858 entered; the result will be converted to an integer if possible,
12859 otherwise @value{GDBN} will report an error. A subsequent
12860 @code{tvariable} command specifying the same name does not create a
12861 variable, but instead assigns the supplied initial value to the
12862 existing variable of that name, overwriting any previous initial
12863 value. The default initial value is 0.
12865 @item info tvariables
12866 @kindex info tvariables
12867 List all the trace state variables along with their initial values.
12868 Their current values may also be displayed, if the trace experiment is
12871 @item delete tvariable @r{[} $@var{name} @dots{} @r{]}
12872 @kindex delete tvariable
12873 Delete the given trace state variables, or all of them if no arguments
12878 @node Tracepoint Actions
12879 @subsection Tracepoint Action Lists
12883 @cindex tracepoint actions
12884 @item actions @r{[}@var{num}@r{]}
12885 This command will prompt for a list of actions to be taken when the
12886 tracepoint is hit. If the tracepoint number @var{num} is not
12887 specified, this command sets the actions for the one that was most
12888 recently defined (so that you can define a tracepoint and then say
12889 @code{actions} without bothering about its number). You specify the
12890 actions themselves on the following lines, one action at a time, and
12891 terminate the actions list with a line containing just @code{end}. So
12892 far, the only defined actions are @code{collect}, @code{teval}, and
12893 @code{while-stepping}.
12895 @code{actions} is actually equivalent to @code{commands} (@pxref{Break
12896 Commands, ,Breakpoint Command Lists}), except that only the defined
12897 actions are allowed; any other @value{GDBN} command is rejected.
12899 @cindex remove actions from a tracepoint
12900 To remove all actions from a tracepoint, type @samp{actions @var{num}}
12901 and follow it immediately with @samp{end}.
12904 (@value{GDBP}) @b{collect @var{data}} // collect some data
12906 (@value{GDBP}) @b{while-stepping 5} // single-step 5 times, collect data
12908 (@value{GDBP}) @b{end} // signals the end of actions.
12911 In the following example, the action list begins with @code{collect}
12912 commands indicating the things to be collected when the tracepoint is
12913 hit. Then, in order to single-step and collect additional data
12914 following the tracepoint, a @code{while-stepping} command is used,
12915 followed by the list of things to be collected after each step in a
12916 sequence of single steps. The @code{while-stepping} command is
12917 terminated by its own separate @code{end} command. Lastly, the action
12918 list is terminated by an @code{end} command.
12921 (@value{GDBP}) @b{trace foo}
12922 (@value{GDBP}) @b{actions}
12923 Enter actions for tracepoint 1, one per line:
12926 > while-stepping 12
12927 > collect $pc, arr[i]
12932 @kindex collect @r{(tracepoints)}
12933 @item collect@r{[}/@var{mods}@r{]} @var{expr1}, @var{expr2}, @dots{}
12934 Collect values of the given expressions when the tracepoint is hit.
12935 This command accepts a comma-separated list of any valid expressions.
12936 In addition to global, static, or local variables, the following
12937 special arguments are supported:
12941 Collect all registers.
12944 Collect all function arguments.
12947 Collect all local variables.
12950 Collect the return address. This is helpful if you want to see more
12953 @emph{Note:} The return address location can not always be reliably
12954 determined up front, and the wrong address / registers may end up
12955 collected instead. On some architectures the reliability is higher
12956 for tracepoints at function entry, while on others it's the opposite.
12957 When this happens, backtracing will stop because the return address is
12958 found unavailable (unless another collect rule happened to match it).
12961 Collects the number of arguments from the static probe at which the
12962 tracepoint is located.
12963 @xref{Static Probe Points}.
12965 @item $_probe_arg@var{n}
12966 @var{n} is an integer between 0 and 11. Collects the @var{n}th argument
12967 from the static probe at which the tracepoint is located.
12968 @xref{Static Probe Points}.
12971 @vindex $_sdata@r{, collect}
12972 Collect static tracepoint marker specific data. Only available for
12973 static tracepoints. @xref{Tracepoint Actions,,Tracepoint Action
12974 Lists}. On the UST static tracepoints library backend, an
12975 instrumentation point resembles a @code{printf} function call. The
12976 tracing library is able to collect user specified data formatted to a
12977 character string using the format provided by the programmer that
12978 instrumented the program. Other backends have similar mechanisms.
12979 Here's an example of a UST marker call:
12982 const char master_name[] = "$your_name";
12983 trace_mark(channel1, marker1, "hello %s", master_name)
12986 In this case, collecting @code{$_sdata} collects the string
12987 @samp{hello $yourname}. When analyzing the trace buffer, you can
12988 inspect @samp{$_sdata} like any other variable available to
12992 You can give several consecutive @code{collect} commands, each one
12993 with a single argument, or one @code{collect} command with several
12994 arguments separated by commas; the effect is the same.
12996 The optional @var{mods} changes the usual handling of the arguments.
12997 @code{s} requests that pointers to chars be handled as strings, in
12998 particular collecting the contents of the memory being pointed at, up
12999 to the first zero. The upper bound is by default the value of the
13000 @code{print elements} variable; if @code{s} is followed by a decimal
13001 number, that is the upper bound instead. So for instance
13002 @samp{collect/s25 mystr} collects as many as 25 characters at
13005 The command @code{info scope} (@pxref{Symbols, info scope}) is
13006 particularly useful for figuring out what data to collect.
13008 @kindex teval @r{(tracepoints)}
13009 @item teval @var{expr1}, @var{expr2}, @dots{}
13010 Evaluate the given expressions when the tracepoint is hit. This
13011 command accepts a comma-separated list of expressions. The results
13012 are discarded, so this is mainly useful for assigning values to trace
13013 state variables (@pxref{Trace State Variables}) without adding those
13014 values to the trace buffer, as would be the case if the @code{collect}
13017 @kindex while-stepping @r{(tracepoints)}
13018 @item while-stepping @var{n}
13019 Perform @var{n} single-step instruction traces after the tracepoint,
13020 collecting new data after each step. The @code{while-stepping}
13021 command is followed by the list of what to collect while stepping
13022 (followed by its own @code{end} command):
13025 > while-stepping 12
13026 > collect $regs, myglobal
13032 Note that @code{$pc} is not automatically collected by
13033 @code{while-stepping}; you need to explicitly collect that register if
13034 you need it. You may abbreviate @code{while-stepping} as @code{ws} or
13037 @item set default-collect @var{expr1}, @var{expr2}, @dots{}
13038 @kindex set default-collect
13039 @cindex default collection action
13040 This variable is a list of expressions to collect at each tracepoint
13041 hit. It is effectively an additional @code{collect} action prepended
13042 to every tracepoint action list. The expressions are parsed
13043 individually for each tracepoint, so for instance a variable named
13044 @code{xyz} may be interpreted as a global for one tracepoint, and a
13045 local for another, as appropriate to the tracepoint's location.
13047 @item show default-collect
13048 @kindex show default-collect
13049 Show the list of expressions that are collected by default at each
13054 @node Listing Tracepoints
13055 @subsection Listing Tracepoints
13058 @kindex info tracepoints @r{[}@var{n}@dots{}@r{]}
13059 @kindex info tp @r{[}@var{n}@dots{}@r{]}
13060 @cindex information about tracepoints
13061 @item info tracepoints @r{[}@var{num}@dots{}@r{]}
13062 Display information about the tracepoint @var{num}. If you don't
13063 specify a tracepoint number, displays information about all the
13064 tracepoints defined so far. The format is similar to that used for
13065 @code{info breakpoints}; in fact, @code{info tracepoints} is the same
13066 command, simply restricting itself to tracepoints.
13068 A tracepoint's listing may include additional information specific to
13073 its passcount as given by the @code{passcount @var{n}} command
13076 the state about installed on target of each location
13080 (@value{GDBP}) @b{info trace}
13081 Num Type Disp Enb Address What
13082 1 tracepoint keep y 0x0804ab57 in foo() at main.cxx:7
13084 collect globfoo, $regs
13089 2 tracepoint keep y <MULTIPLE>
13091 2.1 y 0x0804859c in func4 at change-loc.h:35
13092 installed on target
13093 2.2 y 0xb7ffc480 in func4 at change-loc.h:35
13094 installed on target
13095 2.3 y <PENDING> set_tracepoint
13096 3 tracepoint keep y 0x080485b1 in foo at change-loc.c:29
13097 not installed on target
13102 This command can be abbreviated @code{info tp}.
13105 @node Listing Static Tracepoint Markers
13106 @subsection Listing Static Tracepoint Markers
13109 @kindex info static-tracepoint-markers
13110 @cindex information about static tracepoint markers
13111 @item info static-tracepoint-markers
13112 Display information about all static tracepoint markers defined in the
13115 For each marker, the following columns are printed:
13119 An incrementing counter, output to help readability. This is not a
13122 The marker ID, as reported by the target.
13123 @item Enabled or Disabled
13124 Probed markers are tagged with @samp{y}. @samp{n} identifies marks
13125 that are not enabled.
13127 Where the marker is in your program, as a memory address.
13129 Where the marker is in the source for your program, as a file and line
13130 number. If the debug information included in the program does not
13131 allow @value{GDBN} to locate the source of the marker, this column
13132 will be left blank.
13136 In addition, the following information may be printed for each marker:
13140 User data passed to the tracing library by the marker call. In the
13141 UST backend, this is the format string passed as argument to the
13143 @item Static tracepoints probing the marker
13144 The list of static tracepoints attached to the marker.
13148 (@value{GDBP}) info static-tracepoint-markers
13149 Cnt ID Enb Address What
13150 1 ust/bar2 y 0x0000000000400e1a in main at stexample.c:25
13151 Data: number1 %d number2 %d
13152 Probed by static tracepoints: #2
13153 2 ust/bar33 n 0x0000000000400c87 in main at stexample.c:24
13159 @node Starting and Stopping Trace Experiments
13160 @subsection Starting and Stopping Trace Experiments
13163 @kindex tstart [ @var{notes} ]
13164 @cindex start a new trace experiment
13165 @cindex collected data discarded
13167 This command starts the trace experiment, and begins collecting data.
13168 It has the side effect of discarding all the data collected in the
13169 trace buffer during the previous trace experiment. If any arguments
13170 are supplied, they are taken as a note and stored with the trace
13171 experiment's state. The notes may be arbitrary text, and are
13172 especially useful with disconnected tracing in a multi-user context;
13173 the notes can explain what the trace is doing, supply user contact
13174 information, and so forth.
13176 @kindex tstop [ @var{notes} ]
13177 @cindex stop a running trace experiment
13179 This command stops the trace experiment. If any arguments are
13180 supplied, they are recorded with the experiment as a note. This is
13181 useful if you are stopping a trace started by someone else, for
13182 instance if the trace is interfering with the system's behavior and
13183 needs to be stopped quickly.
13185 @strong{Note}: a trace experiment and data collection may stop
13186 automatically if any tracepoint's passcount is reached
13187 (@pxref{Tracepoint Passcounts}), or if the trace buffer becomes full.
13190 @cindex status of trace data collection
13191 @cindex trace experiment, status of
13193 This command displays the status of the current trace data
13197 Here is an example of the commands we described so far:
13200 (@value{GDBP}) @b{trace gdb_c_test}
13201 (@value{GDBP}) @b{actions}
13202 Enter actions for tracepoint #1, one per line.
13203 > collect $regs,$locals,$args
13204 > while-stepping 11
13208 (@value{GDBP}) @b{tstart}
13209 [time passes @dots{}]
13210 (@value{GDBP}) @b{tstop}
13213 @anchor{disconnected tracing}
13214 @cindex disconnected tracing
13215 You can choose to continue running the trace experiment even if
13216 @value{GDBN} disconnects from the target, voluntarily or
13217 involuntarily. For commands such as @code{detach}, the debugger will
13218 ask what you want to do with the trace. But for unexpected
13219 terminations (@value{GDBN} crash, network outage), it would be
13220 unfortunate to lose hard-won trace data, so the variable
13221 @code{disconnected-tracing} lets you decide whether the trace should
13222 continue running without @value{GDBN}.
13225 @item set disconnected-tracing on
13226 @itemx set disconnected-tracing off
13227 @kindex set disconnected-tracing
13228 Choose whether a tracing run should continue to run if @value{GDBN}
13229 has disconnected from the target. Note that @code{detach} or
13230 @code{quit} will ask you directly what to do about a running trace no
13231 matter what this variable's setting, so the variable is mainly useful
13232 for handling unexpected situations, such as loss of the network.
13234 @item show disconnected-tracing
13235 @kindex show disconnected-tracing
13236 Show the current choice for disconnected tracing.
13240 When you reconnect to the target, the trace experiment may or may not
13241 still be running; it might have filled the trace buffer in the
13242 meantime, or stopped for one of the other reasons. If it is running,
13243 it will continue after reconnection.
13245 Upon reconnection, the target will upload information about the
13246 tracepoints in effect. @value{GDBN} will then compare that
13247 information to the set of tracepoints currently defined, and attempt
13248 to match them up, allowing for the possibility that the numbers may
13249 have changed due to creation and deletion in the meantime. If one of
13250 the target's tracepoints does not match any in @value{GDBN}, the
13251 debugger will create a new tracepoint, so that you have a number with
13252 which to specify that tracepoint. This matching-up process is
13253 necessarily heuristic, and it may result in useless tracepoints being
13254 created; you may simply delete them if they are of no use.
13256 @cindex circular trace buffer
13257 If your target agent supports a @dfn{circular trace buffer}, then you
13258 can run a trace experiment indefinitely without filling the trace
13259 buffer; when space runs out, the agent deletes already-collected trace
13260 frames, oldest first, until there is enough room to continue
13261 collecting. This is especially useful if your tracepoints are being
13262 hit too often, and your trace gets terminated prematurely because the
13263 buffer is full. To ask for a circular trace buffer, simply set
13264 @samp{circular-trace-buffer} to on. You can set this at any time,
13265 including during tracing; if the agent can do it, it will change
13266 buffer handling on the fly, otherwise it will not take effect until
13270 @item set circular-trace-buffer on
13271 @itemx set circular-trace-buffer off
13272 @kindex set circular-trace-buffer
13273 Choose whether a tracing run should use a linear or circular buffer
13274 for trace data. A linear buffer will not lose any trace data, but may
13275 fill up prematurely, while a circular buffer will discard old trace
13276 data, but it will have always room for the latest tracepoint hits.
13278 @item show circular-trace-buffer
13279 @kindex show circular-trace-buffer
13280 Show the current choice for the trace buffer. Note that this may not
13281 match the agent's current buffer handling, nor is it guaranteed to
13282 match the setting that might have been in effect during a past run,
13283 for instance if you are looking at frames from a trace file.
13288 @item set trace-buffer-size @var{n}
13289 @itemx set trace-buffer-size unlimited
13290 @kindex set trace-buffer-size
13291 Request that the target use a trace buffer of @var{n} bytes. Not all
13292 targets will honor the request; they may have a compiled-in size for
13293 the trace buffer, or some other limitation. Set to a value of
13294 @code{unlimited} or @code{-1} to let the target use whatever size it
13295 likes. This is also the default.
13297 @item show trace-buffer-size
13298 @kindex show trace-buffer-size
13299 Show the current requested size for the trace buffer. Note that this
13300 will only match the actual size if the target supports size-setting,
13301 and was able to handle the requested size. For instance, if the
13302 target can only change buffer size between runs, this variable will
13303 not reflect the change until the next run starts. Use @code{tstatus}
13304 to get a report of the actual buffer size.
13308 @item set trace-user @var{text}
13309 @kindex set trace-user
13311 @item show trace-user
13312 @kindex show trace-user
13314 @item set trace-notes @var{text}
13315 @kindex set trace-notes
13316 Set the trace run's notes.
13318 @item show trace-notes
13319 @kindex show trace-notes
13320 Show the trace run's notes.
13322 @item set trace-stop-notes @var{text}
13323 @kindex set trace-stop-notes
13324 Set the trace run's stop notes. The handling of the note is as for
13325 @code{tstop} arguments; the set command is convenient way to fix a
13326 stop note that is mistaken or incomplete.
13328 @item show trace-stop-notes
13329 @kindex show trace-stop-notes
13330 Show the trace run's stop notes.
13334 @node Tracepoint Restrictions
13335 @subsection Tracepoint Restrictions
13337 @cindex tracepoint restrictions
13338 There are a number of restrictions on the use of tracepoints. As
13339 described above, tracepoint data gathering occurs on the target
13340 without interaction from @value{GDBN}. Thus the full capabilities of
13341 the debugger are not available during data gathering, and then at data
13342 examination time, you will be limited by only having what was
13343 collected. The following items describe some common problems, but it
13344 is not exhaustive, and you may run into additional difficulties not
13350 Tracepoint expressions are intended to gather objects (lvalues). Thus
13351 the full flexibility of GDB's expression evaluator is not available.
13352 You cannot call functions, cast objects to aggregate types, access
13353 convenience variables or modify values (except by assignment to trace
13354 state variables). Some language features may implicitly call
13355 functions (for instance Objective-C fields with accessors), and therefore
13356 cannot be collected either.
13359 Collection of local variables, either individually or in bulk with
13360 @code{$locals} or @code{$args}, during @code{while-stepping} may
13361 behave erratically. The stepping action may enter a new scope (for
13362 instance by stepping into a function), or the location of the variable
13363 may change (for instance it is loaded into a register). The
13364 tracepoint data recorded uses the location information for the
13365 variables that is correct for the tracepoint location. When the
13366 tracepoint is created, it is not possible, in general, to determine
13367 where the steps of a @code{while-stepping} sequence will advance the
13368 program---particularly if a conditional branch is stepped.
13371 Collection of an incompletely-initialized or partially-destroyed object
13372 may result in something that @value{GDBN} cannot display, or displays
13373 in a misleading way.
13376 When @value{GDBN} displays a pointer to character it automatically
13377 dereferences the pointer to also display characters of the string
13378 being pointed to. However, collecting the pointer during tracing does
13379 not automatically collect the string. You need to explicitly
13380 dereference the pointer and provide size information if you want to
13381 collect not only the pointer, but the memory pointed to. For example,
13382 @code{*ptr@@50} can be used to collect the 50 element array pointed to
13386 It is not possible to collect a complete stack backtrace at a
13387 tracepoint. Instead, you may collect the registers and a few hundred
13388 bytes from the stack pointer with something like @code{*(unsigned char *)$esp@@300}
13389 (adjust to use the name of the actual stack pointer register on your
13390 target architecture, and the amount of stack you wish to capture).
13391 Then the @code{backtrace} command will show a partial backtrace when
13392 using a trace frame. The number of stack frames that can be examined
13393 depends on the sizes of the frames in the collected stack. Note that
13394 if you ask for a block so large that it goes past the bottom of the
13395 stack, the target agent may report an error trying to read from an
13399 If you do not collect registers at a tracepoint, @value{GDBN} can
13400 infer that the value of @code{$pc} must be the same as the address of
13401 the tracepoint and use that when you are looking at a trace frame
13402 for that tracepoint. However, this cannot work if the tracepoint has
13403 multiple locations (for instance if it was set in a function that was
13404 inlined), or if it has a @code{while-stepping} loop. In those cases
13405 @value{GDBN} will warn you that it can't infer @code{$pc}, and default
13410 @node Analyze Collected Data
13411 @section Using the Collected Data
13413 After the tracepoint experiment ends, you use @value{GDBN} commands
13414 for examining the trace data. The basic idea is that each tracepoint
13415 collects a trace @dfn{snapshot} every time it is hit and another
13416 snapshot every time it single-steps. All these snapshots are
13417 consecutively numbered from zero and go into a buffer, and you can
13418 examine them later. The way you examine them is to @dfn{focus} on a
13419 specific trace snapshot. When the remote stub is focused on a trace
13420 snapshot, it will respond to all @value{GDBN} requests for memory and
13421 registers by reading from the buffer which belongs to that snapshot,
13422 rather than from @emph{real} memory or registers of the program being
13423 debugged. This means that @strong{all} @value{GDBN} commands
13424 (@code{print}, @code{info registers}, @code{backtrace}, etc.) will
13425 behave as if we were currently debugging the program state as it was
13426 when the tracepoint occurred. Any requests for data that are not in
13427 the buffer will fail.
13430 * tfind:: How to select a trace snapshot
13431 * tdump:: How to display all data for a snapshot
13432 * save tracepoints:: How to save tracepoints for a future run
13436 @subsection @code{tfind @var{n}}
13439 @cindex select trace snapshot
13440 @cindex find trace snapshot
13441 The basic command for selecting a trace snapshot from the buffer is
13442 @code{tfind @var{n}}, which finds trace snapshot number @var{n},
13443 counting from zero. If no argument @var{n} is given, the next
13444 snapshot is selected.
13446 Here are the various forms of using the @code{tfind} command.
13450 Find the first snapshot in the buffer. This is a synonym for
13451 @code{tfind 0} (since 0 is the number of the first snapshot).
13454 Stop debugging trace snapshots, resume @emph{live} debugging.
13457 Same as @samp{tfind none}.
13460 No argument means find the next trace snapshot or find the first
13461 one if no trace snapshot is selected.
13464 Find the previous trace snapshot before the current one. This permits
13465 retracing earlier steps.
13467 @item tfind tracepoint @var{num}
13468 Find the next snapshot associated with tracepoint @var{num}. Search
13469 proceeds forward from the last examined trace snapshot. If no
13470 argument @var{num} is given, it means find the next snapshot collected
13471 for the same tracepoint as the current snapshot.
13473 @item tfind pc @var{addr}
13474 Find the next snapshot associated with the value @var{addr} of the
13475 program counter. Search proceeds forward from the last examined trace
13476 snapshot. If no argument @var{addr} is given, it means find the next
13477 snapshot with the same value of PC as the current snapshot.
13479 @item tfind outside @var{addr1}, @var{addr2}
13480 Find the next snapshot whose PC is outside the given range of
13481 addresses (exclusive).
13483 @item tfind range @var{addr1}, @var{addr2}
13484 Find the next snapshot whose PC is between @var{addr1} and
13485 @var{addr2} (inclusive).
13487 @item tfind line @r{[}@var{file}:@r{]}@var{n}
13488 Find the next snapshot associated with the source line @var{n}. If
13489 the optional argument @var{file} is given, refer to line @var{n} in
13490 that source file. Search proceeds forward from the last examined
13491 trace snapshot. If no argument @var{n} is given, it means find the
13492 next line other than the one currently being examined; thus saying
13493 @code{tfind line} repeatedly can appear to have the same effect as
13494 stepping from line to line in a @emph{live} debugging session.
13497 The default arguments for the @code{tfind} commands are specifically
13498 designed to make it easy to scan through the trace buffer. For
13499 instance, @code{tfind} with no argument selects the next trace
13500 snapshot, and @code{tfind -} with no argument selects the previous
13501 trace snapshot. So, by giving one @code{tfind} command, and then
13502 simply hitting @key{RET} repeatedly you can examine all the trace
13503 snapshots in order. Or, by saying @code{tfind -} and then hitting
13504 @key{RET} repeatedly you can examine the snapshots in reverse order.
13505 The @code{tfind line} command with no argument selects the snapshot
13506 for the next source line executed. The @code{tfind pc} command with
13507 no argument selects the next snapshot with the same program counter
13508 (PC) as the current frame. The @code{tfind tracepoint} command with
13509 no argument selects the next trace snapshot collected by the same
13510 tracepoint as the current one.
13512 In addition to letting you scan through the trace buffer manually,
13513 these commands make it easy to construct @value{GDBN} scripts that
13514 scan through the trace buffer and print out whatever collected data
13515 you are interested in. Thus, if we want to examine the PC, FP, and SP
13516 registers from each trace frame in the buffer, we can say this:
13519 (@value{GDBP}) @b{tfind start}
13520 (@value{GDBP}) @b{while ($trace_frame != -1)}
13521 > printf "Frame %d, PC = %08X, SP = %08X, FP = %08X\n", \
13522 $trace_frame, $pc, $sp, $fp
13526 Frame 0, PC = 0020DC64, SP = 0030BF3C, FP = 0030BF44
13527 Frame 1, PC = 0020DC6C, SP = 0030BF38, FP = 0030BF44
13528 Frame 2, PC = 0020DC70, SP = 0030BF34, FP = 0030BF44
13529 Frame 3, PC = 0020DC74, SP = 0030BF30, FP = 0030BF44
13530 Frame 4, PC = 0020DC78, SP = 0030BF2C, FP = 0030BF44
13531 Frame 5, PC = 0020DC7C, SP = 0030BF28, FP = 0030BF44
13532 Frame 6, PC = 0020DC80, SP = 0030BF24, FP = 0030BF44
13533 Frame 7, PC = 0020DC84, SP = 0030BF20, FP = 0030BF44
13534 Frame 8, PC = 0020DC88, SP = 0030BF1C, FP = 0030BF44
13535 Frame 9, PC = 0020DC8E, SP = 0030BF18, FP = 0030BF44
13536 Frame 10, PC = 00203F6C, SP = 0030BE3C, FP = 0030BF14
13539 Or, if we want to examine the variable @code{X} at each source line in
13543 (@value{GDBP}) @b{tfind start}
13544 (@value{GDBP}) @b{while ($trace_frame != -1)}
13545 > printf "Frame %d, X == %d\n", $trace_frame, X
13555 @subsection @code{tdump}
13557 @cindex dump all data collected at tracepoint
13558 @cindex tracepoint data, display
13560 This command takes no arguments. It prints all the data collected at
13561 the current trace snapshot.
13564 (@value{GDBP}) @b{trace 444}
13565 (@value{GDBP}) @b{actions}
13566 Enter actions for tracepoint #2, one per line:
13567 > collect $regs, $locals, $args, gdb_long_test
13570 (@value{GDBP}) @b{tstart}
13572 (@value{GDBP}) @b{tfind line 444}
13573 #0 gdb_test (p1=0x11, p2=0x22, p3=0x33, p4=0x44, p5=0x55, p6=0x66)
13575 444 printp( "%s: arguments = 0x%X 0x%X 0x%X 0x%X 0x%X 0x%X\n", )
13577 (@value{GDBP}) @b{tdump}
13578 Data collected at tracepoint 2, trace frame 1:
13579 d0 0xc4aa0085 -995491707
13583 d4 0x71aea3d 119204413
13586 d7 0x380035 3670069
13587 a0 0x19e24a 1696330
13588 a1 0x3000668 50333288
13590 a3 0x322000 3284992
13591 a4 0x3000698 50333336
13592 a5 0x1ad3cc 1758156
13593 fp 0x30bf3c 0x30bf3c
13594 sp 0x30bf34 0x30bf34
13596 pc 0x20b2c8 0x20b2c8
13600 p = 0x20e5b4 "gdb-test"
13607 gdb_long_test = 17 '\021'
13612 @code{tdump} works by scanning the tracepoint's current collection
13613 actions and printing the value of each expression listed. So
13614 @code{tdump} can fail, if after a run, you change the tracepoint's
13615 actions to mention variables that were not collected during the run.
13617 Also, for tracepoints with @code{while-stepping} loops, @code{tdump}
13618 uses the collected value of @code{$pc} to distinguish between trace
13619 frames that were collected at the tracepoint hit, and frames that were
13620 collected while stepping. This allows it to correctly choose whether
13621 to display the basic list of collections, or the collections from the
13622 body of the while-stepping loop. However, if @code{$pc} was not collected,
13623 then @code{tdump} will always attempt to dump using the basic collection
13624 list, and may fail if a while-stepping frame does not include all the
13625 same data that is collected at the tracepoint hit.
13626 @c This is getting pretty arcane, example would be good.
13628 @node save tracepoints
13629 @subsection @code{save tracepoints @var{filename}}
13630 @kindex save tracepoints
13631 @kindex save-tracepoints
13632 @cindex save tracepoints for future sessions
13634 This command saves all current tracepoint definitions together with
13635 their actions and passcounts, into a file @file{@var{filename}}
13636 suitable for use in a later debugging session. To read the saved
13637 tracepoint definitions, use the @code{source} command (@pxref{Command
13638 Files}). The @w{@code{save-tracepoints}} command is a deprecated
13639 alias for @w{@code{save tracepoints}}
13641 @node Tracepoint Variables
13642 @section Convenience Variables for Tracepoints
13643 @cindex tracepoint variables
13644 @cindex convenience variables for tracepoints
13647 @vindex $trace_frame
13648 @item (int) $trace_frame
13649 The current trace snapshot (a.k.a.@: @dfn{frame}) number, or -1 if no
13650 snapshot is selected.
13652 @vindex $tracepoint
13653 @item (int) $tracepoint
13654 The tracepoint for the current trace snapshot.
13656 @vindex $trace_line
13657 @item (int) $trace_line
13658 The line number for the current trace snapshot.
13660 @vindex $trace_file
13661 @item (char []) $trace_file
13662 The source file for the current trace snapshot.
13664 @vindex $trace_func
13665 @item (char []) $trace_func
13666 The name of the function containing @code{$tracepoint}.
13669 Note: @code{$trace_file} is not suitable for use in @code{printf},
13670 use @code{output} instead.
13672 Here's a simple example of using these convenience variables for
13673 stepping through all the trace snapshots and printing some of their
13674 data. Note that these are not the same as trace state variables,
13675 which are managed by the target.
13678 (@value{GDBP}) @b{tfind start}
13680 (@value{GDBP}) @b{while $trace_frame != -1}
13681 > output $trace_file
13682 > printf ", line %d (tracepoint #%d)\n", $trace_line, $tracepoint
13688 @section Using Trace Files
13689 @cindex trace files
13691 In some situations, the target running a trace experiment may no
13692 longer be available; perhaps it crashed, or the hardware was needed
13693 for a different activity. To handle these cases, you can arrange to
13694 dump the trace data into a file, and later use that file as a source
13695 of trace data, via the @code{target tfile} command.
13700 @item tsave [ -r ] @var{filename}
13701 @itemx tsave [-ctf] @var{dirname}
13702 Save the trace data to @var{filename}. By default, this command
13703 assumes that @var{filename} refers to the host filesystem, so if
13704 necessary @value{GDBN} will copy raw trace data up from the target and
13705 then save it. If the target supports it, you can also supply the
13706 optional argument @code{-r} (``remote'') to direct the target to save
13707 the data directly into @var{filename} in its own filesystem, which may be
13708 more efficient if the trace buffer is very large. (Note, however, that
13709 @code{target tfile} can only read from files accessible to the host.)
13710 By default, this command will save trace frame in tfile format.
13711 You can supply the optional argument @code{-ctf} to save data in CTF
13712 format. The @dfn{Common Trace Format} (CTF) is proposed as a trace format
13713 that can be shared by multiple debugging and tracing tools. Please go to
13714 @indicateurl{http://www.efficios.com/ctf} to get more information.
13716 @kindex target tfile
13720 @item target tfile @var{filename}
13721 @itemx target ctf @var{dirname}
13722 Use the file named @var{filename} or directory named @var{dirname} as
13723 a source of trace data. Commands that examine data work as they do with
13724 a live target, but it is not possible to run any new trace experiments.
13725 @code{tstatus} will report the state of the trace run at the moment
13726 the data was saved, as well as the current trace frame you are examining.
13727 Both @var{filename} and @var{dirname} must be on a filesystem accessible to
13731 (@value{GDBP}) target ctf ctf.ctf
13732 (@value{GDBP}) tfind
13733 Found trace frame 0, tracepoint 2
13734 39 ++a; /* set tracepoint 1 here */
13735 (@value{GDBP}) tdump
13736 Data collected at tracepoint 2, trace frame 0:
13740 c = @{"123", "456", "789", "123", "456", "789"@}
13741 d = @{@{@{a = 1, b = 2@}, @{a = 3, b = 4@}@}, @{@{a = 5, b = 6@}, @{a = 7, b = 8@}@}@}
13749 @chapter Debugging Programs That Use Overlays
13752 If your program is too large to fit completely in your target system's
13753 memory, you can sometimes use @dfn{overlays} to work around this
13754 problem. @value{GDBN} provides some support for debugging programs that
13758 * How Overlays Work:: A general explanation of overlays.
13759 * Overlay Commands:: Managing overlays in @value{GDBN}.
13760 * Automatic Overlay Debugging:: @value{GDBN} can find out which overlays are
13761 mapped by asking the inferior.
13762 * Overlay Sample Program:: A sample program using overlays.
13765 @node How Overlays Work
13766 @section How Overlays Work
13767 @cindex mapped overlays
13768 @cindex unmapped overlays
13769 @cindex load address, overlay's
13770 @cindex mapped address
13771 @cindex overlay area
13773 Suppose you have a computer whose instruction address space is only 64
13774 kilobytes long, but which has much more memory which can be accessed by
13775 other means: special instructions, segment registers, or memory
13776 management hardware, for example. Suppose further that you want to
13777 adapt a program which is larger than 64 kilobytes to run on this system.
13779 One solution is to identify modules of your program which are relatively
13780 independent, and need not call each other directly; call these modules
13781 @dfn{overlays}. Separate the overlays from the main program, and place
13782 their machine code in the larger memory. Place your main program in
13783 instruction memory, but leave at least enough space there to hold the
13784 largest overlay as well.
13786 Now, to call a function located in an overlay, you must first copy that
13787 overlay's machine code from the large memory into the space set aside
13788 for it in the instruction memory, and then jump to its entry point
13791 @c NB: In the below the mapped area's size is greater or equal to the
13792 @c size of all overlays. This is intentional to remind the developer
13793 @c that overlays don't necessarily need to be the same size.
13797 Data Instruction Larger
13798 Address Space Address Space Address Space
13799 +-----------+ +-----------+ +-----------+
13801 +-----------+ +-----------+ +-----------+<-- overlay 1
13802 | program | | main | .----| overlay 1 | load address
13803 | variables | | program | | +-----------+
13804 | and heap | | | | | |
13805 +-----------+ | | | +-----------+<-- overlay 2
13806 | | +-----------+ | | | load address
13807 +-----------+ | | | .-| overlay 2 |
13809 mapped --->+-----------+ | | +-----------+
13810 address | | | | | |
13811 | overlay | <-' | | |
13812 | area | <---' +-----------+<-- overlay 3
13813 | | <---. | | load address
13814 +-----------+ `--| overlay 3 |
13821 @anchor{A code overlay}A code overlay
13825 The diagram (@pxref{A code overlay}) shows a system with separate data
13826 and instruction address spaces. To map an overlay, the program copies
13827 its code from the larger address space to the instruction address space.
13828 Since the overlays shown here all use the same mapped address, only one
13829 may be mapped at a time. For a system with a single address space for
13830 data and instructions, the diagram would be similar, except that the
13831 program variables and heap would share an address space with the main
13832 program and the overlay area.
13834 An overlay loaded into instruction memory and ready for use is called a
13835 @dfn{mapped} overlay; its @dfn{mapped address} is its address in the
13836 instruction memory. An overlay not present (or only partially present)
13837 in instruction memory is called @dfn{unmapped}; its @dfn{load address}
13838 is its address in the larger memory. The mapped address is also called
13839 the @dfn{virtual memory address}, or @dfn{VMA}; the load address is also
13840 called the @dfn{load memory address}, or @dfn{LMA}.
13842 Unfortunately, overlays are not a completely transparent way to adapt a
13843 program to limited instruction memory. They introduce a new set of
13844 global constraints you must keep in mind as you design your program:
13849 Before calling or returning to a function in an overlay, your program
13850 must make sure that overlay is actually mapped. Otherwise, the call or
13851 return will transfer control to the right address, but in the wrong
13852 overlay, and your program will probably crash.
13855 If the process of mapping an overlay is expensive on your system, you
13856 will need to choose your overlays carefully to minimize their effect on
13857 your program's performance.
13860 The executable file you load onto your system must contain each
13861 overlay's instructions, appearing at the overlay's load address, not its
13862 mapped address. However, each overlay's instructions must be relocated
13863 and its symbols defined as if the overlay were at its mapped address.
13864 You can use GNU linker scripts to specify different load and relocation
13865 addresses for pieces of your program; see @ref{Overlay Description,,,
13866 ld.info, Using ld: the GNU linker}.
13869 The procedure for loading executable files onto your system must be able
13870 to load their contents into the larger address space as well as the
13871 instruction and data spaces.
13875 The overlay system described above is rather simple, and could be
13876 improved in many ways:
13881 If your system has suitable bank switch registers or memory management
13882 hardware, you could use those facilities to make an overlay's load area
13883 contents simply appear at their mapped address in instruction space.
13884 This would probably be faster than copying the overlay to its mapped
13885 area in the usual way.
13888 If your overlays are small enough, you could set aside more than one
13889 overlay area, and have more than one overlay mapped at a time.
13892 You can use overlays to manage data, as well as instructions. In
13893 general, data overlays are even less transparent to your design than
13894 code overlays: whereas code overlays only require care when you call or
13895 return to functions, data overlays require care every time you access
13896 the data. Also, if you change the contents of a data overlay, you
13897 must copy its contents back out to its load address before you can copy a
13898 different data overlay into the same mapped area.
13903 @node Overlay Commands
13904 @section Overlay Commands
13906 To use @value{GDBN}'s overlay support, each overlay in your program must
13907 correspond to a separate section of the executable file. The section's
13908 virtual memory address and load memory address must be the overlay's
13909 mapped and load addresses. Identifying overlays with sections allows
13910 @value{GDBN} to determine the appropriate address of a function or
13911 variable, depending on whether the overlay is mapped or not.
13913 @value{GDBN}'s overlay commands all start with the word @code{overlay};
13914 you can abbreviate this as @code{ov} or @code{ovly}. The commands are:
13919 Disable @value{GDBN}'s overlay support. When overlay support is
13920 disabled, @value{GDBN} assumes that all functions and variables are
13921 always present at their mapped addresses. By default, @value{GDBN}'s
13922 overlay support is disabled.
13924 @item overlay manual
13925 @cindex manual overlay debugging
13926 Enable @dfn{manual} overlay debugging. In this mode, @value{GDBN}
13927 relies on you to tell it which overlays are mapped, and which are not,
13928 using the @code{overlay map-overlay} and @code{overlay unmap-overlay}
13929 commands described below.
13931 @item overlay map-overlay @var{overlay}
13932 @itemx overlay map @var{overlay}
13933 @cindex map an overlay
13934 Tell @value{GDBN} that @var{overlay} is now mapped; @var{overlay} must
13935 be the name of the object file section containing the overlay. When an
13936 overlay is mapped, @value{GDBN} assumes it can find the overlay's
13937 functions and variables at their mapped addresses. @value{GDBN} assumes
13938 that any other overlays whose mapped ranges overlap that of
13939 @var{overlay} are now unmapped.
13941 @item overlay unmap-overlay @var{overlay}
13942 @itemx overlay unmap @var{overlay}
13943 @cindex unmap an overlay
13944 Tell @value{GDBN} that @var{overlay} is no longer mapped; @var{overlay}
13945 must be the name of the object file section containing the overlay.
13946 When an overlay is unmapped, @value{GDBN} assumes it can find the
13947 overlay's functions and variables at their load addresses.
13950 Enable @dfn{automatic} overlay debugging. In this mode, @value{GDBN}
13951 consults a data structure the overlay manager maintains in the inferior
13952 to see which overlays are mapped. For details, see @ref{Automatic
13953 Overlay Debugging}.
13955 @item overlay load-target
13956 @itemx overlay load
13957 @cindex reloading the overlay table
13958 Re-read the overlay table from the inferior. Normally, @value{GDBN}
13959 re-reads the table @value{GDBN} automatically each time the inferior
13960 stops, so this command should only be necessary if you have changed the
13961 overlay mapping yourself using @value{GDBN}. This command is only
13962 useful when using automatic overlay debugging.
13964 @item overlay list-overlays
13965 @itemx overlay list
13966 @cindex listing mapped overlays
13967 Display a list of the overlays currently mapped, along with their mapped
13968 addresses, load addresses, and sizes.
13972 Normally, when @value{GDBN} prints a code address, it includes the name
13973 of the function the address falls in:
13976 (@value{GDBP}) print main
13977 $3 = @{int ()@} 0x11a0 <main>
13980 When overlay debugging is enabled, @value{GDBN} recognizes code in
13981 unmapped overlays, and prints the names of unmapped functions with
13982 asterisks around them. For example, if @code{foo} is a function in an
13983 unmapped overlay, @value{GDBN} prints it this way:
13986 (@value{GDBP}) overlay list
13987 No sections are mapped.
13988 (@value{GDBP}) print foo
13989 $5 = @{int (int)@} 0x100000 <*foo*>
13992 When @code{foo}'s overlay is mapped, @value{GDBN} prints the function's
13996 (@value{GDBP}) overlay list
13997 Section .ov.foo.text, loaded at 0x100000 - 0x100034,
13998 mapped at 0x1016 - 0x104a
13999 (@value{GDBP}) print foo
14000 $6 = @{int (int)@} 0x1016 <foo>
14003 When overlay debugging is enabled, @value{GDBN} can find the correct
14004 address for functions and variables in an overlay, whether or not the
14005 overlay is mapped. This allows most @value{GDBN} commands, like
14006 @code{break} and @code{disassemble}, to work normally, even on unmapped
14007 code. However, @value{GDBN}'s breakpoint support has some limitations:
14011 @cindex breakpoints in overlays
14012 @cindex overlays, setting breakpoints in
14013 You can set breakpoints in functions in unmapped overlays, as long as
14014 @value{GDBN} can write to the overlay at its load address.
14016 @value{GDBN} can not set hardware or simulator-based breakpoints in
14017 unmapped overlays. However, if you set a breakpoint at the end of your
14018 overlay manager (and tell @value{GDBN} which overlays are now mapped, if
14019 you are using manual overlay management), @value{GDBN} will re-set its
14020 breakpoints properly.
14024 @node Automatic Overlay Debugging
14025 @section Automatic Overlay Debugging
14026 @cindex automatic overlay debugging
14028 @value{GDBN} can automatically track which overlays are mapped and which
14029 are not, given some simple co-operation from the overlay manager in the
14030 inferior. If you enable automatic overlay debugging with the
14031 @code{overlay auto} command (@pxref{Overlay Commands}), @value{GDBN}
14032 looks in the inferior's memory for certain variables describing the
14033 current state of the overlays.
14035 Here are the variables your overlay manager must define to support
14036 @value{GDBN}'s automatic overlay debugging:
14040 @item @code{_ovly_table}:
14041 This variable must be an array of the following structures:
14046 /* The overlay's mapped address. */
14049 /* The size of the overlay, in bytes. */
14050 unsigned long size;
14052 /* The overlay's load address. */
14055 /* Non-zero if the overlay is currently mapped;
14057 unsigned long mapped;
14061 @item @code{_novlys}:
14062 This variable must be a four-byte signed integer, holding the total
14063 number of elements in @code{_ovly_table}.
14067 To decide whether a particular overlay is mapped or not, @value{GDBN}
14068 looks for an entry in @w{@code{_ovly_table}} whose @code{vma} and
14069 @code{lma} members equal the VMA and LMA of the overlay's section in the
14070 executable file. When @value{GDBN} finds a matching entry, it consults
14071 the entry's @code{mapped} member to determine whether the overlay is
14074 In addition, your overlay manager may define a function called
14075 @code{_ovly_debug_event}. If this function is defined, @value{GDBN}
14076 will silently set a breakpoint there. If the overlay manager then
14077 calls this function whenever it has changed the overlay table, this
14078 will enable @value{GDBN} to accurately keep track of which overlays
14079 are in program memory, and update any breakpoints that may be set
14080 in overlays. This will allow breakpoints to work even if the
14081 overlays are kept in ROM or other non-writable memory while they
14082 are not being executed.
14084 @node Overlay Sample Program
14085 @section Overlay Sample Program
14086 @cindex overlay example program
14088 When linking a program which uses overlays, you must place the overlays
14089 at their load addresses, while relocating them to run at their mapped
14090 addresses. To do this, you must write a linker script (@pxref{Overlay
14091 Description,,, ld.info, Using ld: the GNU linker}). Unfortunately,
14092 since linker scripts are specific to a particular host system, target
14093 architecture, and target memory layout, this manual cannot provide
14094 portable sample code demonstrating @value{GDBN}'s overlay support.
14096 However, the @value{GDBN} source distribution does contain an overlaid
14097 program, with linker scripts for a few systems, as part of its test
14098 suite. The program consists of the following files from
14099 @file{gdb/testsuite/gdb.base}:
14103 The main program file.
14105 A simple overlay manager, used by @file{overlays.c}.
14110 Overlay modules, loaded and used by @file{overlays.c}.
14113 Linker scripts for linking the test program on the @code{d10v-elf}
14114 and @code{m32r-elf} targets.
14117 You can build the test program using the @code{d10v-elf} GCC
14118 cross-compiler like this:
14121 $ d10v-elf-gcc -g -c overlays.c
14122 $ d10v-elf-gcc -g -c ovlymgr.c
14123 $ d10v-elf-gcc -g -c foo.c
14124 $ d10v-elf-gcc -g -c bar.c
14125 $ d10v-elf-gcc -g -c baz.c
14126 $ d10v-elf-gcc -g -c grbx.c
14127 $ d10v-elf-gcc -g overlays.o ovlymgr.o foo.o bar.o \
14128 baz.o grbx.o -Wl,-Td10v.ld -o overlays
14131 The build process is identical for any other architecture, except that
14132 you must substitute the appropriate compiler and linker script for the
14133 target system for @code{d10v-elf-gcc} and @code{d10v.ld}.
14137 @chapter Using @value{GDBN} with Different Languages
14140 Although programming languages generally have common aspects, they are
14141 rarely expressed in the same manner. For instance, in ANSI C,
14142 dereferencing a pointer @code{p} is accomplished by @code{*p}, but in
14143 Modula-2, it is accomplished by @code{p^}. Values can also be
14144 represented (and displayed) differently. Hex numbers in C appear as
14145 @samp{0x1ae}, while in Modula-2 they appear as @samp{1AEH}.
14147 @cindex working language
14148 Language-specific information is built into @value{GDBN} for some languages,
14149 allowing you to express operations like the above in your program's
14150 native language, and allowing @value{GDBN} to output values in a manner
14151 consistent with the syntax of your program's native language. The
14152 language you use to build expressions is called the @dfn{working
14156 * Setting:: Switching between source languages
14157 * Show:: Displaying the language
14158 * Checks:: Type and range checks
14159 * Supported Languages:: Supported languages
14160 * Unsupported Languages:: Unsupported languages
14164 @section Switching Between Source Languages
14166 There are two ways to control the working language---either have @value{GDBN}
14167 set it automatically, or select it manually yourself. You can use the
14168 @code{set language} command for either purpose. On startup, @value{GDBN}
14169 defaults to setting the language automatically. The working language is
14170 used to determine how expressions you type are interpreted, how values
14173 In addition to the working language, every source file that
14174 @value{GDBN} knows about has its own working language. For some object
14175 file formats, the compiler might indicate which language a particular
14176 source file is in. However, most of the time @value{GDBN} infers the
14177 language from the name of the file. The language of a source file
14178 controls whether C@t{++} names are demangled---this way @code{backtrace} can
14179 show each frame appropriately for its own language. There is no way to
14180 set the language of a source file from within @value{GDBN}, but you can
14181 set the language associated with a filename extension. @xref{Show, ,
14182 Displaying the Language}.
14184 This is most commonly a problem when you use a program, such
14185 as @code{cfront} or @code{f2c}, that generates C but is written in
14186 another language. In that case, make the
14187 program use @code{#line} directives in its C output; that way
14188 @value{GDBN} will know the correct language of the source code of the original
14189 program, and will display that source code, not the generated C code.
14192 * Filenames:: Filename extensions and languages.
14193 * Manually:: Setting the working language manually
14194 * Automatically:: Having @value{GDBN} infer the source language
14198 @subsection List of Filename Extensions and Languages
14200 If a source file name ends in one of the following extensions, then
14201 @value{GDBN} infers that its language is the one indicated.
14219 C@t{++} source file
14225 Objective-C source file
14229 Fortran source file
14232 Modula-2 source file
14236 Assembler source file. This actually behaves almost like C, but
14237 @value{GDBN} does not skip over function prologues when stepping.
14240 In addition, you may set the language associated with a filename
14241 extension. @xref{Show, , Displaying the Language}.
14244 @subsection Setting the Working Language
14246 If you allow @value{GDBN} to set the language automatically,
14247 expressions are interpreted the same way in your debugging session and
14250 @kindex set language
14251 If you wish, you may set the language manually. To do this, issue the
14252 command @samp{set language @var{lang}}, where @var{lang} is the name of
14253 a language, such as
14254 @code{c} or @code{modula-2}.
14255 For a list of the supported languages, type @samp{set language}.
14257 Setting the language manually prevents @value{GDBN} from updating the working
14258 language automatically. This can lead to confusion if you try
14259 to debug a program when the working language is not the same as the
14260 source language, when an expression is acceptable to both
14261 languages---but means different things. For instance, if the current
14262 source file were written in C, and @value{GDBN} was parsing Modula-2, a
14270 might not have the effect you intended. In C, this means to add
14271 @code{b} and @code{c} and place the result in @code{a}. The result
14272 printed would be the value of @code{a}. In Modula-2, this means to compare
14273 @code{a} to the result of @code{b+c}, yielding a @code{BOOLEAN} value.
14275 @node Automatically
14276 @subsection Having @value{GDBN} Infer the Source Language
14278 To have @value{GDBN} set the working language automatically, use
14279 @samp{set language local} or @samp{set language auto}. @value{GDBN}
14280 then infers the working language. That is, when your program stops in a
14281 frame (usually by encountering a breakpoint), @value{GDBN} sets the
14282 working language to the language recorded for the function in that
14283 frame. If the language for a frame is unknown (that is, if the function
14284 or block corresponding to the frame was defined in a source file that
14285 does not have a recognized extension), the current working language is
14286 not changed, and @value{GDBN} issues a warning.
14288 This may not seem necessary for most programs, which are written
14289 entirely in one source language. However, program modules and libraries
14290 written in one source language can be used by a main program written in
14291 a different source language. Using @samp{set language auto} in this
14292 case frees you from having to set the working language manually.
14295 @section Displaying the Language
14297 The following commands help you find out which language is the
14298 working language, and also what language source files were written in.
14301 @item show language
14302 @anchor{show language}
14303 @kindex show language
14304 Display the current working language. This is the
14305 language you can use with commands such as @code{print} to
14306 build and compute expressions that may involve variables in your program.
14309 @kindex info frame@r{, show the source language}
14310 Display the source language for this frame. This language becomes the
14311 working language if you use an identifier from this frame.
14312 @xref{Frame Info, ,Information about a Frame}, to identify the other
14313 information listed here.
14316 @kindex info source@r{, show the source language}
14317 Display the source language of this source file.
14318 @xref{Symbols, ,Examining the Symbol Table}, to identify the other
14319 information listed here.
14322 In unusual circumstances, you may have source files with extensions
14323 not in the standard list. You can then set the extension associated
14324 with a language explicitly:
14327 @item set extension-language @var{ext} @var{language}
14328 @kindex set extension-language
14329 Tell @value{GDBN} that source files with extension @var{ext} are to be
14330 assumed as written in the source language @var{language}.
14332 @item info extensions
14333 @kindex info extensions
14334 List all the filename extensions and the associated languages.
14338 @section Type and Range Checking
14340 Some languages are designed to guard you against making seemingly common
14341 errors through a series of compile- and run-time checks. These include
14342 checking the type of arguments to functions and operators and making
14343 sure mathematical overflows are caught at run time. Checks such as
14344 these help to ensure a program's correctness once it has been compiled
14345 by eliminating type mismatches and providing active checks for range
14346 errors when your program is running.
14348 By default @value{GDBN} checks for these errors according to the
14349 rules of the current source language. Although @value{GDBN} does not check
14350 the statements in your program, it can check expressions entered directly
14351 into @value{GDBN} for evaluation via the @code{print} command, for example.
14354 * Type Checking:: An overview of type checking
14355 * Range Checking:: An overview of range checking
14358 @cindex type checking
14359 @cindex checks, type
14360 @node Type Checking
14361 @subsection An Overview of Type Checking
14363 Some languages, such as C and C@t{++}, are strongly typed, meaning that the
14364 arguments to operators and functions have to be of the correct type,
14365 otherwise an error occurs. These checks prevent type mismatch
14366 errors from ever causing any run-time problems. For example,
14369 int klass::my_method(char *b) @{ return b ? 1 : 2; @}
14371 (@value{GDBP}) print obj.my_method (0)
14374 (@value{GDBP}) print obj.my_method (0x1234)
14375 Cannot resolve method klass::my_method to any overloaded instance
14378 The second example fails because in C@t{++} the integer constant
14379 @samp{0x1234} is not type-compatible with the pointer parameter type.
14381 For the expressions you use in @value{GDBN} commands, you can tell
14382 @value{GDBN} to not enforce strict type checking or
14383 to treat any mismatches as errors and abandon the expression;
14384 When type checking is disabled, @value{GDBN} successfully evaluates
14385 expressions like the second example above.
14387 Even if type checking is off, there may be other reasons
14388 related to type that prevent @value{GDBN} from evaluating an expression.
14389 For instance, @value{GDBN} does not know how to add an @code{int} and
14390 a @code{struct foo}. These particular type errors have nothing to do
14391 with the language in use and usually arise from expressions which make
14392 little sense to evaluate anyway.
14394 @value{GDBN} provides some additional commands for controlling type checking:
14396 @kindex set check type
14397 @kindex show check type
14399 @item set check type on
14400 @itemx set check type off
14401 Set strict type checking on or off. If any type mismatches occur in
14402 evaluating an expression while type checking is on, @value{GDBN} prints a
14403 message and aborts evaluation of the expression.
14405 @item show check type
14406 Show the current setting of type checking and whether @value{GDBN}
14407 is enforcing strict type checking rules.
14410 @cindex range checking
14411 @cindex checks, range
14412 @node Range Checking
14413 @subsection An Overview of Range Checking
14415 In some languages (such as Modula-2), it is an error to exceed the
14416 bounds of a type; this is enforced with run-time checks. Such range
14417 checking is meant to ensure program correctness by making sure
14418 computations do not overflow, or indices on an array element access do
14419 not exceed the bounds of the array.
14421 For expressions you use in @value{GDBN} commands, you can tell
14422 @value{GDBN} to treat range errors in one of three ways: ignore them,
14423 always treat them as errors and abandon the expression, or issue
14424 warnings but evaluate the expression anyway.
14426 A range error can result from numerical overflow, from exceeding an
14427 array index bound, or when you type a constant that is not a member
14428 of any type. Some languages, however, do not treat overflows as an
14429 error. In many implementations of C, mathematical overflow causes the
14430 result to ``wrap around'' to lower values---for example, if @var{m} is
14431 the largest integer value, and @var{s} is the smallest, then
14434 @var{m} + 1 @result{} @var{s}
14437 This, too, is specific to individual languages, and in some cases
14438 specific to individual compilers or machines. @xref{Supported Languages, ,
14439 Supported Languages}, for further details on specific languages.
14441 @value{GDBN} provides some additional commands for controlling the range checker:
14443 @kindex set check range
14444 @kindex show check range
14446 @item set check range auto
14447 Set range checking on or off based on the current working language.
14448 @xref{Supported Languages, ,Supported Languages}, for the default settings for
14451 @item set check range on
14452 @itemx set check range off
14453 Set range checking on or off, overriding the default setting for the
14454 current working language. A warning is issued if the setting does not
14455 match the language default. If a range error occurs and range checking is on,
14456 then a message is printed and evaluation of the expression is aborted.
14458 @item set check range warn
14459 Output messages when the @value{GDBN} range checker detects a range error,
14460 but attempt to evaluate the expression anyway. Evaluating the
14461 expression may still be impossible for other reasons, such as accessing
14462 memory that the process does not own (a typical example from many Unix
14466 Show the current setting of the range checker, and whether or not it is
14467 being set automatically by @value{GDBN}.
14470 @node Supported Languages
14471 @section Supported Languages
14473 @value{GDBN} supports C, C@t{++}, D, Go, Objective-C, Fortran,
14474 OpenCL C, Pascal, Rust, assembly, Modula-2, and Ada.
14475 @c This is false ...
14476 Some @value{GDBN} features may be used in expressions regardless of the
14477 language you use: the @value{GDBN} @code{@@} and @code{::} operators,
14478 and the @samp{@{type@}addr} construct (@pxref{Expressions,
14479 ,Expressions}) can be used with the constructs of any supported
14482 The following sections detail to what degree each source language is
14483 supported by @value{GDBN}. These sections are not meant to be language
14484 tutorials or references, but serve only as a reference guide to what the
14485 @value{GDBN} expression parser accepts, and what input and output
14486 formats should look like for different languages. There are many good
14487 books written on each of these languages; please look to these for a
14488 language reference or tutorial.
14491 * C:: C and C@t{++}
14494 * Objective-C:: Objective-C
14495 * OpenCL C:: OpenCL C
14496 * Fortran:: Fortran
14499 * Modula-2:: Modula-2
14504 @subsection C and C@t{++}
14506 @cindex C and C@t{++}
14507 @cindex expressions in C or C@t{++}
14509 Since C and C@t{++} are so closely related, many features of @value{GDBN} apply
14510 to both languages. Whenever this is the case, we discuss those languages
14514 @cindex @code{g++}, @sc{gnu} C@t{++} compiler
14515 @cindex @sc{gnu} C@t{++}
14516 The C@t{++} debugging facilities are jointly implemented by the C@t{++}
14517 compiler and @value{GDBN}. Therefore, to debug your C@t{++} code
14518 effectively, you must compile your C@t{++} programs with a supported
14519 C@t{++} compiler, such as @sc{gnu} @code{g++}, or the HP ANSI C@t{++}
14520 compiler (@code{aCC}).
14523 * C Operators:: C and C@t{++} operators
14524 * C Constants:: C and C@t{++} constants
14525 * C Plus Plus Expressions:: C@t{++} expressions
14526 * C Defaults:: Default settings for C and C@t{++}
14527 * C Checks:: C and C@t{++} type and range checks
14528 * Debugging C:: @value{GDBN} and C
14529 * Debugging C Plus Plus:: @value{GDBN} features for C@t{++}
14530 * Decimal Floating Point:: Numbers in Decimal Floating Point format
14534 @subsubsection C and C@t{++} Operators
14536 @cindex C and C@t{++} operators
14538 Operators must be defined on values of specific types. For instance,
14539 @code{+} is defined on numbers, but not on structures. Operators are
14540 often defined on groups of types.
14542 For the purposes of C and C@t{++}, the following definitions hold:
14547 @emph{Integral types} include @code{int} with any of its storage-class
14548 specifiers; @code{char}; @code{enum}; and, for C@t{++}, @code{bool}.
14551 @emph{Floating-point types} include @code{float}, @code{double}, and
14552 @code{long double} (if supported by the target platform).
14555 @emph{Pointer types} include all types defined as @code{(@var{type} *)}.
14558 @emph{Scalar types} include all of the above.
14563 The following operators are supported. They are listed here
14564 in order of increasing precedence:
14568 The comma or sequencing operator. Expressions in a comma-separated list
14569 are evaluated from left to right, with the result of the entire
14570 expression being the last expression evaluated.
14573 Assignment. The value of an assignment expression is the value
14574 assigned. Defined on scalar types.
14577 Used in an expression of the form @w{@code{@var{a} @var{op}= @var{b}}},
14578 and translated to @w{@code{@var{a} = @var{a op b}}}.
14579 @w{@code{@var{op}=}} and @code{=} have the same precedence. The operator
14580 @var{op} is any one of the operators @code{|}, @code{^}, @code{&},
14581 @code{<<}, @code{>>}, @code{+}, @code{-}, @code{*}, @code{/}, @code{%}.
14584 The ternary operator. @code{@var{a} ? @var{b} : @var{c}} can be thought
14585 of as: if @var{a} then @var{b} else @var{c}. The argument @var{a}
14586 should be of an integral type.
14589 Logical @sc{or}. Defined on integral types.
14592 Logical @sc{and}. Defined on integral types.
14595 Bitwise @sc{or}. Defined on integral types.
14598 Bitwise exclusive-@sc{or}. Defined on integral types.
14601 Bitwise @sc{and}. Defined on integral types.
14604 Equality and inequality. Defined on scalar types. The value of these
14605 expressions is 0 for false and non-zero for true.
14607 @item <@r{, }>@r{, }<=@r{, }>=
14608 Less than, greater than, less than or equal, greater than or equal.
14609 Defined on scalar types. The value of these expressions is 0 for false
14610 and non-zero for true.
14613 left shift, and right shift. Defined on integral types.
14616 The @value{GDBN} ``artificial array'' operator (@pxref{Expressions, ,Expressions}).
14619 Addition and subtraction. Defined on integral types, floating-point types and
14622 @item *@r{, }/@r{, }%
14623 Multiplication, division, and modulus. Multiplication and division are
14624 defined on integral and floating-point types. Modulus is defined on
14628 Increment and decrement. When appearing before a variable, the
14629 operation is performed before the variable is used in an expression;
14630 when appearing after it, the variable's value is used before the
14631 operation takes place.
14634 Pointer dereferencing. Defined on pointer types. Same precedence as
14638 Address operator. Defined on variables. Same precedence as @code{++}.
14640 For debugging C@t{++}, @value{GDBN} implements a use of @samp{&} beyond what is
14641 allowed in the C@t{++} language itself: you can use @samp{&(&@var{ref})}
14642 to examine the address
14643 where a C@t{++} reference variable (declared with @samp{&@var{ref}}) is
14647 Negative. Defined on integral and floating-point types. Same
14648 precedence as @code{++}.
14651 Logical negation. Defined on integral types. Same precedence as
14655 Bitwise complement operator. Defined on integral types. Same precedence as
14660 Structure member, and pointer-to-structure member. For convenience,
14661 @value{GDBN} regards the two as equivalent, choosing whether to dereference a
14662 pointer based on the stored type information.
14663 Defined on @code{struct} and @code{union} data.
14666 Dereferences of pointers to members.
14669 Array indexing. @code{@var{a}[@var{i}]} is defined as
14670 @code{*(@var{a}+@var{i})}. Same precedence as @code{->}.
14673 Function parameter list. Same precedence as @code{->}.
14676 C@t{++} scope resolution operator. Defined on @code{struct}, @code{union},
14677 and @code{class} types.
14680 Doubled colons also represent the @value{GDBN} scope operator
14681 (@pxref{Expressions, ,Expressions}). Same precedence as @code{::},
14685 If an operator is redefined in the user code, @value{GDBN} usually
14686 attempts to invoke the redefined version instead of using the operator's
14687 predefined meaning.
14690 @subsubsection C and C@t{++} Constants
14692 @cindex C and C@t{++} constants
14694 @value{GDBN} allows you to express the constants of C and C@t{++} in the
14699 Integer constants are a sequence of digits. Octal constants are
14700 specified by a leading @samp{0} (i.e.@: zero), and hexadecimal constants
14701 by a leading @samp{0x} or @samp{0X}. Constants may also end with a letter
14702 @samp{l}, specifying that the constant should be treated as a
14706 Floating point constants are a sequence of digits, followed by a decimal
14707 point, followed by a sequence of digits, and optionally followed by an
14708 exponent. An exponent is of the form:
14709 @samp{@w{e@r{[[}+@r{]|}-@r{]}@var{nnn}}}, where @var{nnn} is another
14710 sequence of digits. The @samp{+} is optional for positive exponents.
14711 A floating-point constant may also end with a letter @samp{f} or
14712 @samp{F}, specifying that the constant should be treated as being of
14713 the @code{float} (as opposed to the default @code{double}) type; or with
14714 a letter @samp{l} or @samp{L}, which specifies a @code{long double}
14718 Enumerated constants consist of enumerated identifiers, or their
14719 integral equivalents.
14722 Character constants are a single character surrounded by single quotes
14723 (@code{'}), or a number---the ordinal value of the corresponding character
14724 (usually its @sc{ascii} value). Within quotes, the single character may
14725 be represented by a letter or by @dfn{escape sequences}, which are of
14726 the form @samp{\@var{nnn}}, where @var{nnn} is the octal representation
14727 of the character's ordinal value; or of the form @samp{\@var{x}}, where
14728 @samp{@var{x}} is a predefined special character---for example,
14729 @samp{\n} for newline.
14731 Wide character constants can be written by prefixing a character
14732 constant with @samp{L}, as in C. For example, @samp{L'x'} is the wide
14733 form of @samp{x}. The target wide character set is used when
14734 computing the value of this constant (@pxref{Character Sets}).
14737 String constants are a sequence of character constants surrounded by
14738 double quotes (@code{"}). Any valid character constant (as described
14739 above) may appear. Double quotes within the string must be preceded by
14740 a backslash, so for instance @samp{"a\"b'c"} is a string of five
14743 Wide string constants can be written by prefixing a string constant
14744 with @samp{L}, as in C. The target wide character set is used when
14745 computing the value of this constant (@pxref{Character Sets}).
14748 Pointer constants are an integral value. You can also write pointers
14749 to constants using the C operator @samp{&}.
14752 Array constants are comma-separated lists surrounded by braces @samp{@{}
14753 and @samp{@}}; for example, @samp{@{1,2,3@}} is a three-element array of
14754 integers, @samp{@{@{1,2@}, @{3,4@}, @{5,6@}@}} is a three-by-two array,
14755 and @samp{@{&"hi", &"there", &"fred"@}} is a three-element array of pointers.
14758 @node C Plus Plus Expressions
14759 @subsubsection C@t{++} Expressions
14761 @cindex expressions in C@t{++}
14762 @value{GDBN} expression handling can interpret most C@t{++} expressions.
14764 @cindex debugging C@t{++} programs
14765 @cindex C@t{++} compilers
14766 @cindex debug formats and C@t{++}
14767 @cindex @value{NGCC} and C@t{++}
14769 @emph{Warning:} @value{GDBN} can only debug C@t{++} code if you use
14770 the proper compiler and the proper debug format. Currently,
14771 @value{GDBN} works best when debugging C@t{++} code that is compiled
14772 with the most recent version of @value{NGCC} possible. The DWARF
14773 debugging format is preferred; @value{NGCC} defaults to this on most
14774 popular platforms. Other compilers and/or debug formats are likely to
14775 work badly or not at all when using @value{GDBN} to debug C@t{++}
14776 code. @xref{Compilation}.
14781 @cindex member functions
14783 Member function calls are allowed; you can use expressions like
14786 count = aml->GetOriginal(x, y)
14789 @vindex this@r{, inside C@t{++} member functions}
14790 @cindex namespace in C@t{++}
14792 While a member function is active (in the selected stack frame), your
14793 expressions have the same namespace available as the member function;
14794 that is, @value{GDBN} allows implicit references to the class instance
14795 pointer @code{this} following the same rules as C@t{++}. @code{using}
14796 declarations in the current scope are also respected by @value{GDBN}.
14798 @cindex call overloaded functions
14799 @cindex overloaded functions, calling
14800 @cindex type conversions in C@t{++}
14802 You can call overloaded functions; @value{GDBN} resolves the function
14803 call to the right definition, with some restrictions. @value{GDBN} does not
14804 perform overload resolution involving user-defined type conversions,
14805 calls to constructors, or instantiations of templates that do not exist
14806 in the program. It also cannot handle ellipsis argument lists or
14809 It does perform integral conversions and promotions, floating-point
14810 promotions, arithmetic conversions, pointer conversions, conversions of
14811 class objects to base classes, and standard conversions such as those of
14812 functions or arrays to pointers; it requires an exact match on the
14813 number of function arguments.
14815 Overload resolution is always performed, unless you have specified
14816 @code{set overload-resolution off}. @xref{Debugging C Plus Plus,
14817 ,@value{GDBN} Features for C@t{++}}.
14819 You must specify @code{set overload-resolution off} in order to use an
14820 explicit function signature to call an overloaded function, as in
14822 p 'foo(char,int)'('x', 13)
14825 The @value{GDBN} command-completion facility can simplify this;
14826 see @ref{Completion, ,Command Completion}.
14828 @cindex reference declarations
14830 @value{GDBN} understands variables declared as C@t{++} references; you can use
14831 them in expressions just as you do in C@t{++} source---they are automatically
14834 In the parameter list shown when @value{GDBN} displays a frame, the values of
14835 reference variables are not displayed (unlike other variables); this
14836 avoids clutter, since references are often used for large structures.
14837 The @emph{address} of a reference variable is always shown, unless
14838 you have specified @samp{set print address off}.
14841 @value{GDBN} supports the C@t{++} name resolution operator @code{::}---your
14842 expressions can use it just as expressions in your program do. Since
14843 one scope may be defined in another, you can use @code{::} repeatedly if
14844 necessary, for example in an expression like
14845 @samp{@var{scope1}::@var{scope2}::@var{name}}. @value{GDBN} also allows
14846 resolving name scope by reference to source files, in both C and C@t{++}
14847 debugging (@pxref{Variables, ,Program Variables}).
14850 @value{GDBN} performs argument-dependent lookup, following the C@t{++}
14855 @subsubsection C and C@t{++} Defaults
14857 @cindex C and C@t{++} defaults
14859 If you allow @value{GDBN} to set range checking automatically, it
14860 defaults to @code{off} whenever the working language changes to
14861 C or C@t{++}. This happens regardless of whether you or @value{GDBN}
14862 selects the working language.
14864 If you allow @value{GDBN} to set the language automatically, it
14865 recognizes source files whose names end with @file{.c}, @file{.C}, or
14866 @file{.cc}, etc, and when @value{GDBN} enters code compiled from one of
14867 these files, it sets the working language to C or C@t{++}.
14868 @xref{Automatically, ,Having @value{GDBN} Infer the Source Language},
14869 for further details.
14872 @subsubsection C and C@t{++} Type and Range Checks
14874 @cindex C and C@t{++} checks
14876 By default, when @value{GDBN} parses C or C@t{++} expressions, strict type
14877 checking is used. However, if you turn type checking off, @value{GDBN}
14878 will allow certain non-standard conversions, such as promoting integer
14879 constants to pointers.
14881 Range checking, if turned on, is done on mathematical operations. Array
14882 indices are not checked, since they are often used to index a pointer
14883 that is not itself an array.
14886 @subsubsection @value{GDBN} and C
14888 The @code{set print union} and @code{show print union} commands apply to
14889 the @code{union} type. When set to @samp{on}, any @code{union} that is
14890 inside a @code{struct} or @code{class} is also printed. Otherwise, it
14891 appears as @samp{@{...@}}.
14893 The @code{@@} operator aids in the debugging of dynamic arrays, formed
14894 with pointers and a memory allocation function. @xref{Expressions,
14897 @node Debugging C Plus Plus
14898 @subsubsection @value{GDBN} Features for C@t{++}
14900 @cindex commands for C@t{++}
14902 Some @value{GDBN} commands are particularly useful with C@t{++}, and some are
14903 designed specifically for use with C@t{++}. Here is a summary:
14906 @cindex break in overloaded functions
14907 @item @r{breakpoint menus}
14908 When you want a breakpoint in a function whose name is overloaded,
14909 @value{GDBN} has the capability to display a menu of possible breakpoint
14910 locations to help you specify which function definition you want.
14911 @xref{Ambiguous Expressions,,Ambiguous Expressions}.
14913 @cindex overloading in C@t{++}
14914 @item rbreak @var{regex}
14915 Setting breakpoints using regular expressions is helpful for setting
14916 breakpoints on overloaded functions that are not members of any special
14918 @xref{Set Breaks, ,Setting Breakpoints}.
14920 @cindex C@t{++} exception handling
14922 @itemx catch rethrow
14924 Debug C@t{++} exception handling using these commands. @xref{Set
14925 Catchpoints, , Setting Catchpoints}.
14927 @cindex inheritance
14928 @item ptype @var{typename}
14929 Print inheritance relationships as well as other information for type
14931 @xref{Symbols, ,Examining the Symbol Table}.
14933 @item info vtbl @var{expression}.
14934 The @code{info vtbl} command can be used to display the virtual
14935 method tables of the object computed by @var{expression}. This shows
14936 one entry per virtual table; there may be multiple virtual tables when
14937 multiple inheritance is in use.
14939 @cindex C@t{++} demangling
14940 @item demangle @var{name}
14941 Demangle @var{name}.
14942 @xref{Symbols}, for a more complete description of the @code{demangle} command.
14944 @cindex C@t{++} symbol display
14945 @item set print demangle
14946 @itemx show print demangle
14947 @itemx set print asm-demangle
14948 @itemx show print asm-demangle
14949 Control whether C@t{++} symbols display in their source form, both when
14950 displaying code as C@t{++} source and when displaying disassemblies.
14951 @xref{Print Settings, ,Print Settings}.
14953 @item set print object
14954 @itemx show print object
14955 Choose whether to print derived (actual) or declared types of objects.
14956 @xref{Print Settings, ,Print Settings}.
14958 @item set print vtbl
14959 @itemx show print vtbl
14960 Control the format for printing virtual function tables.
14961 @xref{Print Settings, ,Print Settings}.
14962 (The @code{vtbl} commands do not work on programs compiled with the HP
14963 ANSI C@t{++} compiler (@code{aCC}).)
14965 @kindex set overload-resolution
14966 @cindex overloaded functions, overload resolution
14967 @item set overload-resolution on
14968 Enable overload resolution for C@t{++} expression evaluation. The default
14969 is on. For overloaded functions, @value{GDBN} evaluates the arguments
14970 and searches for a function whose signature matches the argument types,
14971 using the standard C@t{++} conversion rules (see @ref{C Plus Plus
14972 Expressions, ,C@t{++} Expressions}, for details).
14973 If it cannot find a match, it emits a message.
14975 @item set overload-resolution off
14976 Disable overload resolution for C@t{++} expression evaluation. For
14977 overloaded functions that are not class member functions, @value{GDBN}
14978 chooses the first function of the specified name that it finds in the
14979 symbol table, whether or not its arguments are of the correct type. For
14980 overloaded functions that are class member functions, @value{GDBN}
14981 searches for a function whose signature @emph{exactly} matches the
14984 @kindex show overload-resolution
14985 @item show overload-resolution
14986 Show the current setting of overload resolution.
14988 @item @r{Overloaded symbol names}
14989 You can specify a particular definition of an overloaded symbol, using
14990 the same notation that is used to declare such symbols in C@t{++}: type
14991 @code{@var{symbol}(@var{types})} rather than just @var{symbol}. You can
14992 also use the @value{GDBN} command-line word completion facilities to list the
14993 available choices, or to finish the type list for you.
14994 @xref{Completion,, Command Completion}, for details on how to do this.
14997 @node Decimal Floating Point
14998 @subsubsection Decimal Floating Point format
14999 @cindex decimal floating point format
15001 @value{GDBN} can examine, set and perform computations with numbers in
15002 decimal floating point format, which in the C language correspond to the
15003 @code{_Decimal32}, @code{_Decimal64} and @code{_Decimal128} types as
15004 specified by the extension to support decimal floating-point arithmetic.
15006 There are two encodings in use, depending on the architecture: BID (Binary
15007 Integer Decimal) for x86 and x86-64, and DPD (Densely Packed Decimal) for
15008 PowerPC and S/390. @value{GDBN} will use the appropriate encoding for the
15011 Because of a limitation in @file{libdecnumber}, the library used by @value{GDBN}
15012 to manipulate decimal floating point numbers, it is not possible to convert
15013 (using a cast, for example) integers wider than 32-bit to decimal float.
15015 In addition, in order to imitate @value{GDBN}'s behaviour with binary floating
15016 point computations, error checking in decimal float operations ignores
15017 underflow, overflow and divide by zero exceptions.
15019 In the PowerPC architecture, @value{GDBN} provides a set of pseudo-registers
15020 to inspect @code{_Decimal128} values stored in floating point registers.
15021 See @ref{PowerPC,,PowerPC} for more details.
15027 @value{GDBN} can be used to debug programs written in D and compiled with
15028 GDC, LDC or DMD compilers. Currently @value{GDBN} supports only one D
15029 specific feature --- dynamic arrays.
15034 @cindex Go (programming language)
15035 @value{GDBN} can be used to debug programs written in Go and compiled with
15036 @file{gccgo} or @file{6g} compilers.
15038 Here is a summary of the Go-specific features and restrictions:
15041 @cindex current Go package
15042 @item The current Go package
15043 The name of the current package does not need to be specified when
15044 specifying global variables and functions.
15046 For example, given the program:
15050 var myglob = "Shall we?"
15056 When stopped inside @code{main} either of these work:
15060 (gdb) p main.myglob
15063 @cindex builtin Go types
15064 @item Builtin Go types
15065 The @code{string} type is recognized by @value{GDBN} and is printed
15068 @cindex builtin Go functions
15069 @item Builtin Go functions
15070 The @value{GDBN} expression parser recognizes the @code{unsafe.Sizeof}
15071 function and handles it internally.
15073 @cindex restrictions on Go expressions
15074 @item Restrictions on Go expressions
15075 All Go operators are supported except @code{&^}.
15076 The Go @code{_} ``blank identifier'' is not supported.
15077 Automatic dereferencing of pointers is not supported.
15081 @subsection Objective-C
15083 @cindex Objective-C
15084 This section provides information about some commands and command
15085 options that are useful for debugging Objective-C code. See also
15086 @ref{Symbols, info classes}, and @ref{Symbols, info selectors}, for a
15087 few more commands specific to Objective-C support.
15090 * Method Names in Commands::
15091 * The Print Command with Objective-C::
15094 @node Method Names in Commands
15095 @subsubsection Method Names in Commands
15097 The following commands have been extended to accept Objective-C method
15098 names as line specifications:
15100 @kindex clear@r{, and Objective-C}
15101 @kindex break@r{, and Objective-C}
15102 @kindex info line@r{, and Objective-C}
15103 @kindex jump@r{, and Objective-C}
15104 @kindex list@r{, and Objective-C}
15108 @item @code{info line}
15113 A fully qualified Objective-C method name is specified as
15116 -[@var{Class} @var{methodName}]
15119 where the minus sign is used to indicate an instance method and a
15120 plus sign (not shown) is used to indicate a class method. The class
15121 name @var{Class} and method name @var{methodName} are enclosed in
15122 brackets, similar to the way messages are specified in Objective-C
15123 source code. For example, to set a breakpoint at the @code{create}
15124 instance method of class @code{Fruit} in the program currently being
15128 break -[Fruit create]
15131 To list ten program lines around the @code{initialize} class method,
15135 list +[NSText initialize]
15138 In the current version of @value{GDBN}, the plus or minus sign is
15139 required. In future versions of @value{GDBN}, the plus or minus
15140 sign will be optional, but you can use it to narrow the search. It
15141 is also possible to specify just a method name:
15147 You must specify the complete method name, including any colons. If
15148 your program's source files contain more than one @code{create} method,
15149 you'll be presented with a numbered list of classes that implement that
15150 method. Indicate your choice by number, or type @samp{0} to exit if
15153 As another example, to clear a breakpoint established at the
15154 @code{makeKeyAndOrderFront:} method of the @code{NSWindow} class, enter:
15157 clear -[NSWindow makeKeyAndOrderFront:]
15160 @node The Print Command with Objective-C
15161 @subsubsection The Print Command With Objective-C
15162 @cindex Objective-C, print objects
15163 @kindex print-object
15164 @kindex po @r{(@code{print-object})}
15166 The print command has also been extended to accept methods. For example:
15169 print -[@var{object} hash]
15172 @cindex print an Objective-C object description
15173 @cindex @code{_NSPrintForDebugger}, and printing Objective-C objects
15175 will tell @value{GDBN} to send the @code{hash} message to @var{object}
15176 and print the result. Also, an additional command has been added,
15177 @code{print-object} or @code{po} for short, which is meant to print
15178 the description of an object. However, this command may only work
15179 with certain Objective-C libraries that have a particular hook
15180 function, @code{_NSPrintForDebugger}, defined.
15183 @subsection OpenCL C
15186 This section provides information about @value{GDBN}s OpenCL C support.
15189 * OpenCL C Datatypes::
15190 * OpenCL C Expressions::
15191 * OpenCL C Operators::
15194 @node OpenCL C Datatypes
15195 @subsubsection OpenCL C Datatypes
15197 @cindex OpenCL C Datatypes
15198 @value{GDBN} supports the builtin scalar and vector datatypes specified
15199 by OpenCL 1.1. In addition the half- and double-precision floating point
15200 data types of the @code{cl_khr_fp16} and @code{cl_khr_fp64} OpenCL
15201 extensions are also known to @value{GDBN}.
15203 @node OpenCL C Expressions
15204 @subsubsection OpenCL C Expressions
15206 @cindex OpenCL C Expressions
15207 @value{GDBN} supports accesses to vector components including the access as
15208 lvalue where possible. Since OpenCL C is based on C99 most C expressions
15209 supported by @value{GDBN} can be used as well.
15211 @node OpenCL C Operators
15212 @subsubsection OpenCL C Operators
15214 @cindex OpenCL C Operators
15215 @value{GDBN} supports the operators specified by OpenCL 1.1 for scalar and
15219 @subsection Fortran
15220 @cindex Fortran-specific support in @value{GDBN}
15222 @value{GDBN} can be used to debug programs written in Fortran, but it
15223 currently supports only the features of Fortran 77 language.
15225 @cindex trailing underscore, in Fortran symbols
15226 Some Fortran compilers (@sc{gnu} Fortran 77 and Fortran 95 compilers
15227 among them) append an underscore to the names of variables and
15228 functions. When you debug programs compiled by those compilers, you
15229 will need to refer to variables and functions with a trailing
15233 * Fortran Operators:: Fortran operators and expressions
15234 * Fortran Defaults:: Default settings for Fortran
15235 * Special Fortran Commands:: Special @value{GDBN} commands for Fortran
15238 @node Fortran Operators
15239 @subsubsection Fortran Operators and Expressions
15241 @cindex Fortran operators and expressions
15243 Operators must be defined on values of specific types. For instance,
15244 @code{+} is defined on numbers, but not on characters or other non-
15245 arithmetic types. Operators are often defined on groups of types.
15249 The exponentiation operator. It raises the first operand to the power
15253 The range operator. Normally used in the form of array(low:high) to
15254 represent a section of array.
15257 The access component operator. Normally used to access elements in derived
15258 types. Also suitable for unions. As unions aren't part of regular Fortran,
15259 this can only happen when accessing a register that uses a gdbarch-defined
15263 @node Fortran Defaults
15264 @subsubsection Fortran Defaults
15266 @cindex Fortran Defaults
15268 Fortran symbols are usually case-insensitive, so @value{GDBN} by
15269 default uses case-insensitive matches for Fortran symbols. You can
15270 change that with the @samp{set case-insensitive} command, see
15271 @ref{Symbols}, for the details.
15273 @node Special Fortran Commands
15274 @subsubsection Special Fortran Commands
15276 @cindex Special Fortran commands
15278 @value{GDBN} has some commands to support Fortran-specific features,
15279 such as displaying common blocks.
15282 @cindex @code{COMMON} blocks, Fortran
15283 @kindex info common
15284 @item info common @r{[}@var{common-name}@r{]}
15285 This command prints the values contained in the Fortran @code{COMMON}
15286 block whose name is @var{common-name}. With no argument, the names of
15287 all @code{COMMON} blocks visible at the current program location are
15294 @cindex Pascal support in @value{GDBN}, limitations
15295 Debugging Pascal programs which use sets, subranges, file variables, or
15296 nested functions does not currently work. @value{GDBN} does not support
15297 entering expressions, printing values, or similar features using Pascal
15300 The Pascal-specific command @code{set print pascal_static-members}
15301 controls whether static members of Pascal objects are displayed.
15302 @xref{Print Settings, pascal_static-members}.
15307 @value{GDBN} supports the @url{https://www.rust-lang.org/, Rust
15308 Programming Language}. Type- and value-printing, and expression
15309 parsing, are reasonably complete. However, there are a few
15310 peculiarities and holes to be aware of.
15314 Linespecs (@pxref{Specify Location}) are never relative to the current
15315 crate. Instead, they act as if there were a global namespace of
15316 crates, somewhat similar to the way @code{extern crate} behaves.
15318 That is, if @value{GDBN} is stopped at a breakpoint in a function in
15319 crate @samp{A}, module @samp{B}, then @code{break B::f} will attempt
15320 to set a breakpoint in a function named @samp{f} in a crate named
15323 As a consequence of this approach, linespecs also cannot refer to
15324 items using @samp{self::} or @samp{super::}.
15327 Because @value{GDBN} implements Rust name-lookup semantics in
15328 expressions, it will sometimes prepend the current crate to a name.
15329 For example, if @value{GDBN} is stopped at a breakpoint in the crate
15330 @samp{K}, then @code{print ::x::y} will try to find the symbol
15333 However, since it is useful to be able to refer to other crates when
15334 debugging, @value{GDBN} provides the @code{extern} extension to
15335 circumvent this. To use the extension, just put @code{extern} before
15336 a path expression to refer to the otherwise unavailable ``global''
15339 In the above example, if you wanted to refer to the symbol @samp{y} in
15340 the crate @samp{x}, you would use @code{print extern x::y}.
15343 The Rust expression evaluator does not support ``statement-like''
15344 expressions such as @code{if} or @code{match}, or lambda expressions.
15347 Tuple expressions are not implemented.
15350 The Rust expression evaluator does not currently implement the
15351 @code{Drop} trait. Objects that may be created by the evaluator will
15352 never be destroyed.
15355 @value{GDBN} does not implement type inference for generics. In order
15356 to call generic functions or otherwise refer to generic items, you
15357 will have to specify the type parameters manually.
15360 @value{GDBN} currently uses the C@t{++} demangler for Rust. In most
15361 cases this does not cause any problems. However, in an expression
15362 context, completing a generic function name will give syntactically
15363 invalid results. This happens because Rust requires the @samp{::}
15364 operator between the function name and its generic arguments. For
15365 example, @value{GDBN} might provide a completion like
15366 @code{crate::f<u32>}, where the parser would require
15367 @code{crate::f::<u32>}.
15370 As of this writing, the Rust compiler (version 1.8) has a few holes in
15371 the debugging information it generates. These holes prevent certain
15372 features from being implemented by @value{GDBN}:
15376 Method calls cannot be made via traits.
15379 Trait objects cannot be created or inspected.
15382 Operator overloading is not implemented.
15385 When debugging in a monomorphized function, you cannot use the generic
15389 The type @code{Self} is not available.
15392 @code{use} statements are not available, so some names may not be
15393 available in the crate.
15398 @subsection Modula-2
15400 @cindex Modula-2, @value{GDBN} support
15402 The extensions made to @value{GDBN} to support Modula-2 only support
15403 output from the @sc{gnu} Modula-2 compiler (which is currently being
15404 developed). Other Modula-2 compilers are not currently supported, and
15405 attempting to debug executables produced by them is most likely
15406 to give an error as @value{GDBN} reads in the executable's symbol
15409 @cindex expressions in Modula-2
15411 * M2 Operators:: Built-in operators
15412 * Built-In Func/Proc:: Built-in functions and procedures
15413 * M2 Constants:: Modula-2 constants
15414 * M2 Types:: Modula-2 types
15415 * M2 Defaults:: Default settings for Modula-2
15416 * Deviations:: Deviations from standard Modula-2
15417 * M2 Checks:: Modula-2 type and range checks
15418 * M2 Scope:: The scope operators @code{::} and @code{.}
15419 * GDB/M2:: @value{GDBN} and Modula-2
15423 @subsubsection Operators
15424 @cindex Modula-2 operators
15426 Operators must be defined on values of specific types. For instance,
15427 @code{+} is defined on numbers, but not on structures. Operators are
15428 often defined on groups of types. For the purposes of Modula-2, the
15429 following definitions hold:
15434 @emph{Integral types} consist of @code{INTEGER}, @code{CARDINAL}, and
15438 @emph{Character types} consist of @code{CHAR} and its subranges.
15441 @emph{Floating-point types} consist of @code{REAL}.
15444 @emph{Pointer types} consist of anything declared as @code{POINTER TO
15448 @emph{Scalar types} consist of all of the above.
15451 @emph{Set types} consist of @code{SET} and @code{BITSET} types.
15454 @emph{Boolean types} consist of @code{BOOLEAN}.
15458 The following operators are supported, and appear in order of
15459 increasing precedence:
15463 Function argument or array index separator.
15466 Assignment. The value of @var{var} @code{:=} @var{value} is
15470 Less than, greater than on integral, floating-point, or enumerated
15474 Less than or equal to, greater than or equal to
15475 on integral, floating-point and enumerated types, or set inclusion on
15476 set types. Same precedence as @code{<}.
15478 @item =@r{, }<>@r{, }#
15479 Equality and two ways of expressing inequality, valid on scalar types.
15480 Same precedence as @code{<}. In @value{GDBN} scripts, only @code{<>} is
15481 available for inequality, since @code{#} conflicts with the script
15485 Set membership. Defined on set types and the types of their members.
15486 Same precedence as @code{<}.
15489 Boolean disjunction. Defined on boolean types.
15492 Boolean conjunction. Defined on boolean types.
15495 The @value{GDBN} ``artificial array'' operator (@pxref{Expressions, ,Expressions}).
15498 Addition and subtraction on integral and floating-point types, or union
15499 and difference on set types.
15502 Multiplication on integral and floating-point types, or set intersection
15506 Division on floating-point types, or symmetric set difference on set
15507 types. Same precedence as @code{*}.
15510 Integer division and remainder. Defined on integral types. Same
15511 precedence as @code{*}.
15514 Negative. Defined on @code{INTEGER} and @code{REAL} data.
15517 Pointer dereferencing. Defined on pointer types.
15520 Boolean negation. Defined on boolean types. Same precedence as
15524 @code{RECORD} field selector. Defined on @code{RECORD} data. Same
15525 precedence as @code{^}.
15528 Array indexing. Defined on @code{ARRAY} data. Same precedence as @code{^}.
15531 Procedure argument list. Defined on @code{PROCEDURE} objects. Same precedence
15535 @value{GDBN} and Modula-2 scope operators.
15539 @emph{Warning:} Set expressions and their operations are not yet supported, so @value{GDBN}
15540 treats the use of the operator @code{IN}, or the use of operators
15541 @code{+}, @code{-}, @code{*}, @code{/}, @code{=}, , @code{<>}, @code{#},
15542 @code{<=}, and @code{>=} on sets as an error.
15546 @node Built-In Func/Proc
15547 @subsubsection Built-in Functions and Procedures
15548 @cindex Modula-2 built-ins
15550 Modula-2 also makes available several built-in procedures and functions.
15551 In describing these, the following metavariables are used:
15556 represents an @code{ARRAY} variable.
15559 represents a @code{CHAR} constant or variable.
15562 represents a variable or constant of integral type.
15565 represents an identifier that belongs to a set. Generally used in the
15566 same function with the metavariable @var{s}. The type of @var{s} should
15567 be @code{SET OF @var{mtype}} (where @var{mtype} is the type of @var{m}).
15570 represents a variable or constant of integral or floating-point type.
15573 represents a variable or constant of floating-point type.
15579 represents a variable.
15582 represents a variable or constant of one of many types. See the
15583 explanation of the function for details.
15586 All Modula-2 built-in procedures also return a result, described below.
15590 Returns the absolute value of @var{n}.
15593 If @var{c} is a lower case letter, it returns its upper case
15594 equivalent, otherwise it returns its argument.
15597 Returns the character whose ordinal value is @var{i}.
15600 Decrements the value in the variable @var{v} by one. Returns the new value.
15602 @item DEC(@var{v},@var{i})
15603 Decrements the value in the variable @var{v} by @var{i}. Returns the
15606 @item EXCL(@var{m},@var{s})
15607 Removes the element @var{m} from the set @var{s}. Returns the new
15610 @item FLOAT(@var{i})
15611 Returns the floating point equivalent of the integer @var{i}.
15613 @item HIGH(@var{a})
15614 Returns the index of the last member of @var{a}.
15617 Increments the value in the variable @var{v} by one. Returns the new value.
15619 @item INC(@var{v},@var{i})
15620 Increments the value in the variable @var{v} by @var{i}. Returns the
15623 @item INCL(@var{m},@var{s})
15624 Adds the element @var{m} to the set @var{s} if it is not already
15625 there. Returns the new set.
15628 Returns the maximum value of the type @var{t}.
15631 Returns the minimum value of the type @var{t}.
15634 Returns boolean TRUE if @var{i} is an odd number.
15637 Returns the ordinal value of its argument. For example, the ordinal
15638 value of a character is its @sc{ascii} value (on machines supporting
15639 the @sc{ascii} character set). The argument @var{x} must be of an
15640 ordered type, which include integral, character and enumerated types.
15642 @item SIZE(@var{x})
15643 Returns the size of its argument. The argument @var{x} can be a
15644 variable or a type.
15646 @item TRUNC(@var{r})
15647 Returns the integral part of @var{r}.
15649 @item TSIZE(@var{x})
15650 Returns the size of its argument. The argument @var{x} can be a
15651 variable or a type.
15653 @item VAL(@var{t},@var{i})
15654 Returns the member of the type @var{t} whose ordinal value is @var{i}.
15658 @emph{Warning:} Sets and their operations are not yet supported, so
15659 @value{GDBN} treats the use of procedures @code{INCL} and @code{EXCL} as
15663 @cindex Modula-2 constants
15665 @subsubsection Constants
15667 @value{GDBN} allows you to express the constants of Modula-2 in the following
15673 Integer constants are simply a sequence of digits. When used in an
15674 expression, a constant is interpreted to be type-compatible with the
15675 rest of the expression. Hexadecimal integers are specified by a
15676 trailing @samp{H}, and octal integers by a trailing @samp{B}.
15679 Floating point constants appear as a sequence of digits, followed by a
15680 decimal point and another sequence of digits. An optional exponent can
15681 then be specified, in the form @samp{E@r{[}+@r{|}-@r{]}@var{nnn}}, where
15682 @samp{@r{[}+@r{|}-@r{]}@var{nnn}} is the desired exponent. All of the
15683 digits of the floating point constant must be valid decimal (base 10)
15687 Character constants consist of a single character enclosed by a pair of
15688 like quotes, either single (@code{'}) or double (@code{"}). They may
15689 also be expressed by their ordinal value (their @sc{ascii} value, usually)
15690 followed by a @samp{C}.
15693 String constants consist of a sequence of characters enclosed by a
15694 pair of like quotes, either single (@code{'}) or double (@code{"}).
15695 Escape sequences in the style of C are also allowed. @xref{C
15696 Constants, ,C and C@t{++} Constants}, for a brief explanation of escape
15700 Enumerated constants consist of an enumerated identifier.
15703 Boolean constants consist of the identifiers @code{TRUE} and
15707 Pointer constants consist of integral values only.
15710 Set constants are not yet supported.
15714 @subsubsection Modula-2 Types
15715 @cindex Modula-2 types
15717 Currently @value{GDBN} can print the following data types in Modula-2
15718 syntax: array types, record types, set types, pointer types, procedure
15719 types, enumerated types, subrange types and base types. You can also
15720 print the contents of variables declared using these type.
15721 This section gives a number of simple source code examples together with
15722 sample @value{GDBN} sessions.
15724 The first example contains the following section of code:
15733 and you can request @value{GDBN} to interrogate the type and value of
15734 @code{r} and @code{s}.
15737 (@value{GDBP}) print s
15739 (@value{GDBP}) ptype s
15741 (@value{GDBP}) print r
15743 (@value{GDBP}) ptype r
15748 Likewise if your source code declares @code{s} as:
15752 s: SET ['A'..'Z'] ;
15756 then you may query the type of @code{s} by:
15759 (@value{GDBP}) ptype s
15760 type = SET ['A'..'Z']
15764 Note that at present you cannot interactively manipulate set
15765 expressions using the debugger.
15767 The following example shows how you might declare an array in Modula-2
15768 and how you can interact with @value{GDBN} to print its type and contents:
15772 s: ARRAY [-10..10] OF CHAR ;
15776 (@value{GDBP}) ptype s
15777 ARRAY [-10..10] OF CHAR
15780 Note that the array handling is not yet complete and although the type
15781 is printed correctly, expression handling still assumes that all
15782 arrays have a lower bound of zero and not @code{-10} as in the example
15785 Here are some more type related Modula-2 examples:
15789 colour = (blue, red, yellow, green) ;
15790 t = [blue..yellow] ;
15798 The @value{GDBN} interaction shows how you can query the data type
15799 and value of a variable.
15802 (@value{GDBP}) print s
15804 (@value{GDBP}) ptype t
15805 type = [blue..yellow]
15809 In this example a Modula-2 array is declared and its contents
15810 displayed. Observe that the contents are written in the same way as
15811 their @code{C} counterparts.
15815 s: ARRAY [1..5] OF CARDINAL ;
15821 (@value{GDBP}) print s
15822 $1 = @{1, 0, 0, 0, 0@}
15823 (@value{GDBP}) ptype s
15824 type = ARRAY [1..5] OF CARDINAL
15827 The Modula-2 language interface to @value{GDBN} also understands
15828 pointer types as shown in this example:
15832 s: POINTER TO ARRAY [1..5] OF CARDINAL ;
15839 and you can request that @value{GDBN} describes the type of @code{s}.
15842 (@value{GDBP}) ptype s
15843 type = POINTER TO ARRAY [1..5] OF CARDINAL
15846 @value{GDBN} handles compound types as we can see in this example.
15847 Here we combine array types, record types, pointer types and subrange
15858 myarray = ARRAY myrange OF CARDINAL ;
15859 myrange = [-2..2] ;
15861 s: POINTER TO ARRAY myrange OF foo ;
15865 and you can ask @value{GDBN} to describe the type of @code{s} as shown
15869 (@value{GDBP}) ptype s
15870 type = POINTER TO ARRAY [-2..2] OF foo = RECORD
15873 f3 : ARRAY [-2..2] OF CARDINAL;
15878 @subsubsection Modula-2 Defaults
15879 @cindex Modula-2 defaults
15881 If type and range checking are set automatically by @value{GDBN}, they
15882 both default to @code{on} whenever the working language changes to
15883 Modula-2. This happens regardless of whether you or @value{GDBN}
15884 selected the working language.
15886 If you allow @value{GDBN} to set the language automatically, then entering
15887 code compiled from a file whose name ends with @file{.mod} sets the
15888 working language to Modula-2. @xref{Automatically, ,Having @value{GDBN}
15889 Infer the Source Language}, for further details.
15892 @subsubsection Deviations from Standard Modula-2
15893 @cindex Modula-2, deviations from
15895 A few changes have been made to make Modula-2 programs easier to debug.
15896 This is done primarily via loosening its type strictness:
15900 Unlike in standard Modula-2, pointer constants can be formed by
15901 integers. This allows you to modify pointer variables during
15902 debugging. (In standard Modula-2, the actual address contained in a
15903 pointer variable is hidden from you; it can only be modified
15904 through direct assignment to another pointer variable or expression that
15905 returned a pointer.)
15908 C escape sequences can be used in strings and characters to represent
15909 non-printable characters. @value{GDBN} prints out strings with these
15910 escape sequences embedded. Single non-printable characters are
15911 printed using the @samp{CHR(@var{nnn})} format.
15914 The assignment operator (@code{:=}) returns the value of its right-hand
15918 All built-in procedures both modify @emph{and} return their argument.
15922 @subsubsection Modula-2 Type and Range Checks
15923 @cindex Modula-2 checks
15926 @emph{Warning:} in this release, @value{GDBN} does not yet perform type or
15929 @c FIXME remove warning when type/range checks added
15931 @value{GDBN} considers two Modula-2 variables type equivalent if:
15935 They are of types that have been declared equivalent via a @code{TYPE
15936 @var{t1} = @var{t2}} statement
15939 They have been declared on the same line. (Note: This is true of the
15940 @sc{gnu} Modula-2 compiler, but it may not be true of other compilers.)
15943 As long as type checking is enabled, any attempt to combine variables
15944 whose types are not equivalent is an error.
15946 Range checking is done on all mathematical operations, assignment, array
15947 index bounds, and all built-in functions and procedures.
15950 @subsubsection The Scope Operators @code{::} and @code{.}
15952 @cindex @code{.}, Modula-2 scope operator
15953 @cindex colon, doubled as scope operator
15955 @vindex colon-colon@r{, in Modula-2}
15956 @c Info cannot handle :: but TeX can.
15959 @vindex ::@r{, in Modula-2}
15962 There are a few subtle differences between the Modula-2 scope operator
15963 (@code{.}) and the @value{GDBN} scope operator (@code{::}). The two have
15968 @var{module} . @var{id}
15969 @var{scope} :: @var{id}
15973 where @var{scope} is the name of a module or a procedure,
15974 @var{module} the name of a module, and @var{id} is any declared
15975 identifier within your program, except another module.
15977 Using the @code{::} operator makes @value{GDBN} search the scope
15978 specified by @var{scope} for the identifier @var{id}. If it is not
15979 found in the specified scope, then @value{GDBN} searches all scopes
15980 enclosing the one specified by @var{scope}.
15982 Using the @code{.} operator makes @value{GDBN} search the current scope for
15983 the identifier specified by @var{id} that was imported from the
15984 definition module specified by @var{module}. With this operator, it is
15985 an error if the identifier @var{id} was not imported from definition
15986 module @var{module}, or if @var{id} is not an identifier in
15990 @subsubsection @value{GDBN} and Modula-2
15992 Some @value{GDBN} commands have little use when debugging Modula-2 programs.
15993 Five subcommands of @code{set print} and @code{show print} apply
15994 specifically to C and C@t{++}: @samp{vtbl}, @samp{demangle},
15995 @samp{asm-demangle}, @samp{object}, and @samp{union}. The first four
15996 apply to C@t{++}, and the last to the C @code{union} type, which has no direct
15997 analogue in Modula-2.
15999 The @code{@@} operator (@pxref{Expressions, ,Expressions}), while available
16000 with any language, is not useful with Modula-2. Its
16001 intent is to aid the debugging of @dfn{dynamic arrays}, which cannot be
16002 created in Modula-2 as they can in C or C@t{++}. However, because an
16003 address can be specified by an integral constant, the construct
16004 @samp{@{@var{type}@}@var{adrexp}} is still useful.
16006 @cindex @code{#} in Modula-2
16007 In @value{GDBN} scripts, the Modula-2 inequality operator @code{#} is
16008 interpreted as the beginning of a comment. Use @code{<>} instead.
16014 The extensions made to @value{GDBN} for Ada only support
16015 output from the @sc{gnu} Ada (GNAT) compiler.
16016 Other Ada compilers are not currently supported, and
16017 attempting to debug executables produced by them is most likely
16021 @cindex expressions in Ada
16023 * Ada Mode Intro:: General remarks on the Ada syntax
16024 and semantics supported by Ada mode
16026 * Omissions from Ada:: Restrictions on the Ada expression syntax.
16027 * Additions to Ada:: Extensions of the Ada expression syntax.
16028 * Overloading support for Ada:: Support for expressions involving overloaded
16030 * Stopping Before Main Program:: Debugging the program during elaboration.
16031 * Ada Exceptions:: Ada Exceptions
16032 * Ada Tasks:: Listing and setting breakpoints in tasks.
16033 * Ada Tasks and Core Files:: Tasking Support when Debugging Core Files
16034 * Ravenscar Profile:: Tasking Support when using the Ravenscar
16036 * Ada Glitches:: Known peculiarities of Ada mode.
16039 @node Ada Mode Intro
16040 @subsubsection Introduction
16041 @cindex Ada mode, general
16043 The Ada mode of @value{GDBN} supports a fairly large subset of Ada expression
16044 syntax, with some extensions.
16045 The philosophy behind the design of this subset is
16049 That @value{GDBN} should provide basic literals and access to operations for
16050 arithmetic, dereferencing, field selection, indexing, and subprogram calls,
16051 leaving more sophisticated computations to subprograms written into the
16052 program (which therefore may be called from @value{GDBN}).
16055 That type safety and strict adherence to Ada language restrictions
16056 are not particularly important to the @value{GDBN} user.
16059 That brevity is important to the @value{GDBN} user.
16062 Thus, for brevity, the debugger acts as if all names declared in
16063 user-written packages are directly visible, even if they are not visible
16064 according to Ada rules, thus making it unnecessary to fully qualify most
16065 names with their packages, regardless of context. Where this causes
16066 ambiguity, @value{GDBN} asks the user's intent.
16068 The debugger will start in Ada mode if it detects an Ada main program.
16069 As for other languages, it will enter Ada mode when stopped in a program that
16070 was translated from an Ada source file.
16072 While in Ada mode, you may use `@t{--}' for comments. This is useful
16073 mostly for documenting command files. The standard @value{GDBN} comment
16074 (@samp{#}) still works at the beginning of a line in Ada mode, but not in the
16075 middle (to allow based literals).
16077 @node Omissions from Ada
16078 @subsubsection Omissions from Ada
16079 @cindex Ada, omissions from
16081 Here are the notable omissions from the subset:
16085 Only a subset of the attributes are supported:
16089 @t{'First}, @t{'Last}, and @t{'Length}
16090 on array objects (not on types and subtypes).
16093 @t{'Min} and @t{'Max}.
16096 @t{'Pos} and @t{'Val}.
16102 @t{'Range} on array objects (not subtypes), but only as the right
16103 operand of the membership (@code{in}) operator.
16106 @t{'Access}, @t{'Unchecked_Access}, and
16107 @t{'Unrestricted_Access} (a GNAT extension).
16115 @code{Characters.Latin_1} are not available and
16116 concatenation is not implemented. Thus, escape characters in strings are
16117 not currently available.
16120 Equality tests (@samp{=} and @samp{/=}) on arrays test for bitwise
16121 equality of representations. They will generally work correctly
16122 for strings and arrays whose elements have integer or enumeration types.
16123 They may not work correctly for arrays whose element
16124 types have user-defined equality, for arrays of real values
16125 (in particular, IEEE-conformant floating point, because of negative
16126 zeroes and NaNs), and for arrays whose elements contain unused bits with
16127 indeterminate values.
16130 The other component-by-component array operations (@code{and}, @code{or},
16131 @code{xor}, @code{not}, and relational tests other than equality)
16132 are not implemented.
16135 @cindex array aggregates (Ada)
16136 @cindex record aggregates (Ada)
16137 @cindex aggregates (Ada)
16138 There is limited support for array and record aggregates. They are
16139 permitted only on the right sides of assignments, as in these examples:
16142 (@value{GDBP}) set An_Array := (1, 2, 3, 4, 5, 6)
16143 (@value{GDBP}) set An_Array := (1, others => 0)
16144 (@value{GDBP}) set An_Array := (0|4 => 1, 1..3 => 2, 5 => 6)
16145 (@value{GDBP}) set A_2D_Array := ((1, 2, 3), (4, 5, 6), (7, 8, 9))
16146 (@value{GDBP}) set A_Record := (1, "Peter", True);
16147 (@value{GDBP}) set A_Record := (Name => "Peter", Id => 1, Alive => True)
16151 discriminant's value by assigning an aggregate has an
16152 undefined effect if that discriminant is used within the record.
16153 However, you can first modify discriminants by directly assigning to
16154 them (which normally would not be allowed in Ada), and then performing an
16155 aggregate assignment. For example, given a variable @code{A_Rec}
16156 declared to have a type such as:
16159 type Rec (Len : Small_Integer := 0) is record
16161 Vals : IntArray (1 .. Len);
16165 you can assign a value with a different size of @code{Vals} with two
16169 (@value{GDBP}) set A_Rec.Len := 4
16170 (@value{GDBP}) set A_Rec := (Id => 42, Vals => (1, 2, 3, 4))
16173 As this example also illustrates, @value{GDBN} is very loose about the usual
16174 rules concerning aggregates. You may leave out some of the
16175 components of an array or record aggregate (such as the @code{Len}
16176 component in the assignment to @code{A_Rec} above); they will retain their
16177 original values upon assignment. You may freely use dynamic values as
16178 indices in component associations. You may even use overlapping or
16179 redundant component associations, although which component values are
16180 assigned in such cases is not defined.
16183 Calls to dispatching subprograms are not implemented.
16186 The overloading algorithm is much more limited (i.e., less selective)
16187 than that of real Ada. It makes only limited use of the context in
16188 which a subexpression appears to resolve its meaning, and it is much
16189 looser in its rules for allowing type matches. As a result, some
16190 function calls will be ambiguous, and the user will be asked to choose
16191 the proper resolution.
16194 The @code{new} operator is not implemented.
16197 Entry calls are not implemented.
16200 Aside from printing, arithmetic operations on the native VAX floating-point
16201 formats are not supported.
16204 It is not possible to slice a packed array.
16207 The names @code{True} and @code{False}, when not part of a qualified name,
16208 are interpreted as if implicitly prefixed by @code{Standard}, regardless of
16210 Should your program
16211 redefine these names in a package or procedure (at best a dubious practice),
16212 you will have to use fully qualified names to access their new definitions.
16215 @node Additions to Ada
16216 @subsubsection Additions to Ada
16217 @cindex Ada, deviations from
16219 As it does for other languages, @value{GDBN} makes certain generic
16220 extensions to Ada (@pxref{Expressions}):
16224 If the expression @var{E} is a variable residing in memory (typically
16225 a local variable or array element) and @var{N} is a positive integer,
16226 then @code{@var{E}@@@var{N}} displays the values of @var{E} and the
16227 @var{N}-1 adjacent variables following it in memory as an array. In
16228 Ada, this operator is generally not necessary, since its prime use is
16229 in displaying parts of an array, and slicing will usually do this in
16230 Ada. However, there are occasional uses when debugging programs in
16231 which certain debugging information has been optimized away.
16234 @code{@var{B}::@var{var}} means ``the variable named @var{var} that
16235 appears in function or file @var{B}.'' When @var{B} is a file name,
16236 you must typically surround it in single quotes.
16239 The expression @code{@{@var{type}@} @var{addr}} means ``the variable of type
16240 @var{type} that appears at address @var{addr}.''
16243 A name starting with @samp{$} is a convenience variable
16244 (@pxref{Convenience Vars}) or a machine register (@pxref{Registers}).
16247 In addition, @value{GDBN} provides a few other shortcuts and outright
16248 additions specific to Ada:
16252 The assignment statement is allowed as an expression, returning
16253 its right-hand operand as its value. Thus, you may enter
16256 (@value{GDBP}) set x := y + 3
16257 (@value{GDBP}) print A(tmp := y + 1)
16261 The semicolon is allowed as an ``operator,'' returning as its value
16262 the value of its right-hand operand.
16263 This allows, for example,
16264 complex conditional breaks:
16267 (@value{GDBP}) break f
16268 (@value{GDBP}) condition 1 (report(i); k += 1; A(k) > 100)
16272 Rather than use catenation and symbolic character names to introduce special
16273 characters into strings, one may instead use a special bracket notation,
16274 which is also used to print strings. A sequence of characters of the form
16275 @samp{["@var{XX}"]} within a string or character literal denotes the
16276 (single) character whose numeric encoding is @var{XX} in hexadecimal. The
16277 sequence of characters @samp{["""]} also denotes a single quotation mark
16278 in strings. For example,
16280 "One line.["0a"]Next line.["0a"]"
16283 contains an ASCII newline character (@code{Ada.Characters.Latin_1.LF})
16287 The subtype used as a prefix for the attributes @t{'Pos}, @t{'Min}, and
16288 @t{'Max} is optional (and is ignored in any case). For example, it is valid
16292 (@value{GDBP}) print 'max(x, y)
16296 When printing arrays, @value{GDBN} uses positional notation when the
16297 array has a lower bound of 1, and uses a modified named notation otherwise.
16298 For example, a one-dimensional array of three integers with a lower bound
16299 of 3 might print as
16306 That is, in contrast to valid Ada, only the first component has a @code{=>}
16310 You may abbreviate attributes in expressions with any unique,
16311 multi-character subsequence of
16312 their names (an exact match gets preference).
16313 For example, you may use @t{a'len}, @t{a'gth}, or @t{a'lh}
16314 in place of @t{a'length}.
16317 @cindex quoting Ada internal identifiers
16318 Since Ada is case-insensitive, the debugger normally maps identifiers you type
16319 to lower case. The GNAT compiler uses upper-case characters for
16320 some of its internal identifiers, which are normally of no interest to users.
16321 For the rare occasions when you actually have to look at them,
16322 enclose them in angle brackets to avoid the lower-case mapping.
16325 (@value{GDBP}) print <JMPBUF_SAVE>[0]
16329 Printing an object of class-wide type or dereferencing an
16330 access-to-class-wide value will display all the components of the object's
16331 specific type (as indicated by its run-time tag). Likewise, component
16332 selection on such a value will operate on the specific type of the
16337 @node Overloading support for Ada
16338 @subsubsection Overloading support for Ada
16339 @cindex overloading, Ada
16341 The debugger supports limited overloading. Given a subprogram call in which
16342 the function symbol has multiple definitions, it will use the number of
16343 actual parameters and some information about their types to attempt to narrow
16344 the set of definitions. It also makes very limited use of context, preferring
16345 procedures to functions in the context of the @code{call} command, and
16346 functions to procedures elsewhere.
16348 If, after narrowing, the set of matching definitions still contains more than
16349 one definition, @value{GDBN} will display a menu to query which one it should
16353 (@value{GDBP}) print f(1)
16354 Multiple matches for f
16356 [1] foo.f (integer) return boolean at foo.adb:23
16357 [2] foo.f (foo.new_integer) return boolean at foo.adb:28
16361 In this case, just select one menu entry either to cancel expression evaluation
16362 (type @kbd{0} and press @key{RET}) or to continue evaluation with a specific
16363 instance (type the corresponding number and press @key{RET}).
16365 Here are a couple of commands to customize @value{GDBN}'s behavior in this
16370 @kindex set ada print-signatures
16371 @item set ada print-signatures
16372 Control whether parameter types and return types are displayed in overloads
16373 selection menus. It is @code{on} by default.
16374 @xref{Overloading support for Ada}.
16376 @kindex show ada print-signatures
16377 @item show ada print-signatures
16378 Show the current setting for displaying parameter types and return types in
16379 overloads selection menu.
16380 @xref{Overloading support for Ada}.
16384 @node Stopping Before Main Program
16385 @subsubsection Stopping at the Very Beginning
16387 @cindex breakpointing Ada elaboration code
16388 It is sometimes necessary to debug the program during elaboration, and
16389 before reaching the main procedure.
16390 As defined in the Ada Reference
16391 Manual, the elaboration code is invoked from a procedure called
16392 @code{adainit}. To run your program up to the beginning of
16393 elaboration, simply use the following two commands:
16394 @code{tbreak adainit} and @code{run}.
16396 @node Ada Exceptions
16397 @subsubsection Ada Exceptions
16399 A command is provided to list all Ada exceptions:
16402 @kindex info exceptions
16403 @item info exceptions
16404 @itemx info exceptions @var{regexp}
16405 The @code{info exceptions} command allows you to list all Ada exceptions
16406 defined within the program being debugged, as well as their addresses.
16407 With a regular expression, @var{regexp}, as argument, only those exceptions
16408 whose names match @var{regexp} are listed.
16411 Below is a small example, showing how the command can be used, first
16412 without argument, and next with a regular expression passed as an
16416 (@value{GDBP}) info exceptions
16417 All defined Ada exceptions:
16418 constraint_error: 0x613da0
16419 program_error: 0x613d20
16420 storage_error: 0x613ce0
16421 tasking_error: 0x613ca0
16422 const.aint_global_e: 0x613b00
16423 (@value{GDBP}) info exceptions const.aint
16424 All Ada exceptions matching regular expression "const.aint":
16425 constraint_error: 0x613da0
16426 const.aint_global_e: 0x613b00
16429 It is also possible to ask @value{GDBN} to stop your program's execution
16430 when an exception is raised. For more details, see @ref{Set Catchpoints}.
16433 @subsubsection Extensions for Ada Tasks
16434 @cindex Ada, tasking
16436 Support for Ada tasks is analogous to that for threads (@pxref{Threads}).
16437 @value{GDBN} provides the following task-related commands:
16442 This command shows a list of current Ada tasks, as in the following example:
16449 (@value{GDBP}) info tasks
16450 ID TID P-ID Pri State Name
16451 1 8088000 0 15 Child Activation Wait main_task
16452 2 80a4000 1 15 Accept Statement b
16453 3 809a800 1 15 Child Activation Wait a
16454 * 4 80ae800 3 15 Runnable c
16459 In this listing, the asterisk before the last task indicates it to be the
16460 task currently being inspected.
16464 Represents @value{GDBN}'s internal task number.
16470 The parent's task ID (@value{GDBN}'s internal task number).
16473 The base priority of the task.
16476 Current state of the task.
16480 The task has been created but has not been activated. It cannot be
16484 The task is not blocked for any reason known to Ada. (It may be waiting
16485 for a mutex, though.) It is conceptually "executing" in normal mode.
16488 The task is terminated, in the sense of ARM 9.3 (5). Any dependents
16489 that were waiting on terminate alternatives have been awakened and have
16490 terminated themselves.
16492 @item Child Activation Wait
16493 The task is waiting for created tasks to complete activation.
16495 @item Accept Statement
16496 The task is waiting on an accept or selective wait statement.
16498 @item Waiting on entry call
16499 The task is waiting on an entry call.
16501 @item Async Select Wait
16502 The task is waiting to start the abortable part of an asynchronous
16506 The task is waiting on a select statement with only a delay
16509 @item Child Termination Wait
16510 The task is sleeping having completed a master within itself, and is
16511 waiting for the tasks dependent on that master to become terminated or
16512 waiting on a terminate Phase.
16514 @item Wait Child in Term Alt
16515 The task is sleeping waiting for tasks on terminate alternatives to
16516 finish terminating.
16518 @item Accepting RV with @var{taskno}
16519 The task is accepting a rendez-vous with the task @var{taskno}.
16523 Name of the task in the program.
16527 @kindex info task @var{taskno}
16528 @item info task @var{taskno}
16529 This command shows detailled informations on the specified task, as in
16530 the following example:
16535 (@value{GDBP}) info tasks
16536 ID TID P-ID Pri State Name
16537 1 8077880 0 15 Child Activation Wait main_task
16538 * 2 807c468 1 15 Runnable task_1
16539 (@value{GDBP}) info task 2
16540 Ada Task: 0x807c468
16543 Parent: 1 (main_task)
16549 @kindex task@r{ (Ada)}
16550 @cindex current Ada task ID
16551 This command prints the ID of the current task.
16557 (@value{GDBP}) info tasks
16558 ID TID P-ID Pri State Name
16559 1 8077870 0 15 Child Activation Wait main_task
16560 * 2 807c458 1 15 Runnable t
16561 (@value{GDBP}) task
16562 [Current task is 2]
16565 @item task @var{taskno}
16566 @cindex Ada task switching
16567 This command is like the @code{thread @var{thread-id}}
16568 command (@pxref{Threads}). It switches the context of debugging
16569 from the current task to the given task.
16575 (@value{GDBP}) info tasks
16576 ID TID P-ID Pri State Name
16577 1 8077870 0 15 Child Activation Wait main_task
16578 * 2 807c458 1 15 Runnable t
16579 (@value{GDBP}) task 1
16580 [Switching to task 1]
16581 #0 0x8067726 in pthread_cond_wait ()
16583 #0 0x8067726 in pthread_cond_wait ()
16584 #1 0x8056714 in system.os_interface.pthread_cond_wait ()
16585 #2 0x805cb63 in system.task_primitives.operations.sleep ()
16586 #3 0x806153e in system.tasking.stages.activate_tasks ()
16587 #4 0x804aacc in un () at un.adb:5
16590 @item break @var{location} task @var{taskno}
16591 @itemx break @var{location} task @var{taskno} if @dots{}
16592 @cindex breakpoints and tasks, in Ada
16593 @cindex task breakpoints, in Ada
16594 @kindex break @dots{} task @var{taskno}@r{ (Ada)}
16595 These commands are like the @code{break @dots{} thread @dots{}}
16596 command (@pxref{Thread Stops}). The
16597 @var{location} argument specifies source lines, as described
16598 in @ref{Specify Location}.
16600 Use the qualifier @samp{task @var{taskno}} with a breakpoint command
16601 to specify that you only want @value{GDBN} to stop the program when a
16602 particular Ada task reaches this breakpoint. The @var{taskno} is one of the
16603 numeric task identifiers assigned by @value{GDBN}, shown in the first
16604 column of the @samp{info tasks} display.
16606 If you do not specify @samp{task @var{taskno}} when you set a
16607 breakpoint, the breakpoint applies to @emph{all} tasks of your
16610 You can use the @code{task} qualifier on conditional breakpoints as
16611 well; in this case, place @samp{task @var{taskno}} before the
16612 breakpoint condition (before the @code{if}).
16620 (@value{GDBP}) info tasks
16621 ID TID P-ID Pri State Name
16622 1 140022020 0 15 Child Activation Wait main_task
16623 2 140045060 1 15 Accept/Select Wait t2
16624 3 140044840 1 15 Runnable t1
16625 * 4 140056040 1 15 Runnable t3
16626 (@value{GDBP}) b 15 task 2
16627 Breakpoint 5 at 0x120044cb0: file test_task_debug.adb, line 15.
16628 (@value{GDBP}) cont
16633 Breakpoint 5, test_task_debug () at test_task_debug.adb:15
16635 (@value{GDBP}) info tasks
16636 ID TID P-ID Pri State Name
16637 1 140022020 0 15 Child Activation Wait main_task
16638 * 2 140045060 1 15 Runnable t2
16639 3 140044840 1 15 Runnable t1
16640 4 140056040 1 15 Delay Sleep t3
16644 @node Ada Tasks and Core Files
16645 @subsubsection Tasking Support when Debugging Core Files
16646 @cindex Ada tasking and core file debugging
16648 When inspecting a core file, as opposed to debugging a live program,
16649 tasking support may be limited or even unavailable, depending on
16650 the platform being used.
16651 For instance, on x86-linux, the list of tasks is available, but task
16652 switching is not supported.
16654 On certain platforms, the debugger needs to perform some
16655 memory writes in order to provide Ada tasking support. When inspecting
16656 a core file, this means that the core file must be opened with read-write
16657 privileges, using the command @samp{"set write on"} (@pxref{Patching}).
16658 Under these circumstances, you should make a backup copy of the core
16659 file before inspecting it with @value{GDBN}.
16661 @node Ravenscar Profile
16662 @subsubsection Tasking Support when using the Ravenscar Profile
16663 @cindex Ravenscar Profile
16665 The @dfn{Ravenscar Profile} is a subset of the Ada tasking features,
16666 specifically designed for systems with safety-critical real-time
16670 @kindex set ravenscar task-switching on
16671 @cindex task switching with program using Ravenscar Profile
16672 @item set ravenscar task-switching on
16673 Allows task switching when debugging a program that uses the Ravenscar
16674 Profile. This is the default.
16676 @kindex set ravenscar task-switching off
16677 @item set ravenscar task-switching off
16678 Turn off task switching when debugging a program that uses the Ravenscar
16679 Profile. This is mostly intended to disable the code that adds support
16680 for the Ravenscar Profile, in case a bug in either @value{GDBN} or in
16681 the Ravenscar runtime is preventing @value{GDBN} from working properly.
16682 To be effective, this command should be run before the program is started.
16684 @kindex show ravenscar task-switching
16685 @item show ravenscar task-switching
16686 Show whether it is possible to switch from task to task in a program
16687 using the Ravenscar Profile.
16692 @subsubsection Known Peculiarities of Ada Mode
16693 @cindex Ada, problems
16695 Besides the omissions listed previously (@pxref{Omissions from Ada}),
16696 we know of several problems with and limitations of Ada mode in
16698 some of which will be fixed with planned future releases of the debugger
16699 and the GNU Ada compiler.
16703 Static constants that the compiler chooses not to materialize as objects in
16704 storage are invisible to the debugger.
16707 Named parameter associations in function argument lists are ignored (the
16708 argument lists are treated as positional).
16711 Many useful library packages are currently invisible to the debugger.
16714 Fixed-point arithmetic, conversions, input, and output is carried out using
16715 floating-point arithmetic, and may give results that only approximate those on
16719 The GNAT compiler never generates the prefix @code{Standard} for any of
16720 the standard symbols defined by the Ada language. @value{GDBN} knows about
16721 this: it will strip the prefix from names when you use it, and will never
16722 look for a name you have so qualified among local symbols, nor match against
16723 symbols in other packages or subprograms. If you have
16724 defined entities anywhere in your program other than parameters and
16725 local variables whose simple names match names in @code{Standard},
16726 GNAT's lack of qualification here can cause confusion. When this happens,
16727 you can usually resolve the confusion
16728 by qualifying the problematic names with package
16729 @code{Standard} explicitly.
16732 Older versions of the compiler sometimes generate erroneous debugging
16733 information, resulting in the debugger incorrectly printing the value
16734 of affected entities. In some cases, the debugger is able to work
16735 around an issue automatically. In other cases, the debugger is able
16736 to work around the issue, but the work-around has to be specifically
16739 @kindex set ada trust-PAD-over-XVS
16740 @kindex show ada trust-PAD-over-XVS
16743 @item set ada trust-PAD-over-XVS on
16744 Configure GDB to strictly follow the GNAT encoding when computing the
16745 value of Ada entities, particularly when @code{PAD} and @code{PAD___XVS}
16746 types are involved (see @code{ada/exp_dbug.ads} in the GCC sources for
16747 a complete description of the encoding used by the GNAT compiler).
16748 This is the default.
16750 @item set ada trust-PAD-over-XVS off
16751 This is related to the encoding using by the GNAT compiler. If @value{GDBN}
16752 sometimes prints the wrong value for certain entities, changing @code{ada
16753 trust-PAD-over-XVS} to @code{off} activates a work-around which may fix
16754 the issue. It is always safe to set @code{ada trust-PAD-over-XVS} to
16755 @code{off}, but this incurs a slight performance penalty, so it is
16756 recommended to leave this setting to @code{on} unless necessary.
16760 @cindex GNAT descriptive types
16761 @cindex GNAT encoding
16762 Internally, the debugger also relies on the compiler following a number
16763 of conventions known as the @samp{GNAT Encoding}, all documented in
16764 @file{gcc/ada/exp_dbug.ads} in the GCC sources. This encoding describes
16765 how the debugging information should be generated for certain types.
16766 In particular, this convention makes use of @dfn{descriptive types},
16767 which are artificial types generated purely to help the debugger.
16769 These encodings were defined at a time when the debugging information
16770 format used was not powerful enough to describe some of the more complex
16771 types available in Ada. Since DWARF allows us to express nearly all
16772 Ada features, the long-term goal is to slowly replace these descriptive
16773 types by their pure DWARF equivalent. To facilitate that transition,
16774 a new maintenance option is available to force the debugger to ignore
16775 those descriptive types. It allows the user to quickly evaluate how
16776 well @value{GDBN} works without them.
16780 @kindex maint ada set ignore-descriptive-types
16781 @item maintenance ada set ignore-descriptive-types [on|off]
16782 Control whether the debugger should ignore descriptive types.
16783 The default is not to ignore descriptives types (@code{off}).
16785 @kindex maint ada show ignore-descriptive-types
16786 @item maintenance ada show ignore-descriptive-types
16787 Show if descriptive types are ignored by @value{GDBN}.
16791 @node Unsupported Languages
16792 @section Unsupported Languages
16794 @cindex unsupported languages
16795 @cindex minimal language
16796 In addition to the other fully-supported programming languages,
16797 @value{GDBN} also provides a pseudo-language, called @code{minimal}.
16798 It does not represent a real programming language, but provides a set
16799 of capabilities close to what the C or assembly languages provide.
16800 This should allow most simple operations to be performed while debugging
16801 an application that uses a language currently not supported by @value{GDBN}.
16803 If the language is set to @code{auto}, @value{GDBN} will automatically
16804 select this language if the current frame corresponds to an unsupported
16808 @chapter Examining the Symbol Table
16810 The commands described in this chapter allow you to inquire about the
16811 symbols (names of variables, functions and types) defined in your
16812 program. This information is inherent in the text of your program and
16813 does not change as your program executes. @value{GDBN} finds it in your
16814 program's symbol table, in the file indicated when you started @value{GDBN}
16815 (@pxref{File Options, ,Choosing Files}), or by one of the
16816 file-management commands (@pxref{Files, ,Commands to Specify Files}).
16818 @cindex symbol names
16819 @cindex names of symbols
16820 @cindex quoting names
16821 Occasionally, you may need to refer to symbols that contain unusual
16822 characters, which @value{GDBN} ordinarily treats as word delimiters. The
16823 most frequent case is in referring to static variables in other
16824 source files (@pxref{Variables,,Program Variables}). File names
16825 are recorded in object files as debugging symbols, but @value{GDBN} would
16826 ordinarily parse a typical file name, like @file{foo.c}, as the three words
16827 @samp{foo} @samp{.} @samp{c}. To allow @value{GDBN} to recognize
16828 @samp{foo.c} as a single symbol, enclose it in single quotes; for example,
16835 looks up the value of @code{x} in the scope of the file @file{foo.c}.
16838 @cindex case-insensitive symbol names
16839 @cindex case sensitivity in symbol names
16840 @kindex set case-sensitive
16841 @item set case-sensitive on
16842 @itemx set case-sensitive off
16843 @itemx set case-sensitive auto
16844 Normally, when @value{GDBN} looks up symbols, it matches their names
16845 with case sensitivity determined by the current source language.
16846 Occasionally, you may wish to control that. The command @code{set
16847 case-sensitive} lets you do that by specifying @code{on} for
16848 case-sensitive matches or @code{off} for case-insensitive ones. If
16849 you specify @code{auto}, case sensitivity is reset to the default
16850 suitable for the source language. The default is case-sensitive
16851 matches for all languages except for Fortran, for which the default is
16852 case-insensitive matches.
16854 @kindex show case-sensitive
16855 @item show case-sensitive
16856 This command shows the current setting of case sensitivity for symbols
16859 @kindex set print type methods
16860 @item set print type methods
16861 @itemx set print type methods on
16862 @itemx set print type methods off
16863 Normally, when @value{GDBN} prints a class, it displays any methods
16864 declared in that class. You can control this behavior either by
16865 passing the appropriate flag to @code{ptype}, or using @command{set
16866 print type methods}. Specifying @code{on} will cause @value{GDBN} to
16867 display the methods; this is the default. Specifying @code{off} will
16868 cause @value{GDBN} to omit the methods.
16870 @kindex show print type methods
16871 @item show print type methods
16872 This command shows the current setting of method display when printing
16875 @kindex set print type typedefs
16876 @item set print type typedefs
16877 @itemx set print type typedefs on
16878 @itemx set print type typedefs off
16880 Normally, when @value{GDBN} prints a class, it displays any typedefs
16881 defined in that class. You can control this behavior either by
16882 passing the appropriate flag to @code{ptype}, or using @command{set
16883 print type typedefs}. Specifying @code{on} will cause @value{GDBN} to
16884 display the typedef definitions; this is the default. Specifying
16885 @code{off} will cause @value{GDBN} to omit the typedef definitions.
16886 Note that this controls whether the typedef definition itself is
16887 printed, not whether typedef names are substituted when printing other
16890 @kindex show print type typedefs
16891 @item show print type typedefs
16892 This command shows the current setting of typedef display when
16895 @kindex info address
16896 @cindex address of a symbol
16897 @item info address @var{symbol}
16898 Describe where the data for @var{symbol} is stored. For a register
16899 variable, this says which register it is kept in. For a non-register
16900 local variable, this prints the stack-frame offset at which the variable
16903 Note the contrast with @samp{print &@var{symbol}}, which does not work
16904 at all for a register variable, and for a stack local variable prints
16905 the exact address of the current instantiation of the variable.
16907 @kindex info symbol
16908 @cindex symbol from address
16909 @cindex closest symbol and offset for an address
16910 @item info symbol @var{addr}
16911 Print the name of a symbol which is stored at the address @var{addr}.
16912 If no symbol is stored exactly at @var{addr}, @value{GDBN} prints the
16913 nearest symbol and an offset from it:
16916 (@value{GDBP}) info symbol 0x54320
16917 _initialize_vx + 396 in section .text
16921 This is the opposite of the @code{info address} command. You can use
16922 it to find out the name of a variable or a function given its address.
16924 For dynamically linked executables, the name of executable or shared
16925 library containing the symbol is also printed:
16928 (@value{GDBP}) info symbol 0x400225
16929 _start + 5 in section .text of /tmp/a.out
16930 (@value{GDBP}) info symbol 0x2aaaac2811cf
16931 __read_nocancel + 6 in section .text of /usr/lib64/libc.so.6
16936 @item demangle @r{[}-l @var{language}@r{]} @r{[}@var{--}@r{]} @var{name}
16937 Demangle @var{name}.
16938 If @var{language} is provided it is the name of the language to demangle
16939 @var{name} in. Otherwise @var{name} is demangled in the current language.
16941 The @samp{--} option specifies the end of options,
16942 and is useful when @var{name} begins with a dash.
16944 The parameter @code{demangle-style} specifies how to interpret the kind
16945 of mangling used. @xref{Print Settings}.
16948 @item whatis[/@var{flags}] [@var{arg}]
16949 Print the data type of @var{arg}, which can be either an expression
16950 or a name of a data type. With no argument, print the data type of
16951 @code{$}, the last value in the value history.
16953 If @var{arg} is an expression (@pxref{Expressions, ,Expressions}), it
16954 is not actually evaluated, and any side-effecting operations (such as
16955 assignments or function calls) inside it do not take place.
16957 If @var{arg} is a variable or an expression, @code{whatis} prints its
16958 literal type as it is used in the source code. If the type was
16959 defined using a @code{typedef}, @code{whatis} will @emph{not} print
16960 the data type underlying the @code{typedef}. If the type of the
16961 variable or the expression is a compound data type, such as
16962 @code{struct} or @code{class}, @code{whatis} never prints their
16963 fields or methods. It just prints the @code{struct}/@code{class}
16964 name (a.k.a.@: its @dfn{tag}). If you want to see the members of
16965 such a compound data type, use @code{ptype}.
16967 If @var{arg} is a type name that was defined using @code{typedef},
16968 @code{whatis} @dfn{unrolls} only one level of that @code{typedef}.
16969 Unrolling means that @code{whatis} will show the underlying type used
16970 in the @code{typedef} declaration of @var{arg}. However, if that
16971 underlying type is also a @code{typedef}, @code{whatis} will not
16974 For C code, the type names may also have the form @samp{class
16975 @var{class-name}}, @samp{struct @var{struct-tag}}, @samp{union
16976 @var{union-tag}} or @samp{enum @var{enum-tag}}.
16978 @var{flags} can be used to modify how the type is displayed.
16979 Available flags are:
16983 Display in ``raw'' form. Normally, @value{GDBN} substitutes template
16984 parameters and typedefs defined in a class when printing the class'
16985 members. The @code{/r} flag disables this.
16988 Do not print methods defined in the class.
16991 Print methods defined in the class. This is the default, but the flag
16992 exists in case you change the default with @command{set print type methods}.
16995 Do not print typedefs defined in the class. Note that this controls
16996 whether the typedef definition itself is printed, not whether typedef
16997 names are substituted when printing other types.
17000 Print typedefs defined in the class. This is the default, but the flag
17001 exists in case you change the default with @command{set print type typedefs}.
17005 @item ptype[/@var{flags}] [@var{arg}]
17006 @code{ptype} accepts the same arguments as @code{whatis}, but prints a
17007 detailed description of the type, instead of just the name of the type.
17008 @xref{Expressions, ,Expressions}.
17010 Contrary to @code{whatis}, @code{ptype} always unrolls any
17011 @code{typedef}s in its argument declaration, whether the argument is
17012 a variable, expression, or a data type. This means that @code{ptype}
17013 of a variable or an expression will not print literally its type as
17014 present in the source code---use @code{whatis} for that. @code{typedef}s at
17015 the pointer or reference targets are also unrolled. Only @code{typedef}s of
17016 fields, methods and inner @code{class typedef}s of @code{struct}s,
17017 @code{class}es and @code{union}s are not unrolled even with @code{ptype}.
17019 For example, for this variable declaration:
17022 typedef double real_t;
17023 struct complex @{ real_t real; double imag; @};
17024 typedef struct complex complex_t;
17026 real_t *real_pointer_var;
17030 the two commands give this output:
17034 (@value{GDBP}) whatis var
17036 (@value{GDBP}) ptype var
17037 type = struct complex @{
17041 (@value{GDBP}) whatis complex_t
17042 type = struct complex
17043 (@value{GDBP}) whatis struct complex
17044 type = struct complex
17045 (@value{GDBP}) ptype struct complex
17046 type = struct complex @{
17050 (@value{GDBP}) whatis real_pointer_var
17052 (@value{GDBP}) ptype real_pointer_var
17058 As with @code{whatis}, using @code{ptype} without an argument refers to
17059 the type of @code{$}, the last value in the value history.
17061 @cindex incomplete type
17062 Sometimes, programs use opaque data types or incomplete specifications
17063 of complex data structure. If the debug information included in the
17064 program does not allow @value{GDBN} to display a full declaration of
17065 the data type, it will say @samp{<incomplete type>}. For example,
17066 given these declarations:
17070 struct foo *fooptr;
17074 but no definition for @code{struct foo} itself, @value{GDBN} will say:
17077 (@value{GDBP}) ptype foo
17078 $1 = <incomplete type>
17082 ``Incomplete type'' is C terminology for data types that are not
17083 completely specified.
17086 @item info types @var{regexp}
17088 Print a brief description of all types whose names match the regular
17089 expression @var{regexp} (or all types in your program, if you supply
17090 no argument). Each complete typename is matched as though it were a
17091 complete line; thus, @samp{i type value} gives information on all
17092 types in your program whose names include the string @code{value}, but
17093 @samp{i type ^value$} gives information only on types whose complete
17094 name is @code{value}.
17096 This command differs from @code{ptype} in two ways: first, like
17097 @code{whatis}, it does not print a detailed description; second, it
17098 lists all source files where a type is defined.
17100 @kindex info type-printers
17101 @item info type-printers
17102 Versions of @value{GDBN} that ship with Python scripting enabled may
17103 have ``type printers'' available. When using @command{ptype} or
17104 @command{whatis}, these printers are consulted when the name of a type
17105 is needed. @xref{Type Printing API}, for more information on writing
17108 @code{info type-printers} displays all the available type printers.
17110 @kindex enable type-printer
17111 @kindex disable type-printer
17112 @item enable type-printer @var{name}@dots{}
17113 @item disable type-printer @var{name}@dots{}
17114 These commands can be used to enable or disable type printers.
17117 @cindex local variables
17118 @item info scope @var{location}
17119 List all the variables local to a particular scope. This command
17120 accepts a @var{location} argument---a function name, a source line, or
17121 an address preceded by a @samp{*}, and prints all the variables local
17122 to the scope defined by that location. (@xref{Specify Location}, for
17123 details about supported forms of @var{location}.) For example:
17126 (@value{GDBP}) @b{info scope command_line_handler}
17127 Scope for command_line_handler:
17128 Symbol rl is an argument at stack/frame offset 8, length 4.
17129 Symbol linebuffer is in static storage at address 0x150a18, length 4.
17130 Symbol linelength is in static storage at address 0x150a1c, length 4.
17131 Symbol p is a local variable in register $esi, length 4.
17132 Symbol p1 is a local variable in register $ebx, length 4.
17133 Symbol nline is a local variable in register $edx, length 4.
17134 Symbol repeat is a local variable at frame offset -8, length 4.
17138 This command is especially useful for determining what data to collect
17139 during a @dfn{trace experiment}, see @ref{Tracepoint Actions,
17142 @kindex info source
17144 Show information about the current source file---that is, the source file for
17145 the function containing the current point of execution:
17148 the name of the source file, and the directory containing it,
17150 the directory it was compiled in,
17152 its length, in lines,
17154 which programming language it is written in,
17156 if the debug information provides it, the program that compiled the file
17157 (which may include, e.g., the compiler version and command line arguments),
17159 whether the executable includes debugging information for that file, and
17160 if so, what format the information is in (e.g., STABS, Dwarf 2, etc.), and
17162 whether the debugging information includes information about
17163 preprocessor macros.
17167 @kindex info sources
17169 Print the names of all source files in your program for which there is
17170 debugging information, organized into two lists: files whose symbols
17171 have already been read, and files whose symbols will be read when needed.
17173 @kindex info functions
17174 @item info functions
17175 Print the names and data types of all defined functions.
17177 @item info functions @var{regexp}
17178 Print the names and data types of all defined functions
17179 whose names contain a match for regular expression @var{regexp}.
17180 Thus, @samp{info fun step} finds all functions whose names
17181 include @code{step}; @samp{info fun ^step} finds those whose names
17182 start with @code{step}. If a function name contains characters
17183 that conflict with the regular expression language (e.g.@:
17184 @samp{operator*()}), they may be quoted with a backslash.
17186 @kindex info variables
17187 @item info variables
17188 Print the names and data types of all variables that are defined
17189 outside of functions (i.e.@: excluding local variables).
17191 @item info variables @var{regexp}
17192 Print the names and data types of all variables (except for local
17193 variables) whose names contain a match for regular expression
17196 @kindex info classes
17197 @cindex Objective-C, classes and selectors
17199 @itemx info classes @var{regexp}
17200 Display all Objective-C classes in your program, or
17201 (with the @var{regexp} argument) all those matching a particular regular
17204 @kindex info selectors
17205 @item info selectors
17206 @itemx info selectors @var{regexp}
17207 Display all Objective-C selectors in your program, or
17208 (with the @var{regexp} argument) all those matching a particular regular
17212 This was never implemented.
17213 @kindex info methods
17215 @itemx info methods @var{regexp}
17216 The @code{info methods} command permits the user to examine all defined
17217 methods within C@t{++} program, or (with the @var{regexp} argument) a
17218 specific set of methods found in the various C@t{++} classes. Many
17219 C@t{++} classes provide a large number of methods. Thus, the output
17220 from the @code{ptype} command can be overwhelming and hard to use. The
17221 @code{info-methods} command filters the methods, printing only those
17222 which match the regular-expression @var{regexp}.
17225 @cindex opaque data types
17226 @kindex set opaque-type-resolution
17227 @item set opaque-type-resolution on
17228 Tell @value{GDBN} to resolve opaque types. An opaque type is a type
17229 declared as a pointer to a @code{struct}, @code{class}, or
17230 @code{union}---for example, @code{struct MyType *}---that is used in one
17231 source file although the full declaration of @code{struct MyType} is in
17232 another source file. The default is on.
17234 A change in the setting of this subcommand will not take effect until
17235 the next time symbols for a file are loaded.
17237 @item set opaque-type-resolution off
17238 Tell @value{GDBN} not to resolve opaque types. In this case, the type
17239 is printed as follows:
17241 @{<no data fields>@}
17244 @kindex show opaque-type-resolution
17245 @item show opaque-type-resolution
17246 Show whether opaque types are resolved or not.
17248 @kindex set print symbol-loading
17249 @cindex print messages when symbols are loaded
17250 @item set print symbol-loading
17251 @itemx set print symbol-loading full
17252 @itemx set print symbol-loading brief
17253 @itemx set print symbol-loading off
17254 The @code{set print symbol-loading} command allows you to control the
17255 printing of messages when @value{GDBN} loads symbol information.
17256 By default a message is printed for the executable and one for each
17257 shared library, and normally this is what you want. However, when
17258 debugging apps with large numbers of shared libraries these messages
17260 When set to @code{brief} a message is printed for each executable,
17261 and when @value{GDBN} loads a collection of shared libraries at once
17262 it will only print one message regardless of the number of shared
17263 libraries. When set to @code{off} no messages are printed.
17265 @kindex show print symbol-loading
17266 @item show print symbol-loading
17267 Show whether messages will be printed when a @value{GDBN} command
17268 entered from the keyboard causes symbol information to be loaded.
17270 @kindex maint print symbols
17271 @cindex symbol dump
17272 @kindex maint print psymbols
17273 @cindex partial symbol dump
17274 @kindex maint print msymbols
17275 @cindex minimal symbol dump
17276 @item maint print symbols @r{[}-pc @var{address}@r{]} @r{[}@var{filename}@r{]}
17277 @itemx maint print symbols @r{[}-objfile @var{objfile}@r{]} @r{[}-source @var{source}@r{]} @r{[}--@r{]} @r{[}@var{filename}@r{]}
17278 @itemx maint print psymbols @r{[}-objfile @var{objfile}@r{]} @r{[}-pc @var{address}@r{]} @r{[}--@r{]} @r{[}@var{filename}@r{]}
17279 @itemx maint print psymbols @r{[}-objfile @var{objfile}@r{]} @r{[}-source @var{source}@r{]} @r{[}--@r{]} @r{[}@var{filename}@r{]}
17280 @itemx maint print msymbols @r{[}-objfile @var{objfile}@r{]} @r{[}--@r{]} @r{[}@var{filename}@r{]}
17281 Write a dump of debugging symbol data into the file @var{filename} or
17282 the terminal if @var{filename} is unspecified.
17283 If @code{-objfile @var{objfile}} is specified, only dump symbols for
17285 If @code{-pc @var{address}} is specified, only dump symbols for the file
17286 with code at that address. Note that @var{address} may be a symbol like
17288 If @code{-source @var{source}} is specified, only dump symbols for that
17291 These commands are used to debug the @value{GDBN} symbol-reading code.
17292 These commands do not modify internal @value{GDBN} state, therefore
17293 @samp{maint print symbols} will only print symbols for already expanded symbol
17295 You can use the command @code{info sources} to find out which files these are.
17296 If you use @samp{maint print psymbols} instead, the dump shows information
17297 about symbols that @value{GDBN} only knows partially---that is, symbols
17298 defined in files that @value{GDBN} has skimmed, but not yet read completely.
17299 Finally, @samp{maint print msymbols} just dumps ``minimal symbols'', e.g.,
17302 @xref{Files, ,Commands to Specify Files}, for a discussion of how
17303 @value{GDBN} reads symbols (in the description of @code{symbol-file}).
17305 @kindex maint info symtabs
17306 @kindex maint info psymtabs
17307 @cindex listing @value{GDBN}'s internal symbol tables
17308 @cindex symbol tables, listing @value{GDBN}'s internal
17309 @cindex full symbol tables, listing @value{GDBN}'s internal
17310 @cindex partial symbol tables, listing @value{GDBN}'s internal
17311 @item maint info symtabs @r{[} @var{regexp} @r{]}
17312 @itemx maint info psymtabs @r{[} @var{regexp} @r{]}
17314 List the @code{struct symtab} or @code{struct partial_symtab}
17315 structures whose names match @var{regexp}. If @var{regexp} is not
17316 given, list them all. The output includes expressions which you can
17317 copy into a @value{GDBN} debugging this one to examine a particular
17318 structure in more detail. For example:
17321 (@value{GDBP}) maint info psymtabs dwarf2read
17322 @{ objfile /home/gnu/build/gdb/gdb
17323 ((struct objfile *) 0x82e69d0)
17324 @{ psymtab /home/gnu/src/gdb/dwarf2read.c
17325 ((struct partial_symtab *) 0x8474b10)
17328 text addresses 0x814d3c8 -- 0x8158074
17329 globals (* (struct partial_symbol **) 0x8507a08 @@ 9)
17330 statics (* (struct partial_symbol **) 0x40e95b78 @@ 2882)
17331 dependencies (none)
17334 (@value{GDBP}) maint info symtabs
17338 We see that there is one partial symbol table whose filename contains
17339 the string @samp{dwarf2read}, belonging to the @samp{gdb} executable;
17340 and we see that @value{GDBN} has not read in any symtabs yet at all.
17341 If we set a breakpoint on a function, that will cause @value{GDBN} to
17342 read the symtab for the compilation unit containing that function:
17345 (@value{GDBP}) break dwarf2_psymtab_to_symtab
17346 Breakpoint 1 at 0x814e5da: file /home/gnu/src/gdb/dwarf2read.c,
17348 (@value{GDBP}) maint info symtabs
17349 @{ objfile /home/gnu/build/gdb/gdb
17350 ((struct objfile *) 0x82e69d0)
17351 @{ symtab /home/gnu/src/gdb/dwarf2read.c
17352 ((struct symtab *) 0x86c1f38)
17355 blockvector ((struct blockvector *) 0x86c1bd0) (primary)
17356 linetable ((struct linetable *) 0x8370fa0)
17357 debugformat DWARF 2
17363 @kindex maint info line-table
17364 @cindex listing @value{GDBN}'s internal line tables
17365 @cindex line tables, listing @value{GDBN}'s internal
17366 @item maint info line-table @r{[} @var{regexp} @r{]}
17368 List the @code{struct linetable} from all @code{struct symtab}
17369 instances whose name matches @var{regexp}. If @var{regexp} is not
17370 given, list the @code{struct linetable} from all @code{struct symtab}.
17372 @kindex maint set symbol-cache-size
17373 @cindex symbol cache size
17374 @item maint set symbol-cache-size @var{size}
17375 Set the size of the symbol cache to @var{size}.
17376 The default size is intended to be good enough for debugging
17377 most applications. This option exists to allow for experimenting
17378 with different sizes.
17380 @kindex maint show symbol-cache-size
17381 @item maint show symbol-cache-size
17382 Show the size of the symbol cache.
17384 @kindex maint print symbol-cache
17385 @cindex symbol cache, printing its contents
17386 @item maint print symbol-cache
17387 Print the contents of the symbol cache.
17388 This is useful when debugging symbol cache issues.
17390 @kindex maint print symbol-cache-statistics
17391 @cindex symbol cache, printing usage statistics
17392 @item maint print symbol-cache-statistics
17393 Print symbol cache usage statistics.
17394 This helps determine how well the cache is being utilized.
17396 @kindex maint flush-symbol-cache
17397 @cindex symbol cache, flushing
17398 @item maint flush-symbol-cache
17399 Flush the contents of the symbol cache, all entries are removed.
17400 This command is useful when debugging the symbol cache.
17401 It is also useful when collecting performance data.
17406 @chapter Altering Execution
17408 Once you think you have found an error in your program, you might want to
17409 find out for certain whether correcting the apparent error would lead to
17410 correct results in the rest of the run. You can find the answer by
17411 experiment, using the @value{GDBN} features for altering execution of the
17414 For example, you can store new values into variables or memory
17415 locations, give your program a signal, restart it at a different
17416 address, or even return prematurely from a function.
17419 * Assignment:: Assignment to variables
17420 * Jumping:: Continuing at a different address
17421 * Signaling:: Giving your program a signal
17422 * Returning:: Returning from a function
17423 * Calling:: Calling your program's functions
17424 * Patching:: Patching your program
17425 * Compiling and Injecting Code:: Compiling and injecting code in @value{GDBN}
17429 @section Assignment to Variables
17432 @cindex setting variables
17433 To alter the value of a variable, evaluate an assignment expression.
17434 @xref{Expressions, ,Expressions}. For example,
17441 stores the value 4 into the variable @code{x}, and then prints the
17442 value of the assignment expression (which is 4).
17443 @xref{Languages, ,Using @value{GDBN} with Different Languages}, for more
17444 information on operators in supported languages.
17446 @kindex set variable
17447 @cindex variables, setting
17448 If you are not interested in seeing the value of the assignment, use the
17449 @code{set} command instead of the @code{print} command. @code{set} is
17450 really the same as @code{print} except that the expression's value is
17451 not printed and is not put in the value history (@pxref{Value History,
17452 ,Value History}). The expression is evaluated only for its effects.
17454 If the beginning of the argument string of the @code{set} command
17455 appears identical to a @code{set} subcommand, use the @code{set
17456 variable} command instead of just @code{set}. This command is identical
17457 to @code{set} except for its lack of subcommands. For example, if your
17458 program has a variable @code{width}, you get an error if you try to set
17459 a new value with just @samp{set width=13}, because @value{GDBN} has the
17460 command @code{set width}:
17463 (@value{GDBP}) whatis width
17465 (@value{GDBP}) p width
17467 (@value{GDBP}) set width=47
17468 Invalid syntax in expression.
17472 The invalid expression, of course, is @samp{=47}. In
17473 order to actually set the program's variable @code{width}, use
17476 (@value{GDBP}) set var width=47
17479 Because the @code{set} command has many subcommands that can conflict
17480 with the names of program variables, it is a good idea to use the
17481 @code{set variable} command instead of just @code{set}. For example, if
17482 your program has a variable @code{g}, you run into problems if you try
17483 to set a new value with just @samp{set g=4}, because @value{GDBN} has
17484 the command @code{set gnutarget}, abbreviated @code{set g}:
17488 (@value{GDBP}) whatis g
17492 (@value{GDBP}) set g=4
17496 The program being debugged has been started already.
17497 Start it from the beginning? (y or n) y
17498 Starting program: /home/smith/cc_progs/a.out
17499 "/home/smith/cc_progs/a.out": can't open to read symbols:
17500 Invalid bfd target.
17501 (@value{GDBP}) show g
17502 The current BFD target is "=4".
17507 The program variable @code{g} did not change, and you silently set the
17508 @code{gnutarget} to an invalid value. In order to set the variable
17512 (@value{GDBP}) set var g=4
17515 @value{GDBN} allows more implicit conversions in assignments than C; you can
17516 freely store an integer value into a pointer variable or vice versa,
17517 and you can convert any structure to any other structure that is the
17518 same length or shorter.
17519 @comment FIXME: how do structs align/pad in these conversions?
17520 @comment /doc@cygnus.com 18dec1990
17522 To store values into arbitrary places in memory, use the @samp{@{@dots{}@}}
17523 construct to generate a value of specified type at a specified address
17524 (@pxref{Expressions, ,Expressions}). For example, @code{@{int@}0x83040} refers
17525 to memory location @code{0x83040} as an integer (which implies a certain size
17526 and representation in memory), and
17529 set @{int@}0x83040 = 4
17533 stores the value 4 into that memory location.
17536 @section Continuing at a Different Address
17538 Ordinarily, when you continue your program, you do so at the place where
17539 it stopped, with the @code{continue} command. You can instead continue at
17540 an address of your own choosing, with the following commands:
17544 @kindex j @r{(@code{jump})}
17545 @item jump @var{location}
17546 @itemx j @var{location}
17547 Resume execution at @var{location}. Execution stops again immediately
17548 if there is a breakpoint there. @xref{Specify Location}, for a description
17549 of the different forms of @var{location}. It is common
17550 practice to use the @code{tbreak} command in conjunction with
17551 @code{jump}. @xref{Set Breaks, ,Setting Breakpoints}.
17553 The @code{jump} command does not change the current stack frame, or
17554 the stack pointer, or the contents of any memory location or any
17555 register other than the program counter. If @var{location} is in
17556 a different function from the one currently executing, the results may
17557 be bizarre if the two functions expect different patterns of arguments or
17558 of local variables. For this reason, the @code{jump} command requests
17559 confirmation if the specified line is not in the function currently
17560 executing. However, even bizarre results are predictable if you are
17561 well acquainted with the machine-language code of your program.
17564 On many systems, you can get much the same effect as the @code{jump}
17565 command by storing a new value into the register @code{$pc}. The
17566 difference is that this does not start your program running; it only
17567 changes the address of where it @emph{will} run when you continue. For
17575 makes the next @code{continue} command or stepping command execute at
17576 address @code{0x485}, rather than at the address where your program stopped.
17577 @xref{Continuing and Stepping, ,Continuing and Stepping}.
17579 The most common occasion to use the @code{jump} command is to back
17580 up---perhaps with more breakpoints set---over a portion of a program
17581 that has already executed, in order to examine its execution in more
17586 @section Giving your Program a Signal
17587 @cindex deliver a signal to a program
17591 @item signal @var{signal}
17592 Resume execution where your program is stopped, but immediately give it the
17593 signal @var{signal}. The @var{signal} can be the name or the number of a
17594 signal. For example, on many systems @code{signal 2} and @code{signal
17595 SIGINT} are both ways of sending an interrupt signal.
17597 Alternatively, if @var{signal} is zero, continue execution without
17598 giving a signal. This is useful when your program stopped on account of
17599 a signal and would ordinarily see the signal when resumed with the
17600 @code{continue} command; @samp{signal 0} causes it to resume without a
17603 @emph{Note:} When resuming a multi-threaded program, @var{signal} is
17604 delivered to the currently selected thread, not the thread that last
17605 reported a stop. This includes the situation where a thread was
17606 stopped due to a signal. So if you want to continue execution
17607 suppressing the signal that stopped a thread, you should select that
17608 same thread before issuing the @samp{signal 0} command. If you issue
17609 the @samp{signal 0} command with another thread as the selected one,
17610 @value{GDBN} detects that and asks for confirmation.
17612 Invoking the @code{signal} command is not the same as invoking the
17613 @code{kill} utility from the shell. Sending a signal with @code{kill}
17614 causes @value{GDBN} to decide what to do with the signal depending on
17615 the signal handling tables (@pxref{Signals}). The @code{signal} command
17616 passes the signal directly to your program.
17618 @code{signal} does not repeat when you press @key{RET} a second time
17619 after executing the command.
17621 @kindex queue-signal
17622 @item queue-signal @var{signal}
17623 Queue @var{signal} to be delivered immediately to the current thread
17624 when execution of the thread resumes. The @var{signal} can be the name or
17625 the number of a signal. For example, on many systems @code{signal 2} and
17626 @code{signal SIGINT} are both ways of sending an interrupt signal.
17627 The handling of the signal must be set to pass the signal to the program,
17628 otherwise @value{GDBN} will report an error.
17629 You can control the handling of signals from @value{GDBN} with the
17630 @code{handle} command (@pxref{Signals}).
17632 Alternatively, if @var{signal} is zero, any currently queued signal
17633 for the current thread is discarded and when execution resumes no signal
17634 will be delivered. This is useful when your program stopped on account
17635 of a signal and would ordinarily see the signal when resumed with the
17636 @code{continue} command.
17638 This command differs from the @code{signal} command in that the signal
17639 is just queued, execution is not resumed. And @code{queue-signal} cannot
17640 be used to pass a signal whose handling state has been set to @code{nopass}
17645 @xref{stepping into signal handlers}, for information on how stepping
17646 commands behave when the thread has a signal queued.
17649 @section Returning from a Function
17652 @cindex returning from a function
17655 @itemx return @var{expression}
17656 You can cancel execution of a function call with the @code{return}
17657 command. If you give an
17658 @var{expression} argument, its value is used as the function's return
17662 When you use @code{return}, @value{GDBN} discards the selected stack frame
17663 (and all frames within it). You can think of this as making the
17664 discarded frame return prematurely. If you wish to specify a value to
17665 be returned, give that value as the argument to @code{return}.
17667 This pops the selected stack frame (@pxref{Selection, ,Selecting a
17668 Frame}), and any other frames inside of it, leaving its caller as the
17669 innermost remaining frame. That frame becomes selected. The
17670 specified value is stored in the registers used for returning values
17673 The @code{return} command does not resume execution; it leaves the
17674 program stopped in the state that would exist if the function had just
17675 returned. In contrast, the @code{finish} command (@pxref{Continuing
17676 and Stepping, ,Continuing and Stepping}) resumes execution until the
17677 selected stack frame returns naturally.
17679 @value{GDBN} needs to know how the @var{expression} argument should be set for
17680 the inferior. The concrete registers assignment depends on the OS ABI and the
17681 type being returned by the selected stack frame. For example it is common for
17682 OS ABI to return floating point values in FPU registers while integer values in
17683 CPU registers. Still some ABIs return even floating point values in CPU
17684 registers. Larger integer widths (such as @code{long long int}) also have
17685 specific placement rules. @value{GDBN} already knows the OS ABI from its
17686 current target so it needs to find out also the type being returned to make the
17687 assignment into the right register(s).
17689 Normally, the selected stack frame has debug info. @value{GDBN} will always
17690 use the debug info instead of the implicit type of @var{expression} when the
17691 debug info is available. For example, if you type @kbd{return -1}, and the
17692 function in the current stack frame is declared to return a @code{long long
17693 int}, @value{GDBN} transparently converts the implicit @code{int} value of -1
17694 into a @code{long long int}:
17697 Breakpoint 1, func () at gdb.base/return-nodebug.c:29
17699 (@value{GDBP}) return -1
17700 Make func return now? (y or n) y
17701 #0 0x004004f6 in main () at gdb.base/return-nodebug.c:43
17702 43 printf ("result=%lld\n", func ());
17706 However, if the selected stack frame does not have a debug info, e.g., if the
17707 function was compiled without debug info, @value{GDBN} has to find out the type
17708 to return from user. Specifying a different type by mistake may set the value
17709 in different inferior registers than the caller code expects. For example,
17710 typing @kbd{return -1} with its implicit type @code{int} would set only a part
17711 of a @code{long long int} result for a debug info less function (on 32-bit
17712 architectures). Therefore the user is required to specify the return type by
17713 an appropriate cast explicitly:
17716 Breakpoint 2, 0x0040050b in func ()
17717 (@value{GDBP}) return -1
17718 Return value type not available for selected stack frame.
17719 Please use an explicit cast of the value to return.
17720 (@value{GDBP}) return (long long int) -1
17721 Make selected stack frame return now? (y or n) y
17722 #0 0x00400526 in main ()
17727 @section Calling Program Functions
17730 @cindex calling functions
17731 @cindex inferior functions, calling
17732 @item print @var{expr}
17733 Evaluate the expression @var{expr} and display the resulting value.
17734 The expression may include calls to functions in the program being
17738 @item call @var{expr}
17739 Evaluate the expression @var{expr} without displaying @code{void}
17742 You can use this variant of the @code{print} command if you want to
17743 execute a function from your program that does not return anything
17744 (a.k.a.@: @dfn{a void function}), but without cluttering the output
17745 with @code{void} returned values that @value{GDBN} will otherwise
17746 print. If the result is not void, it is printed and saved in the
17750 It is possible for the function you call via the @code{print} or
17751 @code{call} command to generate a signal (e.g., if there's a bug in
17752 the function, or if you passed it incorrect arguments). What happens
17753 in that case is controlled by the @code{set unwindonsignal} command.
17755 Similarly, with a C@t{++} program it is possible for the function you
17756 call via the @code{print} or @code{call} command to generate an
17757 exception that is not handled due to the constraints of the dummy
17758 frame. In this case, any exception that is raised in the frame, but has
17759 an out-of-frame exception handler will not be found. GDB builds a
17760 dummy-frame for the inferior function call, and the unwinder cannot
17761 seek for exception handlers outside of this dummy-frame. What happens
17762 in that case is controlled by the
17763 @code{set unwind-on-terminating-exception} command.
17766 @item set unwindonsignal
17767 @kindex set unwindonsignal
17768 @cindex unwind stack in called functions
17769 @cindex call dummy stack unwinding
17770 Set unwinding of the stack if a signal is received while in a function
17771 that @value{GDBN} called in the program being debugged. If set to on,
17772 @value{GDBN} unwinds the stack it created for the call and restores
17773 the context to what it was before the call. If set to off (the
17774 default), @value{GDBN} stops in the frame where the signal was
17777 @item show unwindonsignal
17778 @kindex show unwindonsignal
17779 Show the current setting of stack unwinding in the functions called by
17782 @item set unwind-on-terminating-exception
17783 @kindex set unwind-on-terminating-exception
17784 @cindex unwind stack in called functions with unhandled exceptions
17785 @cindex call dummy stack unwinding on unhandled exception.
17786 Set unwinding of the stack if a C@t{++} exception is raised, but left
17787 unhandled while in a function that @value{GDBN} called in the program being
17788 debugged. If set to on (the default), @value{GDBN} unwinds the stack
17789 it created for the call and restores the context to what it was before
17790 the call. If set to off, @value{GDBN} the exception is delivered to
17791 the default C@t{++} exception handler and the inferior terminated.
17793 @item show unwind-on-terminating-exception
17794 @kindex show unwind-on-terminating-exception
17795 Show the current setting of stack unwinding in the functions called by
17800 @cindex weak alias functions
17801 Sometimes, a function you wish to call is actually a @dfn{weak alias}
17802 for another function. In such case, @value{GDBN} might not pick up
17803 the type information, including the types of the function arguments,
17804 which causes @value{GDBN} to call the inferior function incorrectly.
17805 As a result, the called function will function erroneously and may
17806 even crash. A solution to that is to use the name of the aliased
17810 @section Patching Programs
17812 @cindex patching binaries
17813 @cindex writing into executables
17814 @cindex writing into corefiles
17816 By default, @value{GDBN} opens the file containing your program's
17817 executable code (or the corefile) read-only. This prevents accidental
17818 alterations to machine code; but it also prevents you from intentionally
17819 patching your program's binary.
17821 If you'd like to be able to patch the binary, you can specify that
17822 explicitly with the @code{set write} command. For example, you might
17823 want to turn on internal debugging flags, or even to make emergency
17829 @itemx set write off
17830 If you specify @samp{set write on}, @value{GDBN} opens executable and
17831 core files for both reading and writing; if you specify @kbd{set write
17832 off} (the default), @value{GDBN} opens them read-only.
17834 If you have already loaded a file, you must load it again (using the
17835 @code{exec-file} or @code{core-file} command) after changing @code{set
17836 write}, for your new setting to take effect.
17840 Display whether executable files and core files are opened for writing
17841 as well as reading.
17844 @node Compiling and Injecting Code
17845 @section Compiling and injecting code in @value{GDBN}
17846 @cindex injecting code
17847 @cindex writing into executables
17848 @cindex compiling code
17850 @value{GDBN} supports on-demand compilation and code injection into
17851 programs running under @value{GDBN}. GCC 5.0 or higher built with
17852 @file{libcc1.so} must be installed for this functionality to be enabled.
17853 This functionality is implemented with the following commands.
17856 @kindex compile code
17857 @item compile code @var{source-code}
17858 @itemx compile code -raw @var{--} @var{source-code}
17859 Compile @var{source-code} with the compiler language found as the current
17860 language in @value{GDBN} (@pxref{Languages}). If compilation and
17861 injection is not supported with the current language specified in
17862 @value{GDBN}, or the compiler does not support this feature, an error
17863 message will be printed. If @var{source-code} compiles and links
17864 successfully, @value{GDBN} will load the object-code emitted,
17865 and execute it within the context of the currently selected inferior.
17866 It is important to note that the compiled code is executed immediately.
17867 After execution, the compiled code is removed from @value{GDBN} and any
17868 new types or variables you have defined will be deleted.
17870 The command allows you to specify @var{source-code} in two ways.
17871 The simplest method is to provide a single line of code to the command.
17875 compile code printf ("hello world\n");
17878 If you specify options on the command line as well as source code, they
17879 may conflict. The @samp{--} delimiter can be used to separate options
17880 from actual source code. E.g.:
17883 compile code -r -- printf ("hello world\n");
17886 Alternatively you can enter source code as multiple lines of text. To
17887 enter this mode, invoke the @samp{compile code} command without any text
17888 following the command. This will start the multiple-line editor and
17889 allow you to type as many lines of source code as required. When you
17890 have completed typing, enter @samp{end} on its own line to exit the
17895 >printf ("hello\n");
17896 >printf ("world\n");
17900 Specifying @samp{-raw}, prohibits @value{GDBN} from wrapping the
17901 provided @var{source-code} in a callable scope. In this case, you must
17902 specify the entry point of the code by defining a function named
17903 @code{_gdb_expr_}. The @samp{-raw} code cannot access variables of the
17904 inferior. Using @samp{-raw} option may be needed for example when
17905 @var{source-code} requires @samp{#include} lines which may conflict with
17906 inferior symbols otherwise.
17908 @kindex compile file
17909 @item compile file @var{filename}
17910 @itemx compile file -raw @var{filename}
17911 Like @code{compile code}, but take the source code from @var{filename}.
17914 compile file /home/user/example.c
17919 @item compile print @var{expr}
17920 @itemx compile print /@var{f} @var{expr}
17921 Compile and execute @var{expr} with the compiler language found as the
17922 current language in @value{GDBN} (@pxref{Languages}). By default the
17923 value of @var{expr} is printed in a format appropriate to its data type;
17924 you can choose a different format by specifying @samp{/@var{f}}, where
17925 @var{f} is a letter specifying the format; see @ref{Output Formats,,Output
17928 @item compile print
17929 @itemx compile print /@var{f}
17930 @cindex reprint the last value
17931 Alternatively you can enter the expression (source code producing it) as
17932 multiple lines of text. To enter this mode, invoke the @samp{compile print}
17933 command without any text following the command. This will start the
17934 multiple-line editor.
17938 The process of compiling and injecting the code can be inspected using:
17941 @anchor{set debug compile}
17942 @item set debug compile
17943 @cindex compile command debugging info
17944 Turns on or off display of @value{GDBN} process of compiling and
17945 injecting the code. The default is off.
17947 @item show debug compile
17948 Displays the current state of displaying @value{GDBN} process of
17949 compiling and injecting the code.
17952 @subsection Compilation options for the @code{compile} command
17954 @value{GDBN} needs to specify the right compilation options for the code
17955 to be injected, in part to make its ABI compatible with the inferior
17956 and in part to make the injected code compatible with @value{GDBN}'s
17960 The options used, in increasing precedence:
17963 @item target architecture and OS options (@code{gdbarch})
17964 These options depend on target processor type and target operating
17965 system, usually they specify at least 32-bit (@code{-m32}) or 64-bit
17966 (@code{-m64}) compilation option.
17968 @item compilation options recorded in the target
17969 @value{NGCC} (since version 4.7) stores the options used for compilation
17970 into @code{DW_AT_producer} part of DWARF debugging information according
17971 to the @value{NGCC} option @code{-grecord-gcc-switches}. One has to
17972 explicitly specify @code{-g} during inferior compilation otherwise
17973 @value{NGCC} produces no DWARF. This feature is only relevant for
17974 platforms where @code{-g} produces DWARF by default, otherwise one may
17975 try to enforce DWARF by using @code{-gdwarf-4}.
17977 @item compilation options set by @code{set compile-args}
17981 You can override compilation options using the following command:
17984 @item set compile-args
17985 @cindex compile command options override
17986 Set compilation options used for compiling and injecting code with the
17987 @code{compile} commands. These options override any conflicting ones
17988 from the target architecture and/or options stored during inferior
17991 @item show compile-args
17992 Displays the current state of compilation options override.
17993 This does not show all the options actually used during compilation,
17994 use @ref{set debug compile} for that.
17997 @subsection Caveats when using the @code{compile} command
17999 There are a few caveats to keep in mind when using the @code{compile}
18000 command. As the caveats are different per language, the table below
18001 highlights specific issues on a per language basis.
18004 @item C code examples and caveats
18005 When the language in @value{GDBN} is set to @samp{C}, the compiler will
18006 attempt to compile the source code with a @samp{C} compiler. The source
18007 code provided to the @code{compile} command will have much the same
18008 access to variables and types as it normally would if it were part of
18009 the program currently being debugged in @value{GDBN}.
18011 Below is a sample program that forms the basis of the examples that
18012 follow. This program has been compiled and loaded into @value{GDBN},
18013 much like any other normal debugging session.
18016 void function1 (void)
18019 printf ("function 1\n");
18022 void function2 (void)
18037 For the purposes of the examples in this section, the program above has
18038 been compiled, loaded into @value{GDBN}, stopped at the function
18039 @code{main}, and @value{GDBN} is awaiting input from the user.
18041 To access variables and types for any program in @value{GDBN}, the
18042 program must be compiled and packaged with debug information. The
18043 @code{compile} command is not an exception to this rule. Without debug
18044 information, you can still use the @code{compile} command, but you will
18045 be very limited in what variables and types you can access.
18047 So with that in mind, the example above has been compiled with debug
18048 information enabled. The @code{compile} command will have access to
18049 all variables and types (except those that may have been optimized
18050 out). Currently, as @value{GDBN} has stopped the program in the
18051 @code{main} function, the @code{compile} command would have access to
18052 the variable @code{k}. You could invoke the @code{compile} command
18053 and type some source code to set the value of @code{k}. You can also
18054 read it, or do anything with that variable you would normally do in
18055 @code{C}. Be aware that changes to inferior variables in the
18056 @code{compile} command are persistent. In the following example:
18059 compile code k = 3;
18063 the variable @code{k} is now 3. It will retain that value until
18064 something else in the example program changes it, or another
18065 @code{compile} command changes it.
18067 Normal scope and access rules apply to source code compiled and
18068 injected by the @code{compile} command. In the example, the variables
18069 @code{j} and @code{k} are not accessible yet, because the program is
18070 currently stopped in the @code{main} function, where these variables
18071 are not in scope. Therefore, the following command
18074 compile code j = 3;
18078 will result in a compilation error message.
18080 Once the program is continued, execution will bring these variables in
18081 scope, and they will become accessible; then the code you specify via
18082 the @code{compile} command will be able to access them.
18084 You can create variables and types with the @code{compile} command as
18085 part of your source code. Variables and types that are created as part
18086 of the @code{compile} command are not visible to the rest of the program for
18087 the duration of its run. This example is valid:
18090 compile code int ff = 5; printf ("ff is %d\n", ff);
18093 However, if you were to type the following into @value{GDBN} after that
18094 command has completed:
18097 compile code printf ("ff is %d\n'', ff);
18101 a compiler error would be raised as the variable @code{ff} no longer
18102 exists. Object code generated and injected by the @code{compile}
18103 command is removed when its execution ends. Caution is advised
18104 when assigning to program variables values of variables created by the
18105 code submitted to the @code{compile} command. This example is valid:
18108 compile code int ff = 5; k = ff;
18111 The value of the variable @code{ff} is assigned to @code{k}. The variable
18112 @code{k} does not require the existence of @code{ff} to maintain the value
18113 it has been assigned. However, pointers require particular care in
18114 assignment. If the source code compiled with the @code{compile} command
18115 changed the address of a pointer in the example program, perhaps to a
18116 variable created in the @code{compile} command, that pointer would point
18117 to an invalid location when the command exits. The following example
18118 would likely cause issues with your debugged program:
18121 compile code int ff = 5; p = &ff;
18124 In this example, @code{p} would point to @code{ff} when the
18125 @code{compile} command is executing the source code provided to it.
18126 However, as variables in the (example) program persist with their
18127 assigned values, the variable @code{p} would point to an invalid
18128 location when the command exists. A general rule should be followed
18129 in that you should either assign @code{NULL} to any assigned pointers,
18130 or restore a valid location to the pointer before the command exits.
18132 Similar caution must be exercised with any structs, unions, and typedefs
18133 defined in @code{compile} command. Types defined in the @code{compile}
18134 command will no longer be available in the next @code{compile} command.
18135 Therefore, if you cast a variable to a type defined in the
18136 @code{compile} command, care must be taken to ensure that any future
18137 need to resolve the type can be achieved.
18140 (gdb) compile code static struct a @{ int a; @} v = @{ 42 @}; argv = &v;
18141 (gdb) compile code printf ("%d\n", ((struct a *) argv)->a);
18142 gdb command line:1:36: error: dereferencing pointer to incomplete type ‘struct a’
18143 Compilation failed.
18144 (gdb) compile code struct a @{ int a; @}; printf ("%d\n", ((struct a *) argv)->a);
18148 Variables that have been optimized away by the compiler are not
18149 accessible to the code submitted to the @code{compile} command.
18150 Access to those variables will generate a compiler error which @value{GDBN}
18151 will print to the console.
18154 @subsection Compiler search for the @code{compile} command
18156 @value{GDBN} needs to find @value{NGCC} for the inferior being debugged which
18157 may not be obvious for remote targets of different architecture than where
18158 @value{GDBN} is running. Environment variable @code{PATH} (@code{PATH} from
18159 shell that executed @value{GDBN}, not the one set by @value{GDBN}
18160 command @code{set environment}). @xref{Environment}. @code{PATH} on
18161 @value{GDBN} host is searched for @value{NGCC} binary matching the
18162 target architecture and operating system.
18164 Specifically @code{PATH} is searched for binaries matching regular expression
18165 @code{@var{arch}(-[^-]*)?-@var{os}-gcc} according to the inferior target being
18166 debugged. @var{arch} is processor name --- multiarch is supported, so for
18167 example both @code{i386} and @code{x86_64} targets look for pattern
18168 @code{(x86_64|i.86)} and both @code{s390} and @code{s390x} targets look
18169 for pattern @code{s390x?}. @var{os} is currently supported only for
18170 pattern @code{linux(-gnu)?}.
18173 @chapter @value{GDBN} Files
18175 @value{GDBN} needs to know the file name of the program to be debugged,
18176 both in order to read its symbol table and in order to start your
18177 program. To debug a core dump of a previous run, you must also tell
18178 @value{GDBN} the name of the core dump file.
18181 * Files:: Commands to specify files
18182 * File Caching:: Information about @value{GDBN}'s file caching
18183 * Separate Debug Files:: Debugging information in separate files
18184 * MiniDebugInfo:: Debugging information in a special section
18185 * Index Files:: Index files speed up GDB
18186 * Symbol Errors:: Errors reading symbol files
18187 * Data Files:: GDB data files
18191 @section Commands to Specify Files
18193 @cindex symbol table
18194 @cindex core dump file
18196 You may want to specify executable and core dump file names. The usual
18197 way to do this is at start-up time, using the arguments to
18198 @value{GDBN}'s start-up commands (@pxref{Invocation, , Getting In and
18199 Out of @value{GDBN}}).
18201 Occasionally it is necessary to change to a different file during a
18202 @value{GDBN} session. Or you may run @value{GDBN} and forget to
18203 specify a file you want to use. Or you are debugging a remote target
18204 via @code{gdbserver} (@pxref{Server, file, Using the @code{gdbserver}
18205 Program}). In these situations the @value{GDBN} commands to specify
18206 new files are useful.
18209 @cindex executable file
18211 @item file @var{filename}
18212 Use @var{filename} as the program to be debugged. It is read for its
18213 symbols and for the contents of pure memory. It is also the program
18214 executed when you use the @code{run} command. If you do not specify a
18215 directory and the file is not found in the @value{GDBN} working directory,
18216 @value{GDBN} uses the environment variable @code{PATH} as a list of
18217 directories to search, just as the shell does when looking for a program
18218 to run. You can change the value of this variable, for both @value{GDBN}
18219 and your program, using the @code{path} command.
18221 @cindex unlinked object files
18222 @cindex patching object files
18223 You can load unlinked object @file{.o} files into @value{GDBN} using
18224 the @code{file} command. You will not be able to ``run'' an object
18225 file, but you can disassemble functions and inspect variables. Also,
18226 if the underlying BFD functionality supports it, you could use
18227 @kbd{gdb -write} to patch object files using this technique. Note
18228 that @value{GDBN} can neither interpret nor modify relocations in this
18229 case, so branches and some initialized variables will appear to go to
18230 the wrong place. But this feature is still handy from time to time.
18233 @code{file} with no argument makes @value{GDBN} discard any information it
18234 has on both executable file and the symbol table.
18237 @item exec-file @r{[} @var{filename} @r{]}
18238 Specify that the program to be run (but not the symbol table) is found
18239 in @var{filename}. @value{GDBN} searches the environment variable @code{PATH}
18240 if necessary to locate your program. Omitting @var{filename} means to
18241 discard information on the executable file.
18243 @kindex symbol-file
18244 @item symbol-file @r{[} @var{filename} @r{]}
18245 Read symbol table information from file @var{filename}. @code{PATH} is
18246 searched when necessary. Use the @code{file} command to get both symbol
18247 table and program to run from the same file.
18249 @code{symbol-file} with no argument clears out @value{GDBN} information on your
18250 program's symbol table.
18252 The @code{symbol-file} command causes @value{GDBN} to forget the contents of
18253 some breakpoints and auto-display expressions. This is because they may
18254 contain pointers to the internal data recording symbols and data types,
18255 which are part of the old symbol table data being discarded inside
18258 @code{symbol-file} does not repeat if you press @key{RET} again after
18261 When @value{GDBN} is configured for a particular environment, it
18262 understands debugging information in whatever format is the standard
18263 generated for that environment; you may use either a @sc{gnu} compiler, or
18264 other compilers that adhere to the local conventions.
18265 Best results are usually obtained from @sc{gnu} compilers; for example,
18266 using @code{@value{NGCC}} you can generate debugging information for
18269 For most kinds of object files, with the exception of old SVR3 systems
18270 using COFF, the @code{symbol-file} command does not normally read the
18271 symbol table in full right away. Instead, it scans the symbol table
18272 quickly to find which source files and which symbols are present. The
18273 details are read later, one source file at a time, as they are needed.
18275 The purpose of this two-stage reading strategy is to make @value{GDBN}
18276 start up faster. For the most part, it is invisible except for
18277 occasional pauses while the symbol table details for a particular source
18278 file are being read. (The @code{set verbose} command can turn these
18279 pauses into messages if desired. @xref{Messages/Warnings, ,Optional
18280 Warnings and Messages}.)
18282 We have not implemented the two-stage strategy for COFF yet. When the
18283 symbol table is stored in COFF format, @code{symbol-file} reads the
18284 symbol table data in full right away. Note that ``stabs-in-COFF''
18285 still does the two-stage strategy, since the debug info is actually
18289 @cindex reading symbols immediately
18290 @cindex symbols, reading immediately
18291 @item symbol-file @r{[} -readnow @r{]} @var{filename}
18292 @itemx file @r{[} -readnow @r{]} @var{filename}
18293 You can override the @value{GDBN} two-stage strategy for reading symbol
18294 tables by using the @samp{-readnow} option with any of the commands that
18295 load symbol table information, if you want to be sure @value{GDBN} has the
18296 entire symbol table available.
18298 @c FIXME: for now no mention of directories, since this seems to be in
18299 @c flux. 13mar1992 status is that in theory GDB would look either in
18300 @c current dir or in same dir as myprog; but issues like competing
18301 @c GDB's, or clutter in system dirs, mean that in practice right now
18302 @c only current dir is used. FFish says maybe a special GDB hierarchy
18303 @c (eg rooted in val of env var GDBSYMS) could exist for mappable symbol
18307 @item core-file @r{[}@var{filename}@r{]}
18309 Specify the whereabouts of a core dump file to be used as the ``contents
18310 of memory''. Traditionally, core files contain only some parts of the
18311 address space of the process that generated them; @value{GDBN} can access the
18312 executable file itself for other parts.
18314 @code{core-file} with no argument specifies that no core file is
18317 Note that the core file is ignored when your program is actually running
18318 under @value{GDBN}. So, if you have been running your program and you
18319 wish to debug a core file instead, you must kill the subprocess in which
18320 the program is running. To do this, use the @code{kill} command
18321 (@pxref{Kill Process, ,Killing the Child Process}).
18323 @kindex add-symbol-file
18324 @cindex dynamic linking
18325 @item add-symbol-file @var{filename} @var{address}
18326 @itemx add-symbol-file @var{filename} @var{address} @r{[} -readnow @r{]}
18327 @itemx add-symbol-file @var{filename} @var{address} -s @var{section} @var{address} @dots{}
18328 The @code{add-symbol-file} command reads additional symbol table
18329 information from the file @var{filename}. You would use this command
18330 when @var{filename} has been dynamically loaded (by some other means)
18331 into the program that is running. The @var{address} should give the memory
18332 address at which the file has been loaded; @value{GDBN} cannot figure
18333 this out for itself. You can additionally specify an arbitrary number
18334 of @samp{-s @var{section} @var{address}} pairs, to give an explicit
18335 section name and base address for that section. You can specify any
18336 @var{address} as an expression.
18338 The symbol table of the file @var{filename} is added to the symbol table
18339 originally read with the @code{symbol-file} command. You can use the
18340 @code{add-symbol-file} command any number of times; the new symbol data
18341 thus read is kept in addition to the old.
18343 Changes can be reverted using the command @code{remove-symbol-file}.
18345 @cindex relocatable object files, reading symbols from
18346 @cindex object files, relocatable, reading symbols from
18347 @cindex reading symbols from relocatable object files
18348 @cindex symbols, reading from relocatable object files
18349 @cindex @file{.o} files, reading symbols from
18350 Although @var{filename} is typically a shared library file, an
18351 executable file, or some other object file which has been fully
18352 relocated for loading into a process, you can also load symbolic
18353 information from relocatable @file{.o} files, as long as:
18357 the file's symbolic information refers only to linker symbols defined in
18358 that file, not to symbols defined by other object files,
18360 every section the file's symbolic information refers to has actually
18361 been loaded into the inferior, as it appears in the file, and
18363 you can determine the address at which every section was loaded, and
18364 provide these to the @code{add-symbol-file} command.
18368 Some embedded operating systems, like Sun Chorus and VxWorks, can load
18369 relocatable files into an already running program; such systems
18370 typically make the requirements above easy to meet. However, it's
18371 important to recognize that many native systems use complex link
18372 procedures (@code{.linkonce} section factoring and C@t{++} constructor table
18373 assembly, for example) that make the requirements difficult to meet. In
18374 general, one cannot assume that using @code{add-symbol-file} to read a
18375 relocatable object file's symbolic information will have the same effect
18376 as linking the relocatable object file into the program in the normal
18379 @code{add-symbol-file} does not repeat if you press @key{RET} after using it.
18381 @kindex remove-symbol-file
18382 @item remove-symbol-file @var{filename}
18383 @item remove-symbol-file -a @var{address}
18384 Remove a symbol file added via the @code{add-symbol-file} command. The
18385 file to remove can be identified by its @var{filename} or by an @var{address}
18386 that lies within the boundaries of this symbol file in memory. Example:
18389 (gdb) add-symbol-file /home/user/gdb/mylib.so 0x7ffff7ff9480
18390 add symbol table from file "/home/user/gdb/mylib.so" at
18391 .text_addr = 0x7ffff7ff9480
18393 Reading symbols from /home/user/gdb/mylib.so...done.
18394 (gdb) remove-symbol-file -a 0x7ffff7ff9480
18395 Remove symbol table from file "/home/user/gdb/mylib.so"? (y or n) y
18400 @code{remove-symbol-file} does not repeat if you press @key{RET} after using it.
18402 @kindex add-symbol-file-from-memory
18403 @cindex @code{syscall DSO}
18404 @cindex load symbols from memory
18405 @item add-symbol-file-from-memory @var{address}
18406 Load symbols from the given @var{address} in a dynamically loaded
18407 object file whose image is mapped directly into the inferior's memory.
18408 For example, the Linux kernel maps a @code{syscall DSO} into each
18409 process's address space; this DSO provides kernel-specific code for
18410 some system calls. The argument can be any expression whose
18411 evaluation yields the address of the file's shared object file header.
18412 For this command to work, you must have used @code{symbol-file} or
18413 @code{exec-file} commands in advance.
18416 @item section @var{section} @var{addr}
18417 The @code{section} command changes the base address of the named
18418 @var{section} of the exec file to @var{addr}. This can be used if the
18419 exec file does not contain section addresses, (such as in the
18420 @code{a.out} format), or when the addresses specified in the file
18421 itself are wrong. Each section must be changed separately. The
18422 @code{info files} command, described below, lists all the sections and
18426 @kindex info target
18429 @code{info files} and @code{info target} are synonymous; both print the
18430 current target (@pxref{Targets, ,Specifying a Debugging Target}),
18431 including the names of the executable and core dump files currently in
18432 use by @value{GDBN}, and the files from which symbols were loaded. The
18433 command @code{help target} lists all possible targets rather than
18436 @kindex maint info sections
18437 @item maint info sections
18438 Another command that can give you extra information about program sections
18439 is @code{maint info sections}. In addition to the section information
18440 displayed by @code{info files}, this command displays the flags and file
18441 offset of each section in the executable and core dump files. In addition,
18442 @code{maint info sections} provides the following command options (which
18443 may be arbitrarily combined):
18447 Display sections for all loaded object files, including shared libraries.
18448 @item @var{sections}
18449 Display info only for named @var{sections}.
18450 @item @var{section-flags}
18451 Display info only for sections for which @var{section-flags} are true.
18452 The section flags that @value{GDBN} currently knows about are:
18455 Section will have space allocated in the process when loaded.
18456 Set for all sections except those containing debug information.
18458 Section will be loaded from the file into the child process memory.
18459 Set for pre-initialized code and data, clear for @code{.bss} sections.
18461 Section needs to be relocated before loading.
18463 Section cannot be modified by the child process.
18465 Section contains executable code only.
18467 Section contains data only (no executable code).
18469 Section will reside in ROM.
18471 Section contains data for constructor/destructor lists.
18473 Section is not empty.
18475 An instruction to the linker to not output the section.
18476 @item COFF_SHARED_LIBRARY
18477 A notification to the linker that the section contains
18478 COFF shared library information.
18480 Section contains common symbols.
18483 @kindex set trust-readonly-sections
18484 @cindex read-only sections
18485 @item set trust-readonly-sections on
18486 Tell @value{GDBN} that readonly sections in your object file
18487 really are read-only (i.e.@: that their contents will not change).
18488 In that case, @value{GDBN} can fetch values from these sections
18489 out of the object file, rather than from the target program.
18490 For some targets (notably embedded ones), this can be a significant
18491 enhancement to debugging performance.
18493 The default is off.
18495 @item set trust-readonly-sections off
18496 Tell @value{GDBN} not to trust readonly sections. This means that
18497 the contents of the section might change while the program is running,
18498 and must therefore be fetched from the target when needed.
18500 @item show trust-readonly-sections
18501 Show the current setting of trusting readonly sections.
18504 All file-specifying commands allow both absolute and relative file names
18505 as arguments. @value{GDBN} always converts the file name to an absolute file
18506 name and remembers it that way.
18508 @cindex shared libraries
18509 @anchor{Shared Libraries}
18510 @value{GDBN} supports @sc{gnu}/Linux, MS-Windows, SunOS,
18511 Darwin/Mach-O, SVr4, IBM RS/6000 AIX, QNX Neutrino, FDPIC (FR-V), and
18512 DSBT (TIC6X) shared libraries.
18514 On MS-Windows @value{GDBN} must be linked with the Expat library to support
18515 shared libraries. @xref{Expat}.
18517 @value{GDBN} automatically loads symbol definitions from shared libraries
18518 when you use the @code{run} command, or when you examine a core file.
18519 (Before you issue the @code{run} command, @value{GDBN} does not understand
18520 references to a function in a shared library, however---unless you are
18521 debugging a core file).
18523 @c FIXME: some @value{GDBN} release may permit some refs to undef
18524 @c FIXME...symbols---eg in a break cmd---assuming they are from a shared
18525 @c FIXME...lib; check this from time to time when updating manual
18527 There are times, however, when you may wish to not automatically load
18528 symbol definitions from shared libraries, such as when they are
18529 particularly large or there are many of them.
18531 To control the automatic loading of shared library symbols, use the
18535 @kindex set auto-solib-add
18536 @item set auto-solib-add @var{mode}
18537 If @var{mode} is @code{on}, symbols from all shared object libraries
18538 will be loaded automatically when the inferior begins execution, you
18539 attach to an independently started inferior, or when the dynamic linker
18540 informs @value{GDBN} that a new library has been loaded. If @var{mode}
18541 is @code{off}, symbols must be loaded manually, using the
18542 @code{sharedlibrary} command. The default value is @code{on}.
18544 @cindex memory used for symbol tables
18545 If your program uses lots of shared libraries with debug info that
18546 takes large amounts of memory, you can decrease the @value{GDBN}
18547 memory footprint by preventing it from automatically loading the
18548 symbols from shared libraries. To that end, type @kbd{set
18549 auto-solib-add off} before running the inferior, then load each
18550 library whose debug symbols you do need with @kbd{sharedlibrary
18551 @var{regexp}}, where @var{regexp} is a regular expression that matches
18552 the libraries whose symbols you want to be loaded.
18554 @kindex show auto-solib-add
18555 @item show auto-solib-add
18556 Display the current autoloading mode.
18559 @cindex load shared library
18560 To explicitly load shared library symbols, use the @code{sharedlibrary}
18564 @kindex info sharedlibrary
18566 @item info share @var{regex}
18567 @itemx info sharedlibrary @var{regex}
18568 Print the names of the shared libraries which are currently loaded
18569 that match @var{regex}. If @var{regex} is omitted then print
18570 all shared libraries that are loaded.
18573 @item info dll @var{regex}
18574 This is an alias of @code{info sharedlibrary}.
18576 @kindex sharedlibrary
18578 @item sharedlibrary @var{regex}
18579 @itemx share @var{regex}
18580 Load shared object library symbols for files matching a
18581 Unix regular expression.
18582 As with files loaded automatically, it only loads shared libraries
18583 required by your program for a core file or after typing @code{run}. If
18584 @var{regex} is omitted all shared libraries required by your program are
18587 @item nosharedlibrary
18588 @kindex nosharedlibrary
18589 @cindex unload symbols from shared libraries
18590 Unload all shared object library symbols. This discards all symbols
18591 that have been loaded from all shared libraries. Symbols from shared
18592 libraries that were loaded by explicit user requests are not
18596 Sometimes you may wish that @value{GDBN} stops and gives you control
18597 when any of shared library events happen. The best way to do this is
18598 to use @code{catch load} and @code{catch unload} (@pxref{Set
18601 @value{GDBN} also supports the the @code{set stop-on-solib-events}
18602 command for this. This command exists for historical reasons. It is
18603 less useful than setting a catchpoint, because it does not allow for
18604 conditions or commands as a catchpoint does.
18607 @item set stop-on-solib-events
18608 @kindex set stop-on-solib-events
18609 This command controls whether @value{GDBN} should give you control
18610 when the dynamic linker notifies it about some shared library event.
18611 The most common event of interest is loading or unloading of a new
18614 @item show stop-on-solib-events
18615 @kindex show stop-on-solib-events
18616 Show whether @value{GDBN} stops and gives you control when shared
18617 library events happen.
18620 Shared libraries are also supported in many cross or remote debugging
18621 configurations. @value{GDBN} needs to have access to the target's libraries;
18622 this can be accomplished either by providing copies of the libraries
18623 on the host system, or by asking @value{GDBN} to automatically retrieve the
18624 libraries from the target. If copies of the target libraries are
18625 provided, they need to be the same as the target libraries, although the
18626 copies on the target can be stripped as long as the copies on the host are
18629 @cindex where to look for shared libraries
18630 For remote debugging, you need to tell @value{GDBN} where the target
18631 libraries are, so that it can load the correct copies---otherwise, it
18632 may try to load the host's libraries. @value{GDBN} has two variables
18633 to specify the search directories for target libraries.
18636 @cindex prefix for executable and shared library file names
18637 @cindex system root, alternate
18638 @kindex set solib-absolute-prefix
18639 @kindex set sysroot
18640 @item set sysroot @var{path}
18641 Use @var{path} as the system root for the program being debugged. Any
18642 absolute shared library paths will be prefixed with @var{path}; many
18643 runtime loaders store the absolute paths to the shared library in the
18644 target program's memory. When starting processes remotely, and when
18645 attaching to already-running processes (local or remote), their
18646 executable filenames will be prefixed with @var{path} if reported to
18647 @value{GDBN} as absolute by the operating system. If you use
18648 @code{set sysroot} to find executables and shared libraries, they need
18649 to be laid out in the same way that they are on the target, with
18650 e.g.@: a @file{/bin}, @file{/lib} and @file{/usr/lib} hierarchy under
18653 If @var{path} starts with the sequence @file{target:} and the target
18654 system is remote then @value{GDBN} will retrieve the target binaries
18655 from the remote system. This is only supported when using a remote
18656 target that supports the @code{remote get} command (@pxref{File
18657 Transfer,,Sending files to a remote system}). The part of @var{path}
18658 following the initial @file{target:} (if present) is used as system
18659 root prefix on the remote file system. If @var{path} starts with the
18660 sequence @file{remote:} this is converted to the sequence
18661 @file{target:} by @code{set sysroot}@footnote{Historically the
18662 functionality to retrieve binaries from the remote system was
18663 provided by prefixing @var{path} with @file{remote:}}. If you want
18664 to specify a local system root using a directory that happens to be
18665 named @file{target:} or @file{remote:}, you need to use some
18666 equivalent variant of the name like @file{./target:}.
18668 For targets with an MS-DOS based filesystem, such as MS-Windows and
18669 SymbianOS, @value{GDBN} tries prefixing a few variants of the target
18670 absolute file name with @var{path}. But first, on Unix hosts,
18671 @value{GDBN} converts all backslash directory separators into forward
18672 slashes, because the backslash is not a directory separator on Unix:
18675 c:\foo\bar.dll @result{} c:/foo/bar.dll
18678 Then, @value{GDBN} attempts prefixing the target file name with
18679 @var{path}, and looks for the resulting file name in the host file
18683 c:/foo/bar.dll @result{} /path/to/sysroot/c:/foo/bar.dll
18686 If that does not find the binary, @value{GDBN} tries removing
18687 the @samp{:} character from the drive spec, both for convenience, and,
18688 for the case of the host file system not supporting file names with
18692 c:/foo/bar.dll @result{} /path/to/sysroot/c/foo/bar.dll
18695 This makes it possible to have a system root that mirrors a target
18696 with more than one drive. E.g., you may want to setup your local
18697 copies of the target system shared libraries like so (note @samp{c} vs
18701 @file{/path/to/sysroot/c/sys/bin/foo.dll}
18702 @file{/path/to/sysroot/c/sys/bin/bar.dll}
18703 @file{/path/to/sysroot/z/sys/bin/bar.dll}
18707 and point the system root at @file{/path/to/sysroot}, so that
18708 @value{GDBN} can find the correct copies of both
18709 @file{c:\sys\bin\foo.dll}, and @file{z:\sys\bin\bar.dll}.
18711 If that still does not find the binary, @value{GDBN} tries
18712 removing the whole drive spec from the target file name:
18715 c:/foo/bar.dll @result{} /path/to/sysroot/foo/bar.dll
18718 This last lookup makes it possible to not care about the drive name,
18719 if you don't want or need to.
18721 The @code{set solib-absolute-prefix} command is an alias for @code{set
18724 @cindex default system root
18725 @cindex @samp{--with-sysroot}
18726 You can set the default system root by using the configure-time
18727 @samp{--with-sysroot} option. If the system root is inside
18728 @value{GDBN}'s configured binary prefix (set with @samp{--prefix} or
18729 @samp{--exec-prefix}), then the default system root will be updated
18730 automatically if the installed @value{GDBN} is moved to a new
18733 @kindex show sysroot
18735 Display the current executable and shared library prefix.
18737 @kindex set solib-search-path
18738 @item set solib-search-path @var{path}
18739 If this variable is set, @var{path} is a colon-separated list of
18740 directories to search for shared libraries. @samp{solib-search-path}
18741 is used after @samp{sysroot} fails to locate the library, or if the
18742 path to the library is relative instead of absolute. If you want to
18743 use @samp{solib-search-path} instead of @samp{sysroot}, be sure to set
18744 @samp{sysroot} to a nonexistent directory to prevent @value{GDBN} from
18745 finding your host's libraries. @samp{sysroot} is preferred; setting
18746 it to a nonexistent directory may interfere with automatic loading
18747 of shared library symbols.
18749 @kindex show solib-search-path
18750 @item show solib-search-path
18751 Display the current shared library search path.
18753 @cindex DOS file-name semantics of file names.
18754 @kindex set target-file-system-kind (unix|dos-based|auto)
18755 @kindex show target-file-system-kind
18756 @item set target-file-system-kind @var{kind}
18757 Set assumed file system kind for target reported file names.
18759 Shared library file names as reported by the target system may not
18760 make sense as is on the system @value{GDBN} is running on. For
18761 example, when remote debugging a target that has MS-DOS based file
18762 system semantics, from a Unix host, the target may be reporting to
18763 @value{GDBN} a list of loaded shared libraries with file names such as
18764 @file{c:\Windows\kernel32.dll}. On Unix hosts, there's no concept of
18765 drive letters, so the @samp{c:\} prefix is not normally understood as
18766 indicating an absolute file name, and neither is the backslash
18767 normally considered a directory separator character. In that case,
18768 the native file system would interpret this whole absolute file name
18769 as a relative file name with no directory components. This would make
18770 it impossible to point @value{GDBN} at a copy of the remote target's
18771 shared libraries on the host using @code{set sysroot}, and impractical
18772 with @code{set solib-search-path}. Setting
18773 @code{target-file-system-kind} to @code{dos-based} tells @value{GDBN}
18774 to interpret such file names similarly to how the target would, and to
18775 map them to file names valid on @value{GDBN}'s native file system
18776 semantics. The value of @var{kind} can be @code{"auto"}, in addition
18777 to one of the supported file system kinds. In that case, @value{GDBN}
18778 tries to determine the appropriate file system variant based on the
18779 current target's operating system (@pxref{ABI, ,Configuring the
18780 Current ABI}). The supported file system settings are:
18784 Instruct @value{GDBN} to assume the target file system is of Unix
18785 kind. Only file names starting the forward slash (@samp{/}) character
18786 are considered absolute, and the directory separator character is also
18790 Instruct @value{GDBN} to assume the target file system is DOS based.
18791 File names starting with either a forward slash, or a drive letter
18792 followed by a colon (e.g., @samp{c:}), are considered absolute, and
18793 both the slash (@samp{/}) and the backslash (@samp{\\}) characters are
18794 considered directory separators.
18797 Instruct @value{GDBN} to use the file system kind associated with the
18798 target operating system (@pxref{ABI, ,Configuring the Current ABI}).
18799 This is the default.
18803 @cindex file name canonicalization
18804 @cindex base name differences
18805 When processing file names provided by the user, @value{GDBN}
18806 frequently needs to compare them to the file names recorded in the
18807 program's debug info. Normally, @value{GDBN} compares just the
18808 @dfn{base names} of the files as strings, which is reasonably fast
18809 even for very large programs. (The base name of a file is the last
18810 portion of its name, after stripping all the leading directories.)
18811 This shortcut in comparison is based upon the assumption that files
18812 cannot have more than one base name. This is usually true, but
18813 references to files that use symlinks or similar filesystem
18814 facilities violate that assumption. If your program records files
18815 using such facilities, or if you provide file names to @value{GDBN}
18816 using symlinks etc., you can set @code{basenames-may-differ} to
18817 @code{true} to instruct @value{GDBN} to completely canonicalize each
18818 pair of file names it needs to compare. This will make file-name
18819 comparisons accurate, but at a price of a significant slowdown.
18822 @item set basenames-may-differ
18823 @kindex set basenames-may-differ
18824 Set whether a source file may have multiple base names.
18826 @item show basenames-may-differ
18827 @kindex show basenames-may-differ
18828 Show whether a source file may have multiple base names.
18832 @section File Caching
18833 @cindex caching of opened files
18834 @cindex caching of bfd objects
18836 To speed up file loading, and reduce memory usage, @value{GDBN} will
18837 reuse the @code{bfd} objects used to track open files. @xref{Top, ,
18838 BFD, bfd, The Binary File Descriptor Library}. The following commands
18839 allow visibility and control of the caching behavior.
18842 @kindex maint info bfds
18843 @item maint info bfds
18844 This prints information about each @code{bfd} object that is known to
18847 @kindex maint set bfd-sharing
18848 @kindex maint show bfd-sharing
18849 @kindex bfd caching
18850 @item maint set bfd-sharing
18851 @item maint show bfd-sharing
18852 Control whether @code{bfd} objects can be shared. When sharing is
18853 enabled @value{GDBN} reuses already open @code{bfd} objects rather
18854 than reopening the same file. Turning sharing off does not cause
18855 already shared @code{bfd} objects to be unshared, but all future files
18856 that are opened will create a new @code{bfd} object. Similarly,
18857 re-enabling sharing does not cause multiple existing @code{bfd}
18858 objects to be collapsed into a single shared @code{bfd} object.
18860 @kindex set debug bfd-cache @var{level}
18861 @kindex bfd caching
18862 @item set debug bfd-cache @var{level}
18863 Turns on debugging of the bfd cache, setting the level to @var{level}.
18865 @kindex show debug bfd-cache
18866 @kindex bfd caching
18867 @item show debug bfd-cache
18868 Show the current debugging level of the bfd cache.
18871 @node Separate Debug Files
18872 @section Debugging Information in Separate Files
18873 @cindex separate debugging information files
18874 @cindex debugging information in separate files
18875 @cindex @file{.debug} subdirectories
18876 @cindex debugging information directory, global
18877 @cindex global debugging information directories
18878 @cindex build ID, and separate debugging files
18879 @cindex @file{.build-id} directory
18881 @value{GDBN} allows you to put a program's debugging information in a
18882 file separate from the executable itself, in a way that allows
18883 @value{GDBN} to find and load the debugging information automatically.
18884 Since debugging information can be very large---sometimes larger
18885 than the executable code itself---some systems distribute debugging
18886 information for their executables in separate files, which users can
18887 install only when they need to debug a problem.
18889 @value{GDBN} supports two ways of specifying the separate debug info
18894 The executable contains a @dfn{debug link} that specifies the name of
18895 the separate debug info file. The separate debug file's name is
18896 usually @file{@var{executable}.debug}, where @var{executable} is the
18897 name of the corresponding executable file without leading directories
18898 (e.g., @file{ls.debug} for @file{/usr/bin/ls}). In addition, the
18899 debug link specifies a 32-bit @dfn{Cyclic Redundancy Check} (CRC)
18900 checksum for the debug file, which @value{GDBN} uses to validate that
18901 the executable and the debug file came from the same build.
18904 The executable contains a @dfn{build ID}, a unique bit string that is
18905 also present in the corresponding debug info file. (This is supported
18906 only on some operating systems, when using the ELF or PE file formats
18907 for binary files and the @sc{gnu} Binutils.) For more details about
18908 this feature, see the description of the @option{--build-id}
18909 command-line option in @ref{Options, , Command Line Options, ld.info,
18910 The GNU Linker}. The debug info file's name is not specified
18911 explicitly by the build ID, but can be computed from the build ID, see
18915 Depending on the way the debug info file is specified, @value{GDBN}
18916 uses two different methods of looking for the debug file:
18920 For the ``debug link'' method, @value{GDBN} looks up the named file in
18921 the directory of the executable file, then in a subdirectory of that
18922 directory named @file{.debug}, and finally under each one of the global debug
18923 directories, in a subdirectory whose name is identical to the leading
18924 directories of the executable's absolute file name.
18927 For the ``build ID'' method, @value{GDBN} looks in the
18928 @file{.build-id} subdirectory of each one of the global debug directories for
18929 a file named @file{@var{nn}/@var{nnnnnnnn}.debug}, where @var{nn} are the
18930 first 2 hex characters of the build ID bit string, and @var{nnnnnnnn}
18931 are the rest of the bit string. (Real build ID strings are 32 or more
18932 hex characters, not 10.)
18935 So, for example, suppose you ask @value{GDBN} to debug
18936 @file{/usr/bin/ls}, which has a debug link that specifies the
18937 file @file{ls.debug}, and a build ID whose value in hex is
18938 @code{abcdef1234}. If the list of the global debug directories includes
18939 @file{/usr/lib/debug}, then @value{GDBN} will look for the following
18940 debug information files, in the indicated order:
18944 @file{/usr/lib/debug/.build-id/ab/cdef1234.debug}
18946 @file{/usr/bin/ls.debug}
18948 @file{/usr/bin/.debug/ls.debug}
18950 @file{/usr/lib/debug/usr/bin/ls.debug}.
18953 @anchor{debug-file-directory}
18954 Global debugging info directories default to what is set by @value{GDBN}
18955 configure option @option{--with-separate-debug-dir}. During @value{GDBN} run
18956 you can also set the global debugging info directories, and view the list
18957 @value{GDBN} is currently using.
18961 @kindex set debug-file-directory
18962 @item set debug-file-directory @var{directories}
18963 Set the directories which @value{GDBN} searches for separate debugging
18964 information files to @var{directory}. Multiple path components can be set
18965 concatenating them by a path separator.
18967 @kindex show debug-file-directory
18968 @item show debug-file-directory
18969 Show the directories @value{GDBN} searches for separate debugging
18974 @cindex @code{.gnu_debuglink} sections
18975 @cindex debug link sections
18976 A debug link is a special section of the executable file named
18977 @code{.gnu_debuglink}. The section must contain:
18981 A filename, with any leading directory components removed, followed by
18984 zero to three bytes of padding, as needed to reach the next four-byte
18985 boundary within the section, and
18987 a four-byte CRC checksum, stored in the same endianness used for the
18988 executable file itself. The checksum is computed on the debugging
18989 information file's full contents by the function given below, passing
18990 zero as the @var{crc} argument.
18993 Any executable file format can carry a debug link, as long as it can
18994 contain a section named @code{.gnu_debuglink} with the contents
18997 @cindex @code{.note.gnu.build-id} sections
18998 @cindex build ID sections
18999 The build ID is a special section in the executable file (and in other
19000 ELF binary files that @value{GDBN} may consider). This section is
19001 often named @code{.note.gnu.build-id}, but that name is not mandatory.
19002 It contains unique identification for the built files---the ID remains
19003 the same across multiple builds of the same build tree. The default
19004 algorithm SHA1 produces 160 bits (40 hexadecimal characters) of the
19005 content for the build ID string. The same section with an identical
19006 value is present in the original built binary with symbols, in its
19007 stripped variant, and in the separate debugging information file.
19009 The debugging information file itself should be an ordinary
19010 executable, containing a full set of linker symbols, sections, and
19011 debugging information. The sections of the debugging information file
19012 should have the same names, addresses, and sizes as the original file,
19013 but they need not contain any data---much like a @code{.bss} section
19014 in an ordinary executable.
19016 The @sc{gnu} binary utilities (Binutils) package includes the
19017 @samp{objcopy} utility that can produce
19018 the separated executable / debugging information file pairs using the
19019 following commands:
19022 @kbd{objcopy --only-keep-debug foo foo.debug}
19027 These commands remove the debugging
19028 information from the executable file @file{foo} and place it in the file
19029 @file{foo.debug}. You can use the first, second or both methods to link the
19034 The debug link method needs the following additional command to also leave
19035 behind a debug link in @file{foo}:
19038 @kbd{objcopy --add-gnu-debuglink=foo.debug foo}
19041 Ulrich Drepper's @file{elfutils} package, starting with version 0.53, contains
19042 a version of the @code{strip} command such that the command @kbd{strip foo -f
19043 foo.debug} has the same functionality as the two @code{objcopy} commands and
19044 the @code{ln -s} command above, together.
19047 Build ID gets embedded into the main executable using @code{ld --build-id} or
19048 the @value{NGCC} counterpart @code{gcc -Wl,--build-id}. Build ID support plus
19049 compatibility fixes for debug files separation are present in @sc{gnu} binary
19050 utilities (Binutils) package since version 2.18.
19055 @cindex CRC algorithm definition
19056 The CRC used in @code{.gnu_debuglink} is the CRC-32 defined in
19057 IEEE 802.3 using the polynomial:
19059 @c TexInfo requires naked braces for multi-digit exponents for Tex
19060 @c output, but this causes HTML output to barf. HTML has to be set using
19061 @c raw commands. So we end up having to specify this equation in 2
19066 <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>
19067 + <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
19073 @math{x^{32} + x^{26} + x^{23} + x^{22} + x^{16} + x^{12} + x^{11}}
19074 @math{+ x^{10} + x^8 + x^7 + x^5 + x^4 + x^2 + x + 1}
19078 The function is computed byte at a time, taking the least
19079 significant bit of each byte first. The initial pattern
19080 @code{0xffffffff} is used, to ensure leading zeros affect the CRC and
19081 the final result is inverted to ensure trailing zeros also affect the
19084 @emph{Note:} This is the same CRC polynomial as used in handling the
19085 @dfn{Remote Serial Protocol} @code{qCRC} packet (@pxref{qCRC packet}).
19086 However in the case of the Remote Serial Protocol, the CRC is computed
19087 @emph{most} significant bit first, and the result is not inverted, so
19088 trailing zeros have no effect on the CRC value.
19090 To complete the description, we show below the code of the function
19091 which produces the CRC used in @code{.gnu_debuglink}. Inverting the
19092 initially supplied @code{crc} argument means that an initial call to
19093 this function passing in zero will start computing the CRC using
19096 @kindex gnu_debuglink_crc32
19099 gnu_debuglink_crc32 (unsigned long crc,
19100 unsigned char *buf, size_t len)
19102 static const unsigned long crc32_table[256] =
19104 0x00000000, 0x77073096, 0xee0e612c, 0x990951ba, 0x076dc419,
19105 0x706af48f, 0xe963a535, 0x9e6495a3, 0x0edb8832, 0x79dcb8a4,
19106 0xe0d5e91e, 0x97d2d988, 0x09b64c2b, 0x7eb17cbd, 0xe7b82d07,
19107 0x90bf1d91, 0x1db71064, 0x6ab020f2, 0xf3b97148, 0x84be41de,
19108 0x1adad47d, 0x6ddde4eb, 0xf4d4b551, 0x83d385c7, 0x136c9856,
19109 0x646ba8c0, 0xfd62f97a, 0x8a65c9ec, 0x14015c4f, 0x63066cd9,
19110 0xfa0f3d63, 0x8d080df5, 0x3b6e20c8, 0x4c69105e, 0xd56041e4,
19111 0xa2677172, 0x3c03e4d1, 0x4b04d447, 0xd20d85fd, 0xa50ab56b,
19112 0x35b5a8fa, 0x42b2986c, 0xdbbbc9d6, 0xacbcf940, 0x32d86ce3,
19113 0x45df5c75, 0xdcd60dcf, 0xabd13d59, 0x26d930ac, 0x51de003a,
19114 0xc8d75180, 0xbfd06116, 0x21b4f4b5, 0x56b3c423, 0xcfba9599,
19115 0xb8bda50f, 0x2802b89e, 0x5f058808, 0xc60cd9b2, 0xb10be924,
19116 0x2f6f7c87, 0x58684c11, 0xc1611dab, 0xb6662d3d, 0x76dc4190,
19117 0x01db7106, 0x98d220bc, 0xefd5102a, 0x71b18589, 0x06b6b51f,
19118 0x9fbfe4a5, 0xe8b8d433, 0x7807c9a2, 0x0f00f934, 0x9609a88e,
19119 0xe10e9818, 0x7f6a0dbb, 0x086d3d2d, 0x91646c97, 0xe6635c01,
19120 0x6b6b51f4, 0x1c6c6162, 0x856530d8, 0xf262004e, 0x6c0695ed,
19121 0x1b01a57b, 0x8208f4c1, 0xf50fc457, 0x65b0d9c6, 0x12b7e950,
19122 0x8bbeb8ea, 0xfcb9887c, 0x62dd1ddf, 0x15da2d49, 0x8cd37cf3,
19123 0xfbd44c65, 0x4db26158, 0x3ab551ce, 0xa3bc0074, 0xd4bb30e2,
19124 0x4adfa541, 0x3dd895d7, 0xa4d1c46d, 0xd3d6f4fb, 0x4369e96a,
19125 0x346ed9fc, 0xad678846, 0xda60b8d0, 0x44042d73, 0x33031de5,
19126 0xaa0a4c5f, 0xdd0d7cc9, 0x5005713c, 0x270241aa, 0xbe0b1010,
19127 0xc90c2086, 0x5768b525, 0x206f85b3, 0xb966d409, 0xce61e49f,
19128 0x5edef90e, 0x29d9c998, 0xb0d09822, 0xc7d7a8b4, 0x59b33d17,
19129 0x2eb40d81, 0xb7bd5c3b, 0xc0ba6cad, 0xedb88320, 0x9abfb3b6,
19130 0x03b6e20c, 0x74b1d29a, 0xead54739, 0x9dd277af, 0x04db2615,
19131 0x73dc1683, 0xe3630b12, 0x94643b84, 0x0d6d6a3e, 0x7a6a5aa8,
19132 0xe40ecf0b, 0x9309ff9d, 0x0a00ae27, 0x7d079eb1, 0xf00f9344,
19133 0x8708a3d2, 0x1e01f268, 0x6906c2fe, 0xf762575d, 0x806567cb,
19134 0x196c3671, 0x6e6b06e7, 0xfed41b76, 0x89d32be0, 0x10da7a5a,
19135 0x67dd4acc, 0xf9b9df6f, 0x8ebeeff9, 0x17b7be43, 0x60b08ed5,
19136 0xd6d6a3e8, 0xa1d1937e, 0x38d8c2c4, 0x4fdff252, 0xd1bb67f1,
19137 0xa6bc5767, 0x3fb506dd, 0x48b2364b, 0xd80d2bda, 0xaf0a1b4c,
19138 0x36034af6, 0x41047a60, 0xdf60efc3, 0xa867df55, 0x316e8eef,
19139 0x4669be79, 0xcb61b38c, 0xbc66831a, 0x256fd2a0, 0x5268e236,
19140 0xcc0c7795, 0xbb0b4703, 0x220216b9, 0x5505262f, 0xc5ba3bbe,
19141 0xb2bd0b28, 0x2bb45a92, 0x5cb36a04, 0xc2d7ffa7, 0xb5d0cf31,
19142 0x2cd99e8b, 0x5bdeae1d, 0x9b64c2b0, 0xec63f226, 0x756aa39c,
19143 0x026d930a, 0x9c0906a9, 0xeb0e363f, 0x72076785, 0x05005713,
19144 0x95bf4a82, 0xe2b87a14, 0x7bb12bae, 0x0cb61b38, 0x92d28e9b,
19145 0xe5d5be0d, 0x7cdcefb7, 0x0bdbdf21, 0x86d3d2d4, 0xf1d4e242,
19146 0x68ddb3f8, 0x1fda836e, 0x81be16cd, 0xf6b9265b, 0x6fb077e1,
19147 0x18b74777, 0x88085ae6, 0xff0f6a70, 0x66063bca, 0x11010b5c,
19148 0x8f659eff, 0xf862ae69, 0x616bffd3, 0x166ccf45, 0xa00ae278,
19149 0xd70dd2ee, 0x4e048354, 0x3903b3c2, 0xa7672661, 0xd06016f7,
19150 0x4969474d, 0x3e6e77db, 0xaed16a4a, 0xd9d65adc, 0x40df0b66,
19151 0x37d83bf0, 0xa9bcae53, 0xdebb9ec5, 0x47b2cf7f, 0x30b5ffe9,
19152 0xbdbdf21c, 0xcabac28a, 0x53b39330, 0x24b4a3a6, 0xbad03605,
19153 0xcdd70693, 0x54de5729, 0x23d967bf, 0xb3667a2e, 0xc4614ab8,
19154 0x5d681b02, 0x2a6f2b94, 0xb40bbe37, 0xc30c8ea1, 0x5a05df1b,
19157 unsigned char *end;
19159 crc = ~crc & 0xffffffff;
19160 for (end = buf + len; buf < end; ++buf)
19161 crc = crc32_table[(crc ^ *buf) & 0xff] ^ (crc >> 8);
19162 return ~crc & 0xffffffff;
19167 This computation does not apply to the ``build ID'' method.
19169 @node MiniDebugInfo
19170 @section Debugging information in a special section
19171 @cindex separate debug sections
19172 @cindex @samp{.gnu_debugdata} section
19174 Some systems ship pre-built executables and libraries that have a
19175 special @samp{.gnu_debugdata} section. This feature is called
19176 @dfn{MiniDebugInfo}. This section holds an LZMA-compressed object and
19177 is used to supply extra symbols for backtraces.
19179 The intent of this section is to provide extra minimal debugging
19180 information for use in simple backtraces. It is not intended to be a
19181 replacement for full separate debugging information (@pxref{Separate
19182 Debug Files}). The example below shows the intended use; however,
19183 @value{GDBN} does not currently put restrictions on what sort of
19184 debugging information might be included in the section.
19186 @value{GDBN} has support for this extension. If the section exists,
19187 then it is used provided that no other source of debugging information
19188 can be found, and that @value{GDBN} was configured with LZMA support.
19190 This section can be easily created using @command{objcopy} and other
19191 standard utilities:
19194 # Extract the dynamic symbols from the main binary, there is no need
19195 # to also have these in the normal symbol table.
19196 nm -D @var{binary} --format=posix --defined-only \
19197 | awk '@{ print $1 @}' | sort > dynsyms
19199 # Extract all the text (i.e. function) symbols from the debuginfo.
19200 # (Note that we actually also accept "D" symbols, for the benefit
19201 # of platforms like PowerPC64 that use function descriptors.)
19202 nm @var{binary} --format=posix --defined-only \
19203 | awk '@{ if ($2 == "T" || $2 == "t" || $2 == "D") print $1 @}' \
19206 # Keep all the function symbols not already in the dynamic symbol
19208 comm -13 dynsyms funcsyms > keep_symbols
19210 # Separate full debug info into debug binary.
19211 objcopy --only-keep-debug @var{binary} debug
19213 # Copy the full debuginfo, keeping only a minimal set of symbols and
19214 # removing some unnecessary sections.
19215 objcopy -S --remove-section .gdb_index --remove-section .comment \
19216 --keep-symbols=keep_symbols debug mini_debuginfo
19218 # Drop the full debug info from the original binary.
19219 strip --strip-all -R .comment @var{binary}
19221 # Inject the compressed data into the .gnu_debugdata section of the
19224 objcopy --add-section .gnu_debugdata=mini_debuginfo.xz @var{binary}
19228 @section Index Files Speed Up @value{GDBN}
19229 @cindex index files
19230 @cindex @samp{.gdb_index} section
19232 When @value{GDBN} finds a symbol file, it scans the symbols in the
19233 file in order to construct an internal symbol table. This lets most
19234 @value{GDBN} operations work quickly---at the cost of a delay early
19235 on. For large programs, this delay can be quite lengthy, so
19236 @value{GDBN} provides a way to build an index, which speeds up
19239 The index is stored as a section in the symbol file. @value{GDBN} can
19240 write the index to a file, then you can put it into the symbol file
19241 using @command{objcopy}.
19243 To create an index file, use the @code{save gdb-index} command:
19246 @item save gdb-index @var{directory}
19247 @kindex save gdb-index
19248 Create an index file for each symbol file currently known by
19249 @value{GDBN}. Each file is named after its corresponding symbol file,
19250 with @samp{.gdb-index} appended, and is written into the given
19254 Once you have created an index file you can merge it into your symbol
19255 file, here named @file{symfile}, using @command{objcopy}:
19258 $ objcopy --add-section .gdb_index=symfile.gdb-index \
19259 --set-section-flags .gdb_index=readonly symfile symfile
19262 @value{GDBN} will normally ignore older versions of @file{.gdb_index}
19263 sections that have been deprecated. Usually they are deprecated because
19264 they are missing a new feature or have performance issues.
19265 To tell @value{GDBN} to use a deprecated index section anyway
19266 specify @code{set use-deprecated-index-sections on}.
19267 The default is @code{off}.
19268 This can speed up startup, but may result in some functionality being lost.
19269 @xref{Index Section Format}.
19271 @emph{Warning:} Setting @code{use-deprecated-index-sections} to @code{on}
19272 must be done before gdb reads the file. The following will not work:
19275 $ gdb -ex "set use-deprecated-index-sections on" <program>
19278 Instead you must do, for example,
19281 $ gdb -iex "set use-deprecated-index-sections on" <program>
19284 There are currently some limitation on indices. They only work when
19285 for DWARF debugging information, not stabs. And, they do not
19286 currently work for programs using Ada.
19288 @node Symbol Errors
19289 @section Errors Reading Symbol Files
19291 While reading a symbol file, @value{GDBN} occasionally encounters problems,
19292 such as symbol types it does not recognize, or known bugs in compiler
19293 output. By default, @value{GDBN} does not notify you of such problems, since
19294 they are relatively common and primarily of interest to people
19295 debugging compilers. If you are interested in seeing information
19296 about ill-constructed symbol tables, you can either ask @value{GDBN} to print
19297 only one message about each such type of problem, no matter how many
19298 times the problem occurs; or you can ask @value{GDBN} to print more messages,
19299 to see how many times the problems occur, with the @code{set
19300 complaints} command (@pxref{Messages/Warnings, ,Optional Warnings and
19303 The messages currently printed, and their meanings, include:
19306 @item inner block not inside outer block in @var{symbol}
19308 The symbol information shows where symbol scopes begin and end
19309 (such as at the start of a function or a block of statements). This
19310 error indicates that an inner scope block is not fully contained
19311 in its outer scope blocks.
19313 @value{GDBN} circumvents the problem by treating the inner block as if it had
19314 the same scope as the outer block. In the error message, @var{symbol}
19315 may be shown as ``@code{(don't know)}'' if the outer block is not a
19318 @item block at @var{address} out of order
19320 The symbol information for symbol scope blocks should occur in
19321 order of increasing addresses. This error indicates that it does not
19324 @value{GDBN} does not circumvent this problem, and has trouble
19325 locating symbols in the source file whose symbols it is reading. (You
19326 can often determine what source file is affected by specifying
19327 @code{set verbose on}. @xref{Messages/Warnings, ,Optional Warnings and
19330 @item bad block start address patched
19332 The symbol information for a symbol scope block has a start address
19333 smaller than the address of the preceding source line. This is known
19334 to occur in the SunOS 4.1.1 (and earlier) C compiler.
19336 @value{GDBN} circumvents the problem by treating the symbol scope block as
19337 starting on the previous source line.
19339 @item bad string table offset in symbol @var{n}
19342 Symbol number @var{n} contains a pointer into the string table which is
19343 larger than the size of the string table.
19345 @value{GDBN} circumvents the problem by considering the symbol to have the
19346 name @code{foo}, which may cause other problems if many symbols end up
19349 @item unknown symbol type @code{0x@var{nn}}
19351 The symbol information contains new data types that @value{GDBN} does
19352 not yet know how to read. @code{0x@var{nn}} is the symbol type of the
19353 uncomprehended information, in hexadecimal.
19355 @value{GDBN} circumvents the error by ignoring this symbol information.
19356 This usually allows you to debug your program, though certain symbols
19357 are not accessible. If you encounter such a problem and feel like
19358 debugging it, you can debug @code{@value{GDBP}} with itself, breakpoint
19359 on @code{complain}, then go up to the function @code{read_dbx_symtab}
19360 and examine @code{*bufp} to see the symbol.
19362 @item stub type has NULL name
19364 @value{GDBN} could not find the full definition for a struct or class.
19366 @item const/volatile indicator missing (ok if using g++ v1.x), got@dots{}
19367 The symbol information for a C@t{++} member function is missing some
19368 information that recent versions of the compiler should have output for
19371 @item info mismatch between compiler and debugger
19373 @value{GDBN} could not parse a type specification output by the compiler.
19378 @section GDB Data Files
19380 @cindex prefix for data files
19381 @value{GDBN} will sometimes read an auxiliary data file. These files
19382 are kept in a directory known as the @dfn{data directory}.
19384 You can set the data directory's name, and view the name @value{GDBN}
19385 is currently using.
19388 @kindex set data-directory
19389 @item set data-directory @var{directory}
19390 Set the directory which @value{GDBN} searches for auxiliary data files
19391 to @var{directory}.
19393 @kindex show data-directory
19394 @item show data-directory
19395 Show the directory @value{GDBN} searches for auxiliary data files.
19398 @cindex default data directory
19399 @cindex @samp{--with-gdb-datadir}
19400 You can set the default data directory by using the configure-time
19401 @samp{--with-gdb-datadir} option. If the data directory is inside
19402 @value{GDBN}'s configured binary prefix (set with @samp{--prefix} or
19403 @samp{--exec-prefix}), then the default data directory will be updated
19404 automatically if the installed @value{GDBN} is moved to a new
19407 The data directory may also be specified with the
19408 @code{--data-directory} command line option.
19409 @xref{Mode Options}.
19412 @chapter Specifying a Debugging Target
19414 @cindex debugging target
19415 A @dfn{target} is the execution environment occupied by your program.
19417 Often, @value{GDBN} runs in the same host environment as your program;
19418 in that case, the debugging target is specified as a side effect when
19419 you use the @code{file} or @code{core} commands. When you need more
19420 flexibility---for example, running @value{GDBN} on a physically separate
19421 host, or controlling a standalone system over a serial port or a
19422 realtime system over a TCP/IP connection---you can use the @code{target}
19423 command to specify one of the target types configured for @value{GDBN}
19424 (@pxref{Target Commands, ,Commands for Managing Targets}).
19426 @cindex target architecture
19427 It is possible to build @value{GDBN} for several different @dfn{target
19428 architectures}. When @value{GDBN} is built like that, you can choose
19429 one of the available architectures with the @kbd{set architecture}
19433 @kindex set architecture
19434 @kindex show architecture
19435 @item set architecture @var{arch}
19436 This command sets the current target architecture to @var{arch}. The
19437 value of @var{arch} can be @code{"auto"}, in addition to one of the
19438 supported architectures.
19440 @item show architecture
19441 Show the current target architecture.
19443 @item set processor
19445 @kindex set processor
19446 @kindex show processor
19447 These are alias commands for, respectively, @code{set architecture}
19448 and @code{show architecture}.
19452 * Active Targets:: Active targets
19453 * Target Commands:: Commands for managing targets
19454 * Byte Order:: Choosing target byte order
19457 @node Active Targets
19458 @section Active Targets
19460 @cindex stacking targets
19461 @cindex active targets
19462 @cindex multiple targets
19464 There are multiple classes of targets such as: processes, executable files or
19465 recording sessions. Core files belong to the process class, making core file
19466 and process mutually exclusive. Otherwise, @value{GDBN} can work concurrently
19467 on multiple active targets, one in each class. This allows you to (for
19468 example) start a process and inspect its activity, while still having access to
19469 the executable file after the process finishes. Or if you start process
19470 recording (@pxref{Reverse Execution}) and @code{reverse-step} there, you are
19471 presented a virtual layer of the recording target, while the process target
19472 remains stopped at the chronologically last point of the process execution.
19474 Use the @code{core-file} and @code{exec-file} commands to select a new core
19475 file or executable target (@pxref{Files, ,Commands to Specify Files}). To
19476 specify as a target a process that is already running, use the @code{attach}
19477 command (@pxref{Attach, ,Debugging an Already-running Process}).
19479 @node Target Commands
19480 @section Commands for Managing Targets
19483 @item target @var{type} @var{parameters}
19484 Connects the @value{GDBN} host environment to a target machine or
19485 process. A target is typically a protocol for talking to debugging
19486 facilities. You use the argument @var{type} to specify the type or
19487 protocol of the target machine.
19489 Further @var{parameters} are interpreted by the target protocol, but
19490 typically include things like device names or host names to connect
19491 with, process numbers, and baud rates.
19493 The @code{target} command does not repeat if you press @key{RET} again
19494 after executing the command.
19496 @kindex help target
19498 Displays the names of all targets available. To display targets
19499 currently selected, use either @code{info target} or @code{info files}
19500 (@pxref{Files, ,Commands to Specify Files}).
19502 @item help target @var{name}
19503 Describe a particular target, including any parameters necessary to
19506 @kindex set gnutarget
19507 @item set gnutarget @var{args}
19508 @value{GDBN} uses its own library BFD to read your files. @value{GDBN}
19509 knows whether it is reading an @dfn{executable},
19510 a @dfn{core}, or a @dfn{.o} file; however, you can specify the file format
19511 with the @code{set gnutarget} command. Unlike most @code{target} commands,
19512 with @code{gnutarget} the @code{target} refers to a program, not a machine.
19515 @emph{Warning:} To specify a file format with @code{set gnutarget},
19516 you must know the actual BFD name.
19520 @xref{Files, , Commands to Specify Files}.
19522 @kindex show gnutarget
19523 @item show gnutarget
19524 Use the @code{show gnutarget} command to display what file format
19525 @code{gnutarget} is set to read. If you have not set @code{gnutarget},
19526 @value{GDBN} will determine the file format for each file automatically,
19527 and @code{show gnutarget} displays @samp{The current BFD target is "auto"}.
19530 @cindex common targets
19531 Here are some common targets (available, or not, depending on the GDB
19536 @item target exec @var{program}
19537 @cindex executable file target
19538 An executable file. @samp{target exec @var{program}} is the same as
19539 @samp{exec-file @var{program}}.
19541 @item target core @var{filename}
19542 @cindex core dump file target
19543 A core dump file. @samp{target core @var{filename}} is the same as
19544 @samp{core-file @var{filename}}.
19546 @item target remote @var{medium}
19547 @cindex remote target
19548 A remote system connected to @value{GDBN} via a serial line or network
19549 connection. This command tells @value{GDBN} to use its own remote
19550 protocol over @var{medium} for debugging. @xref{Remote Debugging}.
19552 For example, if you have a board connected to @file{/dev/ttya} on the
19553 machine running @value{GDBN}, you could say:
19556 target remote /dev/ttya
19559 @code{target remote} supports the @code{load} command. This is only
19560 useful if you have some other way of getting the stub to the target
19561 system, and you can put it somewhere in memory where it won't get
19562 clobbered by the download.
19564 @item target sim @r{[}@var{simargs}@r{]} @dots{}
19565 @cindex built-in simulator target
19566 Builtin CPU simulator. @value{GDBN} includes simulators for most architectures.
19574 works; however, you cannot assume that a specific memory map, device
19575 drivers, or even basic I/O is available, although some simulators do
19576 provide these. For info about any processor-specific simulator details,
19577 see the appropriate section in @ref{Embedded Processors, ,Embedded
19580 @item target native
19581 @cindex native target
19582 Setup for local/native process debugging. Useful to make the
19583 @code{run} command spawn native processes (likewise @code{attach},
19584 etc.@:) even when @code{set auto-connect-native-target} is @code{off}
19585 (@pxref{set auto-connect-native-target}).
19589 Different targets are available on different configurations of @value{GDBN};
19590 your configuration may have more or fewer targets.
19592 Many remote targets require you to download the executable's code once
19593 you've successfully established a connection. You may wish to control
19594 various aspects of this process.
19599 @kindex set hash@r{, for remote monitors}
19600 @cindex hash mark while downloading
19601 This command controls whether a hash mark @samp{#} is displayed while
19602 downloading a file to the remote monitor. If on, a hash mark is
19603 displayed after each S-record is successfully downloaded to the
19607 @kindex show hash@r{, for remote monitors}
19608 Show the current status of displaying the hash mark.
19610 @item set debug monitor
19611 @kindex set debug monitor
19612 @cindex display remote monitor communications
19613 Enable or disable display of communications messages between
19614 @value{GDBN} and the remote monitor.
19616 @item show debug monitor
19617 @kindex show debug monitor
19618 Show the current status of displaying communications between
19619 @value{GDBN} and the remote monitor.
19624 @kindex load @var{filename} @var{offset}
19625 @item load @var{filename} @var{offset}
19627 Depending on what remote debugging facilities are configured into
19628 @value{GDBN}, the @code{load} command may be available. Where it exists, it
19629 is meant to make @var{filename} (an executable) available for debugging
19630 on the remote system---by downloading, or dynamic linking, for example.
19631 @code{load} also records the @var{filename} symbol table in @value{GDBN}, like
19632 the @code{add-symbol-file} command.
19634 If your @value{GDBN} does not have a @code{load} command, attempting to
19635 execute it gets the error message ``@code{You can't do that when your
19636 target is @dots{}}''
19638 The file is loaded at whatever address is specified in the executable.
19639 For some object file formats, you can specify the load address when you
19640 link the program; for other formats, like a.out, the object file format
19641 specifies a fixed address.
19642 @c FIXME! This would be a good place for an xref to the GNU linker doc.
19644 It is also possible to tell @value{GDBN} to load the executable file at a
19645 specific offset described by the optional argument @var{offset}. When
19646 @var{offset} is provided, @var{filename} must also be provided.
19648 Depending on the remote side capabilities, @value{GDBN} may be able to
19649 load programs into flash memory.
19651 @code{load} does not repeat if you press @key{RET} again after using it.
19656 @kindex flash-erase
19658 @anchor{flash-erase}
19660 Erases all known flash memory regions on the target.
19665 @section Choosing Target Byte Order
19667 @cindex choosing target byte order
19668 @cindex target byte order
19670 Some types of processors, such as the @acronym{MIPS}, PowerPC, and Renesas SH,
19671 offer the ability to run either big-endian or little-endian byte
19672 orders. Usually the executable or symbol will include a bit to
19673 designate the endian-ness, and you will not need to worry about
19674 which to use. However, you may still find it useful to adjust
19675 @value{GDBN}'s idea of processor endian-ness manually.
19679 @item set endian big
19680 Instruct @value{GDBN} to assume the target is big-endian.
19682 @item set endian little
19683 Instruct @value{GDBN} to assume the target is little-endian.
19685 @item set endian auto
19686 Instruct @value{GDBN} to use the byte order associated with the
19690 Display @value{GDBN}'s current idea of the target byte order.
19694 Note that these commands merely adjust interpretation of symbolic
19695 data on the host, and that they have absolutely no effect on the
19699 @node Remote Debugging
19700 @chapter Debugging Remote Programs
19701 @cindex remote debugging
19703 If you are trying to debug a program running on a machine that cannot run
19704 @value{GDBN} in the usual way, it is often useful to use remote debugging.
19705 For example, you might use remote debugging on an operating system kernel,
19706 or on a small system which does not have a general purpose operating system
19707 powerful enough to run a full-featured debugger.
19709 Some configurations of @value{GDBN} have special serial or TCP/IP interfaces
19710 to make this work with particular debugging targets. In addition,
19711 @value{GDBN} comes with a generic serial protocol (specific to @value{GDBN},
19712 but not specific to any particular target system) which you can use if you
19713 write the remote stubs---the code that runs on the remote system to
19714 communicate with @value{GDBN}.
19716 Other remote targets may be available in your
19717 configuration of @value{GDBN}; use @code{help target} to list them.
19720 * Connecting:: Connecting to a remote target
19721 * File Transfer:: Sending files to a remote system
19722 * Server:: Using the gdbserver program
19723 * Remote Configuration:: Remote configuration
19724 * Remote Stub:: Implementing a remote stub
19728 @section Connecting to a Remote Target
19729 @cindex remote debugging, connecting
19730 @cindex @code{gdbserver}, connecting
19731 @cindex remote debugging, types of connections
19732 @cindex @code{gdbserver}, types of connections
19733 @cindex @code{gdbserver}, @code{target remote} mode
19734 @cindex @code{gdbserver}, @code{target extended-remote} mode
19736 This section describes how to connect to a remote target, including the
19737 types of connections and their differences, how to set up executable and
19738 symbol files on the host and target, and the commands used for
19739 connecting to and disconnecting from the remote target.
19741 @subsection Types of Remote Connections
19743 @value{GDBN} supports two types of remote connections, @code{target remote}
19744 mode and @code{target extended-remote} mode. Note that many remote targets
19745 support only @code{target remote} mode. There are several major
19746 differences between the two types of connections, enumerated here:
19750 @cindex remote debugging, detach and program exit
19751 @item Result of detach or program exit
19752 @strong{With target remote mode:} When the debugged program exits or you
19753 detach from it, @value{GDBN} disconnects from the target. When using
19754 @code{gdbserver}, @code{gdbserver} will exit.
19756 @strong{With target extended-remote mode:} When the debugged program exits or
19757 you detach from it, @value{GDBN} remains connected to the target, even
19758 though no program is running. You can rerun the program, attach to a
19759 running program, or use @code{monitor} commands specific to the target.
19761 When using @code{gdbserver} in this case, it does not exit unless it was
19762 invoked using the @option{--once} option. If the @option{--once} option
19763 was not used, you can ask @code{gdbserver} to exit using the
19764 @code{monitor exit} command (@pxref{Monitor Commands for gdbserver}).
19766 @item Specifying the program to debug
19767 For both connection types you use the @code{file} command to specify the
19768 program on the host system. If you are using @code{gdbserver} there are
19769 some differences in how to specify the location of the program on the
19772 @strong{With target remote mode:} You must either specify the program to debug
19773 on the @code{gdbserver} command line or use the @option{--attach} option
19774 (@pxref{Attaching to a program,,Attaching to a Running Program}).
19776 @cindex @option{--multi}, @code{gdbserver} option
19777 @strong{With target extended-remote mode:} You may specify the program to debug
19778 on the @code{gdbserver} command line, or you can load the program or attach
19779 to it using @value{GDBN} commands after connecting to @code{gdbserver}.
19781 @anchor{--multi Option in Types of Remote Connnections}
19782 You can start @code{gdbserver} without supplying an initial command to run
19783 or process ID to attach. To do this, use the @option{--multi} command line
19784 option. Then you can connect using @code{target extended-remote} and start
19785 the program you want to debug (see below for details on using the
19786 @code{run} command in this scenario). Note that the conditions under which
19787 @code{gdbserver} terminates depend on how @value{GDBN} connects to it
19788 (@code{target remote} or @code{target extended-remote}). The
19789 @option{--multi} option to @code{gdbserver} has no influence on that.
19791 @item The @code{run} command
19792 @strong{With target remote mode:} The @code{run} command is not
19793 supported. Once a connection has been established, you can use all
19794 the usual @value{GDBN} commands to examine and change data. The
19795 remote program is already running, so you can use commands like
19796 @kbd{step} and @kbd{continue}.
19798 @strong{With target extended-remote mode:} The @code{run} command is
19799 supported. The @code{run} command uses the value set by
19800 @code{set remote exec-file} (@pxref{set remote exec-file}) to select
19801 the program to run. Command line arguments are supported, except for
19802 wildcard expansion and I/O redirection (@pxref{Arguments}).
19804 If you specify the program to debug on the command line, then the
19805 @code{run} command is not required to start execution, and you can
19806 resume using commands like @kbd{step} and @kbd{continue} as with
19807 @code{target remote} mode.
19809 @anchor{Attaching in Types of Remote Connections}
19811 @strong{With target remote mode:} The @value{GDBN} command @code{attach} is
19812 not supported. To attach to a running program using @code{gdbserver}, you
19813 must use the @option{--attach} option (@pxref{Running gdbserver}).
19815 @strong{With target extended-remote mode:} To attach to a running program,
19816 you may use the @code{attach} command after the connection has been
19817 established. If you are using @code{gdbserver}, you may also invoke
19818 @code{gdbserver} using the @option{--attach} option
19819 (@pxref{Running gdbserver}).
19823 @anchor{Host and target files}
19824 @subsection Host and Target Files
19825 @cindex remote debugging, symbol files
19826 @cindex symbol files, remote debugging
19828 @value{GDBN}, running on the host, needs access to symbol and debugging
19829 information for your program running on the target. This requires
19830 access to an unstripped copy of your program, and possibly any associated
19831 symbol files. Note that this section applies equally to both @code{target
19832 remote} mode and @code{target extended-remote} mode.
19834 Some remote targets (@pxref{qXfer executable filename read}, and
19835 @pxref{Host I/O Packets}) allow @value{GDBN} to access program files over
19836 the same connection used to communicate with @value{GDBN}. With such a
19837 target, if the remote program is unstripped, the only command you need is
19838 @code{target remote} (or @code{target extended-remote}).
19840 If the remote program is stripped, or the target does not support remote
19841 program file access, start up @value{GDBN} using the name of the local
19842 unstripped copy of your program as the first argument, or use the
19843 @code{file} command. Use @code{set sysroot} to specify the location (on
19844 the host) of target libraries (unless your @value{GDBN} was compiled with
19845 the correct sysroot using @code{--with-sysroot}). Alternatively, you
19846 may use @code{set solib-search-path} to specify how @value{GDBN} locates
19849 The symbol file and target libraries must exactly match the executable
19850 and libraries on the target, with one exception: the files on the host
19851 system should not be stripped, even if the files on the target system
19852 are. Mismatched or missing files will lead to confusing results
19853 during debugging. On @sc{gnu}/Linux targets, mismatched or missing
19854 files may also prevent @code{gdbserver} from debugging multi-threaded
19857 @subsection Remote Connection Commands
19858 @cindex remote connection commands
19859 @value{GDBN} can communicate with the target over a serial line, or
19860 over an @acronym{IP} network using @acronym{TCP} or @acronym{UDP}. In
19861 each case, @value{GDBN} uses the same protocol for debugging your
19862 program; only the medium carrying the debugging packets varies. The
19863 @code{target remote} and @code{target extended-remote} commands
19864 establish a connection to the target. Both commands accept the same
19865 arguments, which indicate the medium to use:
19869 @item target remote @var{serial-device}
19870 @itemx target extended-remote @var{serial-device}
19871 @cindex serial line, @code{target remote}
19872 Use @var{serial-device} to communicate with the target. For example,
19873 to use a serial line connected to the device named @file{/dev/ttyb}:
19876 target remote /dev/ttyb
19879 If you're using a serial line, you may want to give @value{GDBN} the
19880 @samp{--baud} option, or use the @code{set serial baud} command
19881 (@pxref{Remote Configuration, set serial baud}) before the
19882 @code{target} command.
19884 @item target remote @code{@var{host}:@var{port}}
19885 @itemx target remote @code{tcp:@var{host}:@var{port}}
19886 @itemx target extended-remote @code{@var{host}:@var{port}}
19887 @itemx target extended-remote @code{tcp:@var{host}:@var{port}}
19888 @cindex @acronym{TCP} port, @code{target remote}
19889 Debug using a @acronym{TCP} connection to @var{port} on @var{host}.
19890 The @var{host} may be either a host name or a numeric @acronym{IP}
19891 address; @var{port} must be a decimal number. The @var{host} could be
19892 the target machine itself, if it is directly connected to the net, or
19893 it might be a terminal server which in turn has a serial line to the
19896 For example, to connect to port 2828 on a terminal server named
19900 target remote manyfarms:2828
19903 If your remote target is actually running on the same machine as your
19904 debugger session (e.g.@: a simulator for your target running on the
19905 same host), you can omit the hostname. For example, to connect to
19906 port 1234 on your local machine:
19909 target remote :1234
19913 Note that the colon is still required here.
19915 @item target remote @code{udp:@var{host}:@var{port}}
19916 @itemx target extended-remote @code{udp:@var{host}:@var{port}}
19917 @cindex @acronym{UDP} port, @code{target remote}
19918 Debug using @acronym{UDP} packets to @var{port} on @var{host}. For example, to
19919 connect to @acronym{UDP} port 2828 on a terminal server named @code{manyfarms}:
19922 target remote udp:manyfarms:2828
19925 When using a @acronym{UDP} connection for remote debugging, you should
19926 keep in mind that the `U' stands for ``Unreliable''. @acronym{UDP}
19927 can silently drop packets on busy or unreliable networks, which will
19928 cause havoc with your debugging session.
19930 @item target remote | @var{command}
19931 @itemx target extended-remote | @var{command}
19932 @cindex pipe, @code{target remote} to
19933 Run @var{command} in the background and communicate with it using a
19934 pipe. The @var{command} is a shell command, to be parsed and expanded
19935 by the system's command shell, @code{/bin/sh}; it should expect remote
19936 protocol packets on its standard input, and send replies on its
19937 standard output. You could use this to run a stand-alone simulator
19938 that speaks the remote debugging protocol, to make net connections
19939 using programs like @code{ssh}, or for other similar tricks.
19941 If @var{command} closes its standard output (perhaps by exiting),
19942 @value{GDBN} will try to send it a @code{SIGTERM} signal. (If the
19943 program has already exited, this will have no effect.)
19947 @cindex interrupting remote programs
19948 @cindex remote programs, interrupting
19949 Whenever @value{GDBN} is waiting for the remote program, if you type the
19950 interrupt character (often @kbd{Ctrl-c}), @value{GDBN} attempts to stop the
19951 program. This may or may not succeed, depending in part on the hardware
19952 and the serial drivers the remote system uses. If you type the
19953 interrupt character once again, @value{GDBN} displays this prompt:
19956 Interrupted while waiting for the program.
19957 Give up (and stop debugging it)? (y or n)
19960 In @code{target remote} mode, if you type @kbd{y}, @value{GDBN} abandons
19961 the remote debugging session. (If you decide you want to try again later,
19962 you can use @kbd{target remote} again to connect once more.) If you type
19963 @kbd{n}, @value{GDBN} goes back to waiting.
19965 In @code{target extended-remote} mode, typing @kbd{n} will leave
19966 @value{GDBN} connected to the target.
19969 @kindex detach (remote)
19971 When you have finished debugging the remote program, you can use the
19972 @code{detach} command to release it from @value{GDBN} control.
19973 Detaching from the target normally resumes its execution, but the results
19974 will depend on your particular remote stub. After the @code{detach}
19975 command in @code{target remote} mode, @value{GDBN} is free to connect to
19976 another target. In @code{target extended-remote} mode, @value{GDBN} is
19977 still connected to the target.
19981 The @code{disconnect} command closes the connection to the target, and
19982 the target is generally not resumed. It will wait for @value{GDBN}
19983 (this instance or another one) to connect and continue debugging. After
19984 the @code{disconnect} command, @value{GDBN} is again free to connect to
19987 @cindex send command to remote monitor
19988 @cindex extend @value{GDBN} for remote targets
19989 @cindex add new commands for external monitor
19991 @item monitor @var{cmd}
19992 This command allows you to send arbitrary commands directly to the
19993 remote monitor. Since @value{GDBN} doesn't care about the commands it
19994 sends like this, this command is the way to extend @value{GDBN}---you
19995 can add new commands that only the external monitor will understand
19999 @node File Transfer
20000 @section Sending files to a remote system
20001 @cindex remote target, file transfer
20002 @cindex file transfer
20003 @cindex sending files to remote systems
20005 Some remote targets offer the ability to transfer files over the same
20006 connection used to communicate with @value{GDBN}. This is convenient
20007 for targets accessible through other means, e.g.@: @sc{gnu}/Linux systems
20008 running @code{gdbserver} over a network interface. For other targets,
20009 e.g.@: embedded devices with only a single serial port, this may be
20010 the only way to upload or download files.
20012 Not all remote targets support these commands.
20016 @item remote put @var{hostfile} @var{targetfile}
20017 Copy file @var{hostfile} from the host system (the machine running
20018 @value{GDBN}) to @var{targetfile} on the target system.
20021 @item remote get @var{targetfile} @var{hostfile}
20022 Copy file @var{targetfile} from the target system to @var{hostfile}
20023 on the host system.
20025 @kindex remote delete
20026 @item remote delete @var{targetfile}
20027 Delete @var{targetfile} from the target system.
20032 @section Using the @code{gdbserver} Program
20035 @cindex remote connection without stubs
20036 @code{gdbserver} is a control program for Unix-like systems, which
20037 allows you to connect your program with a remote @value{GDBN} via
20038 @code{target remote} or @code{target extended-remote}---but without
20039 linking in the usual debugging stub.
20041 @code{gdbserver} is not a complete replacement for the debugging stubs,
20042 because it requires essentially the same operating-system facilities
20043 that @value{GDBN} itself does. In fact, a system that can run
20044 @code{gdbserver} to connect to a remote @value{GDBN} could also run
20045 @value{GDBN} locally! @code{gdbserver} is sometimes useful nevertheless,
20046 because it is a much smaller program than @value{GDBN} itself. It is
20047 also easier to port than all of @value{GDBN}, so you may be able to get
20048 started more quickly on a new system by using @code{gdbserver}.
20049 Finally, if you develop code for real-time systems, you may find that
20050 the tradeoffs involved in real-time operation make it more convenient to
20051 do as much development work as possible on another system, for example
20052 by cross-compiling. You can use @code{gdbserver} to make a similar
20053 choice for debugging.
20055 @value{GDBN} and @code{gdbserver} communicate via either a serial line
20056 or a TCP connection, using the standard @value{GDBN} remote serial
20060 @emph{Warning:} @code{gdbserver} does not have any built-in security.
20061 Do not run @code{gdbserver} connected to any public network; a
20062 @value{GDBN} connection to @code{gdbserver} provides access to the
20063 target system with the same privileges as the user running
20067 @anchor{Running gdbserver}
20068 @subsection Running @code{gdbserver}
20069 @cindex arguments, to @code{gdbserver}
20070 @cindex @code{gdbserver}, command-line arguments
20072 Run @code{gdbserver} on the target system. You need a copy of the
20073 program you want to debug, including any libraries it requires.
20074 @code{gdbserver} does not need your program's symbol table, so you can
20075 strip the program if necessary to save space. @value{GDBN} on the host
20076 system does all the symbol handling.
20078 To use the server, you must tell it how to communicate with @value{GDBN};
20079 the name of your program; and the arguments for your program. The usual
20083 target> gdbserver @var{comm} @var{program} [ @var{args} @dots{} ]
20086 @var{comm} is either a device name (to use a serial line), or a TCP
20087 hostname and portnumber, or @code{-} or @code{stdio} to use
20088 stdin/stdout of @code{gdbserver}.
20089 For example, to debug Emacs with the argument
20090 @samp{foo.txt} and communicate with @value{GDBN} over the serial port
20094 target> gdbserver /dev/com1 emacs foo.txt
20097 @code{gdbserver} waits passively for the host @value{GDBN} to communicate
20100 To use a TCP connection instead of a serial line:
20103 target> gdbserver host:2345 emacs foo.txt
20106 The only difference from the previous example is the first argument,
20107 specifying that you are communicating with the host @value{GDBN} via
20108 TCP. The @samp{host:2345} argument means that @code{gdbserver} is to
20109 expect a TCP connection from machine @samp{host} to local TCP port 2345.
20110 (Currently, the @samp{host} part is ignored.) You can choose any number
20111 you want for the port number as long as it does not conflict with any
20112 TCP ports already in use on the target system (for example, @code{23} is
20113 reserved for @code{telnet}).@footnote{If you choose a port number that
20114 conflicts with another service, @code{gdbserver} prints an error message
20115 and exits.} You must use the same port number with the host @value{GDBN}
20116 @code{target remote} command.
20118 The @code{stdio} connection is useful when starting @code{gdbserver}
20122 (gdb) target remote | ssh -T hostname gdbserver - hello
20125 The @samp{-T} option to ssh is provided because we don't need a remote pty,
20126 and we don't want escape-character handling. Ssh does this by default when
20127 a command is provided, the flag is provided to make it explicit.
20128 You could elide it if you want to.
20130 Programs started with stdio-connected gdbserver have @file{/dev/null} for
20131 @code{stdin}, and @code{stdout},@code{stderr} are sent back to gdb for
20132 display through a pipe connected to gdbserver.
20133 Both @code{stdout} and @code{stderr} use the same pipe.
20135 @anchor{Attaching to a program}
20136 @subsubsection Attaching to a Running Program
20137 @cindex attach to a program, @code{gdbserver}
20138 @cindex @option{--attach}, @code{gdbserver} option
20140 On some targets, @code{gdbserver} can also attach to running programs.
20141 This is accomplished via the @code{--attach} argument. The syntax is:
20144 target> gdbserver --attach @var{comm} @var{pid}
20147 @var{pid} is the process ID of a currently running process. It isn't
20148 necessary to point @code{gdbserver} at a binary for the running process.
20150 In @code{target extended-remote} mode, you can also attach using the
20151 @value{GDBN} attach command
20152 (@pxref{Attaching in Types of Remote Connections}).
20155 You can debug processes by name instead of process ID if your target has the
20156 @code{pidof} utility:
20159 target> gdbserver --attach @var{comm} `pidof @var{program}`
20162 In case more than one copy of @var{program} is running, or @var{program}
20163 has multiple threads, most versions of @code{pidof} support the
20164 @code{-s} option to only return the first process ID.
20166 @subsubsection TCP port allocation lifecycle of @code{gdbserver}
20168 This section applies only when @code{gdbserver} is run to listen on a TCP
20171 @code{gdbserver} normally terminates after all of its debugged processes have
20172 terminated in @kbd{target remote} mode. On the other hand, for @kbd{target
20173 extended-remote}, @code{gdbserver} stays running even with no processes left.
20174 @value{GDBN} normally terminates the spawned debugged process on its exit,
20175 which normally also terminates @code{gdbserver} in the @kbd{target remote}
20176 mode. Therefore, when the connection drops unexpectedly, and @value{GDBN}
20177 cannot ask @code{gdbserver} to kill its debugged processes, @code{gdbserver}
20178 stays running even in the @kbd{target remote} mode.
20180 When @code{gdbserver} stays running, @value{GDBN} can connect to it again later.
20181 Such reconnecting is useful for features like @ref{disconnected tracing}. For
20182 completeness, at most one @value{GDBN} can be connected at a time.
20184 @cindex @option{--once}, @code{gdbserver} option
20185 By default, @code{gdbserver} keeps the listening TCP port open, so that
20186 subsequent connections are possible. However, if you start @code{gdbserver}
20187 with the @option{--once} option, it will stop listening for any further
20188 connection attempts after connecting to the first @value{GDBN} session. This
20189 means no further connections to @code{gdbserver} will be possible after the
20190 first one. It also means @code{gdbserver} will terminate after the first
20191 connection with remote @value{GDBN} has closed, even for unexpectedly closed
20192 connections and even in the @kbd{target extended-remote} mode. The
20193 @option{--once} option allows reusing the same port number for connecting to
20194 multiple instances of @code{gdbserver} running on the same host, since each
20195 instance closes its port after the first connection.
20197 @anchor{Other Command-Line Arguments for gdbserver}
20198 @subsubsection Other Command-Line Arguments for @code{gdbserver}
20200 You can use the @option{--multi} option to start @code{gdbserver} without
20201 specifying a program to debug or a process to attach to. Then you can
20202 attach in @code{target extended-remote} mode and run or attach to a
20203 program. For more information,
20204 @pxref{--multi Option in Types of Remote Connnections}.
20206 @cindex @option{--debug}, @code{gdbserver} option
20207 The @option{--debug} option tells @code{gdbserver} to display extra
20208 status information about the debugging process.
20209 @cindex @option{--remote-debug}, @code{gdbserver} option
20210 The @option{--remote-debug} option tells @code{gdbserver} to display
20211 remote protocol debug output. These options are intended for
20212 @code{gdbserver} development and for bug reports to the developers.
20214 @cindex @option{--debug-format}, @code{gdbserver} option
20215 The @option{--debug-format=option1[,option2,...]} option tells
20216 @code{gdbserver} to include additional information in each output.
20217 Possible options are:
20221 Turn off all extra information in debugging output.
20223 Turn on all extra information in debugging output.
20225 Include a timestamp in each line of debugging output.
20228 Options are processed in order. Thus, for example, if @option{none}
20229 appears last then no additional information is added to debugging output.
20231 @cindex @option{--wrapper}, @code{gdbserver} option
20232 The @option{--wrapper} option specifies a wrapper to launch programs
20233 for debugging. The option should be followed by the name of the
20234 wrapper, then any command-line arguments to pass to the wrapper, then
20235 @kbd{--} indicating the end of the wrapper arguments.
20237 @code{gdbserver} runs the specified wrapper program with a combined
20238 command line including the wrapper arguments, then the name of the
20239 program to debug, then any arguments to the program. The wrapper
20240 runs until it executes your program, and then @value{GDBN} gains control.
20242 You can use any program that eventually calls @code{execve} with
20243 its arguments as a wrapper. Several standard Unix utilities do
20244 this, e.g.@: @code{env} and @code{nohup}. Any Unix shell script ending
20245 with @code{exec "$@@"} will also work.
20247 For example, you can use @code{env} to pass an environment variable to
20248 the debugged program, without setting the variable in @code{gdbserver}'s
20252 $ gdbserver --wrapper env LD_PRELOAD=libtest.so -- :2222 ./testprog
20255 @subsection Connecting to @code{gdbserver}
20257 The basic procedure for connecting to the remote target is:
20261 Run @value{GDBN} on the host system.
20264 Make sure you have the necessary symbol files
20265 (@pxref{Host and target files}).
20266 Load symbols for your application using the @code{file} command before you
20267 connect. Use @code{set sysroot} to locate target libraries (unless your
20268 @value{GDBN} was compiled with the correct sysroot using
20269 @code{--with-sysroot}).
20272 Connect to your target (@pxref{Connecting,,Connecting to a Remote Target}).
20273 For TCP connections, you must start up @code{gdbserver} prior to using
20274 the @code{target} command. Otherwise you may get an error whose
20275 text depends on the host system, but which usually looks something like
20276 @samp{Connection refused}. Don't use the @code{load}
20277 command in @value{GDBN} when using @code{target remote} mode, since the
20278 program is already on the target.
20282 @anchor{Monitor Commands for gdbserver}
20283 @subsection Monitor Commands for @code{gdbserver}
20284 @cindex monitor commands, for @code{gdbserver}
20286 During a @value{GDBN} session using @code{gdbserver}, you can use the
20287 @code{monitor} command to send special requests to @code{gdbserver}.
20288 Here are the available commands.
20292 List the available monitor commands.
20294 @item monitor set debug 0
20295 @itemx monitor set debug 1
20296 Disable or enable general debugging messages.
20298 @item monitor set remote-debug 0
20299 @itemx monitor set remote-debug 1
20300 Disable or enable specific debugging messages associated with the remote
20301 protocol (@pxref{Remote Protocol}).
20303 @item monitor set debug-format option1@r{[},option2,...@r{]}
20304 Specify additional text to add to debugging messages.
20305 Possible options are:
20309 Turn off all extra information in debugging output.
20311 Turn on all extra information in debugging output.
20313 Include a timestamp in each line of debugging output.
20316 Options are processed in order. Thus, for example, if @option{none}
20317 appears last then no additional information is added to debugging output.
20319 @item monitor set libthread-db-search-path [PATH]
20320 @cindex gdbserver, search path for @code{libthread_db}
20321 When this command is issued, @var{path} is a colon-separated list of
20322 directories to search for @code{libthread_db} (@pxref{Threads,,set
20323 libthread-db-search-path}). If you omit @var{path},
20324 @samp{libthread-db-search-path} will be reset to its default value.
20326 The special entry @samp{$pdir} for @samp{libthread-db-search-path} is
20327 not supported in @code{gdbserver}.
20330 Tell gdbserver to exit immediately. This command should be followed by
20331 @code{disconnect} to close the debugging session. @code{gdbserver} will
20332 detach from any attached processes and kill any processes it created.
20333 Use @code{monitor exit} to terminate @code{gdbserver} at the end
20334 of a multi-process mode debug session.
20338 @subsection Tracepoints support in @code{gdbserver}
20339 @cindex tracepoints support in @code{gdbserver}
20341 On some targets, @code{gdbserver} supports tracepoints, fast
20342 tracepoints and static tracepoints.
20344 For fast or static tracepoints to work, a special library called the
20345 @dfn{in-process agent} (IPA), must be loaded in the inferior process.
20346 This library is built and distributed as an integral part of
20347 @code{gdbserver}. In addition, support for static tracepoints
20348 requires building the in-process agent library with static tracepoints
20349 support. At present, the UST (LTTng Userspace Tracer,
20350 @url{http://lttng.org/ust}) tracing engine is supported. This support
20351 is automatically available if UST development headers are found in the
20352 standard include path when @code{gdbserver} is built, or if
20353 @code{gdbserver} was explicitly configured using @option{--with-ust}
20354 to point at such headers. You can explicitly disable the support
20355 using @option{--with-ust=no}.
20357 There are several ways to load the in-process agent in your program:
20360 @item Specifying it as dependency at link time
20362 You can link your program dynamically with the in-process agent
20363 library. On most systems, this is accomplished by adding
20364 @code{-linproctrace} to the link command.
20366 @item Using the system's preloading mechanisms
20368 You can force loading the in-process agent at startup time by using
20369 your system's support for preloading shared libraries. Many Unixes
20370 support the concept of preloading user defined libraries. In most
20371 cases, you do that by specifying @code{LD_PRELOAD=libinproctrace.so}
20372 in the environment. See also the description of @code{gdbserver}'s
20373 @option{--wrapper} command line option.
20375 @item Using @value{GDBN} to force loading the agent at run time
20377 On some systems, you can force the inferior to load a shared library,
20378 by calling a dynamic loader function in the inferior that takes care
20379 of dynamically looking up and loading a shared library. On most Unix
20380 systems, the function is @code{dlopen}. You'll use the @code{call}
20381 command for that. For example:
20384 (@value{GDBP}) call dlopen ("libinproctrace.so", ...)
20387 Note that on most Unix systems, for the @code{dlopen} function to be
20388 available, the program needs to be linked with @code{-ldl}.
20391 On systems that have a userspace dynamic loader, like most Unix
20392 systems, when you connect to @code{gdbserver} using @code{target
20393 remote}, you'll find that the program is stopped at the dynamic
20394 loader's entry point, and no shared library has been loaded in the
20395 program's address space yet, including the in-process agent. In that
20396 case, before being able to use any of the fast or static tracepoints
20397 features, you need to let the loader run and load the shared
20398 libraries. The simplest way to do that is to run the program to the
20399 main procedure. E.g., if debugging a C or C@t{++} program, start
20400 @code{gdbserver} like so:
20403 $ gdbserver :9999 myprogram
20406 Start GDB and connect to @code{gdbserver} like so, and run to main:
20410 (@value{GDBP}) target remote myhost:9999
20411 0x00007f215893ba60 in ?? () from /lib64/ld-linux-x86-64.so.2
20412 (@value{GDBP}) b main
20413 (@value{GDBP}) continue
20416 The in-process tracing agent library should now be loaded into the
20417 process; you can confirm it with the @code{info sharedlibrary}
20418 command, which will list @file{libinproctrace.so} as loaded in the
20419 process. You are now ready to install fast tracepoints, list static
20420 tracepoint markers, probe static tracepoints markers, and start
20423 @node Remote Configuration
20424 @section Remote Configuration
20427 @kindex show remote
20428 This section documents the configuration options available when
20429 debugging remote programs. For the options related to the File I/O
20430 extensions of the remote protocol, see @ref{system,
20431 system-call-allowed}.
20434 @item set remoteaddresssize @var{bits}
20435 @cindex address size for remote targets
20436 @cindex bits in remote address
20437 Set the maximum size of address in a memory packet to the specified
20438 number of bits. @value{GDBN} will mask off the address bits above
20439 that number, when it passes addresses to the remote target. The
20440 default value is the number of bits in the target's address.
20442 @item show remoteaddresssize
20443 Show the current value of remote address size in bits.
20445 @item set serial baud @var{n}
20446 @cindex baud rate for remote targets
20447 Set the baud rate for the remote serial I/O to @var{n} baud. The
20448 value is used to set the speed of the serial port used for debugging
20451 @item show serial baud
20452 Show the current speed of the remote connection.
20454 @item set serial parity @var{parity}
20455 Set the parity for the remote serial I/O. Supported values of @var{parity} are:
20456 @code{even}, @code{none}, and @code{odd}. The default is @code{none}.
20458 @item show serial parity
20459 Show the current parity of the serial port.
20461 @item set remotebreak
20462 @cindex interrupt remote programs
20463 @cindex BREAK signal instead of Ctrl-C
20464 @anchor{set remotebreak}
20465 If set to on, @value{GDBN} sends a @code{BREAK} signal to the remote
20466 when you type @kbd{Ctrl-c} to interrupt the program running
20467 on the remote. If set to off, @value{GDBN} sends the @samp{Ctrl-C}
20468 character instead. The default is off, since most remote systems
20469 expect to see @samp{Ctrl-C} as the interrupt signal.
20471 @item show remotebreak
20472 Show whether @value{GDBN} sends @code{BREAK} or @samp{Ctrl-C} to
20473 interrupt the remote program.
20475 @item set remoteflow on
20476 @itemx set remoteflow off
20477 @kindex set remoteflow
20478 Enable or disable hardware flow control (@code{RTS}/@code{CTS})
20479 on the serial port used to communicate to the remote target.
20481 @item show remoteflow
20482 @kindex show remoteflow
20483 Show the current setting of hardware flow control.
20485 @item set remotelogbase @var{base}
20486 Set the base (a.k.a.@: radix) of logging serial protocol
20487 communications to @var{base}. Supported values of @var{base} are:
20488 @code{ascii}, @code{octal}, and @code{hex}. The default is
20491 @item show remotelogbase
20492 Show the current setting of the radix for logging remote serial
20495 @item set remotelogfile @var{file}
20496 @cindex record serial communications on file
20497 Record remote serial communications on the named @var{file}. The
20498 default is not to record at all.
20500 @item show remotelogfile.
20501 Show the current setting of the file name on which to record the
20502 serial communications.
20504 @item set remotetimeout @var{num}
20505 @cindex timeout for serial communications
20506 @cindex remote timeout
20507 Set the timeout limit to wait for the remote target to respond to
20508 @var{num} seconds. The default is 2 seconds.
20510 @item show remotetimeout
20511 Show the current number of seconds to wait for the remote target
20514 @cindex limit hardware breakpoints and watchpoints
20515 @cindex remote target, limit break- and watchpoints
20516 @anchor{set remote hardware-watchpoint-limit}
20517 @anchor{set remote hardware-breakpoint-limit}
20518 @item set remote hardware-watchpoint-limit @var{limit}
20519 @itemx set remote hardware-breakpoint-limit @var{limit}
20520 Restrict @value{GDBN} to using @var{limit} remote hardware breakpoint or
20521 watchpoints. A limit of -1, the default, is treated as unlimited.
20523 @cindex limit hardware watchpoints length
20524 @cindex remote target, limit watchpoints length
20525 @anchor{set remote hardware-watchpoint-length-limit}
20526 @item set remote hardware-watchpoint-length-limit @var{limit}
20527 Restrict @value{GDBN} to using @var{limit} bytes for the maximum length of
20528 a remote hardware watchpoint. A limit of -1, the default, is treated
20531 @item show remote hardware-watchpoint-length-limit
20532 Show the current limit (in bytes) of the maximum length of
20533 a remote hardware watchpoint.
20535 @item set remote exec-file @var{filename}
20536 @itemx show remote exec-file
20537 @anchor{set remote exec-file}
20538 @cindex executable file, for remote target
20539 Select the file used for @code{run} with @code{target
20540 extended-remote}. This should be set to a filename valid on the
20541 target system. If it is not set, the target will use a default
20542 filename (e.g.@: the last program run).
20544 @item set remote interrupt-sequence
20545 @cindex interrupt remote programs
20546 @cindex select Ctrl-C, BREAK or BREAK-g
20547 Allow the user to select one of @samp{Ctrl-C}, a @code{BREAK} or
20548 @samp{BREAK-g} as the
20549 sequence to the remote target in order to interrupt the execution.
20550 @samp{Ctrl-C} is a default. Some system prefers @code{BREAK} which
20551 is high level of serial line for some certain time.
20552 Linux kernel prefers @samp{BREAK-g}, a.k.a Magic SysRq g.
20553 It is @code{BREAK} signal followed by character @code{g}.
20555 @item show interrupt-sequence
20556 Show which of @samp{Ctrl-C}, @code{BREAK} or @code{BREAK-g}
20557 is sent by @value{GDBN} to interrupt the remote program.
20558 @code{BREAK-g} is BREAK signal followed by @code{g} and
20559 also known as Magic SysRq g.
20561 @item set remote interrupt-on-connect
20562 @cindex send interrupt-sequence on start
20563 Specify whether interrupt-sequence is sent to remote target when
20564 @value{GDBN} connects to it. This is mostly needed when you debug
20565 Linux kernel. Linux kernel expects @code{BREAK} followed by @code{g}
20566 which is known as Magic SysRq g in order to connect @value{GDBN}.
20568 @item show interrupt-on-connect
20569 Show whether interrupt-sequence is sent
20570 to remote target when @value{GDBN} connects to it.
20574 @item set tcp auto-retry on
20575 @cindex auto-retry, for remote TCP target
20576 Enable auto-retry for remote TCP connections. This is useful if the remote
20577 debugging agent is launched in parallel with @value{GDBN}; there is a race
20578 condition because the agent may not become ready to accept the connection
20579 before @value{GDBN} attempts to connect. When auto-retry is
20580 enabled, if the initial attempt to connect fails, @value{GDBN} reattempts
20581 to establish the connection using the timeout specified by
20582 @code{set tcp connect-timeout}.
20584 @item set tcp auto-retry off
20585 Do not auto-retry failed TCP connections.
20587 @item show tcp auto-retry
20588 Show the current auto-retry setting.
20590 @item set tcp connect-timeout @var{seconds}
20591 @itemx set tcp connect-timeout unlimited
20592 @cindex connection timeout, for remote TCP target
20593 @cindex timeout, for remote target connection
20594 Set the timeout for establishing a TCP connection to the remote target to
20595 @var{seconds}. The timeout affects both polling to retry failed connections
20596 (enabled by @code{set tcp auto-retry on}) and waiting for connections
20597 that are merely slow to complete, and represents an approximate cumulative
20598 value. If @var{seconds} is @code{unlimited}, there is no timeout and
20599 @value{GDBN} will keep attempting to establish a connection forever,
20600 unless interrupted with @kbd{Ctrl-c}. The default is 15 seconds.
20602 @item show tcp connect-timeout
20603 Show the current connection timeout setting.
20606 @cindex remote packets, enabling and disabling
20607 The @value{GDBN} remote protocol autodetects the packets supported by
20608 your debugging stub. If you need to override the autodetection, you
20609 can use these commands to enable or disable individual packets. Each
20610 packet can be set to @samp{on} (the remote target supports this
20611 packet), @samp{off} (the remote target does not support this packet),
20612 or @samp{auto} (detect remote target support for this packet). They
20613 all default to @samp{auto}. For more information about each packet,
20614 see @ref{Remote Protocol}.
20616 During normal use, you should not have to use any of these commands.
20617 If you do, that may be a bug in your remote debugging stub, or a bug
20618 in @value{GDBN}. You may want to report the problem to the
20619 @value{GDBN} developers.
20621 For each packet @var{name}, the command to enable or disable the
20622 packet is @code{set remote @var{name}-packet}. The available settings
20625 @multitable @columnfractions 0.28 0.32 0.25
20628 @tab Related Features
20630 @item @code{fetch-register}
20632 @tab @code{info registers}
20634 @item @code{set-register}
20638 @item @code{binary-download}
20640 @tab @code{load}, @code{set}
20642 @item @code{read-aux-vector}
20643 @tab @code{qXfer:auxv:read}
20644 @tab @code{info auxv}
20646 @item @code{symbol-lookup}
20647 @tab @code{qSymbol}
20648 @tab Detecting multiple threads
20650 @item @code{attach}
20651 @tab @code{vAttach}
20654 @item @code{verbose-resume}
20656 @tab Stepping or resuming multiple threads
20662 @item @code{software-breakpoint}
20666 @item @code{hardware-breakpoint}
20670 @item @code{write-watchpoint}
20674 @item @code{read-watchpoint}
20678 @item @code{access-watchpoint}
20682 @item @code{pid-to-exec-file}
20683 @tab @code{qXfer:exec-file:read}
20684 @tab @code{attach}, @code{run}
20686 @item @code{target-features}
20687 @tab @code{qXfer:features:read}
20688 @tab @code{set architecture}
20690 @item @code{library-info}
20691 @tab @code{qXfer:libraries:read}
20692 @tab @code{info sharedlibrary}
20694 @item @code{memory-map}
20695 @tab @code{qXfer:memory-map:read}
20696 @tab @code{info mem}
20698 @item @code{read-sdata-object}
20699 @tab @code{qXfer:sdata:read}
20700 @tab @code{print $_sdata}
20702 @item @code{read-spu-object}
20703 @tab @code{qXfer:spu:read}
20704 @tab @code{info spu}
20706 @item @code{write-spu-object}
20707 @tab @code{qXfer:spu:write}
20708 @tab @code{info spu}
20710 @item @code{read-siginfo-object}
20711 @tab @code{qXfer:siginfo:read}
20712 @tab @code{print $_siginfo}
20714 @item @code{write-siginfo-object}
20715 @tab @code{qXfer:siginfo:write}
20716 @tab @code{set $_siginfo}
20718 @item @code{threads}
20719 @tab @code{qXfer:threads:read}
20720 @tab @code{info threads}
20722 @item @code{get-thread-local-@*storage-address}
20723 @tab @code{qGetTLSAddr}
20724 @tab Displaying @code{__thread} variables
20726 @item @code{get-thread-information-block-address}
20727 @tab @code{qGetTIBAddr}
20728 @tab Display MS-Windows Thread Information Block.
20730 @item @code{search-memory}
20731 @tab @code{qSearch:memory}
20734 @item @code{supported-packets}
20735 @tab @code{qSupported}
20736 @tab Remote communications parameters
20738 @item @code{catch-syscalls}
20739 @tab @code{QCatchSyscalls}
20740 @tab @code{catch syscall}
20742 @item @code{pass-signals}
20743 @tab @code{QPassSignals}
20744 @tab @code{handle @var{signal}}
20746 @item @code{program-signals}
20747 @tab @code{QProgramSignals}
20748 @tab @code{handle @var{signal}}
20750 @item @code{hostio-close-packet}
20751 @tab @code{vFile:close}
20752 @tab @code{remote get}, @code{remote put}
20754 @item @code{hostio-open-packet}
20755 @tab @code{vFile:open}
20756 @tab @code{remote get}, @code{remote put}
20758 @item @code{hostio-pread-packet}
20759 @tab @code{vFile:pread}
20760 @tab @code{remote get}, @code{remote put}
20762 @item @code{hostio-pwrite-packet}
20763 @tab @code{vFile:pwrite}
20764 @tab @code{remote get}, @code{remote put}
20766 @item @code{hostio-unlink-packet}
20767 @tab @code{vFile:unlink}
20768 @tab @code{remote delete}
20770 @item @code{hostio-readlink-packet}
20771 @tab @code{vFile:readlink}
20774 @item @code{hostio-fstat-packet}
20775 @tab @code{vFile:fstat}
20778 @item @code{hostio-setfs-packet}
20779 @tab @code{vFile:setfs}
20782 @item @code{noack-packet}
20783 @tab @code{QStartNoAckMode}
20784 @tab Packet acknowledgment
20786 @item @code{osdata}
20787 @tab @code{qXfer:osdata:read}
20788 @tab @code{info os}
20790 @item @code{query-attached}
20791 @tab @code{qAttached}
20792 @tab Querying remote process attach state.
20794 @item @code{trace-buffer-size}
20795 @tab @code{QTBuffer:size}
20796 @tab @code{set trace-buffer-size}
20798 @item @code{trace-status}
20799 @tab @code{qTStatus}
20800 @tab @code{tstatus}
20802 @item @code{traceframe-info}
20803 @tab @code{qXfer:traceframe-info:read}
20804 @tab Traceframe info
20806 @item @code{install-in-trace}
20807 @tab @code{InstallInTrace}
20808 @tab Install tracepoint in tracing
20810 @item @code{disable-randomization}
20811 @tab @code{QDisableRandomization}
20812 @tab @code{set disable-randomization}
20814 @item @code{conditional-breakpoints-packet}
20815 @tab @code{Z0 and Z1}
20816 @tab @code{Support for target-side breakpoint condition evaluation}
20818 @item @code{multiprocess-extensions}
20819 @tab @code{multiprocess extensions}
20820 @tab Debug multiple processes and remote process PID awareness
20822 @item @code{swbreak-feature}
20823 @tab @code{swbreak stop reason}
20826 @item @code{hwbreak-feature}
20827 @tab @code{hwbreak stop reason}
20830 @item @code{fork-event-feature}
20831 @tab @code{fork stop reason}
20834 @item @code{vfork-event-feature}
20835 @tab @code{vfork stop reason}
20838 @item @code{exec-event-feature}
20839 @tab @code{exec stop reason}
20842 @item @code{thread-events}
20843 @tab @code{QThreadEvents}
20844 @tab Tracking thread lifetime.
20846 @item @code{no-resumed-stop-reply}
20847 @tab @code{no resumed thread left stop reply}
20848 @tab Tracking thread lifetime.
20853 @section Implementing a Remote Stub
20855 @cindex debugging stub, example
20856 @cindex remote stub, example
20857 @cindex stub example, remote debugging
20858 The stub files provided with @value{GDBN} implement the target side of the
20859 communication protocol, and the @value{GDBN} side is implemented in the
20860 @value{GDBN} source file @file{remote.c}. Normally, you can simply allow
20861 these subroutines to communicate, and ignore the details. (If you're
20862 implementing your own stub file, you can still ignore the details: start
20863 with one of the existing stub files. @file{sparc-stub.c} is the best
20864 organized, and therefore the easiest to read.)
20866 @cindex remote serial debugging, overview
20867 To debug a program running on another machine (the debugging
20868 @dfn{target} machine), you must first arrange for all the usual
20869 prerequisites for the program to run by itself. For example, for a C
20874 A startup routine to set up the C runtime environment; these usually
20875 have a name like @file{crt0}. The startup routine may be supplied by
20876 your hardware supplier, or you may have to write your own.
20879 A C subroutine library to support your program's
20880 subroutine calls, notably managing input and output.
20883 A way of getting your program to the other machine---for example, a
20884 download program. These are often supplied by the hardware
20885 manufacturer, but you may have to write your own from hardware
20889 The next step is to arrange for your program to use a serial port to
20890 communicate with the machine where @value{GDBN} is running (the @dfn{host}
20891 machine). In general terms, the scheme looks like this:
20895 @value{GDBN} already understands how to use this protocol; when everything
20896 else is set up, you can simply use the @samp{target remote} command
20897 (@pxref{Targets,,Specifying a Debugging Target}).
20899 @item On the target,
20900 you must link with your program a few special-purpose subroutines that
20901 implement the @value{GDBN} remote serial protocol. The file containing these
20902 subroutines is called a @dfn{debugging stub}.
20904 On certain remote targets, you can use an auxiliary program
20905 @code{gdbserver} instead of linking a stub into your program.
20906 @xref{Server,,Using the @code{gdbserver} Program}, for details.
20909 The debugging stub is specific to the architecture of the remote
20910 machine; for example, use @file{sparc-stub.c} to debug programs on
20913 @cindex remote serial stub list
20914 These working remote stubs are distributed with @value{GDBN}:
20919 @cindex @file{i386-stub.c}
20922 For Intel 386 and compatible architectures.
20925 @cindex @file{m68k-stub.c}
20926 @cindex Motorola 680x0
20928 For Motorola 680x0 architectures.
20931 @cindex @file{sh-stub.c}
20934 For Renesas SH architectures.
20937 @cindex @file{sparc-stub.c}
20939 For @sc{sparc} architectures.
20941 @item sparcl-stub.c
20942 @cindex @file{sparcl-stub.c}
20945 For Fujitsu @sc{sparclite} architectures.
20949 The @file{README} file in the @value{GDBN} distribution may list other
20950 recently added stubs.
20953 * Stub Contents:: What the stub can do for you
20954 * Bootstrapping:: What you must do for the stub
20955 * Debug Session:: Putting it all together
20958 @node Stub Contents
20959 @subsection What the Stub Can Do for You
20961 @cindex remote serial stub
20962 The debugging stub for your architecture supplies these three
20966 @item set_debug_traps
20967 @findex set_debug_traps
20968 @cindex remote serial stub, initialization
20969 This routine arranges for @code{handle_exception} to run when your
20970 program stops. You must call this subroutine explicitly in your
20971 program's startup code.
20973 @item handle_exception
20974 @findex handle_exception
20975 @cindex remote serial stub, main routine
20976 This is the central workhorse, but your program never calls it
20977 explicitly---the setup code arranges for @code{handle_exception} to
20978 run when a trap is triggered.
20980 @code{handle_exception} takes control when your program stops during
20981 execution (for example, on a breakpoint), and mediates communications
20982 with @value{GDBN} on the host machine. This is where the communications
20983 protocol is implemented; @code{handle_exception} acts as the @value{GDBN}
20984 representative on the target machine. It begins by sending summary
20985 information on the state of your program, then continues to execute,
20986 retrieving and transmitting any information @value{GDBN} needs, until you
20987 execute a @value{GDBN} command that makes your program resume; at that point,
20988 @code{handle_exception} returns control to your own code on the target
20992 @cindex @code{breakpoint} subroutine, remote
20993 Use this auxiliary subroutine to make your program contain a
20994 breakpoint. Depending on the particular situation, this may be the only
20995 way for @value{GDBN} to get control. For instance, if your target
20996 machine has some sort of interrupt button, you won't need to call this;
20997 pressing the interrupt button transfers control to
20998 @code{handle_exception}---in effect, to @value{GDBN}. On some machines,
20999 simply receiving characters on the serial port may also trigger a trap;
21000 again, in that situation, you don't need to call @code{breakpoint} from
21001 your own program---simply running @samp{target remote} from the host
21002 @value{GDBN} session gets control.
21004 Call @code{breakpoint} if none of these is true, or if you simply want
21005 to make certain your program stops at a predetermined point for the
21006 start of your debugging session.
21009 @node Bootstrapping
21010 @subsection What You Must Do for the Stub
21012 @cindex remote stub, support routines
21013 The debugging stubs that come with @value{GDBN} are set up for a particular
21014 chip architecture, but they have no information about the rest of your
21015 debugging target machine.
21017 First of all you need to tell the stub how to communicate with the
21021 @item int getDebugChar()
21022 @findex getDebugChar
21023 Write this subroutine to read a single character from the serial port.
21024 It may be identical to @code{getchar} for your target system; a
21025 different name is used to allow you to distinguish the two if you wish.
21027 @item void putDebugChar(int)
21028 @findex putDebugChar
21029 Write this subroutine to write a single character to the serial port.
21030 It may be identical to @code{putchar} for your target system; a
21031 different name is used to allow you to distinguish the two if you wish.
21034 @cindex control C, and remote debugging
21035 @cindex interrupting remote targets
21036 If you want @value{GDBN} to be able to stop your program while it is
21037 running, you need to use an interrupt-driven serial driver, and arrange
21038 for it to stop when it receives a @code{^C} (@samp{\003}, the control-C
21039 character). That is the character which @value{GDBN} uses to tell the
21040 remote system to stop.
21042 Getting the debugging target to return the proper status to @value{GDBN}
21043 probably requires changes to the standard stub; one quick and dirty way
21044 is to just execute a breakpoint instruction (the ``dirty'' part is that
21045 @value{GDBN} reports a @code{SIGTRAP} instead of a @code{SIGINT}).
21047 Other routines you need to supply are:
21050 @item void exceptionHandler (int @var{exception_number}, void *@var{exception_address})
21051 @findex exceptionHandler
21052 Write this function to install @var{exception_address} in the exception
21053 handling tables. You need to do this because the stub does not have any
21054 way of knowing what the exception handling tables on your target system
21055 are like (for example, the processor's table might be in @sc{rom},
21056 containing entries which point to a table in @sc{ram}).
21057 The @var{exception_number} specifies the exception which should be changed;
21058 its meaning is architecture-dependent (for example, different numbers
21059 might represent divide by zero, misaligned access, etc). When this
21060 exception occurs, control should be transferred directly to
21061 @var{exception_address}, and the processor state (stack, registers,
21062 and so on) should be just as it is when a processor exception occurs. So if
21063 you want to use a jump instruction to reach @var{exception_address}, it
21064 should be a simple jump, not a jump to subroutine.
21066 For the 386, @var{exception_address} should be installed as an interrupt
21067 gate so that interrupts are masked while the handler runs. The gate
21068 should be at privilege level 0 (the most privileged level). The
21069 @sc{sparc} and 68k stubs are able to mask interrupts themselves without
21070 help from @code{exceptionHandler}.
21072 @item void flush_i_cache()
21073 @findex flush_i_cache
21074 On @sc{sparc} and @sc{sparclite} only, write this subroutine to flush the
21075 instruction cache, if any, on your target machine. If there is no
21076 instruction cache, this subroutine may be a no-op.
21078 On target machines that have instruction caches, @value{GDBN} requires this
21079 function to make certain that the state of your program is stable.
21083 You must also make sure this library routine is available:
21086 @item void *memset(void *, int, int)
21088 This is the standard library function @code{memset} that sets an area of
21089 memory to a known value. If you have one of the free versions of
21090 @code{libc.a}, @code{memset} can be found there; otherwise, you must
21091 either obtain it from your hardware manufacturer, or write your own.
21094 If you do not use the GNU C compiler, you may need other standard
21095 library subroutines as well; this varies from one stub to another,
21096 but in general the stubs are likely to use any of the common library
21097 subroutines which @code{@value{NGCC}} generates as inline code.
21100 @node Debug Session
21101 @subsection Putting it All Together
21103 @cindex remote serial debugging summary
21104 In summary, when your program is ready to debug, you must follow these
21109 Make sure you have defined the supporting low-level routines
21110 (@pxref{Bootstrapping,,What You Must Do for the Stub}):
21112 @code{getDebugChar}, @code{putDebugChar},
21113 @code{flush_i_cache}, @code{memset}, @code{exceptionHandler}.
21117 Insert these lines in your program's startup code, before the main
21118 procedure is called:
21125 On some machines, when a breakpoint trap is raised, the hardware
21126 automatically makes the PC point to the instruction after the
21127 breakpoint. If your machine doesn't do that, you may need to adjust
21128 @code{handle_exception} to arrange for it to return to the instruction
21129 after the breakpoint on this first invocation, so that your program
21130 doesn't keep hitting the initial breakpoint instead of making
21134 For the 680x0 stub only, you need to provide a variable called
21135 @code{exceptionHook}. Normally you just use:
21138 void (*exceptionHook)() = 0;
21142 but if before calling @code{set_debug_traps}, you set it to point to a
21143 function in your program, that function is called when
21144 @code{@value{GDBN}} continues after stopping on a trap (for example, bus
21145 error). The function indicated by @code{exceptionHook} is called with
21146 one parameter: an @code{int} which is the exception number.
21149 Compile and link together: your program, the @value{GDBN} debugging stub for
21150 your target architecture, and the supporting subroutines.
21153 Make sure you have a serial connection between your target machine and
21154 the @value{GDBN} host, and identify the serial port on the host.
21157 @c The "remote" target now provides a `load' command, so we should
21158 @c document that. FIXME.
21159 Download your program to your target machine (or get it there by
21160 whatever means the manufacturer provides), and start it.
21163 Start @value{GDBN} on the host, and connect to the target
21164 (@pxref{Connecting,,Connecting to a Remote Target}).
21168 @node Configurations
21169 @chapter Configuration-Specific Information
21171 While nearly all @value{GDBN} commands are available for all native and
21172 cross versions of the debugger, there are some exceptions. This chapter
21173 describes things that are only available in certain configurations.
21175 There are three major categories of configurations: native
21176 configurations, where the host and target are the same, embedded
21177 operating system configurations, which are usually the same for several
21178 different processor architectures, and bare embedded processors, which
21179 are quite different from each other.
21184 * Embedded Processors::
21191 This section describes details specific to particular native
21195 * BSD libkvm Interface:: Debugging BSD kernel memory images
21196 * SVR4 Process Information:: SVR4 process information
21197 * DJGPP Native:: Features specific to the DJGPP port
21198 * Cygwin Native:: Features specific to the Cygwin port
21199 * Hurd Native:: Features specific to @sc{gnu} Hurd
21200 * Darwin:: Features specific to Darwin
21203 @node BSD libkvm Interface
21204 @subsection BSD libkvm Interface
21207 @cindex kernel memory image
21208 @cindex kernel crash dump
21210 BSD-derived systems (FreeBSD/NetBSD/OpenBSD) have a kernel memory
21211 interface that provides a uniform interface for accessing kernel virtual
21212 memory images, including live systems and crash dumps. @value{GDBN}
21213 uses this interface to allow you to debug live kernels and kernel crash
21214 dumps on many native BSD configurations. This is implemented as a
21215 special @code{kvm} debugging target. For debugging a live system, load
21216 the currently running kernel into @value{GDBN} and connect to the
21220 (@value{GDBP}) @b{target kvm}
21223 For debugging crash dumps, provide the file name of the crash dump as an
21227 (@value{GDBP}) @b{target kvm /var/crash/bsd.0}
21230 Once connected to the @code{kvm} target, the following commands are
21236 Set current context from the @dfn{Process Control Block} (PCB) address.
21239 Set current context from proc address. This command isn't available on
21240 modern FreeBSD systems.
21243 @node SVR4 Process Information
21244 @subsection SVR4 Process Information
21246 @cindex examine process image
21247 @cindex process info via @file{/proc}
21249 Many versions of SVR4 and compatible systems provide a facility called
21250 @samp{/proc} that can be used to examine the image of a running
21251 process using file-system subroutines.
21253 If @value{GDBN} is configured for an operating system with this
21254 facility, the command @code{info proc} is available to report
21255 information about the process running your program, or about any
21256 process running on your system. This includes, as of this writing,
21257 @sc{gnu}/Linux and Solaris, for example.
21259 This command may also work on core files that were created on a system
21260 that has the @samp{/proc} facility.
21266 @itemx info proc @var{process-id}
21267 Summarize available information about any running process. If a
21268 process ID is specified by @var{process-id}, display information about
21269 that process; otherwise display information about the program being
21270 debugged. The summary includes the debugged process ID, the command
21271 line used to invoke it, its current working directory, and its
21272 executable file's absolute file name.
21274 On some systems, @var{process-id} can be of the form
21275 @samp{[@var{pid}]/@var{tid}} which specifies a certain thread ID
21276 within a process. If the optional @var{pid} part is missing, it means
21277 a thread from the process being debugged (the leading @samp{/} still
21278 needs to be present, or else @value{GDBN} will interpret the number as
21279 a process ID rather than a thread ID).
21281 @item info proc cmdline
21282 @cindex info proc cmdline
21283 Show the original command line of the process. This command is
21284 specific to @sc{gnu}/Linux.
21286 @item info proc cwd
21287 @cindex info proc cwd
21288 Show the current working directory of the process. This command is
21289 specific to @sc{gnu}/Linux.
21291 @item info proc exe
21292 @cindex info proc exe
21293 Show the name of executable of the process. This command is specific
21296 @item info proc mappings
21297 @cindex memory address space mappings
21298 Report the memory address space ranges accessible in the program, with
21299 information on whether the process has read, write, or execute access
21300 rights to each range. On @sc{gnu}/Linux systems, each memory range
21301 includes the object file which is mapped to that range, instead of the
21302 memory access rights to that range.
21304 @item info proc stat
21305 @itemx info proc status
21306 @cindex process detailed status information
21307 These subcommands are specific to @sc{gnu}/Linux systems. They show
21308 the process-related information, including the user ID and group ID;
21309 how many threads are there in the process; its virtual memory usage;
21310 the signals that are pending, blocked, and ignored; its TTY; its
21311 consumption of system and user time; its stack size; its @samp{nice}
21312 value; etc. For more information, see the @samp{proc} man page
21313 (type @kbd{man 5 proc} from your shell prompt).
21315 @item info proc all
21316 Show all the information about the process described under all of the
21317 above @code{info proc} subcommands.
21320 @comment These sub-options of 'info proc' were not included when
21321 @comment procfs.c was re-written. Keep their descriptions around
21322 @comment against the day when someone finds the time to put them back in.
21323 @kindex info proc times
21324 @item info proc times
21325 Starting time, user CPU time, and system CPU time for your program and
21328 @kindex info proc id
21330 Report on the process IDs related to your program: its own process ID,
21331 the ID of its parent, the process group ID, and the session ID.
21334 @item set procfs-trace
21335 @kindex set procfs-trace
21336 @cindex @code{procfs} API calls
21337 This command enables and disables tracing of @code{procfs} API calls.
21339 @item show procfs-trace
21340 @kindex show procfs-trace
21341 Show the current state of @code{procfs} API call tracing.
21343 @item set procfs-file @var{file}
21344 @kindex set procfs-file
21345 Tell @value{GDBN} to write @code{procfs} API trace to the named
21346 @var{file}. @value{GDBN} appends the trace info to the previous
21347 contents of the file. The default is to display the trace on the
21350 @item show procfs-file
21351 @kindex show procfs-file
21352 Show the file to which @code{procfs} API trace is written.
21354 @item proc-trace-entry
21355 @itemx proc-trace-exit
21356 @itemx proc-untrace-entry
21357 @itemx proc-untrace-exit
21358 @kindex proc-trace-entry
21359 @kindex proc-trace-exit
21360 @kindex proc-untrace-entry
21361 @kindex proc-untrace-exit
21362 These commands enable and disable tracing of entries into and exits
21363 from the @code{syscall} interface.
21366 @kindex info pidlist
21367 @cindex process list, QNX Neutrino
21368 For QNX Neutrino only, this command displays the list of all the
21369 processes and all the threads within each process.
21372 @kindex info meminfo
21373 @cindex mapinfo list, QNX Neutrino
21374 For QNX Neutrino only, this command displays the list of all mapinfos.
21378 @subsection Features for Debugging @sc{djgpp} Programs
21379 @cindex @sc{djgpp} debugging
21380 @cindex native @sc{djgpp} debugging
21381 @cindex MS-DOS-specific commands
21384 @sc{djgpp} is a port of the @sc{gnu} development tools to MS-DOS and
21385 MS-Windows. @sc{djgpp} programs are 32-bit protected-mode programs
21386 that use the @dfn{DPMI} (DOS Protected-Mode Interface) API to run on
21387 top of real-mode DOS systems and their emulations.
21389 @value{GDBN} supports native debugging of @sc{djgpp} programs, and
21390 defines a few commands specific to the @sc{djgpp} port. This
21391 subsection describes those commands.
21396 This is a prefix of @sc{djgpp}-specific commands which print
21397 information about the target system and important OS structures.
21400 @cindex MS-DOS system info
21401 @cindex free memory information (MS-DOS)
21402 @item info dos sysinfo
21403 This command displays assorted information about the underlying
21404 platform: the CPU type and features, the OS version and flavor, the
21405 DPMI version, and the available conventional and DPMI memory.
21410 @cindex segment descriptor tables
21411 @cindex descriptor tables display
21413 @itemx info dos ldt
21414 @itemx info dos idt
21415 These 3 commands display entries from, respectively, Global, Local,
21416 and Interrupt Descriptor Tables (GDT, LDT, and IDT). The descriptor
21417 tables are data structures which store a descriptor for each segment
21418 that is currently in use. The segment's selector is an index into a
21419 descriptor table; the table entry for that index holds the
21420 descriptor's base address and limit, and its attributes and access
21423 A typical @sc{djgpp} program uses 3 segments: a code segment, a data
21424 segment (used for both data and the stack), and a DOS segment (which
21425 allows access to DOS/BIOS data structures and absolute addresses in
21426 conventional memory). However, the DPMI host will usually define
21427 additional segments in order to support the DPMI environment.
21429 @cindex garbled pointers
21430 These commands allow to display entries from the descriptor tables.
21431 Without an argument, all entries from the specified table are
21432 displayed. An argument, which should be an integer expression, means
21433 display a single entry whose index is given by the argument. For
21434 example, here's a convenient way to display information about the
21435 debugged program's data segment:
21438 @exdent @code{(@value{GDBP}) info dos ldt $ds}
21439 @exdent @code{0x13f: base=0x11970000 limit=0x0009ffff 32-Bit Data (Read/Write, Exp-up)}
21443 This comes in handy when you want to see whether a pointer is outside
21444 the data segment's limit (i.e.@: @dfn{garbled}).
21446 @cindex page tables display (MS-DOS)
21448 @itemx info dos pte
21449 These two commands display entries from, respectively, the Page
21450 Directory and the Page Tables. Page Directories and Page Tables are
21451 data structures which control how virtual memory addresses are mapped
21452 into physical addresses. A Page Table includes an entry for every
21453 page of memory that is mapped into the program's address space; there
21454 may be several Page Tables, each one holding up to 4096 entries. A
21455 Page Directory has up to 4096 entries, one each for every Page Table
21456 that is currently in use.
21458 Without an argument, @kbd{info dos pde} displays the entire Page
21459 Directory, and @kbd{info dos pte} displays all the entries in all of
21460 the Page Tables. An argument, an integer expression, given to the
21461 @kbd{info dos pde} command means display only that entry from the Page
21462 Directory table. An argument given to the @kbd{info dos pte} command
21463 means display entries from a single Page Table, the one pointed to by
21464 the specified entry in the Page Directory.
21466 @cindex direct memory access (DMA) on MS-DOS
21467 These commands are useful when your program uses @dfn{DMA} (Direct
21468 Memory Access), which needs physical addresses to program the DMA
21471 These commands are supported only with some DPMI servers.
21473 @cindex physical address from linear address
21474 @item info dos address-pte @var{addr}
21475 This command displays the Page Table entry for a specified linear
21476 address. The argument @var{addr} is a linear address which should
21477 already have the appropriate segment's base address added to it,
21478 because this command accepts addresses which may belong to @emph{any}
21479 segment. For example, here's how to display the Page Table entry for
21480 the page where a variable @code{i} is stored:
21483 @exdent @code{(@value{GDBP}) info dos address-pte __djgpp_base_address + (char *)&i}
21484 @exdent @code{Page Table entry for address 0x11a00d30:}
21485 @exdent @code{Base=0x02698000 Dirty Acc. Not-Cached Write-Back Usr Read-Write +0xd30}
21489 This says that @code{i} is stored at offset @code{0xd30} from the page
21490 whose physical base address is @code{0x02698000}, and shows all the
21491 attributes of that page.
21493 Note that you must cast the addresses of variables to a @code{char *},
21494 since otherwise the value of @code{__djgpp_base_address}, the base
21495 address of all variables and functions in a @sc{djgpp} program, will
21496 be added using the rules of C pointer arithmetics: if @code{i} is
21497 declared an @code{int}, @value{GDBN} will add 4 times the value of
21498 @code{__djgpp_base_address} to the address of @code{i}.
21500 Here's another example, it displays the Page Table entry for the
21504 @exdent @code{(@value{GDBP}) info dos address-pte *((unsigned *)&_go32_info_block + 3)}
21505 @exdent @code{Page Table entry for address 0x29110:}
21506 @exdent @code{Base=0x00029000 Dirty Acc. Not-Cached Write-Back Usr Read-Write +0x110}
21510 (The @code{+ 3} offset is because the transfer buffer's address is the
21511 3rd member of the @code{_go32_info_block} structure.) The output
21512 clearly shows that this DPMI server maps the addresses in conventional
21513 memory 1:1, i.e.@: the physical (@code{0x00029000} + @code{0x110}) and
21514 linear (@code{0x29110}) addresses are identical.
21516 This command is supported only with some DPMI servers.
21519 @cindex DOS serial data link, remote debugging
21520 In addition to native debugging, the DJGPP port supports remote
21521 debugging via a serial data link. The following commands are specific
21522 to remote serial debugging in the DJGPP port of @value{GDBN}.
21525 @kindex set com1base
21526 @kindex set com1irq
21527 @kindex set com2base
21528 @kindex set com2irq
21529 @kindex set com3base
21530 @kindex set com3irq
21531 @kindex set com4base
21532 @kindex set com4irq
21533 @item set com1base @var{addr}
21534 This command sets the base I/O port address of the @file{COM1} serial
21537 @item set com1irq @var{irq}
21538 This command sets the @dfn{Interrupt Request} (@code{IRQ}) line to use
21539 for the @file{COM1} serial port.
21541 There are similar commands @samp{set com2base}, @samp{set com3irq},
21542 etc.@: for setting the port address and the @code{IRQ} lines for the
21545 @kindex show com1base
21546 @kindex show com1irq
21547 @kindex show com2base
21548 @kindex show com2irq
21549 @kindex show com3base
21550 @kindex show com3irq
21551 @kindex show com4base
21552 @kindex show com4irq
21553 The related commands @samp{show com1base}, @samp{show com1irq} etc.@:
21554 display the current settings of the base address and the @code{IRQ}
21555 lines used by the COM ports.
21558 @kindex info serial
21559 @cindex DOS serial port status
21560 This command prints the status of the 4 DOS serial ports. For each
21561 port, it prints whether it's active or not, its I/O base address and
21562 IRQ number, whether it uses a 16550-style FIFO, its baudrate, and the
21563 counts of various errors encountered so far.
21567 @node Cygwin Native
21568 @subsection Features for Debugging MS Windows PE Executables
21569 @cindex MS Windows debugging
21570 @cindex native Cygwin debugging
21571 @cindex Cygwin-specific commands
21573 @value{GDBN} supports native debugging of MS Windows programs, including
21574 DLLs with and without symbolic debugging information.
21576 @cindex Ctrl-BREAK, MS-Windows
21577 @cindex interrupt debuggee on MS-Windows
21578 MS-Windows programs that call @code{SetConsoleMode} to switch off the
21579 special meaning of the @samp{Ctrl-C} keystroke cannot be interrupted
21580 by typing @kbd{C-c}. For this reason, @value{GDBN} on MS-Windows
21581 supports @kbd{C-@key{BREAK}} as an alternative interrupt key
21582 sequence, which can be used to interrupt the debuggee even if it
21585 There are various additional Cygwin-specific commands, described in
21586 this section. Working with DLLs that have no debugging symbols is
21587 described in @ref{Non-debug DLL Symbols}.
21592 This is a prefix of MS Windows-specific commands which print
21593 information about the target system and important OS structures.
21595 @item info w32 selector
21596 This command displays information returned by
21597 the Win32 API @code{GetThreadSelectorEntry} function.
21598 It takes an optional argument that is evaluated to
21599 a long value to give the information about this given selector.
21600 Without argument, this command displays information
21601 about the six segment registers.
21603 @item info w32 thread-information-block
21604 This command displays thread specific information stored in the
21605 Thread Information Block (readable on the X86 CPU family using @code{$fs}
21606 selector for 32-bit programs and @code{$gs} for 64-bit programs).
21608 @kindex signal-event
21609 @item signal-event @var{id}
21610 This command signals an event with user-provided @var{id}. Used to resume
21611 crashing process when attached to it using MS-Windows JIT debugging (AeDebug).
21613 To use it, create or edit the following keys in
21614 @code{HKLM\SOFTWARE\Microsoft\Windows NT\CurrentVersion\AeDebug} and/or
21615 @code{HKLM\SOFTWARE\Wow6432Node\Microsoft\Windows NT\CurrentVersion\AeDebug}
21616 (for x86_64 versions):
21620 @code{Debugger} (REG_SZ) --- a command to launch the debugger.
21621 Suggested command is: @code{@var{fully-qualified-path-to-gdb.exe} -ex
21622 "attach %ld" -ex "signal-event %ld" -ex "continue"}.
21624 The first @code{%ld} will be replaced by the process ID of the
21625 crashing process, the second @code{%ld} will be replaced by the ID of
21626 the event that blocks the crashing process, waiting for @value{GDBN}
21630 @code{Auto} (REG_SZ) --- either @code{1} or @code{0}. @code{1} will
21631 make the system run debugger specified by the Debugger key
21632 automatically, @code{0} will cause a dialog box with ``OK'' and
21633 ``Cancel'' buttons to appear, which allows the user to either
21634 terminate the crashing process (OK) or debug it (Cancel).
21637 @kindex set cygwin-exceptions
21638 @cindex debugging the Cygwin DLL
21639 @cindex Cygwin DLL, debugging
21640 @item set cygwin-exceptions @var{mode}
21641 If @var{mode} is @code{on}, @value{GDBN} will break on exceptions that
21642 happen inside the Cygwin DLL. If @var{mode} is @code{off},
21643 @value{GDBN} will delay recognition of exceptions, and may ignore some
21644 exceptions which seem to be caused by internal Cygwin DLL
21645 ``bookkeeping''. This option is meant primarily for debugging the
21646 Cygwin DLL itself; the default value is @code{off} to avoid annoying
21647 @value{GDBN} users with false @code{SIGSEGV} signals.
21649 @kindex show cygwin-exceptions
21650 @item show cygwin-exceptions
21651 Displays whether @value{GDBN} will break on exceptions that happen
21652 inside the Cygwin DLL itself.
21654 @kindex set new-console
21655 @item set new-console @var{mode}
21656 If @var{mode} is @code{on} the debuggee will
21657 be started in a new console on next start.
21658 If @var{mode} is @code{off}, the debuggee will
21659 be started in the same console as the debugger.
21661 @kindex show new-console
21662 @item show new-console
21663 Displays whether a new console is used
21664 when the debuggee is started.
21666 @kindex set new-group
21667 @item set new-group @var{mode}
21668 This boolean value controls whether the debuggee should
21669 start a new group or stay in the same group as the debugger.
21670 This affects the way the Windows OS handles
21673 @kindex show new-group
21674 @item show new-group
21675 Displays current value of new-group boolean.
21677 @kindex set debugevents
21678 @item set debugevents
21679 This boolean value adds debug output concerning kernel events related
21680 to the debuggee seen by the debugger. This includes events that
21681 signal thread and process creation and exit, DLL loading and
21682 unloading, console interrupts, and debugging messages produced by the
21683 Windows @code{OutputDebugString} API call.
21685 @kindex set debugexec
21686 @item set debugexec
21687 This boolean value adds debug output concerning execute events
21688 (such as resume thread) seen by the debugger.
21690 @kindex set debugexceptions
21691 @item set debugexceptions
21692 This boolean value adds debug output concerning exceptions in the
21693 debuggee seen by the debugger.
21695 @kindex set debugmemory
21696 @item set debugmemory
21697 This boolean value adds debug output concerning debuggee memory reads
21698 and writes by the debugger.
21702 This boolean values specifies whether the debuggee is called
21703 via a shell or directly (default value is on).
21707 Displays if the debuggee will be started with a shell.
21712 * Non-debug DLL Symbols:: Support for DLLs without debugging symbols
21715 @node Non-debug DLL Symbols
21716 @subsubsection Support for DLLs without Debugging Symbols
21717 @cindex DLLs with no debugging symbols
21718 @cindex Minimal symbols and DLLs
21720 Very often on windows, some of the DLLs that your program relies on do
21721 not include symbolic debugging information (for example,
21722 @file{kernel32.dll}). When @value{GDBN} doesn't recognize any debugging
21723 symbols in a DLL, it relies on the minimal amount of symbolic
21724 information contained in the DLL's export table. This section
21725 describes working with such symbols, known internally to @value{GDBN} as
21726 ``minimal symbols''.
21728 Note that before the debugged program has started execution, no DLLs
21729 will have been loaded. The easiest way around this problem is simply to
21730 start the program --- either by setting a breakpoint or letting the
21731 program run once to completion.
21733 @subsubsection DLL Name Prefixes
21735 In keeping with the naming conventions used by the Microsoft debugging
21736 tools, DLL export symbols are made available with a prefix based on the
21737 DLL name, for instance @code{KERNEL32!CreateFileA}. The plain name is
21738 also entered into the symbol table, so @code{CreateFileA} is often
21739 sufficient. In some cases there will be name clashes within a program
21740 (particularly if the executable itself includes full debugging symbols)
21741 necessitating the use of the fully qualified name when referring to the
21742 contents of the DLL. Use single-quotes around the name to avoid the
21743 exclamation mark (``!'') being interpreted as a language operator.
21745 Note that the internal name of the DLL may be all upper-case, even
21746 though the file name of the DLL is lower-case, or vice-versa. Since
21747 symbols within @value{GDBN} are @emph{case-sensitive} this may cause
21748 some confusion. If in doubt, try the @code{info functions} and
21749 @code{info variables} commands or even @code{maint print msymbols}
21750 (@pxref{Symbols}). Here's an example:
21753 (@value{GDBP}) info function CreateFileA
21754 All functions matching regular expression "CreateFileA":
21756 Non-debugging symbols:
21757 0x77e885f4 CreateFileA
21758 0x77e885f4 KERNEL32!CreateFileA
21762 (@value{GDBP}) info function !
21763 All functions matching regular expression "!":
21765 Non-debugging symbols:
21766 0x6100114c cygwin1!__assert
21767 0x61004034 cygwin1!_dll_crt0@@0
21768 0x61004240 cygwin1!dll_crt0(per_process *)
21772 @subsubsection Working with Minimal Symbols
21774 Symbols extracted from a DLL's export table do not contain very much
21775 type information. All that @value{GDBN} can do is guess whether a symbol
21776 refers to a function or variable depending on the linker section that
21777 contains the symbol. Also note that the actual contents of the memory
21778 contained in a DLL are not available unless the program is running. This
21779 means that you cannot examine the contents of a variable or disassemble
21780 a function within a DLL without a running program.
21782 Variables are generally treated as pointers and dereferenced
21783 automatically. For this reason, it is often necessary to prefix a
21784 variable name with the address-of operator (``&'') and provide explicit
21785 type information in the command. Here's an example of the type of
21789 (@value{GDBP}) print 'cygwin1!__argv'
21794 (@value{GDBP}) x 'cygwin1!__argv'
21795 0x10021610: "\230y\""
21798 And two possible solutions:
21801 (@value{GDBP}) print ((char **)'cygwin1!__argv')[0]
21802 $2 = 0x22fd98 "/cygdrive/c/mydirectory/myprogram"
21806 (@value{GDBP}) x/2x &'cygwin1!__argv'
21807 0x610c0aa8 <cygwin1!__argv>: 0x10021608 0x00000000
21808 (@value{GDBP}) x/x 0x10021608
21809 0x10021608: 0x0022fd98
21810 (@value{GDBP}) x/s 0x0022fd98
21811 0x22fd98: "/cygdrive/c/mydirectory/myprogram"
21814 Setting a break point within a DLL is possible even before the program
21815 starts execution. However, under these circumstances, @value{GDBN} can't
21816 examine the initial instructions of the function in order to skip the
21817 function's frame set-up code. You can work around this by using ``*&''
21818 to set the breakpoint at a raw memory address:
21821 (@value{GDBP}) break *&'python22!PyOS_Readline'
21822 Breakpoint 1 at 0x1e04eff0
21825 The author of these extensions is not entirely convinced that setting a
21826 break point within a shared DLL like @file{kernel32.dll} is completely
21830 @subsection Commands Specific to @sc{gnu} Hurd Systems
21831 @cindex @sc{gnu} Hurd debugging
21833 This subsection describes @value{GDBN} commands specific to the
21834 @sc{gnu} Hurd native debugging.
21839 @kindex set signals@r{, Hurd command}
21840 @kindex set sigs@r{, Hurd command}
21841 This command toggles the state of inferior signal interception by
21842 @value{GDBN}. Mach exceptions, such as breakpoint traps, are not
21843 affected by this command. @code{sigs} is a shorthand alias for
21848 @kindex show signals@r{, Hurd command}
21849 @kindex show sigs@r{, Hurd command}
21850 Show the current state of intercepting inferior's signals.
21852 @item set signal-thread
21853 @itemx set sigthread
21854 @kindex set signal-thread
21855 @kindex set sigthread
21856 This command tells @value{GDBN} which thread is the @code{libc} signal
21857 thread. That thread is run when a signal is delivered to a running
21858 process. @code{set sigthread} is the shorthand alias of @code{set
21861 @item show signal-thread
21862 @itemx show sigthread
21863 @kindex show signal-thread
21864 @kindex show sigthread
21865 These two commands show which thread will run when the inferior is
21866 delivered a signal.
21869 @kindex set stopped@r{, Hurd command}
21870 This commands tells @value{GDBN} that the inferior process is stopped,
21871 as with the @code{SIGSTOP} signal. The stopped process can be
21872 continued by delivering a signal to it.
21875 @kindex show stopped@r{, Hurd command}
21876 This command shows whether @value{GDBN} thinks the debuggee is
21879 @item set exceptions
21880 @kindex set exceptions@r{, Hurd command}
21881 Use this command to turn off trapping of exceptions in the inferior.
21882 When exception trapping is off, neither breakpoints nor
21883 single-stepping will work. To restore the default, set exception
21886 @item show exceptions
21887 @kindex show exceptions@r{, Hurd command}
21888 Show the current state of trapping exceptions in the inferior.
21890 @item set task pause
21891 @kindex set task@r{, Hurd commands}
21892 @cindex task attributes (@sc{gnu} Hurd)
21893 @cindex pause current task (@sc{gnu} Hurd)
21894 This command toggles task suspension when @value{GDBN} has control.
21895 Setting it to on takes effect immediately, and the task is suspended
21896 whenever @value{GDBN} gets control. Setting it to off will take
21897 effect the next time the inferior is continued. If this option is set
21898 to off, you can use @code{set thread default pause on} or @code{set
21899 thread pause on} (see below) to pause individual threads.
21901 @item show task pause
21902 @kindex show task@r{, Hurd commands}
21903 Show the current state of task suspension.
21905 @item set task detach-suspend-count
21906 @cindex task suspend count
21907 @cindex detach from task, @sc{gnu} Hurd
21908 This command sets the suspend count the task will be left with when
21909 @value{GDBN} detaches from it.
21911 @item show task detach-suspend-count
21912 Show the suspend count the task will be left with when detaching.
21914 @item set task exception-port
21915 @itemx set task excp
21916 @cindex task exception port, @sc{gnu} Hurd
21917 This command sets the task exception port to which @value{GDBN} will
21918 forward exceptions. The argument should be the value of the @dfn{send
21919 rights} of the task. @code{set task excp} is a shorthand alias.
21921 @item set noninvasive
21922 @cindex noninvasive task options
21923 This command switches @value{GDBN} to a mode that is the least
21924 invasive as far as interfering with the inferior is concerned. This
21925 is the same as using @code{set task pause}, @code{set exceptions}, and
21926 @code{set signals} to values opposite to the defaults.
21928 @item info send-rights
21929 @itemx info receive-rights
21930 @itemx info port-rights
21931 @itemx info port-sets
21932 @itemx info dead-names
21935 @cindex send rights, @sc{gnu} Hurd
21936 @cindex receive rights, @sc{gnu} Hurd
21937 @cindex port rights, @sc{gnu} Hurd
21938 @cindex port sets, @sc{gnu} Hurd
21939 @cindex dead names, @sc{gnu} Hurd
21940 These commands display information about, respectively, send rights,
21941 receive rights, port rights, port sets, and dead names of a task.
21942 There are also shorthand aliases: @code{info ports} for @code{info
21943 port-rights} and @code{info psets} for @code{info port-sets}.
21945 @item set thread pause
21946 @kindex set thread@r{, Hurd command}
21947 @cindex thread properties, @sc{gnu} Hurd
21948 @cindex pause current thread (@sc{gnu} Hurd)
21949 This command toggles current thread suspension when @value{GDBN} has
21950 control. Setting it to on takes effect immediately, and the current
21951 thread is suspended whenever @value{GDBN} gets control. Setting it to
21952 off will take effect the next time the inferior is continued.
21953 Normally, this command has no effect, since when @value{GDBN} has
21954 control, the whole task is suspended. However, if you used @code{set
21955 task pause off} (see above), this command comes in handy to suspend
21956 only the current thread.
21958 @item show thread pause
21959 @kindex show thread@r{, Hurd command}
21960 This command shows the state of current thread suspension.
21962 @item set thread run
21963 This command sets whether the current thread is allowed to run.
21965 @item show thread run
21966 Show whether the current thread is allowed to run.
21968 @item set thread detach-suspend-count
21969 @cindex thread suspend count, @sc{gnu} Hurd
21970 @cindex detach from thread, @sc{gnu} Hurd
21971 This command sets the suspend count @value{GDBN} will leave on a
21972 thread when detaching. This number is relative to the suspend count
21973 found by @value{GDBN} when it notices the thread; use @code{set thread
21974 takeover-suspend-count} to force it to an absolute value.
21976 @item show thread detach-suspend-count
21977 Show the suspend count @value{GDBN} will leave on the thread when
21980 @item set thread exception-port
21981 @itemx set thread excp
21982 Set the thread exception port to which to forward exceptions. This
21983 overrides the port set by @code{set task exception-port} (see above).
21984 @code{set thread excp} is the shorthand alias.
21986 @item set thread takeover-suspend-count
21987 Normally, @value{GDBN}'s thread suspend counts are relative to the
21988 value @value{GDBN} finds when it notices each thread. This command
21989 changes the suspend counts to be absolute instead.
21991 @item set thread default
21992 @itemx show thread default
21993 @cindex thread default settings, @sc{gnu} Hurd
21994 Each of the above @code{set thread} commands has a @code{set thread
21995 default} counterpart (e.g., @code{set thread default pause}, @code{set
21996 thread default exception-port}, etc.). The @code{thread default}
21997 variety of commands sets the default thread properties for all
21998 threads; you can then change the properties of individual threads with
21999 the non-default commands.
22006 @value{GDBN} provides the following commands specific to the Darwin target:
22009 @item set debug darwin @var{num}
22010 @kindex set debug darwin
22011 When set to a non zero value, enables debugging messages specific to
22012 the Darwin support. Higher values produce more verbose output.
22014 @item show debug darwin
22015 @kindex show debug darwin
22016 Show the current state of Darwin messages.
22018 @item set debug mach-o @var{num}
22019 @kindex set debug mach-o
22020 When set to a non zero value, enables debugging messages while
22021 @value{GDBN} is reading Darwin object files. (@dfn{Mach-O} is the
22022 file format used on Darwin for object and executable files.) Higher
22023 values produce more verbose output. This is a command to diagnose
22024 problems internal to @value{GDBN} and should not be needed in normal
22027 @item show debug mach-o
22028 @kindex show debug mach-o
22029 Show the current state of Mach-O file messages.
22031 @item set mach-exceptions on
22032 @itemx set mach-exceptions off
22033 @kindex set mach-exceptions
22034 On Darwin, faults are first reported as a Mach exception and are then
22035 mapped to a Posix signal. Use this command to turn on trapping of
22036 Mach exceptions in the inferior. This might be sometimes useful to
22037 better understand the cause of a fault. The default is off.
22039 @item show mach-exceptions
22040 @kindex show mach-exceptions
22041 Show the current state of exceptions trapping.
22046 @section Embedded Operating Systems
22048 This section describes configurations involving the debugging of
22049 embedded operating systems that are available for several different
22052 @value{GDBN} includes the ability to debug programs running on
22053 various real-time operating systems.
22055 @node Embedded Processors
22056 @section Embedded Processors
22058 This section goes into details specific to particular embedded
22061 @cindex send command to simulator
22062 Whenever a specific embedded processor has a simulator, @value{GDBN}
22063 allows to send an arbitrary command to the simulator.
22066 @item sim @var{command}
22067 @kindex sim@r{, a command}
22068 Send an arbitrary @var{command} string to the simulator. Consult the
22069 documentation for the specific simulator in use for information about
22070 acceptable commands.
22075 * ARC:: Synopsys ARC
22077 * M68K:: Motorola M68K
22078 * MicroBlaze:: Xilinx MicroBlaze
22079 * MIPS Embedded:: MIPS Embedded
22080 * PowerPC Embedded:: PowerPC Embedded
22083 * Super-H:: Renesas Super-H
22087 @subsection Synopsys ARC
22088 @cindex Synopsys ARC
22089 @cindex ARC specific commands
22095 @value{GDBN} provides the following ARC-specific commands:
22098 @item set debug arc
22099 @kindex set debug arc
22100 Control the level of ARC specific debug messages. Use 0 for no messages (the
22101 default) and 1 for debug messages. At present higher values offer no further
22104 @item show debug arc
22105 @kindex show debug arc
22106 Show the level of ARC specific debugging in operation.
22113 @value{GDBN} provides the following ARM-specific commands:
22116 @item set arm disassembler
22118 This commands selects from a list of disassembly styles. The
22119 @code{"std"} style is the standard style.
22121 @item show arm disassembler
22123 Show the current disassembly style.
22125 @item set arm apcs32
22126 @cindex ARM 32-bit mode
22127 This command toggles ARM operation mode between 32-bit and 26-bit.
22129 @item show arm apcs32
22130 Display the current usage of the ARM 32-bit mode.
22132 @item set arm fpu @var{fputype}
22133 This command sets the ARM floating-point unit (FPU) type. The
22134 argument @var{fputype} can be one of these:
22138 Determine the FPU type by querying the OS ABI.
22140 Software FPU, with mixed-endian doubles on little-endian ARM
22143 GCC-compiled FPA co-processor.
22145 Software FPU with pure-endian doubles.
22151 Show the current type of the FPU.
22154 This command forces @value{GDBN} to use the specified ABI.
22157 Show the currently used ABI.
22159 @item set arm fallback-mode (arm|thumb|auto)
22160 @value{GDBN} uses the symbol table, when available, to determine
22161 whether instructions are ARM or Thumb. This command controls
22162 @value{GDBN}'s default behavior when the symbol table is not
22163 available. The default is @samp{auto}, which causes @value{GDBN} to
22164 use the current execution mode (from the @code{T} bit in the @code{CPSR}
22167 @item show arm fallback-mode
22168 Show the current fallback instruction mode.
22170 @item set arm force-mode (arm|thumb|auto)
22171 This command overrides use of the symbol table to determine whether
22172 instructions are ARM or Thumb. The default is @samp{auto}, which
22173 causes @value{GDBN} to use the symbol table and then the setting
22174 of @samp{set arm fallback-mode}.
22176 @item show arm force-mode
22177 Show the current forced instruction mode.
22179 @item set debug arm
22180 Toggle whether to display ARM-specific debugging messages from the ARM
22181 target support subsystem.
22183 @item show debug arm
22184 Show whether ARM-specific debugging messages are enabled.
22188 @item target sim @r{[}@var{simargs}@r{]} @dots{}
22189 The @value{GDBN} ARM simulator accepts the following optional arguments.
22192 @item --swi-support=@var{type}
22193 Tell the simulator which SWI interfaces to support. The argument
22194 @var{type} may be a comma separated list of the following values.
22195 The default value is @code{all}.
22210 The Motorola m68k configuration includes ColdFire support.
22213 @subsection MicroBlaze
22214 @cindex Xilinx MicroBlaze
22215 @cindex XMD, Xilinx Microprocessor Debugger
22217 The MicroBlaze is a soft-core processor supported on various Xilinx
22218 FPGAs, such as Spartan or Virtex series. Boards with these processors
22219 usually have JTAG ports which connect to a host system running the Xilinx
22220 Embedded Development Kit (EDK) or Software Development Kit (SDK).
22221 This host system is used to download the configuration bitstream to
22222 the target FPGA. The Xilinx Microprocessor Debugger (XMD) program
22223 communicates with the target board using the JTAG interface and
22224 presents a @code{gdbserver} interface to the board. By default
22225 @code{xmd} uses port @code{1234}. (While it is possible to change
22226 this default port, it requires the use of undocumented @code{xmd}
22227 commands. Contact Xilinx support if you need to do this.)
22229 Use these GDB commands to connect to the MicroBlaze target processor.
22232 @item target remote :1234
22233 Use this command to connect to the target if you are running @value{GDBN}
22234 on the same system as @code{xmd}.
22236 @item target remote @var{xmd-host}:1234
22237 Use this command to connect to the target if it is connected to @code{xmd}
22238 running on a different system named @var{xmd-host}.
22241 Use this command to download a program to the MicroBlaze target.
22243 @item set debug microblaze @var{n}
22244 Enable MicroBlaze-specific debugging messages if non-zero.
22246 @item show debug microblaze @var{n}
22247 Show MicroBlaze-specific debugging level.
22250 @node MIPS Embedded
22251 @subsection @acronym{MIPS} Embedded
22254 @value{GDBN} supports these special commands for @acronym{MIPS} targets:
22257 @item set mipsfpu double
22258 @itemx set mipsfpu single
22259 @itemx set mipsfpu none
22260 @itemx set mipsfpu auto
22261 @itemx show mipsfpu
22262 @kindex set mipsfpu
22263 @kindex show mipsfpu
22264 @cindex @acronym{MIPS} remote floating point
22265 @cindex floating point, @acronym{MIPS} remote
22266 If your target board does not support the @acronym{MIPS} floating point
22267 coprocessor, you should use the command @samp{set mipsfpu none} (if you
22268 need this, you may wish to put the command in your @value{GDBN} init
22269 file). This tells @value{GDBN} how to find the return value of
22270 functions which return floating point values. It also allows
22271 @value{GDBN} to avoid saving the floating point registers when calling
22272 functions on the board. If you are using a floating point coprocessor
22273 with only single precision floating point support, as on the @sc{r4650}
22274 processor, use the command @samp{set mipsfpu single}. The default
22275 double precision floating point coprocessor may be selected using
22276 @samp{set mipsfpu double}.
22278 In previous versions the only choices were double precision or no
22279 floating point, so @samp{set mipsfpu on} will select double precision
22280 and @samp{set mipsfpu off} will select no floating point.
22282 As usual, you can inquire about the @code{mipsfpu} variable with
22283 @samp{show mipsfpu}.
22286 @node PowerPC Embedded
22287 @subsection PowerPC Embedded
22289 @cindex DVC register
22290 @value{GDBN} supports using the DVC (Data Value Compare) register to
22291 implement in hardware simple hardware watchpoint conditions of the form:
22294 (@value{GDBP}) watch @var{ADDRESS|VARIABLE} \
22295 if @var{ADDRESS|VARIABLE} == @var{CONSTANT EXPRESSION}
22298 The DVC register will be automatically used when @value{GDBN} detects
22299 such pattern in a condition expression, and the created watchpoint uses one
22300 debug register (either the @code{exact-watchpoints} option is on and the
22301 variable is scalar, or the variable has a length of one byte). This feature
22302 is available in native @value{GDBN} running on a Linux kernel version 2.6.34
22305 When running on PowerPC embedded processors, @value{GDBN} automatically uses
22306 ranged hardware watchpoints, unless the @code{exact-watchpoints} option is on,
22307 in which case watchpoints using only one debug register are created when
22308 watching variables of scalar types.
22310 You can create an artificial array to watch an arbitrary memory
22311 region using one of the following commands (@pxref{Expressions}):
22314 (@value{GDBP}) watch *((char *) @var{address})@@@var{length}
22315 (@value{GDBP}) watch @{char[@var{length}]@} @var{address}
22318 PowerPC embedded processors support masked watchpoints. See the discussion
22319 about the @code{mask} argument in @ref{Set Watchpoints}.
22321 @cindex ranged breakpoint
22322 PowerPC embedded processors support hardware accelerated
22323 @dfn{ranged breakpoints}. A ranged breakpoint stops execution of
22324 the inferior whenever it executes an instruction at any address within
22325 the range it specifies. To set a ranged breakpoint in @value{GDBN},
22326 use the @code{break-range} command.
22328 @value{GDBN} provides the following PowerPC-specific commands:
22331 @kindex break-range
22332 @item break-range @var{start-location}, @var{end-location}
22333 Set a breakpoint for an address range given by
22334 @var{start-location} and @var{end-location}, which can specify a function name,
22335 a line number, an offset of lines from the current line or from the start
22336 location, or an address of an instruction (see @ref{Specify Location},
22337 for a list of all the possible ways to specify a @var{location}.)
22338 The breakpoint will stop execution of the inferior whenever it
22339 executes an instruction at any address within the specified range,
22340 (including @var{start-location} and @var{end-location}.)
22342 @kindex set powerpc
22343 @item set powerpc soft-float
22344 @itemx show powerpc soft-float
22345 Force @value{GDBN} to use (or not use) a software floating point calling
22346 convention. By default, @value{GDBN} selects the calling convention based
22347 on the selected architecture and the provided executable file.
22349 @item set powerpc vector-abi
22350 @itemx show powerpc vector-abi
22351 Force @value{GDBN} to use the specified calling convention for vector
22352 arguments and return values. The valid options are @samp{auto};
22353 @samp{generic}, to avoid vector registers even if they are present;
22354 @samp{altivec}, to use AltiVec registers; and @samp{spe} to use SPE
22355 registers. By default, @value{GDBN} selects the calling convention
22356 based on the selected architecture and the provided executable file.
22358 @item set powerpc exact-watchpoints
22359 @itemx show powerpc exact-watchpoints
22360 Allow @value{GDBN} to use only one debug register when watching a variable
22361 of scalar type, thus assuming that the variable is accessed through the
22362 address of its first byte.
22367 @subsection Atmel AVR
22370 When configured for debugging the Atmel AVR, @value{GDBN} supports the
22371 following AVR-specific commands:
22374 @item info io_registers
22375 @kindex info io_registers@r{, AVR}
22376 @cindex I/O registers (Atmel AVR)
22377 This command displays information about the AVR I/O registers. For
22378 each register, @value{GDBN} prints its number and value.
22385 When configured for debugging CRIS, @value{GDBN} provides the
22386 following CRIS-specific commands:
22389 @item set cris-version @var{ver}
22390 @cindex CRIS version
22391 Set the current CRIS version to @var{ver}, either @samp{10} or @samp{32}.
22392 The CRIS version affects register names and sizes. This command is useful in
22393 case autodetection of the CRIS version fails.
22395 @item show cris-version
22396 Show the current CRIS version.
22398 @item set cris-dwarf2-cfi
22399 @cindex DWARF-2 CFI and CRIS
22400 Set the usage of DWARF-2 CFI for CRIS debugging. The default is @samp{on}.
22401 Change to @samp{off} when using @code{gcc-cris} whose version is below
22404 @item show cris-dwarf2-cfi
22405 Show the current state of using DWARF-2 CFI.
22407 @item set cris-mode @var{mode}
22409 Set the current CRIS mode to @var{mode}. It should only be changed when
22410 debugging in guru mode, in which case it should be set to
22411 @samp{guru} (the default is @samp{normal}).
22413 @item show cris-mode
22414 Show the current CRIS mode.
22418 @subsection Renesas Super-H
22421 For the Renesas Super-H processor, @value{GDBN} provides these
22425 @item set sh calling-convention @var{convention}
22426 @kindex set sh calling-convention
22427 Set the calling-convention used when calling functions from @value{GDBN}.
22428 Allowed values are @samp{gcc}, which is the default setting, and @samp{renesas}.
22429 With the @samp{gcc} setting, functions are called using the @value{NGCC} calling
22430 convention. If the DWARF-2 information of the called function specifies
22431 that the function follows the Renesas calling convention, the function
22432 is called using the Renesas calling convention. If the calling convention
22433 is set to @samp{renesas}, the Renesas calling convention is always used,
22434 regardless of the DWARF-2 information. This can be used to override the
22435 default of @samp{gcc} if debug information is missing, or the compiler
22436 does not emit the DWARF-2 calling convention entry for a function.
22438 @item show sh calling-convention
22439 @kindex show sh calling-convention
22440 Show the current calling convention setting.
22445 @node Architectures
22446 @section Architectures
22448 This section describes characteristics of architectures that affect
22449 all uses of @value{GDBN} with the architecture, both native and cross.
22456 * HPPA:: HP PA architecture
22457 * SPU:: Cell Broadband Engine SPU architecture
22463 @subsection AArch64
22464 @cindex AArch64 support
22466 When @value{GDBN} is debugging the AArch64 architecture, it provides the
22467 following special commands:
22470 @item set debug aarch64
22471 @kindex set debug aarch64
22472 This command determines whether AArch64 architecture-specific debugging
22473 messages are to be displayed.
22475 @item show debug aarch64
22476 Show whether AArch64 debugging messages are displayed.
22481 @subsection x86 Architecture-specific Issues
22484 @item set struct-convention @var{mode}
22485 @kindex set struct-convention
22486 @cindex struct return convention
22487 @cindex struct/union returned in registers
22488 Set the convention used by the inferior to return @code{struct}s and
22489 @code{union}s from functions to @var{mode}. Possible values of
22490 @var{mode} are @code{"pcc"}, @code{"reg"}, and @code{"default"} (the
22491 default). @code{"default"} or @code{"pcc"} means that @code{struct}s
22492 are returned on the stack, while @code{"reg"} means that a
22493 @code{struct} or a @code{union} whose size is 1, 2, 4, or 8 bytes will
22494 be returned in a register.
22496 @item show struct-convention
22497 @kindex show struct-convention
22498 Show the current setting of the convention to return @code{struct}s
22503 @subsubsection Intel @dfn{Memory Protection Extensions} (MPX).
22504 @cindex Intel Memory Protection Extensions (MPX).
22506 Memory Protection Extension (MPX) adds the bound registers @samp{BND0}
22507 @footnote{The register named with capital letters represent the architecture
22508 registers.} through @samp{BND3}. Bound registers store a pair of 64-bit values
22509 which are the lower bound and upper bound. Bounds are effective addresses or
22510 memory locations. The upper bounds are architecturally represented in 1's
22511 complement form. A bound having lower bound = 0, and upper bound = 0
22512 (1's complement of all bits set) will allow access to the entire address space.
22514 @samp{BND0} through @samp{BND3} are represented in @value{GDBN} as @samp{bnd0raw}
22515 through @samp{bnd3raw}. Pseudo registers @samp{bnd0} through @samp{bnd3}
22516 display the upper bound performing the complement of one operation on the
22517 upper bound value, i.e.@ when upper bound in @samp{bnd0raw} is 0 in the
22518 @value{GDBN} @samp{bnd0} it will be @code{0xfff@dots{}}. In this sense it
22519 can also be noted that the upper bounds are inclusive.
22521 As an example, assume that the register BND0 holds bounds for a pointer having
22522 access allowed for the range between 0x32 and 0x71. The values present on
22523 bnd0raw and bnd registers are presented as follows:
22526 bnd0raw = @{0x32, 0xffffffff8e@}
22527 bnd0 = @{lbound = 0x32, ubound = 0x71@} : size 64
22530 This way the raw value can be accessed via bnd0raw@dots{}bnd3raw. Any
22531 change on bnd0@dots{}bnd3 or bnd0raw@dots{}bnd3raw is reflect on its
22532 counterpart. When the bnd0@dots{}bnd3 registers are displayed via
22533 Python, the display includes the memory size, in bits, accessible to
22536 Bounds can also be stored in bounds tables, which are stored in
22537 application memory. These tables store bounds for pointers by specifying
22538 the bounds pointer's value along with its bounds. Evaluating and changing
22539 bounds located in bound tables is therefore interesting while investigating
22540 bugs on MPX context. @value{GDBN} provides commands for this purpose:
22543 @item show mpx bound @var{pointer}
22544 @kindex show mpx bound
22545 Display bounds of the given @var{pointer}.
22547 @item set mpx bound @var{pointer}, @var{lbound}, @var{ubound}
22548 @kindex set mpx bound
22549 Set the bounds of a pointer in the bound table.
22550 This command takes three parameters: @var{pointer} is the pointers
22551 whose bounds are to be changed, @var{lbound} and @var{ubound} are new values
22552 for lower and upper bounds respectively.
22555 When you call an inferior function on an Intel MPX enabled program,
22556 GDB sets the inferior's bound registers to the init (disabled) state
22557 before calling the function. As a consequence, bounds checks for the
22558 pointer arguments passed to the function will always pass.
22560 This is necessary because when you call an inferior function, the
22561 program is usually in the middle of the execution of other function.
22562 Since at that point bound registers are in an arbitrary state, not
22563 clearing them would lead to random bound violations in the called
22566 You can still examine the influence of the bound registers on the
22567 execution of the called function by stopping the execution of the
22568 called function at its prologue, setting bound registers, and
22569 continuing the execution. For example:
22573 Breakpoint 2 at 0x4009de: file i386-mpx-call.c, line 47.
22574 $ print upper (a, b, c, d, 1)
22575 Breakpoint 2, upper (a=0x0, b=0x6e0000005b, c=0x0, d=0x0, len=48)....
22577 @{lbound = 0x0, ubound = ffffffff@} : size -1
22580 At this last step the value of bnd0 can be changed for investigation of bound
22581 violations caused along the execution of the call. In order to know how to
22582 set the bound registers or bound table for the call consult the ABI.
22587 See the following section.
22590 @subsection @acronym{MIPS}
22592 @cindex stack on Alpha
22593 @cindex stack on @acronym{MIPS}
22594 @cindex Alpha stack
22595 @cindex @acronym{MIPS} stack
22596 Alpha- and @acronym{MIPS}-based computers use an unusual stack frame, which
22597 sometimes requires @value{GDBN} to search backward in the object code to
22598 find the beginning of a function.
22600 @cindex response time, @acronym{MIPS} debugging
22601 To improve response time (especially for embedded applications, where
22602 @value{GDBN} may be restricted to a slow serial line for this search)
22603 you may want to limit the size of this search, using one of these
22607 @cindex @code{heuristic-fence-post} (Alpha, @acronym{MIPS})
22608 @item set heuristic-fence-post @var{limit}
22609 Restrict @value{GDBN} to examining at most @var{limit} bytes in its
22610 search for the beginning of a function. A value of @var{0} (the
22611 default) means there is no limit. However, except for @var{0}, the
22612 larger the limit the more bytes @code{heuristic-fence-post} must search
22613 and therefore the longer it takes to run. You should only need to use
22614 this command when debugging a stripped executable.
22616 @item show heuristic-fence-post
22617 Display the current limit.
22621 These commands are available @emph{only} when @value{GDBN} is configured
22622 for debugging programs on Alpha or @acronym{MIPS} processors.
22624 Several @acronym{MIPS}-specific commands are available when debugging @acronym{MIPS}
22628 @item set mips abi @var{arg}
22629 @kindex set mips abi
22630 @cindex set ABI for @acronym{MIPS}
22631 Tell @value{GDBN} which @acronym{MIPS} ABI is used by the inferior. Possible
22632 values of @var{arg} are:
22636 The default ABI associated with the current binary (this is the
22646 @item show mips abi
22647 @kindex show mips abi
22648 Show the @acronym{MIPS} ABI used by @value{GDBN} to debug the inferior.
22650 @item set mips compression @var{arg}
22651 @kindex set mips compression
22652 @cindex code compression, @acronym{MIPS}
22653 Tell @value{GDBN} which @acronym{MIPS} compressed
22654 @acronym{ISA, Instruction Set Architecture} encoding is used by the
22655 inferior. @value{GDBN} uses this for code disassembly and other
22656 internal interpretation purposes. This setting is only referred to
22657 when no executable has been associated with the debugging session or
22658 the executable does not provide information about the encoding it uses.
22659 Otherwise this setting is automatically updated from information
22660 provided by the executable.
22662 Possible values of @var{arg} are @samp{mips16} and @samp{micromips}.
22663 The default compressed @acronym{ISA} encoding is @samp{mips16}, as
22664 executables containing @acronym{MIPS16} code frequently are not
22665 identified as such.
22667 This setting is ``sticky''; that is, it retains its value across
22668 debugging sessions until reset either explicitly with this command or
22669 implicitly from an executable.
22671 The compiler and/or assembler typically add symbol table annotations to
22672 identify functions compiled for the @acronym{MIPS16} or
22673 @acronym{microMIPS} @acronym{ISA}s. If these function-scope annotations
22674 are present, @value{GDBN} uses them in preference to the global
22675 compressed @acronym{ISA} encoding setting.
22677 @item show mips compression
22678 @kindex show mips compression
22679 Show the @acronym{MIPS} compressed @acronym{ISA} encoding used by
22680 @value{GDBN} to debug the inferior.
22683 @itemx show mipsfpu
22684 @xref{MIPS Embedded, set mipsfpu}.
22686 @item set mips mask-address @var{arg}
22687 @kindex set mips mask-address
22688 @cindex @acronym{MIPS} addresses, masking
22689 This command determines whether the most-significant 32 bits of 64-bit
22690 @acronym{MIPS} addresses are masked off. The argument @var{arg} can be
22691 @samp{on}, @samp{off}, or @samp{auto}. The latter is the default
22692 setting, which lets @value{GDBN} determine the correct value.
22694 @item show mips mask-address
22695 @kindex show mips mask-address
22696 Show whether the upper 32 bits of @acronym{MIPS} addresses are masked off or
22699 @item set remote-mips64-transfers-32bit-regs
22700 @kindex set remote-mips64-transfers-32bit-regs
22701 This command controls compatibility with 64-bit @acronym{MIPS} targets that
22702 transfer data in 32-bit quantities. If you have an old @acronym{MIPS} 64 target
22703 that transfers 32 bits for some registers, like @sc{sr} and @sc{fsr},
22704 and 64 bits for other registers, set this option to @samp{on}.
22706 @item show remote-mips64-transfers-32bit-regs
22707 @kindex show remote-mips64-transfers-32bit-regs
22708 Show the current setting of compatibility with older @acronym{MIPS} 64 targets.
22710 @item set debug mips
22711 @kindex set debug mips
22712 This command turns on and off debugging messages for the @acronym{MIPS}-specific
22713 target code in @value{GDBN}.
22715 @item show debug mips
22716 @kindex show debug mips
22717 Show the current setting of @acronym{MIPS} debugging messages.
22723 @cindex HPPA support
22725 When @value{GDBN} is debugging the HP PA architecture, it provides the
22726 following special commands:
22729 @item set debug hppa
22730 @kindex set debug hppa
22731 This command determines whether HPPA architecture-specific debugging
22732 messages are to be displayed.
22734 @item show debug hppa
22735 Show whether HPPA debugging messages are displayed.
22737 @item maint print unwind @var{address}
22738 @kindex maint print unwind@r{, HPPA}
22739 This command displays the contents of the unwind table entry at the
22740 given @var{address}.
22746 @subsection Cell Broadband Engine SPU architecture
22747 @cindex Cell Broadband Engine
22750 When @value{GDBN} is debugging the Cell Broadband Engine SPU architecture,
22751 it provides the following special commands:
22754 @item info spu event
22756 Display SPU event facility status. Shows current event mask
22757 and pending event status.
22759 @item info spu signal
22760 Display SPU signal notification facility status. Shows pending
22761 signal-control word and signal notification mode of both signal
22762 notification channels.
22764 @item info spu mailbox
22765 Display SPU mailbox facility status. Shows all pending entries,
22766 in order of processing, in each of the SPU Write Outbound,
22767 SPU Write Outbound Interrupt, and SPU Read Inbound mailboxes.
22770 Display MFC DMA status. Shows all pending commands in the MFC
22771 DMA queue. For each entry, opcode, tag, class IDs, effective
22772 and local store addresses and transfer size are shown.
22774 @item info spu proxydma
22775 Display MFC Proxy-DMA status. Shows all pending commands in the MFC
22776 Proxy-DMA queue. For each entry, opcode, tag, class IDs, effective
22777 and local store addresses and transfer size are shown.
22781 When @value{GDBN} is debugging a combined PowerPC/SPU application
22782 on the Cell Broadband Engine, it provides in addition the following
22786 @item set spu stop-on-load @var{arg}
22788 Set whether to stop for new SPE threads. When set to @code{on}, @value{GDBN}
22789 will give control to the user when a new SPE thread enters its @code{main}
22790 function. The default is @code{off}.
22792 @item show spu stop-on-load
22794 Show whether to stop for new SPE threads.
22796 @item set spu auto-flush-cache @var{arg}
22797 Set whether to automatically flush the software-managed cache. When set to
22798 @code{on}, @value{GDBN} will automatically cause the SPE software-managed
22799 cache to be flushed whenever SPE execution stops. This provides a consistent
22800 view of PowerPC memory that is accessed via the cache. If an application
22801 does not use the software-managed cache, this option has no effect.
22803 @item show spu auto-flush-cache
22804 Show whether to automatically flush the software-managed cache.
22809 @subsection PowerPC
22810 @cindex PowerPC architecture
22812 When @value{GDBN} is debugging the PowerPC architecture, it provides a set of
22813 pseudo-registers to enable inspection of 128-bit wide Decimal Floating Point
22814 numbers stored in the floating point registers. These values must be stored
22815 in two consecutive registers, always starting at an even register like
22816 @code{f0} or @code{f2}.
22818 The pseudo-registers go from @code{$dl0} through @code{$dl15}, and are formed
22819 by joining the even/odd register pairs @code{f0} and @code{f1} for @code{$dl0},
22820 @code{f2} and @code{f3} for @code{$dl1} and so on.
22822 For POWER7 processors, @value{GDBN} provides a set of pseudo-registers, the 64-bit
22823 wide Extended Floating Point Registers (@samp{f32} through @samp{f63}).
22826 @subsection Nios II
22827 @cindex Nios II architecture
22829 When @value{GDBN} is debugging the Nios II architecture,
22830 it provides the following special commands:
22834 @item set debug nios2
22835 @kindex set debug nios2
22836 This command turns on and off debugging messages for the Nios II
22837 target code in @value{GDBN}.
22839 @item show debug nios2
22840 @kindex show debug nios2
22841 Show the current setting of Nios II debugging messages.
22844 @node Controlling GDB
22845 @chapter Controlling @value{GDBN}
22847 You can alter the way @value{GDBN} interacts with you by using the
22848 @code{set} command. For commands controlling how @value{GDBN} displays
22849 data, see @ref{Print Settings, ,Print Settings}. Other settings are
22854 * Editing:: Command editing
22855 * Command History:: Command history
22856 * Screen Size:: Screen size
22857 * Numbers:: Numbers
22858 * ABI:: Configuring the current ABI
22859 * Auto-loading:: Automatically loading associated files
22860 * Messages/Warnings:: Optional warnings and messages
22861 * Debugging Output:: Optional messages about internal happenings
22862 * Other Misc Settings:: Other Miscellaneous Settings
22870 @value{GDBN} indicates its readiness to read a command by printing a string
22871 called the @dfn{prompt}. This string is normally @samp{(@value{GDBP})}. You
22872 can change the prompt string with the @code{set prompt} command. For
22873 instance, when debugging @value{GDBN} with @value{GDBN}, it is useful to change
22874 the prompt in one of the @value{GDBN} sessions so that you can always tell
22875 which one you are talking to.
22877 @emph{Note:} @code{set prompt} does not add a space for you after the
22878 prompt you set. This allows you to set a prompt which ends in a space
22879 or a prompt that does not.
22883 @item set prompt @var{newprompt}
22884 Directs @value{GDBN} to use @var{newprompt} as its prompt string henceforth.
22886 @kindex show prompt
22888 Prints a line of the form: @samp{Gdb's prompt is: @var{your-prompt}}
22891 Versions of @value{GDBN} that ship with Python scripting enabled have
22892 prompt extensions. The commands for interacting with these extensions
22896 @kindex set extended-prompt
22897 @item set extended-prompt @var{prompt}
22898 Set an extended prompt that allows for substitutions.
22899 @xref{gdb.prompt}, for a list of escape sequences that can be used for
22900 substitution. Any escape sequences specified as part of the prompt
22901 string are replaced with the corresponding strings each time the prompt
22907 set extended-prompt Current working directory: \w (gdb)
22910 Note that when an extended-prompt is set, it takes control of the
22911 @var{prompt_hook} hook. @xref{prompt_hook}, for further information.
22913 @kindex show extended-prompt
22914 @item show extended-prompt
22915 Prints the extended prompt. Any escape sequences specified as part of
22916 the prompt string with @code{set extended-prompt}, are replaced with the
22917 corresponding strings each time the prompt is displayed.
22921 @section Command Editing
22923 @cindex command line editing
22925 @value{GDBN} reads its input commands via the @dfn{Readline} interface. This
22926 @sc{gnu} library provides consistent behavior for programs which provide a
22927 command line interface to the user. Advantages are @sc{gnu} Emacs-style
22928 or @dfn{vi}-style inline editing of commands, @code{csh}-like history
22929 substitution, and a storage and recall of command history across
22930 debugging sessions.
22932 You may control the behavior of command line editing in @value{GDBN} with the
22933 command @code{set}.
22936 @kindex set editing
22939 @itemx set editing on
22940 Enable command line editing (enabled by default).
22942 @item set editing off
22943 Disable command line editing.
22945 @kindex show editing
22947 Show whether command line editing is enabled.
22950 @ifset SYSTEM_READLINE
22951 @xref{Command Line Editing, , , rluserman, GNU Readline Library},
22953 @ifclear SYSTEM_READLINE
22954 @xref{Command Line Editing},
22956 for more details about the Readline
22957 interface. Users unfamiliar with @sc{gnu} Emacs or @code{vi} are
22958 encouraged to read that chapter.
22960 @node Command History
22961 @section Command History
22962 @cindex command history
22964 @value{GDBN} can keep track of the commands you type during your
22965 debugging sessions, so that you can be certain of precisely what
22966 happened. Use these commands to manage the @value{GDBN} command
22969 @value{GDBN} uses the @sc{gnu} History library, a part of the Readline
22970 package, to provide the history facility.
22971 @ifset SYSTEM_READLINE
22972 @xref{Using History Interactively, , , history, GNU History Library},
22974 @ifclear SYSTEM_READLINE
22975 @xref{Using History Interactively},
22977 for the detailed description of the History library.
22979 To issue a command to @value{GDBN} without affecting certain aspects of
22980 the state which is seen by users, prefix it with @samp{server }
22981 (@pxref{Server Prefix}). This
22982 means that this command will not affect the command history, nor will it
22983 affect @value{GDBN}'s notion of which command to repeat if @key{RET} is
22984 pressed on a line by itself.
22986 @cindex @code{server}, command prefix
22987 The server prefix does not affect the recording of values into the value
22988 history; to print a value without recording it into the value history,
22989 use the @code{output} command instead of the @code{print} command.
22991 Here is the description of @value{GDBN} commands related to command
22995 @cindex history substitution
22996 @cindex history file
22997 @kindex set history filename
22998 @cindex @env{GDBHISTFILE}, environment variable
22999 @item set history filename @var{fname}
23000 Set the name of the @value{GDBN} command history file to @var{fname}.
23001 This is the file where @value{GDBN} reads an initial command history
23002 list, and where it writes the command history from this session when it
23003 exits. You can access this list through history expansion or through
23004 the history command editing characters listed below. This file defaults
23005 to the value of the environment variable @code{GDBHISTFILE}, or to
23006 @file{./.gdb_history} (@file{./_gdb_history} on MS-DOS) if this variable
23009 @cindex save command history
23010 @kindex set history save
23011 @item set history save
23012 @itemx set history save on
23013 Record command history in a file, whose name may be specified with the
23014 @code{set history filename} command. By default, this option is disabled.
23016 @item set history save off
23017 Stop recording command history in a file.
23019 @cindex history size
23020 @kindex set history size
23021 @cindex @env{GDBHISTSIZE}, environment variable
23022 @item set history size @var{size}
23023 @itemx set history size unlimited
23024 Set the number of commands which @value{GDBN} keeps in its history list.
23025 This defaults to the value of the environment variable @env{GDBHISTSIZE}, or
23026 to 256 if this variable is not set. Non-numeric values of @env{GDBHISTSIZE}
23027 are ignored. If @var{size} is @code{unlimited} or if @env{GDBHISTSIZE} is
23028 either a negative number or the empty string, then the number of commands
23029 @value{GDBN} keeps in the history list is unlimited.
23031 @cindex remove duplicate history
23032 @kindex set history remove-duplicates
23033 @item set history remove-duplicates @var{count}
23034 @itemx set history remove-duplicates unlimited
23035 Control the removal of duplicate history entries in the command history list.
23036 If @var{count} is non-zero, @value{GDBN} will look back at the last @var{count}
23037 history entries and remove the first entry that is a duplicate of the current
23038 entry being added to the command history list. If @var{count} is
23039 @code{unlimited} then this lookbehind is unbounded. If @var{count} is 0, then
23040 removal of duplicate history entries is disabled.
23042 Only history entries added during the current session are considered for
23043 removal. This option is set to 0 by default.
23047 History expansion assigns special meaning to the character @kbd{!}.
23048 @ifset SYSTEM_READLINE
23049 @xref{Event Designators, , , history, GNU History Library},
23051 @ifclear SYSTEM_READLINE
23052 @xref{Event Designators},
23056 @cindex history expansion, turn on/off
23057 Since @kbd{!} is also the logical not operator in C, history expansion
23058 is off by default. If you decide to enable history expansion with the
23059 @code{set history expansion on} command, you may sometimes need to
23060 follow @kbd{!} (when it is used as logical not, in an expression) with
23061 a space or a tab to prevent it from being expanded. The readline
23062 history facilities do not attempt substitution on the strings
23063 @kbd{!=} and @kbd{!(}, even when history expansion is enabled.
23065 The commands to control history expansion are:
23068 @item set history expansion on
23069 @itemx set history expansion
23070 @kindex set history expansion
23071 Enable history expansion. History expansion is off by default.
23073 @item set history expansion off
23074 Disable history expansion.
23077 @kindex show history
23079 @itemx show history filename
23080 @itemx show history save
23081 @itemx show history size
23082 @itemx show history expansion
23083 These commands display the state of the @value{GDBN} history parameters.
23084 @code{show history} by itself displays all four states.
23089 @kindex show commands
23090 @cindex show last commands
23091 @cindex display command history
23092 @item show commands
23093 Display the last ten commands in the command history.
23095 @item show commands @var{n}
23096 Print ten commands centered on command number @var{n}.
23098 @item show commands +
23099 Print ten commands just after the commands last printed.
23103 @section Screen Size
23104 @cindex size of screen
23105 @cindex screen size
23108 @cindex pauses in output
23110 Certain commands to @value{GDBN} may produce large amounts of
23111 information output to the screen. To help you read all of it,
23112 @value{GDBN} pauses and asks you for input at the end of each page of
23113 output. Type @key{RET} when you want to continue the output, or @kbd{q}
23114 to discard the remaining output. Also, the screen width setting
23115 determines when to wrap lines of output. Depending on what is being
23116 printed, @value{GDBN} tries to break the line at a readable place,
23117 rather than simply letting it overflow onto the following line.
23119 Normally @value{GDBN} knows the size of the screen from the terminal
23120 driver software. For example, on Unix @value{GDBN} uses the termcap data base
23121 together with the value of the @code{TERM} environment variable and the
23122 @code{stty rows} and @code{stty cols} settings. If this is not correct,
23123 you can override it with the @code{set height} and @code{set
23130 @kindex show height
23131 @item set height @var{lpp}
23132 @itemx set height unlimited
23134 @itemx set width @var{cpl}
23135 @itemx set width unlimited
23137 These @code{set} commands specify a screen height of @var{lpp} lines and
23138 a screen width of @var{cpl} characters. The associated @code{show}
23139 commands display the current settings.
23141 If you specify a height of either @code{unlimited} or zero lines,
23142 @value{GDBN} does not pause during output no matter how long the
23143 output is. This is useful if output is to a file or to an editor
23146 Likewise, you can specify @samp{set width unlimited} or @samp{set
23147 width 0} to prevent @value{GDBN} from wrapping its output.
23149 @item set pagination on
23150 @itemx set pagination off
23151 @kindex set pagination
23152 Turn the output pagination on or off; the default is on. Turning
23153 pagination off is the alternative to @code{set height unlimited}. Note that
23154 running @value{GDBN} with the @option{--batch} option (@pxref{Mode
23155 Options, -batch}) also automatically disables pagination.
23157 @item show pagination
23158 @kindex show pagination
23159 Show the current pagination mode.
23164 @cindex number representation
23165 @cindex entering numbers
23167 You can always enter numbers in octal, decimal, or hexadecimal in
23168 @value{GDBN} by the usual conventions: octal numbers begin with
23169 @samp{0}, decimal numbers end with @samp{.}, and hexadecimal numbers
23170 begin with @samp{0x}. Numbers that neither begin with @samp{0} or
23171 @samp{0x}, nor end with a @samp{.} are, by default, entered in base
23172 10; likewise, the default display for numbers---when no particular
23173 format is specified---is base 10. You can change the default base for
23174 both input and output with the commands described below.
23177 @kindex set input-radix
23178 @item set input-radix @var{base}
23179 Set the default base for numeric input. Supported choices
23180 for @var{base} are decimal 8, 10, or 16. The base must itself be
23181 specified either unambiguously or using the current input radix; for
23185 set input-radix 012
23186 set input-radix 10.
23187 set input-radix 0xa
23191 sets the input base to decimal. On the other hand, @samp{set input-radix 10}
23192 leaves the input radix unchanged, no matter what it was, since
23193 @samp{10}, being without any leading or trailing signs of its base, is
23194 interpreted in the current radix. Thus, if the current radix is 16,
23195 @samp{10} is interpreted in hex, i.e.@: as 16 decimal, which doesn't
23198 @kindex set output-radix
23199 @item set output-radix @var{base}
23200 Set the default base for numeric display. Supported choices
23201 for @var{base} are decimal 8, 10, or 16. The base must itself be
23202 specified either unambiguously or using the current input radix.
23204 @kindex show input-radix
23205 @item show input-radix
23206 Display the current default base for numeric input.
23208 @kindex show output-radix
23209 @item show output-radix
23210 Display the current default base for numeric display.
23212 @item set radix @r{[}@var{base}@r{]}
23216 These commands set and show the default base for both input and output
23217 of numbers. @code{set radix} sets the radix of input and output to
23218 the same base; without an argument, it resets the radix back to its
23219 default value of 10.
23224 @section Configuring the Current ABI
23226 @value{GDBN} can determine the @dfn{ABI} (Application Binary Interface) of your
23227 application automatically. However, sometimes you need to override its
23228 conclusions. Use these commands to manage @value{GDBN}'s view of the
23234 @cindex Newlib OS ABI and its influence on the longjmp handling
23236 One @value{GDBN} configuration can debug binaries for multiple operating
23237 system targets, either via remote debugging or native emulation.
23238 @value{GDBN} will autodetect the @dfn{OS ABI} (Operating System ABI) in use,
23239 but you can override its conclusion using the @code{set osabi} command.
23240 One example where this is useful is in debugging of binaries which use
23241 an alternate C library (e.g.@: @sc{uClibc} for @sc{gnu}/Linux) which does
23242 not have the same identifying marks that the standard C library for your
23245 When @value{GDBN} is debugging the AArch64 architecture, it provides a
23246 ``Newlib'' OS ABI. This is useful for handling @code{setjmp} and
23247 @code{longjmp} when debugging binaries that use the @sc{newlib} C library.
23248 The ``Newlib'' OS ABI can be selected by @code{set osabi Newlib}.
23252 Show the OS ABI currently in use.
23255 With no argument, show the list of registered available OS ABI's.
23257 @item set osabi @var{abi}
23258 Set the current OS ABI to @var{abi}.
23261 @cindex float promotion
23263 Generally, the way that an argument of type @code{float} is passed to a
23264 function depends on whether the function is prototyped. For a prototyped
23265 (i.e.@: ANSI/ISO style) function, @code{float} arguments are passed unchanged,
23266 according to the architecture's convention for @code{float}. For unprototyped
23267 (i.e.@: K&R style) functions, @code{float} arguments are first promoted to type
23268 @code{double} and then passed.
23270 Unfortunately, some forms of debug information do not reliably indicate whether
23271 a function is prototyped. If @value{GDBN} calls a function that is not marked
23272 as prototyped, it consults @kbd{set coerce-float-to-double}.
23275 @kindex set coerce-float-to-double
23276 @item set coerce-float-to-double
23277 @itemx set coerce-float-to-double on
23278 Arguments of type @code{float} will be promoted to @code{double} when passed
23279 to an unprototyped function. This is the default setting.
23281 @item set coerce-float-to-double off
23282 Arguments of type @code{float} will be passed directly to unprototyped
23285 @kindex show coerce-float-to-double
23286 @item show coerce-float-to-double
23287 Show the current setting of promoting @code{float} to @code{double}.
23291 @kindex show cp-abi
23292 @value{GDBN} needs to know the ABI used for your program's C@t{++}
23293 objects. The correct C@t{++} ABI depends on which C@t{++} compiler was
23294 used to build your application. @value{GDBN} only fully supports
23295 programs with a single C@t{++} ABI; if your program contains code using
23296 multiple C@t{++} ABI's or if @value{GDBN} can not identify your
23297 program's ABI correctly, you can tell @value{GDBN} which ABI to use.
23298 Currently supported ABI's include ``gnu-v2'', for @code{g++} versions
23299 before 3.0, ``gnu-v3'', for @code{g++} versions 3.0 and later, and
23300 ``hpaCC'' for the HP ANSI C@t{++} compiler. Other C@t{++} compilers may
23301 use the ``gnu-v2'' or ``gnu-v3'' ABI's as well. The default setting is
23306 Show the C@t{++} ABI currently in use.
23309 With no argument, show the list of supported C@t{++} ABI's.
23311 @item set cp-abi @var{abi}
23312 @itemx set cp-abi auto
23313 Set the current C@t{++} ABI to @var{abi}, or return to automatic detection.
23317 @section Automatically loading associated files
23318 @cindex auto-loading
23320 @value{GDBN} sometimes reads files with commands and settings automatically,
23321 without being explicitly told so by the user. We call this feature
23322 @dfn{auto-loading}. While auto-loading is useful for automatically adapting
23323 @value{GDBN} to the needs of your project, it can sometimes produce unexpected
23324 results or introduce security risks (e.g., if the file comes from untrusted
23328 * Init File in the Current Directory:: @samp{set/show/info auto-load local-gdbinit}
23329 * libthread_db.so.1 file:: @samp{set/show/info auto-load libthread-db}
23331 * Auto-loading safe path:: @samp{set/show/info auto-load safe-path}
23332 * Auto-loading verbose mode:: @samp{set/show debug auto-load}
23335 There are various kinds of files @value{GDBN} can automatically load.
23336 In addition to these files, @value{GDBN} supports auto-loading code written
23337 in various extension languages. @xref{Auto-loading extensions}.
23339 Note that loading of these associated files (including the local @file{.gdbinit}
23340 file) requires accordingly configured @code{auto-load safe-path}
23341 (@pxref{Auto-loading safe path}).
23343 For these reasons, @value{GDBN} includes commands and options to let you
23344 control when to auto-load files and which files should be auto-loaded.
23347 @anchor{set auto-load off}
23348 @kindex set auto-load off
23349 @item set auto-load off
23350 Globally disable loading of all auto-loaded files.
23351 You may want to use this command with the @samp{-iex} option
23352 (@pxref{Option -init-eval-command}) such as:
23354 $ @kbd{gdb -iex "set auto-load off" untrusted-executable corefile}
23357 Be aware that system init file (@pxref{System-wide configuration})
23358 and init files from your home directory (@pxref{Home Directory Init File})
23359 still get read (as they come from generally trusted directories).
23360 To prevent @value{GDBN} from auto-loading even those init files, use the
23361 @option{-nx} option (@pxref{Mode Options}), in addition to
23362 @code{set auto-load no}.
23364 @anchor{show auto-load}
23365 @kindex show auto-load
23366 @item show auto-load
23367 Show whether auto-loading of each specific @samp{auto-load} file(s) is enabled
23371 (gdb) show auto-load
23372 gdb-scripts: Auto-loading of canned sequences of commands scripts is on.
23373 libthread-db: Auto-loading of inferior specific libthread_db is on.
23374 local-gdbinit: Auto-loading of .gdbinit script from current directory
23376 python-scripts: Auto-loading of Python scripts is on.
23377 safe-path: List of directories from which it is safe to auto-load files
23378 is $debugdir:$datadir/auto-load.
23379 scripts-directory: List of directories from which to load auto-loaded scripts
23380 is $debugdir:$datadir/auto-load.
23383 @anchor{info auto-load}
23384 @kindex info auto-load
23385 @item info auto-load
23386 Print whether each specific @samp{auto-load} file(s) have been auto-loaded or
23390 (gdb) info auto-load
23393 Yes /home/user/gdb/gdb-gdb.gdb
23394 libthread-db: No auto-loaded libthread-db.
23395 local-gdbinit: Local .gdbinit file "/home/user/gdb/.gdbinit" has been
23399 Yes /home/user/gdb/gdb-gdb.py
23403 These are @value{GDBN} control commands for the auto-loading:
23405 @multitable @columnfractions .5 .5
23406 @item @xref{set auto-load off}.
23407 @tab Disable auto-loading globally.
23408 @item @xref{show auto-load}.
23409 @tab Show setting of all kinds of files.
23410 @item @xref{info auto-load}.
23411 @tab Show state of all kinds of files.
23412 @item @xref{set auto-load gdb-scripts}.
23413 @tab Control for @value{GDBN} command scripts.
23414 @item @xref{show auto-load gdb-scripts}.
23415 @tab Show setting of @value{GDBN} command scripts.
23416 @item @xref{info auto-load gdb-scripts}.
23417 @tab Show state of @value{GDBN} command scripts.
23418 @item @xref{set auto-load python-scripts}.
23419 @tab Control for @value{GDBN} Python scripts.
23420 @item @xref{show auto-load python-scripts}.
23421 @tab Show setting of @value{GDBN} Python scripts.
23422 @item @xref{info auto-load python-scripts}.
23423 @tab Show state of @value{GDBN} Python scripts.
23424 @item @xref{set auto-load guile-scripts}.
23425 @tab Control for @value{GDBN} Guile scripts.
23426 @item @xref{show auto-load guile-scripts}.
23427 @tab Show setting of @value{GDBN} Guile scripts.
23428 @item @xref{info auto-load guile-scripts}.
23429 @tab Show state of @value{GDBN} Guile scripts.
23430 @item @xref{set auto-load scripts-directory}.
23431 @tab Control for @value{GDBN} auto-loaded scripts location.
23432 @item @xref{show auto-load scripts-directory}.
23433 @tab Show @value{GDBN} auto-loaded scripts location.
23434 @item @xref{add-auto-load-scripts-directory}.
23435 @tab Add directory for auto-loaded scripts location list.
23436 @item @xref{set auto-load local-gdbinit}.
23437 @tab Control for init file in the current directory.
23438 @item @xref{show auto-load local-gdbinit}.
23439 @tab Show setting of init file in the current directory.
23440 @item @xref{info auto-load local-gdbinit}.
23441 @tab Show state of init file in the current directory.
23442 @item @xref{set auto-load libthread-db}.
23443 @tab Control for thread debugging library.
23444 @item @xref{show auto-load libthread-db}.
23445 @tab Show setting of thread debugging library.
23446 @item @xref{info auto-load libthread-db}.
23447 @tab Show state of thread debugging library.
23448 @item @xref{set auto-load safe-path}.
23449 @tab Control directories trusted for automatic loading.
23450 @item @xref{show auto-load safe-path}.
23451 @tab Show directories trusted for automatic loading.
23452 @item @xref{add-auto-load-safe-path}.
23453 @tab Add directory trusted for automatic loading.
23456 @node Init File in the Current Directory
23457 @subsection Automatically loading init file in the current directory
23458 @cindex auto-loading init file in the current directory
23460 By default, @value{GDBN} reads and executes the canned sequences of commands
23461 from init file (if any) in the current working directory,
23462 see @ref{Init File in the Current Directory during Startup}.
23464 Note that loading of this local @file{.gdbinit} file also requires accordingly
23465 configured @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
23468 @anchor{set auto-load local-gdbinit}
23469 @kindex set auto-load local-gdbinit
23470 @item set auto-load local-gdbinit [on|off]
23471 Enable or disable the auto-loading of canned sequences of commands
23472 (@pxref{Sequences}) found in init file in the current directory.
23474 @anchor{show auto-load local-gdbinit}
23475 @kindex show auto-load local-gdbinit
23476 @item show auto-load local-gdbinit
23477 Show whether auto-loading of canned sequences of commands from init file in the
23478 current directory is enabled or disabled.
23480 @anchor{info auto-load local-gdbinit}
23481 @kindex info auto-load local-gdbinit
23482 @item info auto-load local-gdbinit
23483 Print whether canned sequences of commands from init file in the
23484 current directory have been auto-loaded.
23487 @node libthread_db.so.1 file
23488 @subsection Automatically loading thread debugging library
23489 @cindex auto-loading libthread_db.so.1
23491 This feature is currently present only on @sc{gnu}/Linux native hosts.
23493 @value{GDBN} reads in some cases thread debugging library from places specific
23494 to the inferior (@pxref{set libthread-db-search-path}).
23496 The special @samp{libthread-db-search-path} entry @samp{$sdir} is processed
23497 without checking this @samp{set auto-load libthread-db} switch as system
23498 libraries have to be trusted in general. In all other cases of
23499 @samp{libthread-db-search-path} entries @value{GDBN} checks first if @samp{set
23500 auto-load libthread-db} is enabled before trying to open such thread debugging
23503 Note that loading of this debugging library also requires accordingly configured
23504 @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
23507 @anchor{set auto-load libthread-db}
23508 @kindex set auto-load libthread-db
23509 @item set auto-load libthread-db [on|off]
23510 Enable or disable the auto-loading of inferior specific thread debugging library.
23512 @anchor{show auto-load libthread-db}
23513 @kindex show auto-load libthread-db
23514 @item show auto-load libthread-db
23515 Show whether auto-loading of inferior specific thread debugging library is
23516 enabled or disabled.
23518 @anchor{info auto-load libthread-db}
23519 @kindex info auto-load libthread-db
23520 @item info auto-load libthread-db
23521 Print the list of all loaded inferior specific thread debugging libraries and
23522 for each such library print list of inferior @var{pid}s using it.
23525 @node Auto-loading safe path
23526 @subsection Security restriction for auto-loading
23527 @cindex auto-loading safe-path
23529 As the files of inferior can come from untrusted source (such as submitted by
23530 an application user) @value{GDBN} does not always load any files automatically.
23531 @value{GDBN} provides the @samp{set auto-load safe-path} setting to list
23532 directories trusted for loading files not explicitly requested by user.
23533 Each directory can also be a shell wildcard pattern.
23535 If the path is not set properly you will see a warning and the file will not
23540 Reading symbols from /home/user/gdb/gdb...done.
23541 warning: File "/home/user/gdb/gdb-gdb.gdb" auto-loading has been
23542 declined by your `auto-load safe-path' set
23543 to "$debugdir:$datadir/auto-load".
23544 warning: File "/home/user/gdb/gdb-gdb.py" auto-loading has been
23545 declined by your `auto-load safe-path' set
23546 to "$debugdir:$datadir/auto-load".
23550 To instruct @value{GDBN} to go ahead and use the init files anyway,
23551 invoke @value{GDBN} like this:
23554 $ gdb -q -iex "set auto-load safe-path /home/user/gdb" ./gdb
23557 The list of trusted directories is controlled by the following commands:
23560 @anchor{set auto-load safe-path}
23561 @kindex set auto-load safe-path
23562 @item set auto-load safe-path @r{[}@var{directories}@r{]}
23563 Set the list of directories (and their subdirectories) trusted for automatic
23564 loading and execution of scripts. You can also enter a specific trusted file.
23565 Each directory can also be a shell wildcard pattern; wildcards do not match
23566 directory separator - see @code{FNM_PATHNAME} for system function @code{fnmatch}
23567 (@pxref{Wildcard Matching, fnmatch, , libc, GNU C Library Reference Manual}).
23568 If you omit @var{directories}, @samp{auto-load safe-path} will be reset to
23569 its default value as specified during @value{GDBN} compilation.
23571 The list of directories uses path separator (@samp{:} on GNU and Unix
23572 systems, @samp{;} on MS-Windows and MS-DOS) to separate directories, similarly
23573 to the @env{PATH} environment variable.
23575 @anchor{show auto-load safe-path}
23576 @kindex show auto-load safe-path
23577 @item show auto-load safe-path
23578 Show the list of directories trusted for automatic loading and execution of
23581 @anchor{add-auto-load-safe-path}
23582 @kindex add-auto-load-safe-path
23583 @item add-auto-load-safe-path
23584 Add an entry (or list of entries) to the list of directories trusted for
23585 automatic loading and execution of scripts. Multiple entries may be delimited
23586 by the host platform path separator in use.
23589 This variable defaults to what @code{--with-auto-load-dir} has been configured
23590 to (@pxref{with-auto-load-dir}). @file{$debugdir} and @file{$datadir}
23591 substitution applies the same as for @ref{set auto-load scripts-directory}.
23592 The default @code{set auto-load safe-path} value can be also overriden by
23593 @value{GDBN} configuration option @option{--with-auto-load-safe-path}.
23595 Setting this variable to @file{/} disables this security protection,
23596 corresponding @value{GDBN} configuration option is
23597 @option{--without-auto-load-safe-path}.
23598 This variable is supposed to be set to the system directories writable by the
23599 system superuser only. Users can add their source directories in init files in
23600 their home directories (@pxref{Home Directory Init File}). See also deprecated
23601 init file in the current directory
23602 (@pxref{Init File in the Current Directory during Startup}).
23604 To force @value{GDBN} to load the files it declined to load in the previous
23605 example, you could use one of the following ways:
23608 @item @file{~/.gdbinit}: @samp{add-auto-load-safe-path ~/src/gdb}
23609 Specify this trusted directory (or a file) as additional component of the list.
23610 You have to specify also any existing directories displayed by
23611 by @samp{show auto-load safe-path} (such as @samp{/usr:/bin} in this example).
23613 @item @kbd{gdb -iex "set auto-load safe-path /usr:/bin:~/src/gdb" @dots{}}
23614 Specify this directory as in the previous case but just for a single
23615 @value{GDBN} session.
23617 @item @kbd{gdb -iex "set auto-load safe-path /" @dots{}}
23618 Disable auto-loading safety for a single @value{GDBN} session.
23619 This assumes all the files you debug during this @value{GDBN} session will come
23620 from trusted sources.
23622 @item @kbd{./configure --without-auto-load-safe-path}
23623 During compilation of @value{GDBN} you may disable any auto-loading safety.
23624 This assumes all the files you will ever debug with this @value{GDBN} come from
23628 On the other hand you can also explicitly forbid automatic files loading which
23629 also suppresses any such warning messages:
23632 @item @kbd{gdb -iex "set auto-load no" @dots{}}
23633 You can use @value{GDBN} command-line option for a single @value{GDBN} session.
23635 @item @file{~/.gdbinit}: @samp{set auto-load no}
23636 Disable auto-loading globally for the user
23637 (@pxref{Home Directory Init File}). While it is improbable, you could also
23638 use system init file instead (@pxref{System-wide configuration}).
23641 This setting applies to the file names as entered by user. If no entry matches
23642 @value{GDBN} tries as a last resort to also resolve all the file names into
23643 their canonical form (typically resolving symbolic links) and compare the
23644 entries again. @value{GDBN} already canonicalizes most of the filenames on its
23645 own before starting the comparison so a canonical form of directories is
23646 recommended to be entered.
23648 @node Auto-loading verbose mode
23649 @subsection Displaying files tried for auto-load
23650 @cindex auto-loading verbose mode
23652 For better visibility of all the file locations where you can place scripts to
23653 be auto-loaded with inferior --- or to protect yourself against accidental
23654 execution of untrusted scripts --- @value{GDBN} provides a feature for printing
23655 all the files attempted to be loaded. Both existing and non-existing files may
23658 For example the list of directories from which it is safe to auto-load files
23659 (@pxref{Auto-loading safe path}) applies also to canonicalized filenames which
23660 may not be too obvious while setting it up.
23663 (gdb) set debug auto-load on
23664 (gdb) file ~/src/t/true
23665 auto-load: Loading canned sequences of commands script "/tmp/true-gdb.gdb"
23666 for objfile "/tmp/true".
23667 auto-load: Updating directories of "/usr:/opt".
23668 auto-load: Using directory "/usr".
23669 auto-load: Using directory "/opt".
23670 warning: File "/tmp/true-gdb.gdb" auto-loading has been declined
23671 by your `auto-load safe-path' set to "/usr:/opt".
23675 @anchor{set debug auto-load}
23676 @kindex set debug auto-load
23677 @item set debug auto-load [on|off]
23678 Set whether to print the filenames attempted to be auto-loaded.
23680 @anchor{show debug auto-load}
23681 @kindex show debug auto-load
23682 @item show debug auto-load
23683 Show whether printing of the filenames attempted to be auto-loaded is turned
23687 @node Messages/Warnings
23688 @section Optional Warnings and Messages
23690 @cindex verbose operation
23691 @cindex optional warnings
23692 By default, @value{GDBN} is silent about its inner workings. If you are
23693 running on a slow machine, you may want to use the @code{set verbose}
23694 command. This makes @value{GDBN} tell you when it does a lengthy
23695 internal operation, so you will not think it has crashed.
23697 Currently, the messages controlled by @code{set verbose} are those
23698 which announce that the symbol table for a source file is being read;
23699 see @code{symbol-file} in @ref{Files, ,Commands to Specify Files}.
23702 @kindex set verbose
23703 @item set verbose on
23704 Enables @value{GDBN} output of certain informational messages.
23706 @item set verbose off
23707 Disables @value{GDBN} output of certain informational messages.
23709 @kindex show verbose
23711 Displays whether @code{set verbose} is on or off.
23714 By default, if @value{GDBN} encounters bugs in the symbol table of an
23715 object file, it is silent; but if you are debugging a compiler, you may
23716 find this information useful (@pxref{Symbol Errors, ,Errors Reading
23721 @kindex set complaints
23722 @item set complaints @var{limit}
23723 Permits @value{GDBN} to output @var{limit} complaints about each type of
23724 unusual symbols before becoming silent about the problem. Set
23725 @var{limit} to zero to suppress all complaints; set it to a large number
23726 to prevent complaints from being suppressed.
23728 @kindex show complaints
23729 @item show complaints
23730 Displays how many symbol complaints @value{GDBN} is permitted to produce.
23734 @anchor{confirmation requests}
23735 By default, @value{GDBN} is cautious, and asks what sometimes seems to be a
23736 lot of stupid questions to confirm certain commands. For example, if
23737 you try to run a program which is already running:
23741 The program being debugged has been started already.
23742 Start it from the beginning? (y or n)
23745 If you are willing to unflinchingly face the consequences of your own
23746 commands, you can disable this ``feature'':
23750 @kindex set confirm
23752 @cindex confirmation
23753 @cindex stupid questions
23754 @item set confirm off
23755 Disables confirmation requests. Note that running @value{GDBN} with
23756 the @option{--batch} option (@pxref{Mode Options, -batch}) also
23757 automatically disables confirmation requests.
23759 @item set confirm on
23760 Enables confirmation requests (the default).
23762 @kindex show confirm
23764 Displays state of confirmation requests.
23768 @cindex command tracing
23769 If you need to debug user-defined commands or sourced files you may find it
23770 useful to enable @dfn{command tracing}. In this mode each command will be
23771 printed as it is executed, prefixed with one or more @samp{+} symbols, the
23772 quantity denoting the call depth of each command.
23775 @kindex set trace-commands
23776 @cindex command scripts, debugging
23777 @item set trace-commands on
23778 Enable command tracing.
23779 @item set trace-commands off
23780 Disable command tracing.
23781 @item show trace-commands
23782 Display the current state of command tracing.
23785 @node Debugging Output
23786 @section Optional Messages about Internal Happenings
23787 @cindex optional debugging messages
23789 @value{GDBN} has commands that enable optional debugging messages from
23790 various @value{GDBN} subsystems; normally these commands are of
23791 interest to @value{GDBN} maintainers, or when reporting a bug. This
23792 section documents those commands.
23795 @kindex set exec-done-display
23796 @item set exec-done-display
23797 Turns on or off the notification of asynchronous commands'
23798 completion. When on, @value{GDBN} will print a message when an
23799 asynchronous command finishes its execution. The default is off.
23800 @kindex show exec-done-display
23801 @item show exec-done-display
23802 Displays the current setting of asynchronous command completion
23805 @cindex ARM AArch64
23806 @item set debug aarch64
23807 Turns on or off display of debugging messages related to ARM AArch64.
23808 The default is off.
23810 @item show debug aarch64
23811 Displays the current state of displaying debugging messages related to
23813 @cindex gdbarch debugging info
23814 @cindex architecture debugging info
23815 @item set debug arch
23816 Turns on or off display of gdbarch debugging info. The default is off
23817 @item show debug arch
23818 Displays the current state of displaying gdbarch debugging info.
23819 @item set debug aix-solib
23820 @cindex AIX shared library debugging
23821 Control display of debugging messages from the AIX shared library
23822 support module. The default is off.
23823 @item show debug aix-thread
23824 Show the current state of displaying AIX shared library debugging messages.
23825 @item set debug aix-thread
23826 @cindex AIX threads
23827 Display debugging messages about inner workings of the AIX thread
23829 @item show debug aix-thread
23830 Show the current state of AIX thread debugging info display.
23831 @item set debug check-physname
23833 Check the results of the ``physname'' computation. When reading DWARF
23834 debugging information for C@t{++}, @value{GDBN} attempts to compute
23835 each entity's name. @value{GDBN} can do this computation in two
23836 different ways, depending on exactly what information is present.
23837 When enabled, this setting causes @value{GDBN} to compute the names
23838 both ways and display any discrepancies.
23839 @item show debug check-physname
23840 Show the current state of ``physname'' checking.
23841 @item set debug coff-pe-read
23842 @cindex COFF/PE exported symbols
23843 Control display of debugging messages related to reading of COFF/PE
23844 exported symbols. The default is off.
23845 @item show debug coff-pe-read
23846 Displays the current state of displaying debugging messages related to
23847 reading of COFF/PE exported symbols.
23848 @item set debug dwarf-die
23850 Dump DWARF DIEs after they are read in.
23851 The value is the number of nesting levels to print.
23852 A value of zero turns off the display.
23853 @item show debug dwarf-die
23854 Show the current state of DWARF DIE debugging.
23855 @item set debug dwarf-line
23856 @cindex DWARF Line Tables
23857 Turns on or off display of debugging messages related to reading
23858 DWARF line tables. The default is 0 (off).
23859 A value of 1 provides basic information.
23860 A value greater than 1 provides more verbose information.
23861 @item show debug dwarf-line
23862 Show the current state of DWARF line table debugging.
23863 @item set debug dwarf-read
23864 @cindex DWARF Reading
23865 Turns on or off display of debugging messages related to reading
23866 DWARF debug info. 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-read
23870 Show the current state of DWARF reader debugging.
23871 @item set debug displaced
23872 @cindex displaced stepping debugging info
23873 Turns on or off display of @value{GDBN} debugging info for the
23874 displaced stepping support. The default is off.
23875 @item show debug displaced
23876 Displays the current state of displaying @value{GDBN} debugging info
23877 related to displaced stepping.
23878 @item set debug event
23879 @cindex event debugging info
23880 Turns on or off display of @value{GDBN} event debugging info. The
23882 @item show debug event
23883 Displays the current state of displaying @value{GDBN} event debugging
23885 @item set debug expression
23886 @cindex expression debugging info
23887 Turns on or off display of debugging info about @value{GDBN}
23888 expression parsing. The default is off.
23889 @item show debug expression
23890 Displays the current state of displaying debugging info about
23891 @value{GDBN} expression parsing.
23892 @item set debug fbsd-lwp
23893 @cindex FreeBSD LWP debug messages
23894 Turns on or off debugging messages from the FreeBSD LWP debug support.
23895 @item show debug fbsd-lwp
23896 Show the current state of FreeBSD LWP debugging messages.
23897 @item set debug frame
23898 @cindex frame debugging info
23899 Turns on or off display of @value{GDBN} frame debugging info. The
23901 @item show debug frame
23902 Displays the current state of displaying @value{GDBN} frame debugging
23904 @item set debug gnu-nat
23905 @cindex @sc{gnu}/Hurd debug messages
23906 Turn on or off debugging messages from the @sc{gnu}/Hurd debug support.
23907 @item show debug gnu-nat
23908 Show the current state of @sc{gnu}/Hurd debugging messages.
23909 @item set debug infrun
23910 @cindex inferior debugging info
23911 Turns on or off display of @value{GDBN} debugging info for running the inferior.
23912 The default is off. @file{infrun.c} contains GDB's runtime state machine used
23913 for implementing operations such as single-stepping the inferior.
23914 @item show debug infrun
23915 Displays the current state of @value{GDBN} inferior debugging.
23916 @item set debug jit
23917 @cindex just-in-time compilation, debugging messages
23918 Turn on or off debugging messages from JIT debug support.
23919 @item show debug jit
23920 Displays the current state of @value{GDBN} JIT debugging.
23921 @item set debug lin-lwp
23922 @cindex @sc{gnu}/Linux LWP debug messages
23923 @cindex Linux lightweight processes
23924 Turn on or off debugging messages from the Linux LWP debug support.
23925 @item show debug lin-lwp
23926 Show the current state of Linux LWP debugging messages.
23927 @item set debug linux-namespaces
23928 @cindex @sc{gnu}/Linux namespaces debug messages
23929 Turn on or off debugging messages from the Linux namespaces debug support.
23930 @item show debug linux-namespaces
23931 Show the current state of Linux namespaces debugging messages.
23932 @item set debug mach-o
23933 @cindex Mach-O symbols processing
23934 Control display of debugging messages related to Mach-O symbols
23935 processing. The default is off.
23936 @item show debug mach-o
23937 Displays the current state of displaying debugging messages related to
23938 reading of COFF/PE exported symbols.
23939 @item set debug notification
23940 @cindex remote async notification debugging info
23941 Turn on or off debugging messages about remote async notification.
23942 The default is off.
23943 @item show debug notification
23944 Displays the current state of remote async notification debugging messages.
23945 @item set debug observer
23946 @cindex observer debugging info
23947 Turns on or off display of @value{GDBN} observer debugging. This
23948 includes info such as the notification of observable events.
23949 @item show debug observer
23950 Displays the current state of observer debugging.
23951 @item set debug overload
23952 @cindex C@t{++} overload debugging info
23953 Turns on or off display of @value{GDBN} C@t{++} overload debugging
23954 info. This includes info such as ranking of functions, etc. The default
23956 @item show debug overload
23957 Displays the current state of displaying @value{GDBN} C@t{++} overload
23959 @cindex expression parser, debugging info
23960 @cindex debug expression parser
23961 @item set debug parser
23962 Turns on or off the display of expression parser debugging output.
23963 Internally, this sets the @code{yydebug} variable in the expression
23964 parser. @xref{Tracing, , Tracing Your Parser, bison, Bison}, for
23965 details. The default is off.
23966 @item show debug parser
23967 Show the current state of expression parser debugging.
23968 @cindex packets, reporting on stdout
23969 @cindex serial connections, debugging
23970 @cindex debug remote protocol
23971 @cindex remote protocol debugging
23972 @cindex display remote packets
23973 @item set debug remote
23974 Turns on or off display of reports on all packets sent back and forth across
23975 the serial line to the remote machine. The info is printed on the
23976 @value{GDBN} standard output stream. The default is off.
23977 @item show debug remote
23978 Displays the state of display of remote packets.
23979 @item set debug serial
23980 Turns on or off display of @value{GDBN} serial debugging info. The
23982 @item show debug serial
23983 Displays the current state of displaying @value{GDBN} serial debugging
23985 @item set debug solib-frv
23986 @cindex FR-V shared-library debugging
23987 Turn on or off debugging messages for FR-V shared-library code.
23988 @item show debug solib-frv
23989 Display the current state of FR-V shared-library code debugging
23991 @item set debug symbol-lookup
23992 @cindex symbol lookup
23993 Turns on or off display of debugging messages related to symbol lookup.
23994 The default is 0 (off).
23995 A value of 1 provides basic information.
23996 A value greater than 1 provides more verbose information.
23997 @item show debug symbol-lookup
23998 Show the current state of symbol lookup debugging messages.
23999 @item set debug symfile
24000 @cindex symbol file functions
24001 Turns on or off display of debugging messages related to symbol file functions.
24002 The default is off. @xref{Files}.
24003 @item show debug symfile
24004 Show the current state of symbol file debugging messages.
24005 @item set debug symtab-create
24006 @cindex symbol table creation
24007 Turns on or off display of debugging messages related to symbol table creation.
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 symtab-create
24012 Show the current state of symbol table creation debugging.
24013 @item set debug target
24014 @cindex target debugging info
24015 Turns on or off display of @value{GDBN} target debugging info. This info
24016 includes what is going on at the target level of GDB, as it happens. The
24017 default is 0. Set it to 1 to track events, and to 2 to also track the
24018 value of large memory transfers.
24019 @item show debug target
24020 Displays the current state of displaying @value{GDBN} target debugging
24022 @item set debug timestamp
24023 @cindex timestampping debugging info
24024 Turns on or off display of timestamps with @value{GDBN} debugging info.
24025 When enabled, seconds and microseconds are displayed before each debugging
24027 @item show debug timestamp
24028 Displays the current state of displaying timestamps with @value{GDBN}
24030 @item set debug varobj
24031 @cindex variable object debugging info
24032 Turns on or off display of @value{GDBN} variable object debugging
24033 info. The default is off.
24034 @item show debug varobj
24035 Displays the current state of displaying @value{GDBN} variable object
24037 @item set debug xml
24038 @cindex XML parser debugging
24039 Turn on or off debugging messages for built-in XML parsers.
24040 @item show debug xml
24041 Displays the current state of XML debugging messages.
24044 @node Other Misc Settings
24045 @section Other Miscellaneous Settings
24046 @cindex miscellaneous settings
24049 @kindex set interactive-mode
24050 @item set interactive-mode
24051 If @code{on}, forces @value{GDBN} to assume that GDB was started
24052 in a terminal. In practice, this means that @value{GDBN} should wait
24053 for the user to answer queries generated by commands entered at
24054 the command prompt. If @code{off}, forces @value{GDBN} to operate
24055 in the opposite mode, and it uses the default answers to all queries.
24056 If @code{auto} (the default), @value{GDBN} tries to determine whether
24057 its standard input is a terminal, and works in interactive-mode if it
24058 is, non-interactively otherwise.
24060 In the vast majority of cases, the debugger should be able to guess
24061 correctly which mode should be used. But this setting can be useful
24062 in certain specific cases, such as running a MinGW @value{GDBN}
24063 inside a cygwin window.
24065 @kindex show interactive-mode
24066 @item show interactive-mode
24067 Displays whether the debugger is operating in interactive mode or not.
24070 @node Extending GDB
24071 @chapter Extending @value{GDBN}
24072 @cindex extending GDB
24074 @value{GDBN} provides several mechanisms for extension.
24075 @value{GDBN} also provides the ability to automatically load
24076 extensions when it reads a file for debugging. This allows the
24077 user to automatically customize @value{GDBN} for the program
24081 * Sequences:: Canned Sequences of @value{GDBN} Commands
24082 * Python:: Extending @value{GDBN} using Python
24083 * Guile:: Extending @value{GDBN} using Guile
24084 * Auto-loading extensions:: Automatically loading extensions
24085 * Multiple Extension Languages:: Working with multiple extension languages
24086 * Aliases:: Creating new spellings of existing commands
24089 To facilitate the use of extension languages, @value{GDBN} is capable
24090 of evaluating the contents of a file. When doing so, @value{GDBN}
24091 can recognize which extension language is being used by looking at
24092 the filename extension. Files with an unrecognized filename extension
24093 are always treated as a @value{GDBN} Command Files.
24094 @xref{Command Files,, Command files}.
24096 You can control how @value{GDBN} evaluates these files with the following
24100 @kindex set script-extension
24101 @kindex show script-extension
24102 @item set script-extension off
24103 All scripts are always evaluated as @value{GDBN} Command Files.
24105 @item set script-extension soft
24106 The debugger determines the scripting language based on filename
24107 extension. If this scripting language is supported, @value{GDBN}
24108 evaluates the script using that language. Otherwise, it evaluates
24109 the file as a @value{GDBN} Command File.
24111 @item set script-extension strict
24112 The debugger determines the scripting language based on filename
24113 extension, and evaluates the script using that language. If the
24114 language is not supported, then the evaluation fails.
24116 @item show script-extension
24117 Display the current value of the @code{script-extension} option.
24122 @section Canned Sequences of Commands
24124 Aside from breakpoint commands (@pxref{Break Commands, ,Breakpoint
24125 Command Lists}), @value{GDBN} provides two ways to store sequences of
24126 commands for execution as a unit: user-defined commands and command
24130 * Define:: How to define your own commands
24131 * Hooks:: Hooks for user-defined commands
24132 * Command Files:: How to write scripts of commands to be stored in a file
24133 * Output:: Commands for controlled output
24134 * Auto-loading sequences:: Controlling auto-loaded command files
24138 @subsection User-defined Commands
24140 @cindex user-defined command
24141 @cindex arguments, to user-defined commands
24142 A @dfn{user-defined command} is a sequence of @value{GDBN} commands to
24143 which you assign a new name as a command. This is done with the
24144 @code{define} command. User commands may accept an unlimited number of arguments
24145 separated by whitespace. Arguments are accessed within the user command
24146 via @code{$arg0@dots{}$argN}. A trivial example:
24150 print $arg0 + $arg1 + $arg2
24155 To execute the command use:
24162 This defines the command @code{adder}, which prints the sum of
24163 its three arguments. Note the arguments are text substitutions, so they may
24164 reference variables, use complex expressions, or even perform inferior
24167 @cindex argument count in user-defined commands
24168 @cindex how many arguments (user-defined commands)
24169 In addition, @code{$argc} may be used to find out how many arguments have
24175 print $arg0 + $arg1
24178 print $arg0 + $arg1 + $arg2
24183 Combining with the @code{eval} command (@pxref{eval}) makes it easier
24184 to process a variable number of arguments:
24191 eval "set $sum = $sum + $arg%d", $i
24201 @item define @var{commandname}
24202 Define a command named @var{commandname}. If there is already a command
24203 by that name, you are asked to confirm that you want to redefine it.
24204 The argument @var{commandname} may be a bare command name consisting of letters,
24205 numbers, dashes, and underscores. It may also start with any predefined
24206 prefix command. For example, @samp{define target my-target} creates
24207 a user-defined @samp{target my-target} command.
24209 The definition of the command is made up of other @value{GDBN} command lines,
24210 which are given following the @code{define} command. The end of these
24211 commands is marked by a line containing @code{end}.
24214 @kindex end@r{ (user-defined commands)}
24215 @item document @var{commandname}
24216 Document the user-defined command @var{commandname}, so that it can be
24217 accessed by @code{help}. The command @var{commandname} must already be
24218 defined. This command reads lines of documentation just as @code{define}
24219 reads the lines of the command definition, ending with @code{end}.
24220 After the @code{document} command is finished, @code{help} on command
24221 @var{commandname} displays the documentation you have written.
24223 You may use the @code{document} command again to change the
24224 documentation of a command. Redefining the command with @code{define}
24225 does not change the documentation.
24227 @kindex dont-repeat
24228 @cindex don't repeat command
24230 Used inside a user-defined command, this tells @value{GDBN} that this
24231 command should not be repeated when the user hits @key{RET}
24232 (@pxref{Command Syntax, repeat last command}).
24234 @kindex help user-defined
24235 @item help user-defined
24236 List all user-defined commands and all python commands defined in class
24237 COMAND_USER. The first line of the documentation or docstring is
24242 @itemx show user @var{commandname}
24243 Display the @value{GDBN} commands used to define @var{commandname} (but
24244 not its documentation). If no @var{commandname} is given, display the
24245 definitions for all user-defined commands.
24246 This does not work for user-defined python commands.
24248 @cindex infinite recursion in user-defined commands
24249 @kindex show max-user-call-depth
24250 @kindex set max-user-call-depth
24251 @item show max-user-call-depth
24252 @itemx set max-user-call-depth
24253 The value of @code{max-user-call-depth} controls how many recursion
24254 levels are allowed in user-defined commands before @value{GDBN} suspects an
24255 infinite recursion and aborts the command.
24256 This does not apply to user-defined python commands.
24259 In addition to the above commands, user-defined commands frequently
24260 use control flow commands, described in @ref{Command Files}.
24262 When user-defined commands are executed, the
24263 commands of the definition are not printed. An error in any command
24264 stops execution of the user-defined command.
24266 If used interactively, commands that would ask for confirmation proceed
24267 without asking when used inside a user-defined command. Many @value{GDBN}
24268 commands that normally print messages to say what they are doing omit the
24269 messages when used in a user-defined command.
24272 @subsection User-defined Command Hooks
24273 @cindex command hooks
24274 @cindex hooks, for commands
24275 @cindex hooks, pre-command
24278 You may define @dfn{hooks}, which are a special kind of user-defined
24279 command. Whenever you run the command @samp{foo}, if the user-defined
24280 command @samp{hook-foo} exists, it is executed (with no arguments)
24281 before that command.
24283 @cindex hooks, post-command
24285 A hook may also be defined which is run after the command you executed.
24286 Whenever you run the command @samp{foo}, if the user-defined command
24287 @samp{hookpost-foo} exists, it is executed (with no arguments) after
24288 that command. Post-execution hooks may exist simultaneously with
24289 pre-execution hooks, for the same command.
24291 It is valid for a hook to call the command which it hooks. If this
24292 occurs, the hook is not re-executed, thereby avoiding infinite recursion.
24294 @c It would be nice if hookpost could be passed a parameter indicating
24295 @c if the command it hooks executed properly or not. FIXME!
24297 @kindex stop@r{, a pseudo-command}
24298 In addition, a pseudo-command, @samp{stop} exists. Defining
24299 (@samp{hook-stop}) makes the associated commands execute every time
24300 execution stops in your program: before breakpoint commands are run,
24301 displays are printed, or the stack frame is printed.
24303 For example, to ignore @code{SIGALRM} signals while
24304 single-stepping, but treat them normally during normal execution,
24309 handle SIGALRM nopass
24313 handle SIGALRM pass
24316 define hook-continue
24317 handle SIGALRM pass
24321 As a further example, to hook at the beginning and end of the @code{echo}
24322 command, and to add extra text to the beginning and end of the message,
24330 define hookpost-echo
24334 (@value{GDBP}) echo Hello World
24335 <<<---Hello World--->>>
24340 You can define a hook for any single-word command in @value{GDBN}, but
24341 not for command aliases; you should define a hook for the basic command
24342 name, e.g.@: @code{backtrace} rather than @code{bt}.
24343 @c FIXME! So how does Joe User discover whether a command is an alias
24345 You can hook a multi-word command by adding @code{hook-} or
24346 @code{hookpost-} to the last word of the command, e.g.@:
24347 @samp{define target hook-remote} to add a hook to @samp{target remote}.
24349 If an error occurs during the execution of your hook, execution of
24350 @value{GDBN} commands stops and @value{GDBN} issues a prompt
24351 (before the command that you actually typed had a chance to run).
24353 If you try to define a hook which does not match any known command, you
24354 get a warning from the @code{define} command.
24356 @node Command Files
24357 @subsection Command Files
24359 @cindex command files
24360 @cindex scripting commands
24361 A command file for @value{GDBN} is a text file made of lines that are
24362 @value{GDBN} commands. Comments (lines starting with @kbd{#}) may
24363 also be included. An empty line in a command file does nothing; it
24364 does not mean to repeat the last command, as it would from the
24367 You can request the execution of a command file with the @code{source}
24368 command. Note that the @code{source} command is also used to evaluate
24369 scripts that are not Command Files. The exact behavior can be configured
24370 using the @code{script-extension} setting.
24371 @xref{Extending GDB,, Extending GDB}.
24375 @cindex execute commands from a file
24376 @item source [-s] [-v] @var{filename}
24377 Execute the command file @var{filename}.
24380 The lines in a command file are generally executed sequentially,
24381 unless the order of execution is changed by one of the
24382 @emph{flow-control commands} described below. The commands are not
24383 printed as they are executed. An error in any command terminates
24384 execution of the command file and control is returned to the console.
24386 @value{GDBN} first searches for @var{filename} in the current directory.
24387 If the file is not found there, and @var{filename} does not specify a
24388 directory, then @value{GDBN} also looks for the file on the source search path
24389 (specified with the @samp{directory} command);
24390 except that @file{$cdir} is not searched because the compilation directory
24391 is not relevant to scripts.
24393 If @code{-s} is specified, then @value{GDBN} searches for @var{filename}
24394 on the search path even if @var{filename} specifies a directory.
24395 The search is done by appending @var{filename} to each element of the
24396 search path. So, for example, if @var{filename} is @file{mylib/myscript}
24397 and the search path contains @file{/home/user} then @value{GDBN} will
24398 look for the script @file{/home/user/mylib/myscript}.
24399 The search is also done if @var{filename} is an absolute path.
24400 For example, if @var{filename} is @file{/tmp/myscript} and
24401 the search path contains @file{/home/user} then @value{GDBN} will
24402 look for the script @file{/home/user/tmp/myscript}.
24403 For DOS-like systems, if @var{filename} contains a drive specification,
24404 it is stripped before concatenation. For example, if @var{filename} is
24405 @file{d:myscript} and the search path contains @file{c:/tmp} then @value{GDBN}
24406 will look for the script @file{c:/tmp/myscript}.
24408 If @code{-v}, for verbose mode, is given then @value{GDBN} displays
24409 each command as it is executed. The option must be given before
24410 @var{filename}, and is interpreted as part of the filename anywhere else.
24412 Commands that would ask for confirmation if used interactively proceed
24413 without asking when used in a command file. Many @value{GDBN} commands that
24414 normally print messages to say what they are doing omit the messages
24415 when called from command files.
24417 @value{GDBN} also accepts command input from standard input. In this
24418 mode, normal output goes to standard output and error output goes to
24419 standard error. Errors in a command file supplied on standard input do
24420 not terminate execution of the command file---execution continues with
24424 gdb < cmds > log 2>&1
24427 (The syntax above will vary depending on the shell used.) This example
24428 will execute commands from the file @file{cmds}. All output and errors
24429 would be directed to @file{log}.
24431 Since commands stored on command files tend to be more general than
24432 commands typed interactively, they frequently need to deal with
24433 complicated situations, such as different or unexpected values of
24434 variables and symbols, changes in how the program being debugged is
24435 built, etc. @value{GDBN} provides a set of flow-control commands to
24436 deal with these complexities. Using these commands, you can write
24437 complex scripts that loop over data structures, execute commands
24438 conditionally, etc.
24445 This command allows to include in your script conditionally executed
24446 commands. The @code{if} command takes a single argument, which is an
24447 expression to evaluate. It is followed by a series of commands that
24448 are executed only if the expression is true (its value is nonzero).
24449 There can then optionally be an @code{else} line, followed by a series
24450 of commands that are only executed if the expression was false. The
24451 end of the list is marked by a line containing @code{end}.
24455 This command allows to write loops. Its syntax is similar to
24456 @code{if}: the command takes a single argument, which is an expression
24457 to evaluate, and must be followed by the commands to execute, one per
24458 line, terminated by an @code{end}. These commands are called the
24459 @dfn{body} of the loop. The commands in the body of @code{while} are
24460 executed repeatedly as long as the expression evaluates to true.
24464 This command exits the @code{while} loop in whose body it is included.
24465 Execution of the script continues after that @code{while}s @code{end}
24468 @kindex loop_continue
24469 @item loop_continue
24470 This command skips the execution of the rest of the body of commands
24471 in the @code{while} loop in whose body it is included. Execution
24472 branches to the beginning of the @code{while} loop, where it evaluates
24473 the controlling expression.
24475 @kindex end@r{ (if/else/while commands)}
24477 Terminate the block of commands that are the body of @code{if},
24478 @code{else}, or @code{while} flow-control commands.
24483 @subsection Commands for Controlled Output
24485 During the execution of a command file or a user-defined command, normal
24486 @value{GDBN} output is suppressed; the only output that appears is what is
24487 explicitly printed by the commands in the definition. This section
24488 describes three commands useful for generating exactly the output you
24493 @item echo @var{text}
24494 @c I do not consider backslash-space a standard C escape sequence
24495 @c because it is not in ANSI.
24496 Print @var{text}. Nonprinting characters can be included in
24497 @var{text} using C escape sequences, such as @samp{\n} to print a
24498 newline. @strong{No newline is printed unless you specify one.}
24499 In addition to the standard C escape sequences, a backslash followed
24500 by a space stands for a space. This is useful for displaying a
24501 string with spaces at the beginning or the end, since leading and
24502 trailing spaces are otherwise trimmed from all arguments.
24503 To print @samp{@w{ }and foo =@w{ }}, use the command
24504 @samp{echo \@w{ }and foo = \@w{ }}.
24506 A backslash at the end of @var{text} can be used, as in C, to continue
24507 the command onto subsequent lines. For example,
24510 echo This is some text\n\
24511 which is continued\n\
24512 onto several lines.\n
24515 produces the same output as
24518 echo This is some text\n
24519 echo which is continued\n
24520 echo onto several lines.\n
24524 @item output @var{expression}
24525 Print the value of @var{expression} and nothing but that value: no
24526 newlines, no @samp{$@var{nn} = }. The value is not entered in the
24527 value history either. @xref{Expressions, ,Expressions}, for more information
24530 @item output/@var{fmt} @var{expression}
24531 Print the value of @var{expression} in format @var{fmt}. You can use
24532 the same formats as for @code{print}. @xref{Output Formats,,Output
24533 Formats}, for more information.
24536 @item printf @var{template}, @var{expressions}@dots{}
24537 Print the values of one or more @var{expressions} under the control of
24538 the string @var{template}. To print several values, make
24539 @var{expressions} be a comma-separated list of individual expressions,
24540 which may be either numbers or pointers. Their values are printed as
24541 specified by @var{template}, exactly as a C program would do by
24542 executing the code below:
24545 printf (@var{template}, @var{expressions}@dots{});
24548 As in @code{C} @code{printf}, ordinary characters in @var{template}
24549 are printed verbatim, while @dfn{conversion specification} introduced
24550 by the @samp{%} character cause subsequent @var{expressions} to be
24551 evaluated, their values converted and formatted according to type and
24552 style information encoded in the conversion specifications, and then
24555 For example, you can print two values in hex like this:
24558 printf "foo, bar-foo = 0x%x, 0x%x\n", foo, bar-foo
24561 @code{printf} supports all the standard @code{C} conversion
24562 specifications, including the flags and modifiers between the @samp{%}
24563 character and the conversion letter, with the following exceptions:
24567 The argument-ordering modifiers, such as @samp{2$}, are not supported.
24570 The modifier @samp{*} is not supported for specifying precision or
24574 The @samp{'} flag (for separation of digits into groups according to
24575 @code{LC_NUMERIC'}) is not supported.
24578 The type modifiers @samp{hh}, @samp{j}, @samp{t}, and @samp{z} are not
24582 The conversion letter @samp{n} (as in @samp{%n}) is not supported.
24585 The conversion letters @samp{a} and @samp{A} are not supported.
24589 Note that the @samp{ll} type modifier is supported only if the
24590 underlying @code{C} implementation used to build @value{GDBN} supports
24591 the @code{long long int} type, and the @samp{L} type modifier is
24592 supported only if @code{long double} type is available.
24594 As in @code{C}, @code{printf} supports simple backslash-escape
24595 sequences, such as @code{\n}, @samp{\t}, @samp{\\}, @samp{\"},
24596 @samp{\a}, and @samp{\f}, that consist of backslash followed by a
24597 single character. Octal and hexadecimal escape sequences are not
24600 Additionally, @code{printf} supports conversion specifications for DFP
24601 (@dfn{Decimal Floating Point}) types using the following length modifiers
24602 together with a floating point specifier.
24607 @samp{H} for printing @code{Decimal32} types.
24610 @samp{D} for printing @code{Decimal64} types.
24613 @samp{DD} for printing @code{Decimal128} types.
24616 If the underlying @code{C} implementation used to build @value{GDBN} has
24617 support for the three length modifiers for DFP types, other modifiers
24618 such as width and precision will also be available for @value{GDBN} to use.
24620 In case there is no such @code{C} support, no additional modifiers will be
24621 available and the value will be printed in the standard way.
24623 Here's an example of printing DFP types using the above conversion letters:
24625 printf "D32: %Hf - D64: %Df - D128: %DDf\n",1.2345df,1.2E10dd,1.2E1dl
24630 @item eval @var{template}, @var{expressions}@dots{}
24631 Convert the values of one or more @var{expressions} under the control of
24632 the string @var{template} to a command line, and call it.
24636 @node Auto-loading sequences
24637 @subsection Controlling auto-loading native @value{GDBN} scripts
24638 @cindex native script auto-loading
24640 When a new object file is read (for example, due to the @code{file}
24641 command, or because the inferior has loaded a shared library),
24642 @value{GDBN} will look for the command file @file{@var{objfile}-gdb.gdb}.
24643 @xref{Auto-loading extensions}.
24645 Auto-loading can be enabled or disabled,
24646 and the list of auto-loaded scripts can be printed.
24649 @anchor{set auto-load gdb-scripts}
24650 @kindex set auto-load gdb-scripts
24651 @item set auto-load gdb-scripts [on|off]
24652 Enable or disable the auto-loading of canned sequences of commands scripts.
24654 @anchor{show auto-load gdb-scripts}
24655 @kindex show auto-load gdb-scripts
24656 @item show auto-load gdb-scripts
24657 Show whether auto-loading of canned sequences of commands scripts is enabled or
24660 @anchor{info auto-load gdb-scripts}
24661 @kindex info auto-load gdb-scripts
24662 @cindex print list of auto-loaded canned sequences of commands scripts
24663 @item info auto-load gdb-scripts [@var{regexp}]
24664 Print the list of all canned sequences of commands scripts that @value{GDBN}
24668 If @var{regexp} is supplied only canned sequences of commands scripts with
24669 matching names are printed.
24671 @c Python docs live in a separate file.
24672 @include python.texi
24674 @c Guile docs live in a separate file.
24675 @include guile.texi
24677 @node Auto-loading extensions
24678 @section Auto-loading extensions
24679 @cindex auto-loading extensions
24681 @value{GDBN} provides two mechanisms for automatically loading extensions
24682 when a new object file is read (for example, due to the @code{file}
24683 command, or because the inferior has loaded a shared library):
24684 @file{@var{objfile}-gdb.@var{ext}} and the @code{.debug_gdb_scripts}
24685 section of modern file formats like ELF.
24688 * objfile-gdb.ext file: objfile-gdbdotext file. The @file{@var{objfile}-gdb.@var{ext}} file
24689 * .debug_gdb_scripts section: dotdebug_gdb_scripts section. The @code{.debug_gdb_scripts} section
24690 * Which flavor to choose?::
24693 The auto-loading feature is useful for supplying application-specific
24694 debugging commands and features.
24696 Auto-loading can be enabled or disabled,
24697 and the list of auto-loaded scripts can be printed.
24698 See the @samp{auto-loading} section of each extension language
24699 for more information.
24700 For @value{GDBN} command files see @ref{Auto-loading sequences}.
24701 For Python files see @ref{Python Auto-loading}.
24703 Note that loading of this script file also requires accordingly configured
24704 @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
24706 @node objfile-gdbdotext file
24707 @subsection The @file{@var{objfile}-gdb.@var{ext}} file
24708 @cindex @file{@var{objfile}-gdb.gdb}
24709 @cindex @file{@var{objfile}-gdb.py}
24710 @cindex @file{@var{objfile}-gdb.scm}
24712 When a new object file is read, @value{GDBN} looks for a file named
24713 @file{@var{objfile}-gdb.@var{ext}} (we call it @var{script-name} below),
24714 where @var{objfile} is the object file's name and
24715 where @var{ext} is the file extension for the extension language:
24718 @item @file{@var{objfile}-gdb.gdb}
24719 GDB's own command language
24720 @item @file{@var{objfile}-gdb.py}
24722 @item @file{@var{objfile}-gdb.scm}
24726 @var{script-name} is formed by ensuring that the file name of @var{objfile}
24727 is absolute, following all symlinks, and resolving @code{.} and @code{..}
24728 components, and appending the @file{-gdb.@var{ext}} suffix.
24729 If this file exists and is readable, @value{GDBN} will evaluate it as a
24730 script in the specified extension language.
24732 If this file does not exist, then @value{GDBN} will look for
24733 @var{script-name} file in all of the directories as specified below.
24735 Note that loading of these files requires an accordingly configured
24736 @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
24738 For object files using @file{.exe} suffix @value{GDBN} tries to load first the
24739 scripts normally according to its @file{.exe} filename. But if no scripts are
24740 found @value{GDBN} also tries script filenames matching the object file without
24741 its @file{.exe} suffix. This @file{.exe} stripping is case insensitive and it
24742 is attempted on any platform. This makes the script filenames compatible
24743 between Unix and MS-Windows hosts.
24746 @anchor{set auto-load scripts-directory}
24747 @kindex set auto-load scripts-directory
24748 @item set auto-load scripts-directory @r{[}@var{directories}@r{]}
24749 Control @value{GDBN} auto-loaded scripts location. Multiple directory entries
24750 may be delimited by the host platform path separator in use
24751 (@samp{:} on Unix, @samp{;} on MS-Windows and MS-DOS).
24753 Each entry here needs to be covered also by the security setting
24754 @code{set auto-load safe-path} (@pxref{set auto-load safe-path}).
24756 @anchor{with-auto-load-dir}
24757 This variable defaults to @file{$debugdir:$datadir/auto-load}. The default
24758 @code{set auto-load safe-path} value can be also overriden by @value{GDBN}
24759 configuration option @option{--with-auto-load-dir}.
24761 Any reference to @file{$debugdir} will get replaced by
24762 @var{debug-file-directory} value (@pxref{Separate Debug Files}) and any
24763 reference to @file{$datadir} will get replaced by @var{data-directory} which is
24764 determined at @value{GDBN} startup (@pxref{Data Files}). @file{$debugdir} and
24765 @file{$datadir} must be placed as a directory component --- either alone or
24766 delimited by @file{/} or @file{\} directory separators, depending on the host
24769 The list of directories uses path separator (@samp{:} on GNU and Unix
24770 systems, @samp{;} on MS-Windows and MS-DOS) to separate directories, similarly
24771 to the @env{PATH} environment variable.
24773 @anchor{show auto-load scripts-directory}
24774 @kindex show auto-load scripts-directory
24775 @item show auto-load scripts-directory
24776 Show @value{GDBN} auto-loaded scripts location.
24778 @anchor{add-auto-load-scripts-directory}
24779 @kindex add-auto-load-scripts-directory
24780 @item add-auto-load-scripts-directory @r{[}@var{directories}@dots{}@r{]}
24781 Add an entry (or list of entries) to the list of auto-loaded scripts locations.
24782 Multiple entries may be delimited by the host platform path separator in use.
24785 @value{GDBN} does not track which files it has already auto-loaded this way.
24786 @value{GDBN} will load the associated script every time the corresponding
24787 @var{objfile} is opened.
24788 So your @file{-gdb.@var{ext}} file should be careful to avoid errors if it
24789 is evaluated more than once.
24791 @node dotdebug_gdb_scripts section
24792 @subsection The @code{.debug_gdb_scripts} section
24793 @cindex @code{.debug_gdb_scripts} section
24795 For systems using file formats like ELF and COFF,
24796 when @value{GDBN} loads a new object file
24797 it will look for a special section named @code{.debug_gdb_scripts}.
24798 If this section exists, its contents is a list of null-terminated entries
24799 specifying scripts to load. Each entry begins with a non-null prefix byte that
24800 specifies the kind of entry, typically the extension language and whether the
24801 script is in a file or inlined in @code{.debug_gdb_scripts}.
24803 The following entries are supported:
24806 @item SECTION_SCRIPT_ID_PYTHON_FILE = 1
24807 @item SECTION_SCRIPT_ID_SCHEME_FILE = 3
24808 @item SECTION_SCRIPT_ID_PYTHON_TEXT = 4
24809 @item SECTION_SCRIPT_ID_SCHEME_TEXT = 6
24812 @subsubsection Script File Entries
24814 If the entry specifies a file, @value{GDBN} will look for the file first
24815 in the current directory and then along the source search path
24816 (@pxref{Source Path, ,Specifying Source Directories}),
24817 except that @file{$cdir} is not searched, since the compilation
24818 directory is not relevant to scripts.
24820 File entries can be placed in section @code{.debug_gdb_scripts} with,
24821 for example, this GCC macro for Python scripts.
24824 /* Note: The "MS" section flags are to remove duplicates. */
24825 #define DEFINE_GDB_PY_SCRIPT(script_name) \
24827 .pushsection \".debug_gdb_scripts\", \"MS\",@@progbits,1\n\
24828 .byte 1 /* Python */\n\
24829 .asciz \"" script_name "\"\n\
24835 For Guile scripts, replace @code{.byte 1} with @code{.byte 3}.
24836 Then one can reference the macro in a header or source file like this:
24839 DEFINE_GDB_PY_SCRIPT ("my-app-scripts.py")
24842 The script name may include directories if desired.
24844 Note that loading of this script file also requires accordingly configured
24845 @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
24847 If the macro invocation is put in a header, any application or library
24848 using this header will get a reference to the specified script,
24849 and with the use of @code{"MS"} attributes on the section, the linker
24850 will remove duplicates.
24852 @subsubsection Script Text Entries
24854 Script text entries allow to put the executable script in the entry
24855 itself instead of loading it from a file.
24856 The first line of the entry, everything after the prefix byte and up to
24857 the first newline (@code{0xa}) character, is the script name, and must not
24858 contain any kind of space character, e.g., spaces or tabs.
24859 The rest of the entry, up to the trailing null byte, is the script to
24860 execute in the specified language. The name needs to be unique among
24861 all script names, as @value{GDBN} executes each script only once based
24864 Here is an example from file @file{py-section-script.c} in the @value{GDBN}
24868 #include "symcat.h"
24869 #include "gdb/section-scripts.h"
24871 ".pushsection \".debug_gdb_scripts\", \"MS\",@@progbits,1\n"
24872 ".byte " XSTRING (SECTION_SCRIPT_ID_PYTHON_TEXT) "\n"
24873 ".ascii \"gdb.inlined-script\\n\"\n"
24874 ".ascii \"class test_cmd (gdb.Command):\\n\"\n"
24875 ".ascii \" def __init__ (self):\\n\"\n"
24876 ".ascii \" super (test_cmd, self).__init__ ("
24877 "\\\"test-cmd\\\", gdb.COMMAND_OBSCURE)\\n\"\n"
24878 ".ascii \" def invoke (self, arg, from_tty):\\n\"\n"
24879 ".ascii \" print (\\\"test-cmd output, arg = %s\\\" % arg)\\n\"\n"
24880 ".ascii \"test_cmd ()\\n\"\n"
24886 Loading of inlined scripts requires a properly configured
24887 @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
24888 The path to specify in @code{auto-load safe-path} is the path of the file
24889 containing the @code{.debug_gdb_scripts} section.
24891 @node Which flavor to choose?
24892 @subsection Which flavor to choose?
24894 Given the multiple ways of auto-loading extensions, it might not always
24895 be clear which one to choose. This section provides some guidance.
24898 Benefits of the @file{-gdb.@var{ext}} way:
24902 Can be used with file formats that don't support multiple sections.
24905 Ease of finding scripts for public libraries.
24907 Scripts specified in the @code{.debug_gdb_scripts} section are searched for
24908 in the source search path.
24909 For publicly installed libraries, e.g., @file{libstdc++}, there typically
24910 isn't a source directory in which to find the script.
24913 Doesn't require source code additions.
24917 Benefits of the @code{.debug_gdb_scripts} way:
24921 Works with static linking.
24923 Scripts for libraries done the @file{-gdb.@var{ext}} way require an objfile to
24924 trigger their loading. When an application is statically linked the only
24925 objfile available is the executable, and it is cumbersome to attach all the
24926 scripts from all the input libraries to the executable's
24927 @file{-gdb.@var{ext}} script.
24930 Works with classes that are entirely inlined.
24932 Some classes can be entirely inlined, and thus there may not be an associated
24933 shared library to attach a @file{-gdb.@var{ext}} script to.
24936 Scripts needn't be copied out of the source tree.
24938 In some circumstances, apps can be built out of large collections of internal
24939 libraries, and the build infrastructure necessary to install the
24940 @file{-gdb.@var{ext}} scripts in a place where @value{GDBN} can find them is
24941 cumbersome. It may be easier to specify the scripts in the
24942 @code{.debug_gdb_scripts} section as relative paths, and add a path to the
24943 top of the source tree to the source search path.
24946 @node Multiple Extension Languages
24947 @section Multiple Extension Languages
24949 The Guile and Python extension languages do not share any state,
24950 and generally do not interfere with each other.
24951 There are some things to be aware of, however.
24953 @subsection Python comes first
24955 Python was @value{GDBN}'s first extension language, and to avoid breaking
24956 existing behaviour Python comes first. This is generally solved by the
24957 ``first one wins'' principle. @value{GDBN} maintains a list of enabled
24958 extension languages, and when it makes a call to an extension language,
24959 (say to pretty-print a value), it tries each in turn until an extension
24960 language indicates it has performed the request (e.g., has returned the
24961 pretty-printed form of a value).
24962 This extends to errors while performing such requests: If an error happens
24963 while, for example, trying to pretty-print an object then the error is
24964 reported and any following extension languages are not tried.
24967 @section Creating new spellings of existing commands
24968 @cindex aliases for commands
24970 It is often useful to define alternate spellings of existing commands.
24971 For example, if a new @value{GDBN} command defined in Python has
24972 a long name to type, it is handy to have an abbreviated version of it
24973 that involves less typing.
24975 @value{GDBN} itself uses aliases. For example @samp{s} is an alias
24976 of the @samp{step} command even though it is otherwise an ambiguous
24977 abbreviation of other commands like @samp{set} and @samp{show}.
24979 Aliases are also used to provide shortened or more common versions
24980 of multi-word commands. For example, @value{GDBN} provides the
24981 @samp{tty} alias of the @samp{set inferior-tty} command.
24983 You can define a new alias with the @samp{alias} command.
24988 @item alias [-a] [--] @var{ALIAS} = @var{COMMAND}
24992 @var{ALIAS} specifies the name of the new alias.
24993 Each word of @var{ALIAS} must consist of letters, numbers, dashes and
24996 @var{COMMAND} specifies the name of an existing command
24997 that is being aliased.
24999 The @samp{-a} option specifies that the new alias is an abbreviation
25000 of the command. Abbreviations are not shown in command
25001 lists displayed by the @samp{help} command.
25003 The @samp{--} option specifies the end of options,
25004 and is useful when @var{ALIAS} begins with a dash.
25006 Here is a simple example showing how to make an abbreviation
25007 of a command so that there is less to type.
25008 Suppose you were tired of typing @samp{disas}, the current
25009 shortest unambiguous abbreviation of the @samp{disassemble} command
25010 and you wanted an even shorter version named @samp{di}.
25011 The following will accomplish this.
25014 (gdb) alias -a di = disas
25017 Note that aliases are different from user-defined commands.
25018 With a user-defined command, you also need to write documentation
25019 for it with the @samp{document} command.
25020 An alias automatically picks up the documentation of the existing command.
25022 Here is an example where we make @samp{elms} an abbreviation of
25023 @samp{elements} in the @samp{set print elements} command.
25024 This is to show that you can make an abbreviation of any part
25028 (gdb) alias -a set print elms = set print elements
25029 (gdb) alias -a show print elms = show print elements
25030 (gdb) set p elms 20
25032 Limit on string chars or array elements to print is 200.
25035 Note that if you are defining an alias of a @samp{set} command,
25036 and you want to have an alias for the corresponding @samp{show}
25037 command, then you need to define the latter separately.
25039 Unambiguously abbreviated commands are allowed in @var{COMMAND} and
25040 @var{ALIAS}, just as they are normally.
25043 (gdb) alias -a set pr elms = set p ele
25046 Finally, here is an example showing the creation of a one word
25047 alias for a more complex command.
25048 This creates alias @samp{spe} of the command @samp{set print elements}.
25051 (gdb) alias spe = set print elements
25056 @chapter Command Interpreters
25057 @cindex command interpreters
25059 @value{GDBN} supports multiple command interpreters, and some command
25060 infrastructure to allow users or user interface writers to switch
25061 between interpreters or run commands in other interpreters.
25063 @value{GDBN} currently supports two command interpreters, the console
25064 interpreter (sometimes called the command-line interpreter or @sc{cli})
25065 and the machine interface interpreter (or @sc{gdb/mi}). This manual
25066 describes both of these interfaces in great detail.
25068 By default, @value{GDBN} will start with the console interpreter.
25069 However, the user may choose to start @value{GDBN} with another
25070 interpreter by specifying the @option{-i} or @option{--interpreter}
25071 startup options. Defined interpreters include:
25075 @cindex console interpreter
25076 The traditional console or command-line interpreter. This is the most often
25077 used interpreter with @value{GDBN}. With no interpreter specified at runtime,
25078 @value{GDBN} will use this interpreter.
25081 @cindex mi interpreter
25082 The newest @sc{gdb/mi} interface (currently @code{mi2}). Used primarily
25083 by programs wishing to use @value{GDBN} as a backend for a debugger GUI
25084 or an IDE. For more information, see @ref{GDB/MI, ,The @sc{gdb/mi}
25088 @cindex mi2 interpreter
25089 The current @sc{gdb/mi} interface.
25092 @cindex mi1 interpreter
25093 The @sc{gdb/mi} interface included in @value{GDBN} 5.1, 5.2, and 5.3.
25097 @cindex invoke another interpreter
25099 @kindex interpreter-exec
25100 You may execute commands in any interpreter from the current
25101 interpreter using the appropriate command. If you are running the
25102 console interpreter, simply use the @code{interpreter-exec} command:
25105 interpreter-exec mi "-data-list-register-names"
25108 @sc{gdb/mi} has a similar command, although it is only available in versions of
25109 @value{GDBN} which support @sc{gdb/mi} version 2 (or greater).
25111 Note that @code{interpreter-exec} only changes the interpreter for the
25112 duration of the specified command. It does not change the interpreter
25115 @cindex start a new independent interpreter
25117 Although you may only choose a single interpreter at startup, it is
25118 possible to run an independent interpreter on a specified input/output
25119 device (usually a tty).
25121 For example, consider a debugger GUI or IDE that wants to provide a
25122 @value{GDBN} console view. It may do so by embedding a terminal
25123 emulator widget in its GUI, starting @value{GDBN} in the traditional
25124 command-line mode with stdin/stdout/stderr redirected to that
25125 terminal, and then creating an MI interpreter running on a specified
25126 input/output device. The console interpreter created by @value{GDBN}
25127 at startup handles commands the user types in the terminal widget,
25128 while the GUI controls and synchronizes state with @value{GDBN} using
25129 the separate MI interpreter.
25131 To start a new secondary @dfn{user interface} running MI, use the
25132 @code{new-ui} command:
25135 @cindex new user interface
25137 new-ui @var{interpreter} @var{tty}
25140 The @var{interpreter} parameter specifies the interpreter to run.
25141 This accepts the same values as the @code{interpreter-exec} command.
25142 For example, @samp{console}, @samp{mi}, @samp{mi2}, etc. The
25143 @var{tty} parameter specifies the name of the bidirectional file the
25144 interpreter uses for input/output, usually the name of a
25145 pseudoterminal slave on Unix systems. For example:
25148 (@value{GDBP}) new-ui mi /dev/pts/9
25152 runs an MI interpreter on @file{/dev/pts/9}.
25155 @chapter @value{GDBN} Text User Interface
25157 @cindex Text User Interface
25160 * TUI Overview:: TUI overview
25161 * TUI Keys:: TUI key bindings
25162 * TUI Single Key Mode:: TUI single key mode
25163 * TUI Commands:: TUI-specific commands
25164 * TUI Configuration:: TUI configuration variables
25167 The @value{GDBN} Text User Interface (TUI) is a terminal
25168 interface which uses the @code{curses} library to show the source
25169 file, the assembly output, the program registers and @value{GDBN}
25170 commands in separate text windows. The TUI mode is supported only
25171 on platforms where a suitable version of the @code{curses} library
25174 The TUI mode is enabled by default when you invoke @value{GDBN} as
25175 @samp{@value{GDBP} -tui}.
25176 You can also switch in and out of TUI mode while @value{GDBN} runs by
25177 using various TUI commands and key bindings, such as @command{tui
25178 enable} or @kbd{C-x C-a}. @xref{TUI Commands, ,TUI Commands}, and
25179 @ref{TUI Keys, ,TUI Key Bindings}.
25182 @section TUI Overview
25184 In TUI mode, @value{GDBN} can display several text windows:
25188 This window is the @value{GDBN} command window with the @value{GDBN}
25189 prompt and the @value{GDBN} output. The @value{GDBN} input is still
25190 managed using readline.
25193 The source window shows the source file of the program. The current
25194 line and active breakpoints are displayed in this window.
25197 The assembly window shows the disassembly output of the program.
25200 This window shows the processor registers. Registers are highlighted
25201 when their values change.
25204 The source and assembly windows show the current program position
25205 by highlighting the current line and marking it with a @samp{>} marker.
25206 Breakpoints are indicated with two markers. The first marker
25207 indicates the breakpoint type:
25211 Breakpoint which was hit at least once.
25214 Breakpoint which was never hit.
25217 Hardware breakpoint which was hit at least once.
25220 Hardware breakpoint which was never hit.
25223 The second marker indicates whether the breakpoint is enabled or not:
25227 Breakpoint is enabled.
25230 Breakpoint is disabled.
25233 The source, assembly and register windows are updated when the current
25234 thread changes, when the frame changes, or when the program counter
25237 These windows are not all visible at the same time. The command
25238 window is always visible. The others can be arranged in several
25249 source and assembly,
25252 source and registers, or
25255 assembly and registers.
25258 A status line above the command window shows the following information:
25262 Indicates the current @value{GDBN} target.
25263 (@pxref{Targets, ,Specifying a Debugging Target}).
25266 Gives the current process or thread number.
25267 When no process is being debugged, this field is set to @code{No process}.
25270 Gives the current function name for the selected frame.
25271 The name is demangled if demangling is turned on (@pxref{Print Settings}).
25272 When there is no symbol corresponding to the current program counter,
25273 the string @code{??} is displayed.
25276 Indicates the current line number for the selected frame.
25277 When the current line number is not known, the string @code{??} is displayed.
25280 Indicates the current program counter address.
25284 @section TUI Key Bindings
25285 @cindex TUI key bindings
25287 The TUI installs several key bindings in the readline keymaps
25288 @ifset SYSTEM_READLINE
25289 (@pxref{Command Line Editing, , , rluserman, GNU Readline Library}).
25291 @ifclear SYSTEM_READLINE
25292 (@pxref{Command Line Editing}).
25294 The following key bindings are installed for both TUI mode and the
25295 @value{GDBN} standard mode.
25304 Enter or leave the TUI mode. When leaving the TUI mode,
25305 the curses window management stops and @value{GDBN} operates using
25306 its standard mode, writing on the terminal directly. When reentering
25307 the TUI mode, control is given back to the curses windows.
25308 The screen is then refreshed.
25312 Use a TUI layout with only one window. The layout will
25313 either be @samp{source} or @samp{assembly}. When the TUI mode
25314 is not active, it will switch to the TUI mode.
25316 Think of this key binding as the Emacs @kbd{C-x 1} binding.
25320 Use a TUI layout with at least two windows. When the current
25321 layout already has two windows, the next layout with two windows is used.
25322 When a new layout is chosen, one window will always be common to the
25323 previous layout and the new one.
25325 Think of it as the Emacs @kbd{C-x 2} binding.
25329 Change the active window. The TUI associates several key bindings
25330 (like scrolling and arrow keys) with the active window. This command
25331 gives the focus to the next TUI window.
25333 Think of it as the Emacs @kbd{C-x o} binding.
25337 Switch in and out of the TUI SingleKey mode that binds single
25338 keys to @value{GDBN} commands (@pxref{TUI Single Key Mode}).
25341 The following key bindings only work in the TUI mode:
25346 Scroll the active window one page up.
25350 Scroll the active window one page down.
25354 Scroll the active window one line up.
25358 Scroll the active window one line down.
25362 Scroll the active window one column left.
25366 Scroll the active window one column right.
25370 Refresh the screen.
25373 Because the arrow keys scroll the active window in the TUI mode, they
25374 are not available for their normal use by readline unless the command
25375 window has the focus. When another window is active, you must use
25376 other readline key bindings such as @kbd{C-p}, @kbd{C-n}, @kbd{C-b}
25377 and @kbd{C-f} to control the command window.
25379 @node TUI Single Key Mode
25380 @section TUI Single Key Mode
25381 @cindex TUI single key mode
25383 The TUI also provides a @dfn{SingleKey} mode, which binds several
25384 frequently used @value{GDBN} commands to single keys. Type @kbd{C-x s} to
25385 switch into this mode, where the following key bindings are used:
25388 @kindex c @r{(SingleKey TUI key)}
25392 @kindex d @r{(SingleKey TUI key)}
25396 @kindex f @r{(SingleKey TUI key)}
25400 @kindex n @r{(SingleKey TUI key)}
25404 @kindex q @r{(SingleKey TUI key)}
25406 exit the SingleKey mode.
25408 @kindex r @r{(SingleKey TUI key)}
25412 @kindex s @r{(SingleKey TUI key)}
25416 @kindex u @r{(SingleKey TUI key)}
25420 @kindex v @r{(SingleKey TUI key)}
25424 @kindex w @r{(SingleKey TUI key)}
25429 Other keys temporarily switch to the @value{GDBN} command prompt.
25430 The key that was pressed is inserted in the editing buffer so that
25431 it is possible to type most @value{GDBN} commands without interaction
25432 with the TUI SingleKey mode. Once the command is entered the TUI
25433 SingleKey mode is restored. The only way to permanently leave
25434 this mode is by typing @kbd{q} or @kbd{C-x s}.
25438 @section TUI-specific Commands
25439 @cindex TUI commands
25441 The TUI has specific commands to control the text windows.
25442 These commands are always available, even when @value{GDBN} is not in
25443 the TUI mode. When @value{GDBN} is in the standard mode, most
25444 of these commands will automatically switch to the TUI mode.
25446 Note that if @value{GDBN}'s @code{stdout} is not connected to a
25447 terminal, or @value{GDBN} has been started with the machine interface
25448 interpreter (@pxref{GDB/MI, ,The @sc{gdb/mi} Interface}), most of
25449 these commands will fail with an error, because it would not be
25450 possible or desirable to enable curses window management.
25455 Activate TUI mode. The last active TUI window layout will be used if
25456 TUI mode has prevsiouly been used in the current debugging session,
25457 otherwise a default layout is used.
25460 @kindex tui disable
25461 Disable TUI mode, returning to the console interpreter.
25465 List and give the size of all displayed windows.
25467 @item layout @var{name}
25469 Changes which TUI windows are displayed. In each layout the command
25470 window is always displayed, the @var{name} parameter controls which
25471 additional windows are displayed, and can be any of the following:
25475 Display the next layout.
25478 Display the previous layout.
25481 Display the source and command windows.
25484 Display the assembly and command windows.
25487 Display the source, assembly, and command windows.
25490 When in @code{src} layout display the register, source, and command
25491 windows. When in @code{asm} or @code{split} layout display the
25492 register, assembler, and command windows.
25495 @item focus @var{name}
25497 Changes which TUI window is currently active for scrolling. The
25498 @var{name} parameter can be any of the following:
25502 Make the next window active for scrolling.
25505 Make the previous window active for scrolling.
25508 Make the source window active for scrolling.
25511 Make the assembly window active for scrolling.
25514 Make the register window active for scrolling.
25517 Make the command window active for scrolling.
25522 Refresh the screen. This is similar to typing @kbd{C-L}.
25524 @item tui reg @var{group}
25526 Changes the register group displayed in the tui register window to
25527 @var{group}. If the register window is not currently displayed this
25528 command will cause the register window to be displayed. The list of
25529 register groups, as well as their order is target specific. The
25530 following groups are available on most targets:
25533 Repeatedly selecting this group will cause the display to cycle
25534 through all of the available register groups.
25537 Repeatedly selecting this group will cause the display to cycle
25538 through all of the available register groups in the reverse order to
25542 Display the general registers.
25544 Display the floating point registers.
25546 Display the system registers.
25548 Display the vector registers.
25550 Display all registers.
25555 Update the source window and the current execution point.
25557 @item winheight @var{name} +@var{count}
25558 @itemx winheight @var{name} -@var{count}
25560 Change the height of the window @var{name} by @var{count}
25561 lines. Positive counts increase the height, while negative counts
25562 decrease it. The @var{name} parameter can be one of @code{src} (the
25563 source window), @code{cmd} (the command window), @code{asm} (the
25564 disassembly window), or @code{regs} (the register display window).
25566 @item tabset @var{nchars}
25568 Set the width of tab stops to be @var{nchars} characters. This
25569 setting affects the display of TAB characters in the source and
25573 @node TUI Configuration
25574 @section TUI Configuration Variables
25575 @cindex TUI configuration variables
25577 Several configuration variables control the appearance of TUI windows.
25580 @item set tui border-kind @var{kind}
25581 @kindex set tui border-kind
25582 Select the border appearance for the source, assembly and register windows.
25583 The possible values are the following:
25586 Use a space character to draw the border.
25589 Use @sc{ascii} characters @samp{+}, @samp{-} and @samp{|} to draw the border.
25592 Use the Alternate Character Set to draw the border. The border is
25593 drawn using character line graphics if the terminal supports them.
25596 @item set tui border-mode @var{mode}
25597 @kindex set tui border-mode
25598 @itemx set tui active-border-mode @var{mode}
25599 @kindex set tui active-border-mode
25600 Select the display attributes for the borders of the inactive windows
25601 or the active window. The @var{mode} can be one of the following:
25604 Use normal attributes to display the border.
25610 Use reverse video mode.
25613 Use half bright mode.
25615 @item half-standout
25616 Use half bright and standout mode.
25619 Use extra bright or bold mode.
25621 @item bold-standout
25622 Use extra bright or bold and standout mode.
25627 @chapter Using @value{GDBN} under @sc{gnu} Emacs
25630 @cindex @sc{gnu} Emacs
25631 A special interface allows you to use @sc{gnu} Emacs to view (and
25632 edit) the source files for the program you are debugging with
25635 To use this interface, use the command @kbd{M-x gdb} in Emacs. Give the
25636 executable file you want to debug as an argument. This command starts
25637 @value{GDBN} as a subprocess of Emacs, with input and output through a newly
25638 created Emacs buffer.
25639 @c (Do not use the @code{-tui} option to run @value{GDBN} from Emacs.)
25641 Running @value{GDBN} under Emacs can be just like running @value{GDBN} normally except for two
25646 All ``terminal'' input and output goes through an Emacs buffer, called
25649 This applies both to @value{GDBN} commands and their output, and to the input
25650 and output done by the program you are debugging.
25652 This is useful because it means that you can copy the text of previous
25653 commands and input them again; you can even use parts of the output
25656 All the facilities of Emacs' Shell mode are available for interacting
25657 with your program. In particular, you can send signals the usual
25658 way---for example, @kbd{C-c C-c} for an interrupt, @kbd{C-c C-z} for a
25662 @value{GDBN} displays source code through Emacs.
25664 Each time @value{GDBN} displays a stack frame, Emacs automatically finds the
25665 source file for that frame and puts an arrow (@samp{=>}) at the
25666 left margin of the current line. Emacs uses a separate buffer for
25667 source display, and splits the screen to show both your @value{GDBN} session
25670 Explicit @value{GDBN} @code{list} or search commands still produce output as
25671 usual, but you probably have no reason to use them from Emacs.
25674 We call this @dfn{text command mode}. Emacs 22.1, and later, also uses
25675 a graphical mode, enabled by default, which provides further buffers
25676 that can control the execution and describe the state of your program.
25677 @xref{GDB Graphical Interface,,, Emacs, The @sc{gnu} Emacs Manual}.
25679 If you specify an absolute file name when prompted for the @kbd{M-x
25680 gdb} argument, then Emacs sets your current working directory to where
25681 your program resides. If you only specify the file name, then Emacs
25682 sets your current working directory to the directory associated
25683 with the previous buffer. In this case, @value{GDBN} may find your
25684 program by searching your environment's @code{PATH} variable, but on
25685 some operating systems it might not find the source. So, although the
25686 @value{GDBN} input and output session proceeds normally, the auxiliary
25687 buffer does not display the current source and line of execution.
25689 The initial working directory of @value{GDBN} is printed on the top
25690 line of the GUD buffer and this serves as a default for the commands
25691 that specify files for @value{GDBN} to operate on. @xref{Files,
25692 ,Commands to Specify Files}.
25694 By default, @kbd{M-x gdb} calls the program called @file{gdb}. If you
25695 need to call @value{GDBN} by a different name (for example, if you
25696 keep several configurations around, with different names) you can
25697 customize the Emacs variable @code{gud-gdb-command-name} to run the
25700 In the GUD buffer, you can use these special Emacs commands in
25701 addition to the standard Shell mode commands:
25705 Describe the features of Emacs' GUD Mode.
25708 Execute to another source line, like the @value{GDBN} @code{step} command; also
25709 update the display window to show the current file and location.
25712 Execute to next source line in this function, skipping all function
25713 calls, like the @value{GDBN} @code{next} command. Then update the display window
25714 to show the current file and location.
25717 Execute one instruction, like the @value{GDBN} @code{stepi} command; update
25718 display window accordingly.
25721 Execute until exit from the selected stack frame, like the @value{GDBN}
25722 @code{finish} command.
25725 Continue execution of your program, like the @value{GDBN} @code{continue}
25729 Go up the number of frames indicated by the numeric argument
25730 (@pxref{Arguments, , Numeric Arguments, Emacs, The @sc{gnu} Emacs Manual}),
25731 like the @value{GDBN} @code{up} command.
25734 Go down the number of frames indicated by the numeric argument, like the
25735 @value{GDBN} @code{down} command.
25738 In any source file, the Emacs command @kbd{C-x @key{SPC}} (@code{gud-break})
25739 tells @value{GDBN} to set a breakpoint on the source line point is on.
25741 In text command mode, if you type @kbd{M-x speedbar}, Emacs displays a
25742 separate frame which shows a backtrace when the GUD buffer is current.
25743 Move point to any frame in the stack and type @key{RET} to make it
25744 become the current frame and display the associated source in the
25745 source buffer. Alternatively, click @kbd{Mouse-2} to make the
25746 selected frame become the current one. In graphical mode, the
25747 speedbar displays watch expressions.
25749 If you accidentally delete the source-display buffer, an easy way to get
25750 it back is to type the command @code{f} in the @value{GDBN} buffer, to
25751 request a frame display; when you run under Emacs, this recreates
25752 the source buffer if necessary to show you the context of the current
25755 The source files displayed in Emacs are in ordinary Emacs buffers
25756 which are visiting the source files in the usual way. You can edit
25757 the files with these buffers if you wish; but keep in mind that @value{GDBN}
25758 communicates with Emacs in terms of line numbers. If you add or
25759 delete lines from the text, the line numbers that @value{GDBN} knows cease
25760 to correspond properly with the code.
25762 A more detailed description of Emacs' interaction with @value{GDBN} is
25763 given in the Emacs manual (@pxref{Debuggers,,, Emacs, The @sc{gnu}
25767 @chapter The @sc{gdb/mi} Interface
25769 @unnumberedsec Function and Purpose
25771 @cindex @sc{gdb/mi}, its purpose
25772 @sc{gdb/mi} is a line based machine oriented text interface to
25773 @value{GDBN} and is activated by specifying using the
25774 @option{--interpreter} command line option (@pxref{Mode Options}). It
25775 is specifically intended to support the development of systems which
25776 use the debugger as just one small component of a larger system.
25778 This chapter is a specification of the @sc{gdb/mi} interface. It is written
25779 in the form of a reference manual.
25781 Note that @sc{gdb/mi} is still under construction, so some of the
25782 features described below are incomplete and subject to change
25783 (@pxref{GDB/MI Development and Front Ends, , @sc{gdb/mi} Development and Front Ends}).
25785 @unnumberedsec Notation and Terminology
25787 @cindex notational conventions, for @sc{gdb/mi}
25788 This chapter uses the following notation:
25792 @code{|} separates two alternatives.
25795 @code{[ @var{something} ]} indicates that @var{something} is optional:
25796 it may or may not be given.
25799 @code{( @var{group} )*} means that @var{group} inside the parentheses
25800 may repeat zero or more times.
25803 @code{( @var{group} )+} means that @var{group} inside the parentheses
25804 may repeat one or more times.
25807 @code{"@var{string}"} means a literal @var{string}.
25811 @heading Dependencies
25815 * GDB/MI General Design::
25816 * GDB/MI Command Syntax::
25817 * GDB/MI Compatibility with CLI::
25818 * GDB/MI Development and Front Ends::
25819 * GDB/MI Output Records::
25820 * GDB/MI Simple Examples::
25821 * GDB/MI Command Description Format::
25822 * GDB/MI Breakpoint Commands::
25823 * GDB/MI Catchpoint Commands::
25824 * GDB/MI Program Context::
25825 * GDB/MI Thread Commands::
25826 * GDB/MI Ada Tasking Commands::
25827 * GDB/MI Program Execution::
25828 * GDB/MI Stack Manipulation::
25829 * GDB/MI Variable Objects::
25830 * GDB/MI Data Manipulation::
25831 * GDB/MI Tracepoint Commands::
25832 * GDB/MI Symbol Query::
25833 * GDB/MI File Commands::
25835 * GDB/MI Kod Commands::
25836 * GDB/MI Memory Overlay Commands::
25837 * GDB/MI Signal Handling Commands::
25839 * GDB/MI Target Manipulation::
25840 * GDB/MI File Transfer Commands::
25841 * GDB/MI Ada Exceptions Commands::
25842 * GDB/MI Support Commands::
25843 * GDB/MI Miscellaneous Commands::
25846 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
25847 @node GDB/MI General Design
25848 @section @sc{gdb/mi} General Design
25849 @cindex GDB/MI General Design
25851 Interaction of a @sc{GDB/MI} frontend with @value{GDBN} involves three
25852 parts---commands sent to @value{GDBN}, responses to those commands
25853 and notifications. Each command results in exactly one response,
25854 indicating either successful completion of the command, or an error.
25855 For the commands that do not resume the target, the response contains the
25856 requested information. For the commands that resume the target, the
25857 response only indicates whether the target was successfully resumed.
25858 Notifications is the mechanism for reporting changes in the state of the
25859 target, or in @value{GDBN} state, that cannot conveniently be associated with
25860 a command and reported as part of that command response.
25862 The important examples of notifications are:
25866 Exec notifications. These are used to report changes in
25867 target state---when a target is resumed, or stopped. It would not
25868 be feasible to include this information in response of resuming
25869 commands, because one resume commands can result in multiple events in
25870 different threads. Also, quite some time may pass before any event
25871 happens in the target, while a frontend needs to know whether the resuming
25872 command itself was successfully executed.
25875 Console output, and status notifications. Console output
25876 notifications are used to report output of CLI commands, as well as
25877 diagnostics for other commands. Status notifications are used to
25878 report the progress of a long-running operation. Naturally, including
25879 this information in command response would mean no output is produced
25880 until the command is finished, which is undesirable.
25883 General notifications. Commands may have various side effects on
25884 the @value{GDBN} or target state beyond their official purpose. For example,
25885 a command may change the selected thread. Although such changes can
25886 be included in command response, using notification allows for more
25887 orthogonal frontend design.
25891 There's no guarantee that whenever an MI command reports an error,
25892 @value{GDBN} or the target are in any specific state, and especially,
25893 the state is not reverted to the state before the MI command was
25894 processed. Therefore, whenever an MI command results in an error,
25895 we recommend that the frontend refreshes all the information shown in
25896 the user interface.
25900 * Context management::
25901 * Asynchronous and non-stop modes::
25905 @node Context management
25906 @subsection Context management
25908 @subsubsection Threads and Frames
25910 In most cases when @value{GDBN} accesses the target, this access is
25911 done in context of a specific thread and frame (@pxref{Frames}).
25912 Often, even when accessing global data, the target requires that a thread
25913 be specified. The CLI interface maintains the selected thread and frame,
25914 and supplies them to target on each command. This is convenient,
25915 because a command line user would not want to specify that information
25916 explicitly on each command, and because user interacts with
25917 @value{GDBN} via a single terminal, so no confusion is possible as
25918 to what thread and frame are the current ones.
25920 In the case of MI, the concept of selected thread and frame is less
25921 useful. First, a frontend can easily remember this information
25922 itself. Second, a graphical frontend can have more than one window,
25923 each one used for debugging a different thread, and the frontend might
25924 want to access additional threads for internal purposes. This
25925 increases the risk that by relying on implicitly selected thread, the
25926 frontend may be operating on a wrong one. Therefore, each MI command
25927 should explicitly specify which thread and frame to operate on. To
25928 make it possible, each MI command accepts the @samp{--thread} and
25929 @samp{--frame} options, the value to each is @value{GDBN} global
25930 identifier for thread and frame to operate on.
25932 Usually, each top-level window in a frontend allows the user to select
25933 a thread and a frame, and remembers the user selection for further
25934 operations. However, in some cases @value{GDBN} may suggest that the
25935 current thread or frame be changed. For example, when stopping on a
25936 breakpoint it is reasonable to switch to the thread where breakpoint is
25937 hit. For another example, if the user issues the CLI @samp{thread} or
25938 @samp{frame} commands via the frontend, it is desirable to change the
25939 frontend's selection to the one specified by user. @value{GDBN}
25940 communicates the suggestion to change current thread and frame using the
25941 @samp{=thread-selected} notification.
25943 Note that historically, MI shares the selected thread with CLI, so
25944 frontends used the @code{-thread-select} to execute commands in the
25945 right context. However, getting this to work right is cumbersome. The
25946 simplest way is for frontend to emit @code{-thread-select} command
25947 before every command. This doubles the number of commands that need
25948 to be sent. The alternative approach is to suppress @code{-thread-select}
25949 if the selected thread in @value{GDBN} is supposed to be identical to the
25950 thread the frontend wants to operate on. However, getting this
25951 optimization right can be tricky. In particular, if the frontend
25952 sends several commands to @value{GDBN}, and one of the commands changes the
25953 selected thread, then the behaviour of subsequent commands will
25954 change. So, a frontend should either wait for response from such
25955 problematic commands, or explicitly add @code{-thread-select} for
25956 all subsequent commands. No frontend is known to do this exactly
25957 right, so it is suggested to just always pass the @samp{--thread} and
25958 @samp{--frame} options.
25960 @subsubsection Language
25962 The execution of several commands depends on which language is selected.
25963 By default, the current language (@pxref{show language}) is used.
25964 But for commands known to be language-sensitive, it is recommended
25965 to use the @samp{--language} option. This option takes one argument,
25966 which is the name of the language to use while executing the command.
25970 -data-evaluate-expression --language c "sizeof (void*)"
25975 The valid language names are the same names accepted by the
25976 @samp{set language} command (@pxref{Manually}), excluding @samp{auto},
25977 @samp{local} or @samp{unknown}.
25979 @node Asynchronous and non-stop modes
25980 @subsection Asynchronous command execution and non-stop mode
25982 On some targets, @value{GDBN} is capable of processing MI commands
25983 even while the target is running. This is called @dfn{asynchronous
25984 command execution} (@pxref{Background Execution}). The frontend may
25985 specify a preferrence for asynchronous execution using the
25986 @code{-gdb-set mi-async 1} command, which should be emitted before
25987 either running the executable or attaching to the target. After the
25988 frontend has started the executable or attached to the target, it can
25989 find if asynchronous execution is enabled using the
25990 @code{-list-target-features} command.
25993 @item -gdb-set mi-async on
25994 @item -gdb-set mi-async off
25995 Set whether MI is in asynchronous mode.
25997 When @code{off}, which is the default, MI execution commands (e.g.,
25998 @code{-exec-continue}) are foreground commands, and @value{GDBN} waits
25999 for the program to stop before processing further commands.
26001 When @code{on}, MI execution commands are background execution
26002 commands (e.g., @code{-exec-continue} becomes the equivalent of the
26003 @code{c&} CLI command), and so @value{GDBN} is capable of processing
26004 MI commands even while the target is running.
26006 @item -gdb-show mi-async
26007 Show whether MI asynchronous mode is enabled.
26010 Note: In @value{GDBN} version 7.7 and earlier, this option was called
26011 @code{target-async} instead of @code{mi-async}, and it had the effect
26012 of both putting MI in asynchronous mode and making CLI background
26013 commands possible. CLI background commands are now always possible
26014 ``out of the box'' if the target supports them. The old spelling is
26015 kept as a deprecated alias for backwards compatibility.
26017 Even if @value{GDBN} can accept a command while target is running,
26018 many commands that access the target do not work when the target is
26019 running. Therefore, asynchronous command execution is most useful
26020 when combined with non-stop mode (@pxref{Non-Stop Mode}). Then,
26021 it is possible to examine the state of one thread, while other threads
26024 When a given thread is running, MI commands that try to access the
26025 target in the context of that thread may not work, or may work only on
26026 some targets. In particular, commands that try to operate on thread's
26027 stack will not work, on any target. Commands that read memory, or
26028 modify breakpoints, may work or not work, depending on the target. Note
26029 that even commands that operate on global state, such as @code{print},
26030 @code{set}, and breakpoint commands, still access the target in the
26031 context of a specific thread, so frontend should try to find a
26032 stopped thread and perform the operation on that thread (using the
26033 @samp{--thread} option).
26035 Which commands will work in the context of a running thread is
26036 highly target dependent. However, the two commands
26037 @code{-exec-interrupt}, to stop a thread, and @code{-thread-info},
26038 to find the state of a thread, will always work.
26040 @node Thread groups
26041 @subsection Thread groups
26042 @value{GDBN} may be used to debug several processes at the same time.
26043 On some platfroms, @value{GDBN} may support debugging of several
26044 hardware systems, each one having several cores with several different
26045 processes running on each core. This section describes the MI
26046 mechanism to support such debugging scenarios.
26048 The key observation is that regardless of the structure of the
26049 target, MI can have a global list of threads, because most commands that
26050 accept the @samp{--thread} option do not need to know what process that
26051 thread belongs to. Therefore, it is not necessary to introduce
26052 neither additional @samp{--process} option, nor an notion of the
26053 current process in the MI interface. The only strictly new feature
26054 that is required is the ability to find how the threads are grouped
26057 To allow the user to discover such grouping, and to support arbitrary
26058 hierarchy of machines/cores/processes, MI introduces the concept of a
26059 @dfn{thread group}. Thread group is a collection of threads and other
26060 thread groups. A thread group always has a string identifier, a type,
26061 and may have additional attributes specific to the type. A new
26062 command, @code{-list-thread-groups}, returns the list of top-level
26063 thread groups, which correspond to processes that @value{GDBN} is
26064 debugging at the moment. By passing an identifier of a thread group
26065 to the @code{-list-thread-groups} command, it is possible to obtain
26066 the members of specific thread group.
26068 To allow the user to easily discover processes, and other objects, he
26069 wishes to debug, a concept of @dfn{available thread group} is
26070 introduced. Available thread group is an thread group that
26071 @value{GDBN} is not debugging, but that can be attached to, using the
26072 @code{-target-attach} command. The list of available top-level thread
26073 groups can be obtained using @samp{-list-thread-groups --available}.
26074 In general, the content of a thread group may be only retrieved only
26075 after attaching to that thread group.
26077 Thread groups are related to inferiors (@pxref{Inferiors and
26078 Programs}). Each inferior corresponds to a thread group of a special
26079 type @samp{process}, and some additional operations are permitted on
26080 such thread groups.
26082 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
26083 @node GDB/MI Command Syntax
26084 @section @sc{gdb/mi} Command Syntax
26087 * GDB/MI Input Syntax::
26088 * GDB/MI Output Syntax::
26091 @node GDB/MI Input Syntax
26092 @subsection @sc{gdb/mi} Input Syntax
26094 @cindex input syntax for @sc{gdb/mi}
26095 @cindex @sc{gdb/mi}, input syntax
26097 @item @var{command} @expansion{}
26098 @code{@var{cli-command} | @var{mi-command}}
26100 @item @var{cli-command} @expansion{}
26101 @code{[ @var{token} ] @var{cli-command} @var{nl}}, where
26102 @var{cli-command} is any existing @value{GDBN} CLI command.
26104 @item @var{mi-command} @expansion{}
26105 @code{[ @var{token} ] "-" @var{operation} ( " " @var{option} )*
26106 @code{[} " --" @code{]} ( " " @var{parameter} )* @var{nl}}
26108 @item @var{token} @expansion{}
26109 "any sequence of digits"
26111 @item @var{option} @expansion{}
26112 @code{"-" @var{parameter} [ " " @var{parameter} ]}
26114 @item @var{parameter} @expansion{}
26115 @code{@var{non-blank-sequence} | @var{c-string}}
26117 @item @var{operation} @expansion{}
26118 @emph{any of the operations described in this chapter}
26120 @item @var{non-blank-sequence} @expansion{}
26121 @emph{anything, provided it doesn't contain special characters such as
26122 "-", @var{nl}, """ and of course " "}
26124 @item @var{c-string} @expansion{}
26125 @code{""" @var{seven-bit-iso-c-string-content} """}
26127 @item @var{nl} @expansion{}
26136 The CLI commands are still handled by the @sc{mi} interpreter; their
26137 output is described below.
26140 The @code{@var{token}}, when present, is passed back when the command
26144 Some @sc{mi} commands accept optional arguments as part of the parameter
26145 list. Each option is identified by a leading @samp{-} (dash) and may be
26146 followed by an optional argument parameter. Options occur first in the
26147 parameter list and can be delimited from normal parameters using
26148 @samp{--} (this is useful when some parameters begin with a dash).
26155 We want easy access to the existing CLI syntax (for debugging).
26158 We want it to be easy to spot a @sc{mi} operation.
26161 @node GDB/MI Output Syntax
26162 @subsection @sc{gdb/mi} Output Syntax
26164 @cindex output syntax of @sc{gdb/mi}
26165 @cindex @sc{gdb/mi}, output syntax
26166 The output from @sc{gdb/mi} consists of zero or more out-of-band records
26167 followed, optionally, by a single result record. This result record
26168 is for the most recent command. The sequence of output records is
26169 terminated by @samp{(gdb)}.
26171 If an input command was prefixed with a @code{@var{token}} then the
26172 corresponding output for that command will also be prefixed by that same
26176 @item @var{output} @expansion{}
26177 @code{( @var{out-of-band-record} )* [ @var{result-record} ] "(gdb)" @var{nl}}
26179 @item @var{result-record} @expansion{}
26180 @code{ [ @var{token} ] "^" @var{result-class} ( "," @var{result} )* @var{nl}}
26182 @item @var{out-of-band-record} @expansion{}
26183 @code{@var{async-record} | @var{stream-record}}
26185 @item @var{async-record} @expansion{}
26186 @code{@var{exec-async-output} | @var{status-async-output} | @var{notify-async-output}}
26188 @item @var{exec-async-output} @expansion{}
26189 @code{[ @var{token} ] "*" @var{async-output nl}}
26191 @item @var{status-async-output} @expansion{}
26192 @code{[ @var{token} ] "+" @var{async-output nl}}
26194 @item @var{notify-async-output} @expansion{}
26195 @code{[ @var{token} ] "=" @var{async-output nl}}
26197 @item @var{async-output} @expansion{}
26198 @code{@var{async-class} ( "," @var{result} )*}
26200 @item @var{result-class} @expansion{}
26201 @code{"done" | "running" | "connected" | "error" | "exit"}
26203 @item @var{async-class} @expansion{}
26204 @code{"stopped" | @var{others}} (where @var{others} will be added
26205 depending on the needs---this is still in development).
26207 @item @var{result} @expansion{}
26208 @code{ @var{variable} "=" @var{value}}
26210 @item @var{variable} @expansion{}
26211 @code{ @var{string} }
26213 @item @var{value} @expansion{}
26214 @code{ @var{const} | @var{tuple} | @var{list} }
26216 @item @var{const} @expansion{}
26217 @code{@var{c-string}}
26219 @item @var{tuple} @expansion{}
26220 @code{ "@{@}" | "@{" @var{result} ( "," @var{result} )* "@}" }
26222 @item @var{list} @expansion{}
26223 @code{ "[]" | "[" @var{value} ( "," @var{value} )* "]" | "["
26224 @var{result} ( "," @var{result} )* "]" }
26226 @item @var{stream-record} @expansion{}
26227 @code{@var{console-stream-output} | @var{target-stream-output} | @var{log-stream-output}}
26229 @item @var{console-stream-output} @expansion{}
26230 @code{"~" @var{c-string nl}}
26232 @item @var{target-stream-output} @expansion{}
26233 @code{"@@" @var{c-string nl}}
26235 @item @var{log-stream-output} @expansion{}
26236 @code{"&" @var{c-string nl}}
26238 @item @var{nl} @expansion{}
26241 @item @var{token} @expansion{}
26242 @emph{any sequence of digits}.
26250 All output sequences end in a single line containing a period.
26253 The @code{@var{token}} is from the corresponding request. Note that
26254 for all async output, while the token is allowed by the grammar and
26255 may be output by future versions of @value{GDBN} for select async
26256 output messages, it is generally omitted. Frontends should treat
26257 all async output as reporting general changes in the state of the
26258 target and there should be no need to associate async output to any
26262 @cindex status output in @sc{gdb/mi}
26263 @var{status-async-output} contains on-going status information about the
26264 progress of a slow operation. It can be discarded. All status output is
26265 prefixed by @samp{+}.
26268 @cindex async output in @sc{gdb/mi}
26269 @var{exec-async-output} contains asynchronous state change on the target
26270 (stopped, started, disappeared). All async output is prefixed by
26274 @cindex notify output in @sc{gdb/mi}
26275 @var{notify-async-output} contains supplementary information that the
26276 client should handle (e.g., a new breakpoint information). All notify
26277 output is prefixed by @samp{=}.
26280 @cindex console output in @sc{gdb/mi}
26281 @var{console-stream-output} is output that should be displayed as is in the
26282 console. It is the textual response to a CLI command. All the console
26283 output is prefixed by @samp{~}.
26286 @cindex target output in @sc{gdb/mi}
26287 @var{target-stream-output} is the output produced by the target program.
26288 All the target output is prefixed by @samp{@@}.
26291 @cindex log output in @sc{gdb/mi}
26292 @var{log-stream-output} is output text coming from @value{GDBN}'s internals, for
26293 instance messages that should be displayed as part of an error log. All
26294 the log output is prefixed by @samp{&}.
26297 @cindex list output in @sc{gdb/mi}
26298 New @sc{gdb/mi} commands should only output @var{lists} containing
26304 @xref{GDB/MI Stream Records, , @sc{gdb/mi} Stream Records}, for more
26305 details about the various output records.
26307 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
26308 @node GDB/MI Compatibility with CLI
26309 @section @sc{gdb/mi} Compatibility with CLI
26311 @cindex compatibility, @sc{gdb/mi} and CLI
26312 @cindex @sc{gdb/mi}, compatibility with CLI
26314 For the developers convenience CLI commands can be entered directly,
26315 but there may be some unexpected behaviour. For example, commands
26316 that query the user will behave as if the user replied yes, breakpoint
26317 command lists are not executed and some CLI commands, such as
26318 @code{if}, @code{when} and @code{define}, prompt for further input with
26319 @samp{>}, which is not valid MI output.
26321 This feature may be removed at some stage in the future and it is
26322 recommended that front ends use the @code{-interpreter-exec} command
26323 (@pxref{-interpreter-exec}).
26325 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
26326 @node GDB/MI Development and Front Ends
26327 @section @sc{gdb/mi} Development and Front Ends
26328 @cindex @sc{gdb/mi} development
26330 The application which takes the MI output and presents the state of the
26331 program being debugged to the user is called a @dfn{front end}.
26333 Although @sc{gdb/mi} is still incomplete, it is currently being used
26334 by a variety of front ends to @value{GDBN}. This makes it difficult
26335 to introduce new functionality without breaking existing usage. This
26336 section tries to minimize the problems by describing how the protocol
26339 Some changes in MI need not break a carefully designed front end, and
26340 for these the MI version will remain unchanged. The following is a
26341 list of changes that may occur within one level, so front ends should
26342 parse MI output in a way that can handle them:
26346 New MI commands may be added.
26349 New fields may be added to the output of any MI command.
26352 The range of values for fields with specified values, e.g.,
26353 @code{in_scope} (@pxref{-var-update}) may be extended.
26355 @c The format of field's content e.g type prefix, may change so parse it
26356 @c at your own risk. Yes, in general?
26358 @c The order of fields may change? Shouldn't really matter but it might
26359 @c resolve inconsistencies.
26362 If the changes are likely to break front ends, the MI version level
26363 will be increased by one. This will allow the front end to parse the
26364 output according to the MI version. Apart from mi0, new versions of
26365 @value{GDBN} will not support old versions of MI and it will be the
26366 responsibility of the front end to work with the new one.
26368 @c Starting with mi3, add a new command -mi-version that prints the MI
26371 The best way to avoid unexpected changes in MI that might break your front
26372 end is to make your project known to @value{GDBN} developers and
26373 follow development on @email{gdb@@sourceware.org} and
26374 @email{gdb-patches@@sourceware.org}.
26375 @cindex mailing lists
26377 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
26378 @node GDB/MI Output Records
26379 @section @sc{gdb/mi} Output Records
26382 * GDB/MI Result Records::
26383 * GDB/MI Stream Records::
26384 * GDB/MI Async Records::
26385 * GDB/MI Breakpoint Information::
26386 * GDB/MI Frame Information::
26387 * GDB/MI Thread Information::
26388 * GDB/MI Ada Exception Information::
26391 @node GDB/MI Result Records
26392 @subsection @sc{gdb/mi} Result Records
26394 @cindex result records in @sc{gdb/mi}
26395 @cindex @sc{gdb/mi}, result records
26396 In addition to a number of out-of-band notifications, the response to a
26397 @sc{gdb/mi} command includes one of the following result indications:
26401 @item "^done" [ "," @var{results} ]
26402 The synchronous operation was successful, @code{@var{results}} are the return
26407 This result record is equivalent to @samp{^done}. Historically, it
26408 was output instead of @samp{^done} if the command has resumed the
26409 target. This behaviour is maintained for backward compatibility, but
26410 all frontends should treat @samp{^done} and @samp{^running}
26411 identically and rely on the @samp{*running} output record to determine
26412 which threads are resumed.
26416 @value{GDBN} has connected to a remote target.
26418 @item "^error" "," "msg=" @var{c-string} [ "," "code=" @var{c-string} ]
26420 The operation failed. The @code{msg=@var{c-string}} variable contains
26421 the corresponding error message.
26423 If present, the @code{code=@var{c-string}} variable provides an error
26424 code on which consumers can rely on to detect the corresponding
26425 error condition. At present, only one error code is defined:
26428 @item "undefined-command"
26429 Indicates that the command causing the error does not exist.
26434 @value{GDBN} has terminated.
26438 @node GDB/MI Stream Records
26439 @subsection @sc{gdb/mi} Stream Records
26441 @cindex @sc{gdb/mi}, stream records
26442 @cindex stream records in @sc{gdb/mi}
26443 @value{GDBN} internally maintains a number of output streams: the console, the
26444 target, and the log. The output intended for each of these streams is
26445 funneled through the @sc{gdb/mi} interface using @dfn{stream records}.
26447 Each stream record begins with a unique @dfn{prefix character} which
26448 identifies its stream (@pxref{GDB/MI Output Syntax, , @sc{gdb/mi} Output
26449 Syntax}). In addition to the prefix, each stream record contains a
26450 @code{@var{string-output}}. This is either raw text (with an implicit new
26451 line) or a quoted C string (which does not contain an implicit newline).
26454 @item "~" @var{string-output}
26455 The console output stream contains text that should be displayed in the
26456 CLI console window. It contains the textual responses to CLI commands.
26458 @item "@@" @var{string-output}
26459 The target output stream contains any textual output from the running
26460 target. This is only present when GDB's event loop is truly
26461 asynchronous, which is currently only the case for remote targets.
26463 @item "&" @var{string-output}
26464 The log stream contains debugging messages being produced by @value{GDBN}'s
26468 @node GDB/MI Async Records
26469 @subsection @sc{gdb/mi} Async Records
26471 @cindex async records in @sc{gdb/mi}
26472 @cindex @sc{gdb/mi}, async records
26473 @dfn{Async} records are used to notify the @sc{gdb/mi} client of
26474 additional changes that have occurred. Those changes can either be a
26475 consequence of @sc{gdb/mi} commands (e.g., a breakpoint modified) or a result of
26476 target activity (e.g., target stopped).
26478 The following is the list of possible async records:
26482 @item *running,thread-id="@var{thread}"
26483 The target is now running. The @var{thread} field can be the global
26484 thread ID of the the thread that is now running, and it can be
26485 @samp{all} if all threads are running. The frontend should assume
26486 that no interaction with a running thread is possible after this
26487 notification is produced. The frontend should not assume that this
26488 notification is output only once for any command. @value{GDBN} may
26489 emit this notification several times, either for different threads,
26490 because it cannot resume all threads together, or even for a single
26491 thread, if the thread must be stepped though some code before letting
26494 @item *stopped,reason="@var{reason}",thread-id="@var{id}",stopped-threads="@var{stopped}",core="@var{core}"
26495 The target has stopped. The @var{reason} field can have one of the
26499 @item breakpoint-hit
26500 A breakpoint was reached.
26501 @item watchpoint-trigger
26502 A watchpoint was triggered.
26503 @item read-watchpoint-trigger
26504 A read watchpoint was triggered.
26505 @item access-watchpoint-trigger
26506 An access watchpoint was triggered.
26507 @item function-finished
26508 An -exec-finish or similar CLI command was accomplished.
26509 @item location-reached
26510 An -exec-until or similar CLI command was accomplished.
26511 @item watchpoint-scope
26512 A watchpoint has gone out of scope.
26513 @item end-stepping-range
26514 An -exec-next, -exec-next-instruction, -exec-step, -exec-step-instruction or
26515 similar CLI command was accomplished.
26516 @item exited-signalled
26517 The inferior exited because of a signal.
26519 The inferior exited.
26520 @item exited-normally
26521 The inferior exited normally.
26522 @item signal-received
26523 A signal was received by the inferior.
26525 The inferior has stopped due to a library being loaded or unloaded.
26526 This can happen when @code{stop-on-solib-events} (@pxref{Files}) is
26527 set or when a @code{catch load} or @code{catch unload} catchpoint is
26528 in use (@pxref{Set Catchpoints}).
26530 The inferior has forked. This is reported when @code{catch fork}
26531 (@pxref{Set Catchpoints}) has been used.
26533 The inferior has vforked. This is reported in when @code{catch vfork}
26534 (@pxref{Set Catchpoints}) has been used.
26535 @item syscall-entry
26536 The inferior entered a system call. This is reported when @code{catch
26537 syscall} (@pxref{Set Catchpoints}) has been used.
26538 @item syscall-return
26539 The inferior returned from a system call. This is reported when
26540 @code{catch syscall} (@pxref{Set Catchpoints}) has been used.
26542 The inferior called @code{exec}. This is reported when @code{catch exec}
26543 (@pxref{Set Catchpoints}) has been used.
26546 The @var{id} field identifies the global thread ID of the thread
26547 that directly caused the stop -- for example by hitting a breakpoint.
26548 Depending on whether all-stop
26549 mode is in effect (@pxref{All-Stop Mode}), @value{GDBN} may either
26550 stop all threads, or only the thread that directly triggered the stop.
26551 If all threads are stopped, the @var{stopped} field will have the
26552 value of @code{"all"}. Otherwise, the value of the @var{stopped}
26553 field will be a list of thread identifiers. Presently, this list will
26554 always include a single thread, but frontend should be prepared to see
26555 several threads in the list. The @var{core} field reports the
26556 processor core on which the stop event has happened. This field may be absent
26557 if such information is not available.
26559 @item =thread-group-added,id="@var{id}"
26560 @itemx =thread-group-removed,id="@var{id}"
26561 A thread group was either added or removed. The @var{id} field
26562 contains the @value{GDBN} identifier of the thread group. When a thread
26563 group is added, it generally might not be associated with a running
26564 process. When a thread group is removed, its id becomes invalid and
26565 cannot be used in any way.
26567 @item =thread-group-started,id="@var{id}",pid="@var{pid}"
26568 A thread group became associated with a running program,
26569 either because the program was just started or the thread group
26570 was attached to a program. The @var{id} field contains the
26571 @value{GDBN} identifier of the thread group. The @var{pid} field
26572 contains process identifier, specific to the operating system.
26574 @item =thread-group-exited,id="@var{id}"[,exit-code="@var{code}"]
26575 A thread group is no longer associated with a running program,
26576 either because the program has exited, or because it was detached
26577 from. The @var{id} field contains the @value{GDBN} identifier of the
26578 thread group. The @var{code} field is the exit code of the inferior; it exists
26579 only when the inferior exited with some code.
26581 @item =thread-created,id="@var{id}",group-id="@var{gid}"
26582 @itemx =thread-exited,id="@var{id}",group-id="@var{gid}"
26583 A thread either was created, or has exited. The @var{id} field
26584 contains the global @value{GDBN} identifier of the thread. The @var{gid}
26585 field identifies the thread group this thread belongs to.
26587 @item =thread-selected,id="@var{id}"[,frame="@var{frame}"]
26588 Informs that the selected thread or frame were changed. This notification
26589 is not emitted as result of the @code{-thread-select} or
26590 @code{-stack-select-frame} commands, but is emitted whenever an MI command
26591 that is not documented to change the selected thread and frame actually
26592 changes them. In particular, invoking, directly or indirectly
26593 (via user-defined command), the CLI @code{thread} or @code{frame} commands,
26594 will generate this notification. Changing the thread or frame from another
26595 user interface (see @ref{Interpreters}) will also generate this notification.
26597 The @var{frame} field is only present if the newly selected thread is
26598 stopped. See @ref{GDB/MI Frame Information} for the format of its value.
26600 We suggest that in response to this notification, front ends
26601 highlight the selected thread and cause subsequent commands to apply to
26604 @item =library-loaded,...
26605 Reports that a new library file was loaded by the program. This
26606 notification has 5 fields---@var{id}, @var{target-name},
26607 @var{host-name}, @var{symbols-loaded} and @var{ranges}. The @var{id} field is an
26608 opaque identifier of the library. For remote debugging case,
26609 @var{target-name} and @var{host-name} fields give the name of the
26610 library file on the target, and on the host respectively. For native
26611 debugging, both those fields have the same value. The
26612 @var{symbols-loaded} field is emitted only for backward compatibility
26613 and should not be relied on to convey any useful information. The
26614 @var{thread-group} field, if present, specifies the id of the thread
26615 group in whose context the library was loaded. If the field is
26616 absent, it means the library was loaded in the context of all present
26617 thread groups. The @var{ranges} field specifies the ranges of addresses belonging
26620 @item =library-unloaded,...
26621 Reports that a library was unloaded by the program. This notification
26622 has 3 fields---@var{id}, @var{target-name} and @var{host-name} with
26623 the same meaning as for the @code{=library-loaded} notification.
26624 The @var{thread-group} field, if present, specifies the id of the
26625 thread group in whose context the library was unloaded. If the field is
26626 absent, it means the library was unloaded in the context of all present
26629 @item =traceframe-changed,num=@var{tfnum},tracepoint=@var{tpnum}
26630 @itemx =traceframe-changed,end
26631 Reports that the trace frame was changed and its new number is
26632 @var{tfnum}. The number of the tracepoint associated with this trace
26633 frame is @var{tpnum}.
26635 @item =tsv-created,name=@var{name},initial=@var{initial}
26636 Reports that the new trace state variable @var{name} is created with
26637 initial value @var{initial}.
26639 @item =tsv-deleted,name=@var{name}
26640 @itemx =tsv-deleted
26641 Reports that the trace state variable @var{name} is deleted or all
26642 trace state variables are deleted.
26644 @item =tsv-modified,name=@var{name},initial=@var{initial}[,current=@var{current}]
26645 Reports that the trace state variable @var{name} is modified with
26646 the initial value @var{initial}. The current value @var{current} of
26647 trace state variable is optional and is reported if the current
26648 value of trace state variable is known.
26650 @item =breakpoint-created,bkpt=@{...@}
26651 @itemx =breakpoint-modified,bkpt=@{...@}
26652 @itemx =breakpoint-deleted,id=@var{number}
26653 Reports that a breakpoint was created, modified, or deleted,
26654 respectively. Only user-visible breakpoints are reported to the MI
26657 The @var{bkpt} argument is of the same form as returned by the various
26658 breakpoint commands; @xref{GDB/MI Breakpoint Commands}. The
26659 @var{number} is the ordinal number of the breakpoint.
26661 Note that if a breakpoint is emitted in the result record of a
26662 command, then it will not also be emitted in an async record.
26664 @item =record-started,thread-group="@var{id}",method="@var{method}"[,format="@var{format}"]
26665 @itemx =record-stopped,thread-group="@var{id}"
26666 Execution log recording was either started or stopped on an
26667 inferior. The @var{id} is the @value{GDBN} identifier of the thread
26668 group corresponding to the affected inferior.
26670 The @var{method} field indicates the method used to record execution. If the
26671 method in use supports multiple recording formats, @var{format} will be present
26672 and contain the currently used format. @xref{Process Record and Replay},
26673 for existing method and format values.
26675 @item =cmd-param-changed,param=@var{param},value=@var{value}
26676 Reports that a parameter of the command @code{set @var{param}} is
26677 changed to @var{value}. In the multi-word @code{set} command,
26678 the @var{param} is the whole parameter list to @code{set} command.
26679 For example, In command @code{set check type on}, @var{param}
26680 is @code{check type} and @var{value} is @code{on}.
26682 @item =memory-changed,thread-group=@var{id},addr=@var{addr},len=@var{len}[,type="code"]
26683 Reports that bytes from @var{addr} to @var{data} + @var{len} were
26684 written in an inferior. The @var{id} is the identifier of the
26685 thread group corresponding to the affected inferior. The optional
26686 @code{type="code"} part is reported if the memory written to holds
26690 @node GDB/MI Breakpoint Information
26691 @subsection @sc{gdb/mi} Breakpoint Information
26693 When @value{GDBN} reports information about a breakpoint, a
26694 tracepoint, a watchpoint, or a catchpoint, it uses a tuple with the
26699 The breakpoint number. For a breakpoint that represents one location
26700 of a multi-location breakpoint, this will be a dotted pair, like
26704 The type of the breakpoint. For ordinary breakpoints this will be
26705 @samp{breakpoint}, but many values are possible.
26708 If the type of the breakpoint is @samp{catchpoint}, then this
26709 indicates the exact type of catchpoint.
26712 This is the breakpoint disposition---either @samp{del}, meaning that
26713 the breakpoint will be deleted at the next stop, or @samp{keep},
26714 meaning that the breakpoint will not be deleted.
26717 This indicates whether the breakpoint is enabled, in which case the
26718 value is @samp{y}, or disabled, in which case the value is @samp{n}.
26719 Note that this is not the same as the field @code{enable}.
26722 The address of the breakpoint. This may be a hexidecimal number,
26723 giving the address; or the string @samp{<PENDING>}, for a pending
26724 breakpoint; or the string @samp{<MULTIPLE>}, for a breakpoint with
26725 multiple locations. This field will not be present if no address can
26726 be determined. For example, a watchpoint does not have an address.
26729 If known, the function in which the breakpoint appears.
26730 If not known, this field is not present.
26733 The name of the source file which contains this function, if known.
26734 If not known, this field is not present.
26737 The full file name of the source file which contains this function, if
26738 known. If not known, this field is not present.
26741 The line number at which this breakpoint appears, if known.
26742 If not known, this field is not present.
26745 If the source file is not known, this field may be provided. If
26746 provided, this holds the address of the breakpoint, possibly followed
26750 If this breakpoint is pending, this field is present and holds the
26751 text used to set the breakpoint, as entered by the user.
26754 Where this breakpoint's condition is evaluated, either @samp{host} or
26758 If this is a thread-specific breakpoint, then this identifies the
26759 thread in which the breakpoint can trigger.
26762 If this breakpoint is restricted to a particular Ada task, then this
26763 field will hold the task identifier.
26766 If the breakpoint is conditional, this is the condition expression.
26769 The ignore count of the breakpoint.
26772 The enable count of the breakpoint.
26774 @item traceframe-usage
26777 @item static-tracepoint-marker-string-id
26778 For a static tracepoint, the name of the static tracepoint marker.
26781 For a masked watchpoint, this is the mask.
26784 A tracepoint's pass count.
26786 @item original-location
26787 The location of the breakpoint as originally specified by the user.
26788 This field is optional.
26791 The number of times the breakpoint has been hit.
26794 This field is only given for tracepoints. This is either @samp{y},
26795 meaning that the tracepoint is installed, or @samp{n}, meaning that it
26799 Some extra data, the exact contents of which are type-dependent.
26803 For example, here is what the output of @code{-break-insert}
26804 (@pxref{GDB/MI Breakpoint Commands}) might be:
26807 -> -break-insert main
26808 <- ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
26809 enabled="y",addr="0x08048564",func="main",file="myprog.c",
26810 fullname="/home/nickrob/myprog.c",line="68",thread-groups=["i1"],
26815 @node GDB/MI Frame Information
26816 @subsection @sc{gdb/mi} Frame Information
26818 Response from many MI commands includes an information about stack
26819 frame. This information is a tuple that may have the following
26824 The level of the stack frame. The innermost frame has the level of
26825 zero. This field is always present.
26828 The name of the function corresponding to the frame. This field may
26829 be absent if @value{GDBN} is unable to determine the function name.
26832 The code address for the frame. This field is always present.
26835 The name of the source files that correspond to the frame's code
26836 address. This field may be absent.
26839 The source line corresponding to the frames' code address. This field
26843 The name of the binary file (either executable or shared library) the
26844 corresponds to the frame's code address. This field may be absent.
26848 @node GDB/MI Thread Information
26849 @subsection @sc{gdb/mi} Thread Information
26851 Whenever @value{GDBN} has to report an information about a thread, it
26852 uses a tuple with the following fields:
26856 The global numeric id assigned to the thread by @value{GDBN}. This field is
26860 Target-specific string identifying the thread. This field is always present.
26863 Additional information about the thread provided by the target.
26864 It is supposed to be human-readable and not interpreted by the
26865 frontend. This field is optional.
26868 Either @samp{stopped} or @samp{running}, depending on whether the
26869 thread is presently running. This field is always present.
26872 The value of this field is an integer number of the processor core the
26873 thread was last seen on. This field is optional.
26876 @node GDB/MI Ada Exception Information
26877 @subsection @sc{gdb/mi} Ada Exception Information
26879 Whenever a @code{*stopped} record is emitted because the program
26880 stopped after hitting an exception catchpoint (@pxref{Set Catchpoints}),
26881 @value{GDBN} provides the name of the exception that was raised via
26882 the @code{exception-name} field.
26884 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
26885 @node GDB/MI Simple Examples
26886 @section Simple Examples of @sc{gdb/mi} Interaction
26887 @cindex @sc{gdb/mi}, simple examples
26889 This subsection presents several simple examples of interaction using
26890 the @sc{gdb/mi} interface. In these examples, @samp{->} means that the
26891 following line is passed to @sc{gdb/mi} as input, while @samp{<-} means
26892 the output received from @sc{gdb/mi}.
26894 Note the line breaks shown in the examples are here only for
26895 readability, they don't appear in the real output.
26897 @subheading Setting a Breakpoint
26899 Setting a breakpoint generates synchronous output which contains detailed
26900 information of the breakpoint.
26903 -> -break-insert main
26904 <- ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
26905 enabled="y",addr="0x08048564",func="main",file="myprog.c",
26906 fullname="/home/nickrob/myprog.c",line="68",thread-groups=["i1"],
26911 @subheading Program Execution
26913 Program execution generates asynchronous records and MI gives the
26914 reason that execution stopped.
26920 <- *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",thread-id="0",
26921 frame=@{addr="0x08048564",func="main",
26922 args=[@{name="argc",value="1"@},@{name="argv",value="0xbfc4d4d4"@}],
26923 file="myprog.c",fullname="/home/nickrob/myprog.c",line="68"@}
26928 <- *stopped,reason="exited-normally"
26932 @subheading Quitting @value{GDBN}
26934 Quitting @value{GDBN} just prints the result class @samp{^exit}.
26942 Please note that @samp{^exit} is printed immediately, but it might
26943 take some time for @value{GDBN} to actually exit. During that time, @value{GDBN}
26944 performs necessary cleanups, including killing programs being debugged
26945 or disconnecting from debug hardware, so the frontend should wait till
26946 @value{GDBN} exits and should only forcibly kill @value{GDBN} if it
26947 fails to exit in reasonable time.
26949 @subheading A Bad Command
26951 Here's what happens if you pass a non-existent command:
26955 <- ^error,msg="Undefined MI command: rubbish"
26960 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
26961 @node GDB/MI Command Description Format
26962 @section @sc{gdb/mi} Command Description Format
26964 The remaining sections describe blocks of commands. Each block of
26965 commands is laid out in a fashion similar to this section.
26967 @subheading Motivation
26969 The motivation for this collection of commands.
26971 @subheading Introduction
26973 A brief introduction to this collection of commands as a whole.
26975 @subheading Commands
26977 For each command in the block, the following is described:
26979 @subsubheading Synopsis
26982 -command @var{args}@dots{}
26985 @subsubheading Result
26987 @subsubheading @value{GDBN} Command
26989 The corresponding @value{GDBN} CLI command(s), if any.
26991 @subsubheading Example
26993 Example(s) formatted for readability. Some of the described commands have
26994 not been implemented yet and these are labeled N.A.@: (not available).
26997 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
26998 @node GDB/MI Breakpoint Commands
26999 @section @sc{gdb/mi} Breakpoint Commands
27001 @cindex breakpoint commands for @sc{gdb/mi}
27002 @cindex @sc{gdb/mi}, breakpoint commands
27003 This section documents @sc{gdb/mi} commands for manipulating
27006 @subheading The @code{-break-after} Command
27007 @findex -break-after
27009 @subsubheading Synopsis
27012 -break-after @var{number} @var{count}
27015 The breakpoint number @var{number} is not in effect until it has been
27016 hit @var{count} times. To see how this is reflected in the output of
27017 the @samp{-break-list} command, see the description of the
27018 @samp{-break-list} command below.
27020 @subsubheading @value{GDBN} Command
27022 The corresponding @value{GDBN} command is @samp{ignore}.
27024 @subsubheading Example
27029 ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
27030 enabled="y",addr="0x000100d0",func="main",file="hello.c",
27031 fullname="/home/foo/hello.c",line="5",thread-groups=["i1"],
27039 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
27040 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
27041 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
27042 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
27043 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
27044 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
27045 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
27046 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
27047 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
27048 line="5",thread-groups=["i1"],times="0",ignore="3"@}]@}
27053 @subheading The @code{-break-catch} Command
27054 @findex -break-catch
27057 @subheading The @code{-break-commands} Command
27058 @findex -break-commands
27060 @subsubheading Synopsis
27063 -break-commands @var{number} [ @var{command1} ... @var{commandN} ]
27066 Specifies the CLI commands that should be executed when breakpoint
27067 @var{number} is hit. The parameters @var{command1} to @var{commandN}
27068 are the commands. If no command is specified, any previously-set
27069 commands are cleared. @xref{Break Commands}. Typical use of this
27070 functionality is tracing a program, that is, printing of values of
27071 some variables whenever breakpoint is hit and then continuing.
27073 @subsubheading @value{GDBN} Command
27075 The corresponding @value{GDBN} command is @samp{commands}.
27077 @subsubheading Example
27082 ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
27083 enabled="y",addr="0x000100d0",func="main",file="hello.c",
27084 fullname="/home/foo/hello.c",line="5",thread-groups=["i1"],
27087 -break-commands 1 "print v" "continue"
27092 @subheading The @code{-break-condition} Command
27093 @findex -break-condition
27095 @subsubheading Synopsis
27098 -break-condition @var{number} @var{expr}
27101 Breakpoint @var{number} will stop the program only if the condition in
27102 @var{expr} is true. The condition becomes part of the
27103 @samp{-break-list} output (see the description of the @samp{-break-list}
27106 @subsubheading @value{GDBN} Command
27108 The corresponding @value{GDBN} command is @samp{condition}.
27110 @subsubheading Example
27114 -break-condition 1 1
27118 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
27119 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
27120 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
27121 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
27122 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
27123 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
27124 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
27125 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
27126 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
27127 line="5",cond="1",thread-groups=["i1"],times="0",ignore="3"@}]@}
27131 @subheading The @code{-break-delete} Command
27132 @findex -break-delete
27134 @subsubheading Synopsis
27137 -break-delete ( @var{breakpoint} )+
27140 Delete the breakpoint(s) whose number(s) are specified in the argument
27141 list. This is obviously reflected in the breakpoint list.
27143 @subsubheading @value{GDBN} Command
27145 The corresponding @value{GDBN} command is @samp{delete}.
27147 @subsubheading Example
27155 ^done,BreakpointTable=@{nr_rows="0",nr_cols="6",
27156 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
27157 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
27158 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
27159 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
27160 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
27161 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
27166 @subheading The @code{-break-disable} Command
27167 @findex -break-disable
27169 @subsubheading Synopsis
27172 -break-disable ( @var{breakpoint} )+
27175 Disable the named @var{breakpoint}(s). The field @samp{enabled} in the
27176 break list is now set to @samp{n} for the named @var{breakpoint}(s).
27178 @subsubheading @value{GDBN} Command
27180 The corresponding @value{GDBN} command is @samp{disable}.
27182 @subsubheading Example
27190 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
27191 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
27192 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
27193 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
27194 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
27195 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
27196 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
27197 body=[bkpt=@{number="2",type="breakpoint",disp="keep",enabled="n",
27198 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
27199 line="5",thread-groups=["i1"],times="0"@}]@}
27203 @subheading The @code{-break-enable} Command
27204 @findex -break-enable
27206 @subsubheading Synopsis
27209 -break-enable ( @var{breakpoint} )+
27212 Enable (previously disabled) @var{breakpoint}(s).
27214 @subsubheading @value{GDBN} Command
27216 The corresponding @value{GDBN} command is @samp{enable}.
27218 @subsubheading Example
27226 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
27227 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
27228 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
27229 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
27230 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
27231 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
27232 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
27233 body=[bkpt=@{number="2",type="breakpoint",disp="keep",enabled="y",
27234 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
27235 line="5",thread-groups=["i1"],times="0"@}]@}
27239 @subheading The @code{-break-info} Command
27240 @findex -break-info
27242 @subsubheading Synopsis
27245 -break-info @var{breakpoint}
27249 Get information about a single breakpoint.
27251 The result is a table of breakpoints. @xref{GDB/MI Breakpoint
27252 Information}, for details on the format of each breakpoint in the
27255 @subsubheading @value{GDBN} Command
27257 The corresponding @value{GDBN} command is @samp{info break @var{breakpoint}}.
27259 @subsubheading Example
27262 @subheading The @code{-break-insert} Command
27263 @findex -break-insert
27264 @anchor{-break-insert}
27266 @subsubheading Synopsis
27269 -break-insert [ -t ] [ -h ] [ -f ] [ -d ] [ -a ]
27270 [ -c @var{condition} ] [ -i @var{ignore-count} ]
27271 [ -p @var{thread-id} ] [ @var{location} ]
27275 If specified, @var{location}, can be one of:
27278 @item linespec location
27279 A linespec location. @xref{Linespec Locations}.
27281 @item explicit location
27282 An explicit location. @sc{gdb/mi} explicit locations are
27283 analogous to the CLI's explicit locations using the option names
27284 listed below. @xref{Explicit Locations}.
27287 @item --source @var{filename}
27288 The source file name of the location. This option requires the use
27289 of either @samp{--function} or @samp{--line}.
27291 @item --function @var{function}
27292 The name of a function or method.
27294 @item --label @var{label}
27295 The name of a label.
27297 @item --line @var{lineoffset}
27298 An absolute or relative line offset from the start of the location.
27301 @item address location
27302 An address location, *@var{address}. @xref{Address Locations}.
27306 The possible optional parameters of this command are:
27310 Insert a temporary breakpoint.
27312 Insert a hardware breakpoint.
27314 If @var{location} cannot be parsed (for example if it
27315 refers to unknown files or functions), create a pending
27316 breakpoint. Without this flag, @value{GDBN} will report
27317 an error, and won't create a breakpoint, if @var{location}
27320 Create a disabled breakpoint.
27322 Create a tracepoint. @xref{Tracepoints}. When this parameter
27323 is used together with @samp{-h}, a fast tracepoint is created.
27324 @item -c @var{condition}
27325 Make the breakpoint conditional on @var{condition}.
27326 @item -i @var{ignore-count}
27327 Initialize the @var{ignore-count}.
27328 @item -p @var{thread-id}
27329 Restrict the breakpoint to the thread with the specified global
27333 @subsubheading Result
27335 @xref{GDB/MI Breakpoint Information}, for details on the format of the
27336 resulting breakpoint.
27338 Note: this format is open to change.
27339 @c An out-of-band breakpoint instead of part of the result?
27341 @subsubheading @value{GDBN} Command
27343 The corresponding @value{GDBN} commands are @samp{break}, @samp{tbreak},
27344 @samp{hbreak}, and @samp{thbreak}. @c and @samp{rbreak}.
27346 @subsubheading Example
27351 ^done,bkpt=@{number="1",addr="0x0001072c",file="recursive2.c",
27352 fullname="/home/foo/recursive2.c,line="4",thread-groups=["i1"],
27355 -break-insert -t foo
27356 ^done,bkpt=@{number="2",addr="0x00010774",file="recursive2.c",
27357 fullname="/home/foo/recursive2.c,line="11",thread-groups=["i1"],
27361 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
27362 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
27363 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
27364 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
27365 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
27366 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
27367 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
27368 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
27369 addr="0x0001072c", func="main",file="recursive2.c",
27370 fullname="/home/foo/recursive2.c,"line="4",thread-groups=["i1"],
27372 bkpt=@{number="2",type="breakpoint",disp="del",enabled="y",
27373 addr="0x00010774",func="foo",file="recursive2.c",
27374 fullname="/home/foo/recursive2.c",line="11",thread-groups=["i1"],
27377 @c -break-insert -r foo.*
27378 @c ~int foo(int, int);
27379 @c ^done,bkpt=@{number="3",addr="0x00010774",file="recursive2.c,
27380 @c "fullname="/home/foo/recursive2.c",line="11",thread-groups=["i1"],
27385 @subheading The @code{-dprintf-insert} Command
27386 @findex -dprintf-insert
27388 @subsubheading Synopsis
27391 -dprintf-insert [ -t ] [ -f ] [ -d ]
27392 [ -c @var{condition} ] [ -i @var{ignore-count} ]
27393 [ -p @var{thread-id} ] [ @var{location} ] [ @var{format} ]
27398 If supplied, @var{location} may be specified the same way as for
27399 the @code{-break-insert} command. @xref{-break-insert}.
27401 The possible optional parameters of this command are:
27405 Insert a temporary breakpoint.
27407 If @var{location} cannot be parsed (for example, if it
27408 refers to unknown files or functions), create a pending
27409 breakpoint. Without this flag, @value{GDBN} will report
27410 an error, and won't create a breakpoint, if @var{location}
27413 Create a disabled breakpoint.
27414 @item -c @var{condition}
27415 Make the breakpoint conditional on @var{condition}.
27416 @item -i @var{ignore-count}
27417 Set the ignore count of the breakpoint (@pxref{Conditions, ignore count})
27418 to @var{ignore-count}.
27419 @item -p @var{thread-id}
27420 Restrict the breakpoint to the thread with the specified global
27424 @subsubheading Result
27426 @xref{GDB/MI Breakpoint Information}, for details on the format of the
27427 resulting breakpoint.
27429 @c An out-of-band breakpoint instead of part of the result?
27431 @subsubheading @value{GDBN} Command
27433 The corresponding @value{GDBN} command is @samp{dprintf}.
27435 @subsubheading Example
27439 4-dprintf-insert foo "At foo entry\n"
27440 4^done,bkpt=@{number="1",type="dprintf",disp="keep",enabled="y",
27441 addr="0x000000000040061b",func="foo",file="mi-dprintf.c",
27442 fullname="mi-dprintf.c",line="25",thread-groups=["i1"],
27443 times="0",script=@{"printf \"At foo entry\\n\"","continue"@},
27444 original-location="foo"@}
27446 5-dprintf-insert 26 "arg=%d, g=%d\n" arg g
27447 5^done,bkpt=@{number="2",type="dprintf",disp="keep",enabled="y",
27448 addr="0x000000000040062a",func="foo",file="mi-dprintf.c",
27449 fullname="mi-dprintf.c",line="26",thread-groups=["i1"],
27450 times="0",script=@{"printf \"arg=%d, g=%d\\n\", arg, g","continue"@},
27451 original-location="mi-dprintf.c:26"@}
27455 @subheading The @code{-break-list} Command
27456 @findex -break-list
27458 @subsubheading Synopsis
27464 Displays the list of inserted breakpoints, showing the following fields:
27468 number of the breakpoint
27470 type of the breakpoint: @samp{breakpoint} or @samp{watchpoint}
27472 should the breakpoint be deleted or disabled when it is hit: @samp{keep}
27475 is the breakpoint enabled or no: @samp{y} or @samp{n}
27477 memory location at which the breakpoint is set
27479 logical location of the breakpoint, expressed by function name, file
27481 @item Thread-groups
27482 list of thread groups to which this breakpoint applies
27484 number of times the breakpoint has been hit
27487 If there are no breakpoints or watchpoints, the @code{BreakpointTable}
27488 @code{body} field is an empty list.
27490 @subsubheading @value{GDBN} Command
27492 The corresponding @value{GDBN} command is @samp{info break}.
27494 @subsubheading Example
27499 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
27500 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
27501 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
27502 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
27503 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
27504 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
27505 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
27506 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
27507 addr="0x000100d0",func="main",file="hello.c",line="5",thread-groups=["i1"],
27509 bkpt=@{number="2",type="breakpoint",disp="keep",enabled="y",
27510 addr="0x00010114",func="foo",file="hello.c",fullname="/home/foo/hello.c",
27511 line="13",thread-groups=["i1"],times="0"@}]@}
27515 Here's an example of the result when there are no breakpoints:
27520 ^done,BreakpointTable=@{nr_rows="0",nr_cols="6",
27521 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
27522 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
27523 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
27524 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
27525 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
27526 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
27531 @subheading The @code{-break-passcount} Command
27532 @findex -break-passcount
27534 @subsubheading Synopsis
27537 -break-passcount @var{tracepoint-number} @var{passcount}
27540 Set the passcount for tracepoint @var{tracepoint-number} to
27541 @var{passcount}. If the breakpoint referred to by @var{tracepoint-number}
27542 is not a tracepoint, error is emitted. This corresponds to CLI
27543 command @samp{passcount}.
27545 @subheading The @code{-break-watch} Command
27546 @findex -break-watch
27548 @subsubheading Synopsis
27551 -break-watch [ -a | -r ]
27554 Create a watchpoint. With the @samp{-a} option it will create an
27555 @dfn{access} watchpoint, i.e., a watchpoint that triggers either on a
27556 read from or on a write to the memory location. With the @samp{-r}
27557 option, the watchpoint created is a @dfn{read} watchpoint, i.e., it will
27558 trigger only when the memory location is accessed for reading. Without
27559 either of the options, the watchpoint created is a regular watchpoint,
27560 i.e., it will trigger when the memory location is accessed for writing.
27561 @xref{Set Watchpoints, , Setting Watchpoints}.
27563 Note that @samp{-break-list} will report a single list of watchpoints and
27564 breakpoints inserted.
27566 @subsubheading @value{GDBN} Command
27568 The corresponding @value{GDBN} commands are @samp{watch}, @samp{awatch}, and
27571 @subsubheading Example
27573 Setting a watchpoint on a variable in the @code{main} function:
27578 ^done,wpt=@{number="2",exp="x"@}
27583 *stopped,reason="watchpoint-trigger",wpt=@{number="2",exp="x"@},
27584 value=@{old="-268439212",new="55"@},
27585 frame=@{func="main",args=[],file="recursive2.c",
27586 fullname="/home/foo/bar/recursive2.c",line="5"@}
27590 Setting a watchpoint on a variable local to a function. @value{GDBN} will stop
27591 the program execution twice: first for the variable changing value, then
27592 for the watchpoint going out of scope.
27597 ^done,wpt=@{number="5",exp="C"@}
27602 *stopped,reason="watchpoint-trigger",
27603 wpt=@{number="5",exp="C"@},value=@{old="-276895068",new="3"@},
27604 frame=@{func="callee4",args=[],
27605 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
27606 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="13"@}
27611 *stopped,reason="watchpoint-scope",wpnum="5",
27612 frame=@{func="callee3",args=[@{name="strarg",
27613 value="0x11940 \"A string argument.\""@}],
27614 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
27615 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="18"@}
27619 Listing breakpoints and watchpoints, at different points in the program
27620 execution. Note that once the watchpoint goes out of scope, it is
27626 ^done,wpt=@{number="2",exp="C"@}
27629 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
27630 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
27631 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
27632 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
27633 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
27634 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
27635 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
27636 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
27637 addr="0x00010734",func="callee4",
27638 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
27639 fullname="/home/foo/devo/gdb/testsuite/gdb.mi/basics.c"line="8",thread-groups=["i1"],
27641 bkpt=@{number="2",type="watchpoint",disp="keep",
27642 enabled="y",addr="",what="C",thread-groups=["i1"],times="0"@}]@}
27647 *stopped,reason="watchpoint-trigger",wpt=@{number="2",exp="C"@},
27648 value=@{old="-276895068",new="3"@},
27649 frame=@{func="callee4",args=[],
27650 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
27651 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="13"@}
27654 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
27655 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
27656 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
27657 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
27658 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
27659 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
27660 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
27661 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
27662 addr="0x00010734",func="callee4",
27663 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
27664 fullname="/home/foo/devo/gdb/testsuite/gdb.mi/basics.c",line="8",thread-groups=["i1"],
27666 bkpt=@{number="2",type="watchpoint",disp="keep",
27667 enabled="y",addr="",what="C",thread-groups=["i1"],times="-5"@}]@}
27671 ^done,reason="watchpoint-scope",wpnum="2",
27672 frame=@{func="callee3",args=[@{name="strarg",
27673 value="0x11940 \"A string argument.\""@}],
27674 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
27675 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="18"@}
27678 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
27679 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
27680 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
27681 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
27682 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
27683 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
27684 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
27685 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
27686 addr="0x00010734",func="callee4",
27687 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
27688 fullname="/home/foo/devo/gdb/testsuite/gdb.mi/basics.c",line="8",
27689 thread-groups=["i1"],times="1"@}]@}
27694 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
27695 @node GDB/MI Catchpoint Commands
27696 @section @sc{gdb/mi} Catchpoint Commands
27698 This section documents @sc{gdb/mi} commands for manipulating
27702 * Shared Library GDB/MI Catchpoint Commands::
27703 * Ada Exception GDB/MI Catchpoint Commands::
27706 @node Shared Library GDB/MI Catchpoint Commands
27707 @subsection Shared Library @sc{gdb/mi} Catchpoints
27709 @subheading The @code{-catch-load} Command
27710 @findex -catch-load
27712 @subsubheading Synopsis
27715 -catch-load [ -t ] [ -d ] @var{regexp}
27718 Add a catchpoint for library load events. If the @samp{-t} option is used,
27719 the catchpoint is a temporary one (@pxref{Set Breaks, ,Setting
27720 Breakpoints}). If the @samp{-d} option is used, the catchpoint is created
27721 in a disabled state. The @samp{regexp} argument is a regular
27722 expression used to match the name of the loaded library.
27725 @subsubheading @value{GDBN} Command
27727 The corresponding @value{GDBN} command is @samp{catch load}.
27729 @subsubheading Example
27732 -catch-load -t foo.so
27733 ^done,bkpt=@{number="1",type="catchpoint",disp="del",enabled="y",
27734 what="load of library matching foo.so",catch-type="load",times="0"@}
27739 @subheading The @code{-catch-unload} Command
27740 @findex -catch-unload
27742 @subsubheading Synopsis
27745 -catch-unload [ -t ] [ -d ] @var{regexp}
27748 Add a catchpoint for library unload events. If the @samp{-t} option is
27749 used, the catchpoint is a temporary one (@pxref{Set Breaks, ,Setting
27750 Breakpoints}). If the @samp{-d} option is used, the catchpoint is
27751 created in a disabled state. The @samp{regexp} argument is a regular
27752 expression used to match the name of the unloaded library.
27754 @subsubheading @value{GDBN} Command
27756 The corresponding @value{GDBN} command is @samp{catch unload}.
27758 @subsubheading Example
27761 -catch-unload -d bar.so
27762 ^done,bkpt=@{number="2",type="catchpoint",disp="keep",enabled="n",
27763 what="load of library matching bar.so",catch-type="unload",times="0"@}
27767 @node Ada Exception GDB/MI Catchpoint Commands
27768 @subsection Ada Exception @sc{gdb/mi} Catchpoints
27770 The following @sc{gdb/mi} commands can be used to create catchpoints
27771 that stop the execution when Ada exceptions are being raised.
27773 @subheading The @code{-catch-assert} Command
27774 @findex -catch-assert
27776 @subsubheading Synopsis
27779 -catch-assert [ -c @var{condition}] [ -d ] [ -t ]
27782 Add a catchpoint for failed Ada assertions.
27784 The possible optional parameters for this command are:
27787 @item -c @var{condition}
27788 Make the catchpoint conditional on @var{condition}.
27790 Create a disabled catchpoint.
27792 Create a temporary catchpoint.
27795 @subsubheading @value{GDBN} Command
27797 The corresponding @value{GDBN} command is @samp{catch assert}.
27799 @subsubheading Example
27803 ^done,bkptno="5",bkpt=@{number="5",type="breakpoint",disp="keep",
27804 enabled="y",addr="0x0000000000404888",what="failed Ada assertions",
27805 thread-groups=["i1"],times="0",
27806 original-location="__gnat_debug_raise_assert_failure"@}
27810 @subheading The @code{-catch-exception} Command
27811 @findex -catch-exception
27813 @subsubheading Synopsis
27816 -catch-exception [ -c @var{condition}] [ -d ] [ -e @var{exception-name} ]
27820 Add a catchpoint stopping when Ada exceptions are raised.
27821 By default, the command stops the program when any Ada exception
27822 gets raised. But it is also possible, by using some of the
27823 optional parameters described below, to create more selective
27826 The possible optional parameters for this command are:
27829 @item -c @var{condition}
27830 Make the catchpoint conditional on @var{condition}.
27832 Create a disabled catchpoint.
27833 @item -e @var{exception-name}
27834 Only stop when @var{exception-name} is raised. This option cannot
27835 be used combined with @samp{-u}.
27837 Create a temporary catchpoint.
27839 Stop only when an unhandled exception gets raised. This option
27840 cannot be used combined with @samp{-e}.
27843 @subsubheading @value{GDBN} Command
27845 The corresponding @value{GDBN} commands are @samp{catch exception}
27846 and @samp{catch exception unhandled}.
27848 @subsubheading Example
27851 -catch-exception -e Program_Error
27852 ^done,bkptno="4",bkpt=@{number="4",type="breakpoint",disp="keep",
27853 enabled="y",addr="0x0000000000404874",
27854 what="`Program_Error' Ada exception", thread-groups=["i1"],
27855 times="0",original-location="__gnat_debug_raise_exception"@}
27859 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
27860 @node GDB/MI Program Context
27861 @section @sc{gdb/mi} Program Context
27863 @subheading The @code{-exec-arguments} Command
27864 @findex -exec-arguments
27867 @subsubheading Synopsis
27870 -exec-arguments @var{args}
27873 Set the inferior program arguments, to be used in the next
27876 @subsubheading @value{GDBN} Command
27878 The corresponding @value{GDBN} command is @samp{set args}.
27880 @subsubheading Example
27884 -exec-arguments -v word
27891 @subheading The @code{-exec-show-arguments} Command
27892 @findex -exec-show-arguments
27894 @subsubheading Synopsis
27897 -exec-show-arguments
27900 Print the arguments of the program.
27902 @subsubheading @value{GDBN} Command
27904 The corresponding @value{GDBN} command is @samp{show args}.
27906 @subsubheading Example
27911 @subheading The @code{-environment-cd} Command
27912 @findex -environment-cd
27914 @subsubheading Synopsis
27917 -environment-cd @var{pathdir}
27920 Set @value{GDBN}'s working directory.
27922 @subsubheading @value{GDBN} Command
27924 The corresponding @value{GDBN} command is @samp{cd}.
27926 @subsubheading Example
27930 -environment-cd /kwikemart/marge/ezannoni/flathead-dev/devo/gdb
27936 @subheading The @code{-environment-directory} Command
27937 @findex -environment-directory
27939 @subsubheading Synopsis
27942 -environment-directory [ -r ] [ @var{pathdir} ]+
27945 Add directories @var{pathdir} to beginning of search path for source files.
27946 If the @samp{-r} option is used, the search path is reset to the default
27947 search path. If directories @var{pathdir} are supplied in addition to the
27948 @samp{-r} option, the search path is first reset and then addition
27950 Multiple directories may be specified, separated by blanks. Specifying
27951 multiple directories in a single command
27952 results in the directories added to the beginning of the
27953 search path in the same order they were presented in the command.
27954 If blanks are needed as
27955 part of a directory name, double-quotes should be used around
27956 the name. In the command output, the path will show up separated
27957 by the system directory-separator character. The directory-separator
27958 character must not be used
27959 in any directory name.
27960 If no directories are specified, the current search path is displayed.
27962 @subsubheading @value{GDBN} Command
27964 The corresponding @value{GDBN} command is @samp{dir}.
27966 @subsubheading Example
27970 -environment-directory /kwikemart/marge/ezannoni/flathead-dev/devo/gdb
27971 ^done,source-path="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb:$cdir:$cwd"
27973 -environment-directory ""
27974 ^done,source-path="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb:$cdir:$cwd"
27976 -environment-directory -r /home/jjohnstn/src/gdb /usr/src
27977 ^done,source-path="/home/jjohnstn/src/gdb:/usr/src:$cdir:$cwd"
27979 -environment-directory -r
27980 ^done,source-path="$cdir:$cwd"
27985 @subheading The @code{-environment-path} Command
27986 @findex -environment-path
27988 @subsubheading Synopsis
27991 -environment-path [ -r ] [ @var{pathdir} ]+
27994 Add directories @var{pathdir} to beginning of search path for object files.
27995 If the @samp{-r} option is used, the search path is reset to the original
27996 search path that existed at gdb start-up. If directories @var{pathdir} are
27997 supplied in addition to the
27998 @samp{-r} option, the search path is first reset and then addition
28000 Multiple directories may be specified, separated by blanks. Specifying
28001 multiple directories in a single command
28002 results in the directories added to the beginning of the
28003 search path in the same order they were presented in the command.
28004 If blanks are needed as
28005 part of a directory name, double-quotes should be used around
28006 the name. In the command output, the path will show up separated
28007 by the system directory-separator character. The directory-separator
28008 character must not be used
28009 in any directory name.
28010 If no directories are specified, the current path is displayed.
28013 @subsubheading @value{GDBN} Command
28015 The corresponding @value{GDBN} command is @samp{path}.
28017 @subsubheading Example
28022 ^done,path="/usr/bin"
28024 -environment-path /kwikemart/marge/ezannoni/flathead-dev/ppc-eabi/gdb /bin
28025 ^done,path="/kwikemart/marge/ezannoni/flathead-dev/ppc-eabi/gdb:/bin:/usr/bin"
28027 -environment-path -r /usr/local/bin
28028 ^done,path="/usr/local/bin:/usr/bin"
28033 @subheading The @code{-environment-pwd} Command
28034 @findex -environment-pwd
28036 @subsubheading Synopsis
28042 Show the current working directory.
28044 @subsubheading @value{GDBN} Command
28046 The corresponding @value{GDBN} command is @samp{pwd}.
28048 @subsubheading Example
28053 ^done,cwd="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb"
28057 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
28058 @node GDB/MI Thread Commands
28059 @section @sc{gdb/mi} Thread Commands
28062 @subheading The @code{-thread-info} Command
28063 @findex -thread-info
28065 @subsubheading Synopsis
28068 -thread-info [ @var{thread-id} ]
28071 Reports information about either a specific thread, if the
28072 @var{thread-id} parameter is present, or about all threads.
28073 @var{thread-id} is the thread's global thread ID. When printing
28074 information about all threads, also reports the global ID of the
28077 @subsubheading @value{GDBN} Command
28079 The @samp{info thread} command prints the same information
28082 @subsubheading Result
28084 The result is a list of threads. The following attributes are
28085 defined for a given thread:
28089 This field exists only for the current thread. It has the value @samp{*}.
28092 The global identifier that @value{GDBN} uses to refer to the thread.
28095 The identifier that the target uses to refer to the thread.
28098 Extra information about the thread, in a target-specific format. This
28102 The name of the thread. If the user specified a name using the
28103 @code{thread name} command, then this name is given. Otherwise, if
28104 @value{GDBN} can extract the thread name from the target, then that
28105 name is given. If @value{GDBN} cannot find the thread name, then this
28109 The stack frame currently executing in the thread.
28112 The thread's state. The @samp{state} field may have the following
28117 The thread is stopped. Frame information is available for stopped
28121 The thread is running. There's no frame information for running
28127 If @value{GDBN} can find the CPU core on which this thread is running,
28128 then this field is the core identifier. This field is optional.
28132 @subsubheading Example
28137 @{id="2",target-id="Thread 0xb7e14b90 (LWP 21257)",
28138 frame=@{level="0",addr="0xffffe410",func="__kernel_vsyscall",
28139 args=[]@},state="running"@},
28140 @{id="1",target-id="Thread 0xb7e156b0 (LWP 21254)",
28141 frame=@{level="0",addr="0x0804891f",func="foo",
28142 args=[@{name="i",value="10"@}],
28143 file="/tmp/a.c",fullname="/tmp/a.c",line="158"@},
28144 state="running"@}],
28145 current-thread-id="1"
28149 @subheading The @code{-thread-list-ids} Command
28150 @findex -thread-list-ids
28152 @subsubheading Synopsis
28158 Produces a list of the currently known global @value{GDBN} thread ids.
28159 At the end of the list it also prints the total number of such
28162 This command is retained for historical reasons, the
28163 @code{-thread-info} command should be used instead.
28165 @subsubheading @value{GDBN} Command
28167 Part of @samp{info threads} supplies the same information.
28169 @subsubheading Example
28174 ^done,thread-ids=@{thread-id="3",thread-id="2",thread-id="1"@},
28175 current-thread-id="1",number-of-threads="3"
28180 @subheading The @code{-thread-select} Command
28181 @findex -thread-select
28183 @subsubheading Synopsis
28186 -thread-select @var{thread-id}
28189 Make thread with global thread number @var{thread-id} the current
28190 thread. It prints the number of the new current thread, and the
28191 topmost frame for that thread.
28193 This command is deprecated in favor of explicitly using the
28194 @samp{--thread} option to each command.
28196 @subsubheading @value{GDBN} Command
28198 The corresponding @value{GDBN} command is @samp{thread}.
28200 @subsubheading Example
28207 *stopped,reason="end-stepping-range",thread-id="2",line="187",
28208 file="../../../devo/gdb/testsuite/gdb.threads/linux-dp.c"
28212 thread-ids=@{thread-id="3",thread-id="2",thread-id="1"@},
28213 number-of-threads="3"
28216 ^done,new-thread-id="3",
28217 frame=@{level="0",func="vprintf",
28218 args=[@{name="format",value="0x8048e9c \"%*s%c %d %c\\n\""@},
28219 @{name="arg",value="0x2"@}],file="vprintf.c",line="31"@}
28223 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
28224 @node GDB/MI Ada Tasking Commands
28225 @section @sc{gdb/mi} Ada Tasking Commands
28227 @subheading The @code{-ada-task-info} Command
28228 @findex -ada-task-info
28230 @subsubheading Synopsis
28233 -ada-task-info [ @var{task-id} ]
28236 Reports information about either a specific Ada task, if the
28237 @var{task-id} parameter is present, or about all Ada tasks.
28239 @subsubheading @value{GDBN} Command
28241 The @samp{info tasks} command prints the same information
28242 about all Ada tasks (@pxref{Ada Tasks}).
28244 @subsubheading Result
28246 The result is a table of Ada tasks. The following columns are
28247 defined for each Ada task:
28251 This field exists only for the current thread. It has the value @samp{*}.
28254 The identifier that @value{GDBN} uses to refer to the Ada task.
28257 The identifier that the target uses to refer to the Ada task.
28260 The global thread identifier of the thread corresponding to the Ada
28263 This field should always exist, as Ada tasks are always implemented
28264 on top of a thread. But if @value{GDBN} cannot find this corresponding
28265 thread for any reason, the field is omitted.
28268 This field exists only when the task was created by another task.
28269 In this case, it provides the ID of the parent task.
28272 The base priority of the task.
28275 The current state of the task. For a detailed description of the
28276 possible states, see @ref{Ada Tasks}.
28279 The name of the task.
28283 @subsubheading Example
28287 ^done,tasks=@{nr_rows="3",nr_cols="8",
28288 hdr=[@{width="1",alignment="-1",col_name="current",colhdr=""@},
28289 @{width="3",alignment="1",col_name="id",colhdr="ID"@},
28290 @{width="9",alignment="1",col_name="task-id",colhdr="TID"@},
28291 @{width="4",alignment="1",col_name="thread-id",colhdr=""@},
28292 @{width="4",alignment="1",col_name="parent-id",colhdr="P-ID"@},
28293 @{width="3",alignment="1",col_name="priority",colhdr="Pri"@},
28294 @{width="22",alignment="-1",col_name="state",colhdr="State"@},
28295 @{width="1",alignment="2",col_name="name",colhdr="Name"@}],
28296 body=[@{current="*",id="1",task-id=" 644010",thread-id="1",priority="48",
28297 state="Child Termination Wait",name="main_task"@}]@}
28301 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
28302 @node GDB/MI Program Execution
28303 @section @sc{gdb/mi} Program Execution
28305 These are the asynchronous commands which generate the out-of-band
28306 record @samp{*stopped}. Currently @value{GDBN} only really executes
28307 asynchronously with remote targets and this interaction is mimicked in
28310 @subheading The @code{-exec-continue} Command
28311 @findex -exec-continue
28313 @subsubheading Synopsis
28316 -exec-continue [--reverse] [--all|--thread-group N]
28319 Resumes the execution of the inferior program, which will continue
28320 to execute until it reaches a debugger stop event. If the
28321 @samp{--reverse} option is specified, execution resumes in reverse until
28322 it reaches a stop event. Stop events may include
28325 breakpoints or watchpoints
28327 signals or exceptions
28329 the end of the process (or its beginning under @samp{--reverse})
28331 the end or beginning of a replay log if one is being used.
28333 In all-stop mode (@pxref{All-Stop
28334 Mode}), may resume only one thread, or all threads, depending on the
28335 value of the @samp{scheduler-locking} variable. If @samp{--all} is
28336 specified, all threads (in all inferiors) will be resumed. The @samp{--all} option is
28337 ignored in all-stop mode. If the @samp{--thread-group} options is
28338 specified, then all threads in that thread group are resumed.
28340 @subsubheading @value{GDBN} Command
28342 The corresponding @value{GDBN} corresponding is @samp{continue}.
28344 @subsubheading Example
28351 *stopped,reason="breakpoint-hit",disp="keep",bkptno="2",frame=@{
28352 func="foo",args=[],file="hello.c",fullname="/home/foo/bar/hello.c",
28358 @subheading The @code{-exec-finish} Command
28359 @findex -exec-finish
28361 @subsubheading Synopsis
28364 -exec-finish [--reverse]
28367 Resumes the execution of the inferior program until the current
28368 function is exited. Displays the results returned by the function.
28369 If the @samp{--reverse} option is specified, resumes the reverse
28370 execution of the inferior program until the point where current
28371 function was called.
28373 @subsubheading @value{GDBN} Command
28375 The corresponding @value{GDBN} command is @samp{finish}.
28377 @subsubheading Example
28379 Function returning @code{void}.
28386 *stopped,reason="function-finished",frame=@{func="main",args=[],
28387 file="hello.c",fullname="/home/foo/bar/hello.c",line="7"@}
28391 Function returning other than @code{void}. The name of the internal
28392 @value{GDBN} variable storing the result is printed, together with the
28399 *stopped,reason="function-finished",frame=@{addr="0x000107b0",func="foo",
28400 args=[@{name="a",value="1"],@{name="b",value="9"@}@},
28401 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
28402 gdb-result-var="$1",return-value="0"
28407 @subheading The @code{-exec-interrupt} Command
28408 @findex -exec-interrupt
28410 @subsubheading Synopsis
28413 -exec-interrupt [--all|--thread-group N]
28416 Interrupts the background execution of the target. Note how the token
28417 associated with the stop message is the one for the execution command
28418 that has been interrupted. The token for the interrupt itself only
28419 appears in the @samp{^done} output. If the user is trying to
28420 interrupt a non-running program, an error message will be printed.
28422 Note that when asynchronous execution is enabled, this command is
28423 asynchronous just like other execution commands. That is, first the
28424 @samp{^done} response will be printed, and the target stop will be
28425 reported after that using the @samp{*stopped} notification.
28427 In non-stop mode, only the context thread is interrupted by default.
28428 All threads (in all inferiors) will be interrupted if the
28429 @samp{--all} option is specified. If the @samp{--thread-group}
28430 option is specified, all threads in that group will be interrupted.
28432 @subsubheading @value{GDBN} Command
28434 The corresponding @value{GDBN} command is @samp{interrupt}.
28436 @subsubheading Example
28447 111*stopped,signal-name="SIGINT",signal-meaning="Interrupt",
28448 frame=@{addr="0x00010140",func="foo",args=[],file="try.c",
28449 fullname="/home/foo/bar/try.c",line="13"@}
28454 ^error,msg="mi_cmd_exec_interrupt: Inferior not executing."
28458 @subheading The @code{-exec-jump} Command
28461 @subsubheading Synopsis
28464 -exec-jump @var{location}
28467 Resumes execution of the inferior program at the location specified by
28468 parameter. @xref{Specify Location}, for a description of the
28469 different forms of @var{location}.
28471 @subsubheading @value{GDBN} Command
28473 The corresponding @value{GDBN} command is @samp{jump}.
28475 @subsubheading Example
28478 -exec-jump foo.c:10
28479 *running,thread-id="all"
28484 @subheading The @code{-exec-next} Command
28487 @subsubheading Synopsis
28490 -exec-next [--reverse]
28493 Resumes execution of the inferior program, stopping when the beginning
28494 of the next source line is reached.
28496 If the @samp{--reverse} option is specified, resumes reverse execution
28497 of the inferior program, stopping at the beginning of the previous
28498 source line. If you issue this command on the first line of a
28499 function, it will take you back to the caller of that function, to the
28500 source line where the function was called.
28503 @subsubheading @value{GDBN} Command
28505 The corresponding @value{GDBN} command is @samp{next}.
28507 @subsubheading Example
28513 *stopped,reason="end-stepping-range",line="8",file="hello.c"
28518 @subheading The @code{-exec-next-instruction} Command
28519 @findex -exec-next-instruction
28521 @subsubheading Synopsis
28524 -exec-next-instruction [--reverse]
28527 Executes one machine instruction. If the instruction is a function
28528 call, continues until the function returns. If the program stops at an
28529 instruction in the middle of a source line, the address will be
28532 If the @samp{--reverse} option is specified, resumes reverse execution
28533 of the inferior program, stopping at the previous instruction. If the
28534 previously executed instruction was a return from another function,
28535 it will continue to execute in reverse until the call to that function
28536 (from the current stack frame) is reached.
28538 @subsubheading @value{GDBN} Command
28540 The corresponding @value{GDBN} command is @samp{nexti}.
28542 @subsubheading Example
28546 -exec-next-instruction
28550 *stopped,reason="end-stepping-range",
28551 addr="0x000100d4",line="5",file="hello.c"
28556 @subheading The @code{-exec-return} Command
28557 @findex -exec-return
28559 @subsubheading Synopsis
28565 Makes current function return immediately. Doesn't execute the inferior.
28566 Displays the new current frame.
28568 @subsubheading @value{GDBN} Command
28570 The corresponding @value{GDBN} command is @samp{return}.
28572 @subsubheading Example
28576 200-break-insert callee4
28577 200^done,bkpt=@{number="1",addr="0x00010734",
28578 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",line="8"@}
28583 000*stopped,reason="breakpoint-hit",disp="keep",bkptno="1",
28584 frame=@{func="callee4",args=[],
28585 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
28586 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="8"@}
28592 111^done,frame=@{level="0",func="callee3",
28593 args=[@{name="strarg",
28594 value="0x11940 \"A string argument.\""@}],
28595 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
28596 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="18"@}
28601 @subheading The @code{-exec-run} Command
28604 @subsubheading Synopsis
28607 -exec-run [ --all | --thread-group N ] [ --start ]
28610 Starts execution of the inferior from the beginning. The inferior
28611 executes until either a breakpoint is encountered or the program
28612 exits. In the latter case the output will include an exit code, if
28613 the program has exited exceptionally.
28615 When neither the @samp{--all} nor the @samp{--thread-group} option
28616 is specified, the current inferior is started. If the
28617 @samp{--thread-group} option is specified, it should refer to a thread
28618 group of type @samp{process}, and that thread group will be started.
28619 If the @samp{--all} option is specified, then all inferiors will be started.
28621 Using the @samp{--start} option instructs the debugger to stop
28622 the execution at the start of the inferior's main subprogram,
28623 following the same behavior as the @code{start} command
28624 (@pxref{Starting}).
28626 @subsubheading @value{GDBN} Command
28628 The corresponding @value{GDBN} command is @samp{run}.
28630 @subsubheading Examples
28635 ^done,bkpt=@{number="1",addr="0x0001072c",file="recursive2.c",line="4"@}
28640 *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",
28641 frame=@{func="main",args=[],file="recursive2.c",
28642 fullname="/home/foo/bar/recursive2.c",line="4"@}
28647 Program exited normally:
28655 *stopped,reason="exited-normally"
28660 Program exited exceptionally:
28668 *stopped,reason="exited",exit-code="01"
28672 Another way the program can terminate is if it receives a signal such as
28673 @code{SIGINT}. In this case, @sc{gdb/mi} displays this:
28677 *stopped,reason="exited-signalled",signal-name="SIGINT",
28678 signal-meaning="Interrupt"
28682 @c @subheading -exec-signal
28685 @subheading The @code{-exec-step} Command
28688 @subsubheading Synopsis
28691 -exec-step [--reverse]
28694 Resumes execution of the inferior program, stopping when the beginning
28695 of the next source line is reached, if the next source line is not a
28696 function call. If it is, stop at the first instruction of the called
28697 function. If the @samp{--reverse} option is specified, resumes reverse
28698 execution of the inferior program, stopping at the beginning of the
28699 previously executed source line.
28701 @subsubheading @value{GDBN} Command
28703 The corresponding @value{GDBN} command is @samp{step}.
28705 @subsubheading Example
28707 Stepping into a function:
28713 *stopped,reason="end-stepping-range",
28714 frame=@{func="foo",args=[@{name="a",value="10"@},
28715 @{name="b",value="0"@}],file="recursive2.c",
28716 fullname="/home/foo/bar/recursive2.c",line="11"@}
28726 *stopped,reason="end-stepping-range",line="14",file="recursive2.c"
28731 @subheading The @code{-exec-step-instruction} Command
28732 @findex -exec-step-instruction
28734 @subsubheading Synopsis
28737 -exec-step-instruction [--reverse]
28740 Resumes the inferior which executes one machine instruction. If the
28741 @samp{--reverse} option is specified, resumes reverse execution of the
28742 inferior program, stopping at the previously executed instruction.
28743 The output, once @value{GDBN} has stopped, will vary depending on
28744 whether we have stopped in the middle of a source line or not. In the
28745 former case, the address at which the program stopped will be printed
28748 @subsubheading @value{GDBN} Command
28750 The corresponding @value{GDBN} command is @samp{stepi}.
28752 @subsubheading Example
28756 -exec-step-instruction
28760 *stopped,reason="end-stepping-range",
28761 frame=@{func="foo",args=[],file="try.c",
28762 fullname="/home/foo/bar/try.c",line="10"@}
28764 -exec-step-instruction
28768 *stopped,reason="end-stepping-range",
28769 frame=@{addr="0x000100f4",func="foo",args=[],file="try.c",
28770 fullname="/home/foo/bar/try.c",line="10"@}
28775 @subheading The @code{-exec-until} Command
28776 @findex -exec-until
28778 @subsubheading Synopsis
28781 -exec-until [ @var{location} ]
28784 Executes the inferior until the @var{location} specified in the
28785 argument is reached. If there is no argument, the inferior executes
28786 until a source line greater than the current one is reached. The
28787 reason for stopping in this case will be @samp{location-reached}.
28789 @subsubheading @value{GDBN} Command
28791 The corresponding @value{GDBN} command is @samp{until}.
28793 @subsubheading Example
28797 -exec-until recursive2.c:6
28801 *stopped,reason="location-reached",frame=@{func="main",args=[],
28802 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="6"@}
28807 @subheading -file-clear
28808 Is this going away????
28811 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
28812 @node GDB/MI Stack Manipulation
28813 @section @sc{gdb/mi} Stack Manipulation Commands
28815 @subheading The @code{-enable-frame-filters} Command
28816 @findex -enable-frame-filters
28819 -enable-frame-filters
28822 @value{GDBN} allows Python-based frame filters to affect the output of
28823 the MI commands relating to stack traces. As there is no way to
28824 implement this in a fully backward-compatible way, a front end must
28825 request that this functionality be enabled.
28827 Once enabled, this feature cannot be disabled.
28829 Note that if Python support has not been compiled into @value{GDBN},
28830 this command will still succeed (and do nothing).
28832 @subheading The @code{-stack-info-frame} Command
28833 @findex -stack-info-frame
28835 @subsubheading Synopsis
28841 Get info on the selected frame.
28843 @subsubheading @value{GDBN} Command
28845 The corresponding @value{GDBN} command is @samp{info frame} or @samp{frame}
28846 (without arguments).
28848 @subsubheading Example
28853 ^done,frame=@{level="1",addr="0x0001076c",func="callee3",
28854 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
28855 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="17"@}
28859 @subheading The @code{-stack-info-depth} Command
28860 @findex -stack-info-depth
28862 @subsubheading Synopsis
28865 -stack-info-depth [ @var{max-depth} ]
28868 Return the depth of the stack. If the integer argument @var{max-depth}
28869 is specified, do not count beyond @var{max-depth} frames.
28871 @subsubheading @value{GDBN} Command
28873 There's no equivalent @value{GDBN} command.
28875 @subsubheading Example
28877 For a stack with frame levels 0 through 11:
28884 -stack-info-depth 4
28887 -stack-info-depth 12
28890 -stack-info-depth 11
28893 -stack-info-depth 13
28898 @anchor{-stack-list-arguments}
28899 @subheading The @code{-stack-list-arguments} Command
28900 @findex -stack-list-arguments
28902 @subsubheading Synopsis
28905 -stack-list-arguments [ --no-frame-filters ] [ --skip-unavailable ] @var{print-values}
28906 [ @var{low-frame} @var{high-frame} ]
28909 Display a list of the arguments for the frames between @var{low-frame}
28910 and @var{high-frame} (inclusive). If @var{low-frame} and
28911 @var{high-frame} are not provided, list the arguments for the whole
28912 call stack. If the two arguments are equal, show the single frame
28913 at the corresponding level. It is an error if @var{low-frame} is
28914 larger than the actual number of frames. On the other hand,
28915 @var{high-frame} may be larger than the actual number of frames, in
28916 which case only existing frames will be returned.
28918 If @var{print-values} is 0 or @code{--no-values}, print only the names of
28919 the variables; if it is 1 or @code{--all-values}, print also their
28920 values; and if it is 2 or @code{--simple-values}, print the name,
28921 type and value for simple data types, and the name and type for arrays,
28922 structures and unions. If the option @code{--no-frame-filters} is
28923 supplied, then Python frame filters will not be executed.
28925 If the @code{--skip-unavailable} option is specified, arguments that
28926 are not available are not listed. Partially available arguments
28927 are still displayed, however.
28929 Use of this command to obtain arguments in a single frame is
28930 deprecated in favor of the @samp{-stack-list-variables} command.
28932 @subsubheading @value{GDBN} Command
28934 @value{GDBN} does not have an equivalent command. @code{gdbtk} has a
28935 @samp{gdb_get_args} command which partially overlaps with the
28936 functionality of @samp{-stack-list-arguments}.
28938 @subsubheading Example
28945 frame=@{level="0",addr="0x00010734",func="callee4",
28946 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
28947 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="8"@},
28948 frame=@{level="1",addr="0x0001076c",func="callee3",
28949 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
28950 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="17"@},
28951 frame=@{level="2",addr="0x0001078c",func="callee2",
28952 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
28953 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="22"@},
28954 frame=@{level="3",addr="0x000107b4",func="callee1",
28955 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
28956 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="27"@},
28957 frame=@{level="4",addr="0x000107e0",func="main",
28958 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
28959 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="32"@}]
28961 -stack-list-arguments 0
28964 frame=@{level="0",args=[]@},
28965 frame=@{level="1",args=[name="strarg"]@},
28966 frame=@{level="2",args=[name="intarg",name="strarg"]@},
28967 frame=@{level="3",args=[name="intarg",name="strarg",name="fltarg"]@},
28968 frame=@{level="4",args=[]@}]
28970 -stack-list-arguments 1
28973 frame=@{level="0",args=[]@},
28975 args=[@{name="strarg",value="0x11940 \"A string argument.\""@}]@},
28976 frame=@{level="2",args=[
28977 @{name="intarg",value="2"@},
28978 @{name="strarg",value="0x11940 \"A string argument.\""@}]@},
28979 @{frame=@{level="3",args=[
28980 @{name="intarg",value="2"@},
28981 @{name="strarg",value="0x11940 \"A string argument.\""@},
28982 @{name="fltarg",value="3.5"@}]@},
28983 frame=@{level="4",args=[]@}]
28985 -stack-list-arguments 0 2 2
28986 ^done,stack-args=[frame=@{level="2",args=[name="intarg",name="strarg"]@}]
28988 -stack-list-arguments 1 2 2
28989 ^done,stack-args=[frame=@{level="2",
28990 args=[@{name="intarg",value="2"@},
28991 @{name="strarg",value="0x11940 \"A string argument.\""@}]@}]
28995 @c @subheading -stack-list-exception-handlers
28998 @anchor{-stack-list-frames}
28999 @subheading The @code{-stack-list-frames} Command
29000 @findex -stack-list-frames
29002 @subsubheading Synopsis
29005 -stack-list-frames [ --no-frame-filters @var{low-frame} @var{high-frame} ]
29008 List the frames currently on the stack. For each frame it displays the
29013 The frame number, 0 being the topmost frame, i.e., the innermost function.
29015 The @code{$pc} value for that frame.
29019 File name of the source file where the function lives.
29020 @item @var{fullname}
29021 The full file name of the source file where the function lives.
29023 Line number corresponding to the @code{$pc}.
29025 The shared library where this function is defined. This is only given
29026 if the frame's function is not known.
29029 If invoked without arguments, this command prints a backtrace for the
29030 whole stack. If given two integer arguments, it shows the frames whose
29031 levels are between the two arguments (inclusive). If the two arguments
29032 are equal, it shows the single frame at the corresponding level. It is
29033 an error if @var{low-frame} is larger than the actual number of
29034 frames. On the other hand, @var{high-frame} may be larger than the
29035 actual number of frames, in which case only existing frames will be
29036 returned. If the option @code{--no-frame-filters} is supplied, then
29037 Python frame filters will not be executed.
29039 @subsubheading @value{GDBN} Command
29041 The corresponding @value{GDBN} commands are @samp{backtrace} and @samp{where}.
29043 @subsubheading Example
29045 Full stack backtrace:
29051 [frame=@{level="0",addr="0x0001076c",func="foo",
29052 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="11"@},
29053 frame=@{level="1",addr="0x000107a4",func="foo",
29054 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
29055 frame=@{level="2",addr="0x000107a4",func="foo",
29056 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
29057 frame=@{level="3",addr="0x000107a4",func="foo",
29058 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
29059 frame=@{level="4",addr="0x000107a4",func="foo",
29060 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
29061 frame=@{level="5",addr="0x000107a4",func="foo",
29062 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
29063 frame=@{level="6",addr="0x000107a4",func="foo",
29064 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
29065 frame=@{level="7",addr="0x000107a4",func="foo",
29066 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
29067 frame=@{level="8",addr="0x000107a4",func="foo",
29068 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
29069 frame=@{level="9",addr="0x000107a4",func="foo",
29070 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
29071 frame=@{level="10",addr="0x000107a4",func="foo",
29072 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
29073 frame=@{level="11",addr="0x00010738",func="main",
29074 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="4"@}]
29078 Show frames between @var{low_frame} and @var{high_frame}:
29082 -stack-list-frames 3 5
29084 [frame=@{level="3",addr="0x000107a4",func="foo",
29085 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
29086 frame=@{level="4",addr="0x000107a4",func="foo",
29087 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@},
29088 frame=@{level="5",addr="0x000107a4",func="foo",
29089 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@}]
29093 Show a single frame:
29097 -stack-list-frames 3 3
29099 [frame=@{level="3",addr="0x000107a4",func="foo",
29100 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14"@}]
29105 @subheading The @code{-stack-list-locals} Command
29106 @findex -stack-list-locals
29107 @anchor{-stack-list-locals}
29109 @subsubheading Synopsis
29112 -stack-list-locals [ --no-frame-filters ] [ --skip-unavailable ] @var{print-values}
29115 Display the local variable names for the selected frame. If
29116 @var{print-values} is 0 or @code{--no-values}, print only the names of
29117 the variables; if it is 1 or @code{--all-values}, print also their
29118 values; and if it is 2 or @code{--simple-values}, print the name,
29119 type and value for simple data types, and the name and type for arrays,
29120 structures and unions. In this last case, a frontend can immediately
29121 display the value of simple data types and create variable objects for
29122 other data types when the user wishes to explore their values in
29123 more detail. If the option @code{--no-frame-filters} is supplied, then
29124 Python frame filters will not be executed.
29126 If the @code{--skip-unavailable} option is specified, local variables
29127 that are not available are not listed. Partially available local
29128 variables are still displayed, however.
29130 This command is deprecated in favor of the
29131 @samp{-stack-list-variables} command.
29133 @subsubheading @value{GDBN} Command
29135 @samp{info locals} in @value{GDBN}, @samp{gdb_get_locals} in @code{gdbtk}.
29137 @subsubheading Example
29141 -stack-list-locals 0
29142 ^done,locals=[name="A",name="B",name="C"]
29144 -stack-list-locals --all-values
29145 ^done,locals=[@{name="A",value="1"@},@{name="B",value="2"@},
29146 @{name="C",value="@{1, 2, 3@}"@}]
29147 -stack-list-locals --simple-values
29148 ^done,locals=[@{name="A",type="int",value="1"@},
29149 @{name="B",type="int",value="2"@},@{name="C",type="int [3]"@}]
29153 @anchor{-stack-list-variables}
29154 @subheading The @code{-stack-list-variables} Command
29155 @findex -stack-list-variables
29157 @subsubheading Synopsis
29160 -stack-list-variables [ --no-frame-filters ] [ --skip-unavailable ] @var{print-values}
29163 Display the names of local variables and function arguments for the selected frame. If
29164 @var{print-values} is 0 or @code{--no-values}, print only the names of
29165 the variables; if it is 1 or @code{--all-values}, print also their
29166 values; and if it is 2 or @code{--simple-values}, print the name,
29167 type and value for simple data types, and the name and type for arrays,
29168 structures and unions. If the option @code{--no-frame-filters} is
29169 supplied, then Python frame filters will not be executed.
29171 If the @code{--skip-unavailable} option is specified, local variables
29172 and arguments that are not available are not listed. Partially
29173 available arguments and local variables are still displayed, however.
29175 @subsubheading Example
29179 -stack-list-variables --thread 1 --frame 0 --all-values
29180 ^done,variables=[@{name="x",value="11"@},@{name="s",value="@{a = 1, b = 2@}"@}]
29185 @subheading The @code{-stack-select-frame} Command
29186 @findex -stack-select-frame
29188 @subsubheading Synopsis
29191 -stack-select-frame @var{framenum}
29194 Change the selected frame. Select a different frame @var{framenum} on
29197 This command in deprecated in favor of passing the @samp{--frame}
29198 option to every command.
29200 @subsubheading @value{GDBN} Command
29202 The corresponding @value{GDBN} commands are @samp{frame}, @samp{up},
29203 @samp{down}, @samp{select-frame}, @samp{up-silent}, and @samp{down-silent}.
29205 @subsubheading Example
29209 -stack-select-frame 2
29214 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
29215 @node GDB/MI Variable Objects
29216 @section @sc{gdb/mi} Variable Objects
29220 @subheading Motivation for Variable Objects in @sc{gdb/mi}
29222 For the implementation of a variable debugger window (locals, watched
29223 expressions, etc.), we are proposing the adaptation of the existing code
29224 used by @code{Insight}.
29226 The two main reasons for that are:
29230 It has been proven in practice (it is already on its second generation).
29233 It will shorten development time (needless to say how important it is
29237 The original interface was designed to be used by Tcl code, so it was
29238 slightly changed so it could be used through @sc{gdb/mi}. This section
29239 describes the @sc{gdb/mi} operations that will be available and gives some
29240 hints about their use.
29242 @emph{Note}: In addition to the set of operations described here, we
29243 expect the @sc{gui} implementation of a variable window to require, at
29244 least, the following operations:
29247 @item @code{-gdb-show} @code{output-radix}
29248 @item @code{-stack-list-arguments}
29249 @item @code{-stack-list-locals}
29250 @item @code{-stack-select-frame}
29255 @subheading Introduction to Variable Objects
29257 @cindex variable objects in @sc{gdb/mi}
29259 Variable objects are "object-oriented" MI interface for examining and
29260 changing values of expressions. Unlike some other MI interfaces that
29261 work with expressions, variable objects are specifically designed for
29262 simple and efficient presentation in the frontend. A variable object
29263 is identified by string name. When a variable object is created, the
29264 frontend specifies the expression for that variable object. The
29265 expression can be a simple variable, or it can be an arbitrary complex
29266 expression, and can even involve CPU registers. After creating a
29267 variable object, the frontend can invoke other variable object
29268 operations---for example to obtain or change the value of a variable
29269 object, or to change display format.
29271 Variable objects have hierarchical tree structure. Any variable object
29272 that corresponds to a composite type, such as structure in C, has
29273 a number of child variable objects, for example corresponding to each
29274 element of a structure. A child variable object can itself have
29275 children, recursively. Recursion ends when we reach
29276 leaf variable objects, which always have built-in types. Child variable
29277 objects are created only by explicit request, so if a frontend
29278 is not interested in the children of a particular variable object, no
29279 child will be created.
29281 For a leaf variable object it is possible to obtain its value as a
29282 string, or set the value from a string. String value can be also
29283 obtained for a non-leaf variable object, but it's generally a string
29284 that only indicates the type of the object, and does not list its
29285 contents. Assignment to a non-leaf variable object is not allowed.
29287 A frontend does not need to read the values of all variable objects each time
29288 the program stops. Instead, MI provides an update command that lists all
29289 variable objects whose values has changed since the last update
29290 operation. This considerably reduces the amount of data that must
29291 be transferred to the frontend. As noted above, children variable
29292 objects are created on demand, and only leaf variable objects have a
29293 real value. As result, gdb will read target memory only for leaf
29294 variables that frontend has created.
29296 The automatic update is not always desirable. For example, a frontend
29297 might want to keep a value of some expression for future reference,
29298 and never update it. For another example, fetching memory is
29299 relatively slow for embedded targets, so a frontend might want
29300 to disable automatic update for the variables that are either not
29301 visible on the screen, or ``closed''. This is possible using so
29302 called ``frozen variable objects''. Such variable objects are never
29303 implicitly updated.
29305 Variable objects can be either @dfn{fixed} or @dfn{floating}. For the
29306 fixed variable object, the expression is parsed when the variable
29307 object is created, including associating identifiers to specific
29308 variables. The meaning of expression never changes. For a floating
29309 variable object the values of variables whose names appear in the
29310 expressions are re-evaluated every time in the context of the current
29311 frame. Consider this example:
29316 struct work_state state;
29323 If a fixed variable object for the @code{state} variable is created in
29324 this function, and we enter the recursive call, the variable
29325 object will report the value of @code{state} in the top-level
29326 @code{do_work} invocation. On the other hand, a floating variable
29327 object will report the value of @code{state} in the current frame.
29329 If an expression specified when creating a fixed variable object
29330 refers to a local variable, the variable object becomes bound to the
29331 thread and frame in which the variable object is created. When such
29332 variable object is updated, @value{GDBN} makes sure that the
29333 thread/frame combination the variable object is bound to still exists,
29334 and re-evaluates the variable object in context of that thread/frame.
29336 The following is the complete set of @sc{gdb/mi} operations defined to
29337 access this functionality:
29339 @multitable @columnfractions .4 .6
29340 @item @strong{Operation}
29341 @tab @strong{Description}
29343 @item @code{-enable-pretty-printing}
29344 @tab enable Python-based pretty-printing
29345 @item @code{-var-create}
29346 @tab create a variable object
29347 @item @code{-var-delete}
29348 @tab delete the variable object and/or its children
29349 @item @code{-var-set-format}
29350 @tab set the display format of this variable
29351 @item @code{-var-show-format}
29352 @tab show the display format of this variable
29353 @item @code{-var-info-num-children}
29354 @tab tells how many children this object has
29355 @item @code{-var-list-children}
29356 @tab return a list of the object's children
29357 @item @code{-var-info-type}
29358 @tab show the type of this variable object
29359 @item @code{-var-info-expression}
29360 @tab print parent-relative expression that this variable object represents
29361 @item @code{-var-info-path-expression}
29362 @tab print full expression that this variable object represents
29363 @item @code{-var-show-attributes}
29364 @tab is this variable editable? does it exist here?
29365 @item @code{-var-evaluate-expression}
29366 @tab get the value of this variable
29367 @item @code{-var-assign}
29368 @tab set the value of this variable
29369 @item @code{-var-update}
29370 @tab update the variable and its children
29371 @item @code{-var-set-frozen}
29372 @tab set frozeness attribute
29373 @item @code{-var-set-update-range}
29374 @tab set range of children to display on update
29377 In the next subsection we describe each operation in detail and suggest
29378 how it can be used.
29380 @subheading Description And Use of Operations on Variable Objects
29382 @subheading The @code{-enable-pretty-printing} Command
29383 @findex -enable-pretty-printing
29386 -enable-pretty-printing
29389 @value{GDBN} allows Python-based visualizers to affect the output of the
29390 MI variable object commands. However, because there was no way to
29391 implement this in a fully backward-compatible way, a front end must
29392 request that this functionality be enabled.
29394 Once enabled, this feature cannot be disabled.
29396 Note that if Python support has not been compiled into @value{GDBN},
29397 this command will still succeed (and do nothing).
29399 This feature is currently (as of @value{GDBN} 7.0) experimental, and
29400 may work differently in future versions of @value{GDBN}.
29402 @subheading The @code{-var-create} Command
29403 @findex -var-create
29405 @subsubheading Synopsis
29408 -var-create @{@var{name} | "-"@}
29409 @{@var{frame-addr} | "*" | "@@"@} @var{expression}
29412 This operation creates a variable object, which allows the monitoring of
29413 a variable, the result of an expression, a memory cell or a CPU
29416 The @var{name} parameter is the string by which the object can be
29417 referenced. It must be unique. If @samp{-} is specified, the varobj
29418 system will generate a string ``varNNNNNN'' automatically. It will be
29419 unique provided that one does not specify @var{name} of that format.
29420 The command fails if a duplicate name is found.
29422 The frame under which the expression should be evaluated can be
29423 specified by @var{frame-addr}. A @samp{*} indicates that the current
29424 frame should be used. A @samp{@@} indicates that a floating variable
29425 object must be created.
29427 @var{expression} is any expression valid on the current language set (must not
29428 begin with a @samp{*}), or one of the following:
29432 @samp{*@var{addr}}, where @var{addr} is the address of a memory cell
29435 @samp{*@var{addr}-@var{addr}} --- a memory address range (TBD)
29438 @samp{$@var{regname}} --- a CPU register name
29441 @cindex dynamic varobj
29442 A varobj's contents may be provided by a Python-based pretty-printer. In this
29443 case the varobj is known as a @dfn{dynamic varobj}. Dynamic varobjs
29444 have slightly different semantics in some cases. If the
29445 @code{-enable-pretty-printing} command is not sent, then @value{GDBN}
29446 will never create a dynamic varobj. This ensures backward
29447 compatibility for existing clients.
29449 @subsubheading Result
29451 This operation returns attributes of the newly-created varobj. These
29456 The name of the varobj.
29459 The number of children of the varobj. This number is not necessarily
29460 reliable for a dynamic varobj. Instead, you must examine the
29461 @samp{has_more} attribute.
29464 The varobj's scalar value. For a varobj whose type is some sort of
29465 aggregate (e.g., a @code{struct}), or for a dynamic varobj, this value
29466 will not be interesting.
29469 The varobj's type. This is a string representation of the type, as
29470 would be printed by the @value{GDBN} CLI. If @samp{print object}
29471 (@pxref{Print Settings, set print object}) is set to @code{on}, the
29472 @emph{actual} (derived) type of the object is shown rather than the
29473 @emph{declared} one.
29476 If a variable object is bound to a specific thread, then this is the
29477 thread's global identifier.
29480 For a dynamic varobj, this indicates whether there appear to be any
29481 children available. For a non-dynamic varobj, this will be 0.
29484 This attribute will be present and have the value @samp{1} if the
29485 varobj is a dynamic varobj. If the varobj is not a dynamic varobj,
29486 then this attribute will not be present.
29489 A dynamic varobj can supply a display hint to the front end. The
29490 value comes directly from the Python pretty-printer object's
29491 @code{display_hint} method. @xref{Pretty Printing API}.
29494 Typical output will look like this:
29497 name="@var{name}",numchild="@var{N}",type="@var{type}",thread-id="@var{M}",
29498 has_more="@var{has_more}"
29502 @subheading The @code{-var-delete} Command
29503 @findex -var-delete
29505 @subsubheading Synopsis
29508 -var-delete [ -c ] @var{name}
29511 Deletes a previously created variable object and all of its children.
29512 With the @samp{-c} option, just deletes the children.
29514 Returns an error if the object @var{name} is not found.
29517 @subheading The @code{-var-set-format} Command
29518 @findex -var-set-format
29520 @subsubheading Synopsis
29523 -var-set-format @var{name} @var{format-spec}
29526 Sets the output format for the value of the object @var{name} to be
29529 @anchor{-var-set-format}
29530 The syntax for the @var{format-spec} is as follows:
29533 @var{format-spec} @expansion{}
29534 @{binary | decimal | hexadecimal | octal | natural | zero-hexadecimal@}
29537 The natural format is the default format choosen automatically
29538 based on the variable type (like decimal for an @code{int}, hex
29539 for pointers, etc.).
29541 The zero-hexadecimal format has a representation similar to hexadecimal
29542 but with padding zeroes to the left of the value. For example, a 32-bit
29543 hexadecimal value of 0x1234 would be represented as 0x00001234 in the
29544 zero-hexadecimal format.
29546 For a variable with children, the format is set only on the
29547 variable itself, and the children are not affected.
29549 @subheading The @code{-var-show-format} Command
29550 @findex -var-show-format
29552 @subsubheading Synopsis
29555 -var-show-format @var{name}
29558 Returns the format used to display the value of the object @var{name}.
29561 @var{format} @expansion{}
29566 @subheading The @code{-var-info-num-children} Command
29567 @findex -var-info-num-children
29569 @subsubheading Synopsis
29572 -var-info-num-children @var{name}
29575 Returns the number of children of a variable object @var{name}:
29581 Note that this number is not completely reliable for a dynamic varobj.
29582 It will return the current number of children, but more children may
29586 @subheading The @code{-var-list-children} Command
29587 @findex -var-list-children
29589 @subsubheading Synopsis
29592 -var-list-children [@var{print-values}] @var{name} [@var{from} @var{to}]
29594 @anchor{-var-list-children}
29596 Return a list of the children of the specified variable object and
29597 create variable objects for them, if they do not already exist. With
29598 a single argument or if @var{print-values} has a value of 0 or
29599 @code{--no-values}, print only the names of the variables; if
29600 @var{print-values} is 1 or @code{--all-values}, also print their
29601 values; and if it is 2 or @code{--simple-values} print the name and
29602 value for simple data types and just the name for arrays, structures
29605 @var{from} and @var{to}, if specified, indicate the range of children
29606 to report. If @var{from} or @var{to} is less than zero, the range is
29607 reset and all children will be reported. Otherwise, children starting
29608 at @var{from} (zero-based) and up to and excluding @var{to} will be
29611 If a child range is requested, it will only affect the current call to
29612 @code{-var-list-children}, but not future calls to @code{-var-update}.
29613 For this, you must instead use @code{-var-set-update-range}. The
29614 intent of this approach is to enable a front end to implement any
29615 update approach it likes; for example, scrolling a view may cause the
29616 front end to request more children with @code{-var-list-children}, and
29617 then the front end could call @code{-var-set-update-range} with a
29618 different range to ensure that future updates are restricted to just
29621 For each child the following results are returned:
29626 Name of the variable object created for this child.
29629 The expression to be shown to the user by the front end to designate this child.
29630 For example this may be the name of a structure member.
29632 For a dynamic varobj, this value cannot be used to form an
29633 expression. There is no way to do this at all with a dynamic varobj.
29635 For C/C@t{++} structures there are several pseudo children returned to
29636 designate access qualifiers. For these pseudo children @var{exp} is
29637 @samp{public}, @samp{private}, or @samp{protected}. In this case the
29638 type and value are not present.
29640 A dynamic varobj will not report the access qualifying
29641 pseudo-children, regardless of the language. This information is not
29642 available at all with a dynamic varobj.
29645 Number of children this child has. For a dynamic varobj, this will be
29649 The type of the child. If @samp{print object}
29650 (@pxref{Print Settings, set print object}) is set to @code{on}, the
29651 @emph{actual} (derived) type of the object is shown rather than the
29652 @emph{declared} one.
29655 If values were requested, this is the value.
29658 If this variable object is associated with a thread, this is the
29659 thread's global thread id. Otherwise this result is not present.
29662 If the variable object is frozen, this variable will be present with a value of 1.
29665 A dynamic varobj can supply a display hint to the front end. The
29666 value comes directly from the Python pretty-printer object's
29667 @code{display_hint} method. @xref{Pretty Printing API}.
29670 This attribute will be present and have the value @samp{1} if the
29671 varobj is a dynamic varobj. If the varobj is not a dynamic varobj,
29672 then this attribute will not be present.
29676 The result may have its own attributes:
29680 A dynamic varobj can supply a display hint to the front end. The
29681 value comes directly from the Python pretty-printer object's
29682 @code{display_hint} method. @xref{Pretty Printing API}.
29685 This is an integer attribute which is nonzero if there are children
29686 remaining after the end of the selected range.
29689 @subsubheading Example
29693 -var-list-children n
29694 ^done,numchild=@var{n},children=[child=@{name=@var{name},exp=@var{exp},
29695 numchild=@var{n},type=@var{type}@},@r{(repeats N times)}]
29697 -var-list-children --all-values n
29698 ^done,numchild=@var{n},children=[child=@{name=@var{name},exp=@var{exp},
29699 numchild=@var{n},value=@var{value},type=@var{type}@},@r{(repeats N times)}]
29703 @subheading The @code{-var-info-type} Command
29704 @findex -var-info-type
29706 @subsubheading Synopsis
29709 -var-info-type @var{name}
29712 Returns the type of the specified variable @var{name}. The type is
29713 returned as a string in the same format as it is output by the
29717 type=@var{typename}
29721 @subheading The @code{-var-info-expression} Command
29722 @findex -var-info-expression
29724 @subsubheading Synopsis
29727 -var-info-expression @var{name}
29730 Returns a string that is suitable for presenting this
29731 variable object in user interface. The string is generally
29732 not valid expression in the current language, and cannot be evaluated.
29734 For example, if @code{a} is an array, and variable object
29735 @code{A} was created for @code{a}, then we'll get this output:
29738 (gdb) -var-info-expression A.1
29739 ^done,lang="C",exp="1"
29743 Here, the value of @code{lang} is the language name, which can be
29744 found in @ref{Supported Languages}.
29746 Note that the output of the @code{-var-list-children} command also
29747 includes those expressions, so the @code{-var-info-expression} command
29750 @subheading The @code{-var-info-path-expression} Command
29751 @findex -var-info-path-expression
29753 @subsubheading Synopsis
29756 -var-info-path-expression @var{name}
29759 Returns an expression that can be evaluated in the current
29760 context and will yield the same value that a variable object has.
29761 Compare this with the @code{-var-info-expression} command, which
29762 result can be used only for UI presentation. Typical use of
29763 the @code{-var-info-path-expression} command is creating a
29764 watchpoint from a variable object.
29766 This command is currently not valid for children of a dynamic varobj,
29767 and will give an error when invoked on one.
29769 For example, suppose @code{C} is a C@t{++} class, derived from class
29770 @code{Base}, and that the @code{Base} class has a member called
29771 @code{m_size}. Assume a variable @code{c} is has the type of
29772 @code{C} and a variable object @code{C} was created for variable
29773 @code{c}. Then, we'll get this output:
29775 (gdb) -var-info-path-expression C.Base.public.m_size
29776 ^done,path_expr=((Base)c).m_size)
29779 @subheading The @code{-var-show-attributes} Command
29780 @findex -var-show-attributes
29782 @subsubheading Synopsis
29785 -var-show-attributes @var{name}
29788 List attributes of the specified variable object @var{name}:
29791 status=@var{attr} [ ( ,@var{attr} )* ]
29795 where @var{attr} is @code{@{ @{ editable | noneditable @} | TBD @}}.
29797 @subheading The @code{-var-evaluate-expression} Command
29798 @findex -var-evaluate-expression
29800 @subsubheading Synopsis
29803 -var-evaluate-expression [-f @var{format-spec}] @var{name}
29806 Evaluates the expression that is represented by the specified variable
29807 object and returns its value as a string. The format of the string
29808 can be specified with the @samp{-f} option. The possible values of
29809 this option are the same as for @code{-var-set-format}
29810 (@pxref{-var-set-format}). If the @samp{-f} option is not specified,
29811 the current display format will be used. The current display format
29812 can be changed using the @code{-var-set-format} command.
29818 Note that one must invoke @code{-var-list-children} for a variable
29819 before the value of a child variable can be evaluated.
29821 @subheading The @code{-var-assign} Command
29822 @findex -var-assign
29824 @subsubheading Synopsis
29827 -var-assign @var{name} @var{expression}
29830 Assigns the value of @var{expression} to the variable object specified
29831 by @var{name}. The object must be @samp{editable}. If the variable's
29832 value is altered by the assign, the variable will show up in any
29833 subsequent @code{-var-update} list.
29835 @subsubheading Example
29843 ^done,changelist=[@{name="var1",in_scope="true",type_changed="false"@}]
29847 @subheading The @code{-var-update} Command
29848 @findex -var-update
29850 @subsubheading Synopsis
29853 -var-update [@var{print-values}] @{@var{name} | "*"@}
29856 Reevaluate the expressions corresponding to the variable object
29857 @var{name} and all its direct and indirect children, and return the
29858 list of variable objects whose values have changed; @var{name} must
29859 be a root variable object. Here, ``changed'' means that the result of
29860 @code{-var-evaluate-expression} before and after the
29861 @code{-var-update} is different. If @samp{*} is used as the variable
29862 object names, all existing variable objects are updated, except
29863 for frozen ones (@pxref{-var-set-frozen}). The option
29864 @var{print-values} determines whether both names and values, or just
29865 names are printed. The possible values of this option are the same
29866 as for @code{-var-list-children} (@pxref{-var-list-children}). It is
29867 recommended to use the @samp{--all-values} option, to reduce the
29868 number of MI commands needed on each program stop.
29870 With the @samp{*} parameter, if a variable object is bound to a
29871 currently running thread, it will not be updated, without any
29874 If @code{-var-set-update-range} was previously used on a varobj, then
29875 only the selected range of children will be reported.
29877 @code{-var-update} reports all the changed varobjs in a tuple named
29880 Each item in the change list is itself a tuple holding:
29884 The name of the varobj.
29887 If values were requested for this update, then this field will be
29888 present and will hold the value of the varobj.
29891 @anchor{-var-update}
29892 This field is a string which may take one of three values:
29896 The variable object's current value is valid.
29899 The variable object does not currently hold a valid value but it may
29900 hold one in the future if its associated expression comes back into
29904 The variable object no longer holds a valid value.
29905 This can occur when the executable file being debugged has changed,
29906 either through recompilation or by using the @value{GDBN} @code{file}
29907 command. The front end should normally choose to delete these variable
29911 In the future new values may be added to this list so the front should
29912 be prepared for this possibility. @xref{GDB/MI Development and Front Ends, ,@sc{GDB/MI} Development and Front Ends}.
29915 This is only present if the varobj is still valid. If the type
29916 changed, then this will be the string @samp{true}; otherwise it will
29919 When a varobj's type changes, its children are also likely to have
29920 become incorrect. Therefore, the varobj's children are automatically
29921 deleted when this attribute is @samp{true}. Also, the varobj's update
29922 range, when set using the @code{-var-set-update-range} command, is
29926 If the varobj's type changed, then this field will be present and will
29929 @item new_num_children
29930 For a dynamic varobj, if the number of children changed, or if the
29931 type changed, this will be the new number of children.
29933 The @samp{numchild} field in other varobj responses is generally not
29934 valid for a dynamic varobj -- it will show the number of children that
29935 @value{GDBN} knows about, but because dynamic varobjs lazily
29936 instantiate their children, this will not reflect the number of
29937 children which may be available.
29939 The @samp{new_num_children} attribute only reports changes to the
29940 number of children known by @value{GDBN}. This is the only way to
29941 detect whether an update has removed children (which necessarily can
29942 only happen at the end of the update range).
29945 The display hint, if any.
29948 This is an integer value, which will be 1 if there are more children
29949 available outside the varobj's update range.
29952 This attribute will be present and have the value @samp{1} if the
29953 varobj is a dynamic varobj. If the varobj is not a dynamic varobj,
29954 then this attribute will not be present.
29957 If new children were added to a dynamic varobj within the selected
29958 update range (as set by @code{-var-set-update-range}), then they will
29959 be listed in this attribute.
29962 @subsubheading Example
29969 -var-update --all-values var1
29970 ^done,changelist=[@{name="var1",value="3",in_scope="true",
29971 type_changed="false"@}]
29975 @subheading The @code{-var-set-frozen} Command
29976 @findex -var-set-frozen
29977 @anchor{-var-set-frozen}
29979 @subsubheading Synopsis
29982 -var-set-frozen @var{name} @var{flag}
29985 Set the frozenness flag on the variable object @var{name}. The
29986 @var{flag} parameter should be either @samp{1} to make the variable
29987 frozen or @samp{0} to make it unfrozen. If a variable object is
29988 frozen, then neither itself, nor any of its children, are
29989 implicitly updated by @code{-var-update} of
29990 a parent variable or by @code{-var-update *}. Only
29991 @code{-var-update} of the variable itself will update its value and
29992 values of its children. After a variable object is unfrozen, it is
29993 implicitly updated by all subsequent @code{-var-update} operations.
29994 Unfreezing a variable does not update it, only subsequent
29995 @code{-var-update} does.
29997 @subsubheading Example
30001 -var-set-frozen V 1
30006 @subheading The @code{-var-set-update-range} command
30007 @findex -var-set-update-range
30008 @anchor{-var-set-update-range}
30010 @subsubheading Synopsis
30013 -var-set-update-range @var{name} @var{from} @var{to}
30016 Set the range of children to be returned by future invocations of
30017 @code{-var-update}.
30019 @var{from} and @var{to} indicate the range of children to report. If
30020 @var{from} or @var{to} is less than zero, the range is reset and all
30021 children will be reported. Otherwise, children starting at @var{from}
30022 (zero-based) and up to and excluding @var{to} will be reported.
30024 @subsubheading Example
30028 -var-set-update-range V 1 2
30032 @subheading The @code{-var-set-visualizer} command
30033 @findex -var-set-visualizer
30034 @anchor{-var-set-visualizer}
30036 @subsubheading Synopsis
30039 -var-set-visualizer @var{name} @var{visualizer}
30042 Set a visualizer for the variable object @var{name}.
30044 @var{visualizer} is the visualizer to use. The special value
30045 @samp{None} means to disable any visualizer in use.
30047 If not @samp{None}, @var{visualizer} must be a Python expression.
30048 This expression must evaluate to a callable object which accepts a
30049 single argument. @value{GDBN} will call this object with the value of
30050 the varobj @var{name} as an argument (this is done so that the same
30051 Python pretty-printing code can be used for both the CLI and MI).
30052 When called, this object must return an object which conforms to the
30053 pretty-printing interface (@pxref{Pretty Printing API}).
30055 The pre-defined function @code{gdb.default_visualizer} may be used to
30056 select a visualizer by following the built-in process
30057 (@pxref{Selecting Pretty-Printers}). This is done automatically when
30058 a varobj is created, and so ordinarily is not needed.
30060 This feature is only available if Python support is enabled. The MI
30061 command @code{-list-features} (@pxref{GDB/MI Support Commands})
30062 can be used to check this.
30064 @subsubheading Example
30066 Resetting the visualizer:
30070 -var-set-visualizer V None
30074 Reselecting the default (type-based) visualizer:
30078 -var-set-visualizer V gdb.default_visualizer
30082 Suppose @code{SomeClass} is a visualizer class. A lambda expression
30083 can be used to instantiate this class for a varobj:
30087 -var-set-visualizer V "lambda val: SomeClass()"
30091 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
30092 @node GDB/MI Data Manipulation
30093 @section @sc{gdb/mi} Data Manipulation
30095 @cindex data manipulation, in @sc{gdb/mi}
30096 @cindex @sc{gdb/mi}, data manipulation
30097 This section describes the @sc{gdb/mi} commands that manipulate data:
30098 examine memory and registers, evaluate expressions, etc.
30100 For details about what an addressable memory unit is,
30101 @pxref{addressable memory unit}.
30103 @c REMOVED FROM THE INTERFACE.
30104 @c @subheading -data-assign
30105 @c Change the value of a program variable. Plenty of side effects.
30106 @c @subsubheading GDB Command
30108 @c @subsubheading Example
30111 @subheading The @code{-data-disassemble} Command
30112 @findex -data-disassemble
30114 @subsubheading Synopsis
30118 [ -s @var{start-addr} -e @var{end-addr} ]
30119 | [ -f @var{filename} -l @var{linenum} [ -n @var{lines} ] ]
30127 @item @var{start-addr}
30128 is the beginning address (or @code{$pc})
30129 @item @var{end-addr}
30131 @item @var{filename}
30132 is the name of the file to disassemble
30133 @item @var{linenum}
30134 is the line number to disassemble around
30136 is the number of disassembly lines to be produced. If it is -1,
30137 the whole function will be disassembled, in case no @var{end-addr} is
30138 specified. If @var{end-addr} is specified as a non-zero value, and
30139 @var{lines} is lower than the number of disassembly lines between
30140 @var{start-addr} and @var{end-addr}, only @var{lines} lines are
30141 displayed; if @var{lines} is higher than the number of lines between
30142 @var{start-addr} and @var{end-addr}, only the lines up to @var{end-addr}
30147 @item 0 disassembly only
30148 @item 1 mixed source and disassembly (deprecated)
30149 @item 2 disassembly with raw opcodes
30150 @item 3 mixed source and disassembly with raw opcodes (deprecated)
30151 @item 4 mixed source and disassembly
30152 @item 5 mixed source and disassembly with raw opcodes
30155 Modes 1 and 3 are deprecated. The output is ``source centric''
30156 which hasn't proved useful in practice.
30157 @xref{Machine Code}, for a discussion of the difference between
30158 @code{/m} and @code{/s} output of the @code{disassemble} command.
30161 @subsubheading Result
30163 The result of the @code{-data-disassemble} command will be a list named
30164 @samp{asm_insns}, the contents of this list depend on the @var{mode}
30165 used with the @code{-data-disassemble} command.
30167 For modes 0 and 2 the @samp{asm_insns} list contains tuples with the
30172 The address at which this instruction was disassembled.
30175 The name of the function this instruction is within.
30178 The decimal offset in bytes from the start of @samp{func-name}.
30181 The text disassembly for this @samp{address}.
30184 This field is only present for modes 2, 3 and 5. This contains the raw opcode
30185 bytes for the @samp{inst} field.
30189 For modes 1, 3, 4 and 5 the @samp{asm_insns} list contains tuples named
30190 @samp{src_and_asm_line}, each of which has the following fields:
30194 The line number within @samp{file}.
30197 The file name from the compilation unit. This might be an absolute
30198 file name or a relative file name depending on the compile command
30202 Absolute file name of @samp{file}. It is converted to a canonical form
30203 using the source file search path
30204 (@pxref{Source Path, ,Specifying Source Directories})
30205 and after resolving all the symbolic links.
30207 If the source file is not found this field will contain the path as
30208 present in the debug information.
30210 @item line_asm_insn
30211 This is a list of tuples containing the disassembly for @samp{line} in
30212 @samp{file}. The fields of each tuple are the same as for
30213 @code{-data-disassemble} in @var{mode} 0 and 2, so @samp{address},
30214 @samp{func-name}, @samp{offset}, @samp{inst}, and optionally
30219 Note that whatever included in the @samp{inst} field, is not
30220 manipulated directly by @sc{gdb/mi}, i.e., it is not possible to
30223 @subsubheading @value{GDBN} Command
30225 The corresponding @value{GDBN} command is @samp{disassemble}.
30227 @subsubheading Example
30229 Disassemble from the current value of @code{$pc} to @code{$pc + 20}:
30233 -data-disassemble -s $pc -e "$pc + 20" -- 0
30236 @{address="0x000107c0",func-name="main",offset="4",
30237 inst="mov 2, %o0"@},
30238 @{address="0x000107c4",func-name="main",offset="8",
30239 inst="sethi %hi(0x11800), %o2"@},
30240 @{address="0x000107c8",func-name="main",offset="12",
30241 inst="or %o2, 0x140, %o1\t! 0x11940 <_lib_version+8>"@},
30242 @{address="0x000107cc",func-name="main",offset="16",
30243 inst="sethi %hi(0x11800), %o2"@},
30244 @{address="0x000107d0",func-name="main",offset="20",
30245 inst="or %o2, 0x168, %o4\t! 0x11968 <_lib_version+48>"@}]
30249 Disassemble the whole @code{main} function. Line 32 is part of
30253 -data-disassemble -f basics.c -l 32 -- 0
30255 @{address="0x000107bc",func-name="main",offset="0",
30256 inst="save %sp, -112, %sp"@},
30257 @{address="0x000107c0",func-name="main",offset="4",
30258 inst="mov 2, %o0"@},
30259 @{address="0x000107c4",func-name="main",offset="8",
30260 inst="sethi %hi(0x11800), %o2"@},
30262 @{address="0x0001081c",func-name="main",offset="96",inst="ret "@},
30263 @{address="0x00010820",func-name="main",offset="100",inst="restore "@}]
30267 Disassemble 3 instructions from the start of @code{main}:
30271 -data-disassemble -f basics.c -l 32 -n 3 -- 0
30273 @{address="0x000107bc",func-name="main",offset="0",
30274 inst="save %sp, -112, %sp"@},
30275 @{address="0x000107c0",func-name="main",offset="4",
30276 inst="mov 2, %o0"@},
30277 @{address="0x000107c4",func-name="main",offset="8",
30278 inst="sethi %hi(0x11800), %o2"@}]
30282 Disassemble 3 instructions from the start of @code{main} in mixed mode:
30286 -data-disassemble -f basics.c -l 32 -n 3 -- 1
30288 src_and_asm_line=@{line="31",
30289 file="../../../src/gdb/testsuite/gdb.mi/basics.c",
30290 fullname="/absolute/path/to/src/gdb/testsuite/gdb.mi/basics.c",
30291 line_asm_insn=[@{address="0x000107bc",
30292 func-name="main",offset="0",inst="save %sp, -112, %sp"@}]@},
30293 src_and_asm_line=@{line="32",
30294 file="../../../src/gdb/testsuite/gdb.mi/basics.c",
30295 fullname="/absolute/path/to/src/gdb/testsuite/gdb.mi/basics.c",
30296 line_asm_insn=[@{address="0x000107c0",
30297 func-name="main",offset="4",inst="mov 2, %o0"@},
30298 @{address="0x000107c4",func-name="main",offset="8",
30299 inst="sethi %hi(0x11800), %o2"@}]@}]
30304 @subheading The @code{-data-evaluate-expression} Command
30305 @findex -data-evaluate-expression
30307 @subsubheading Synopsis
30310 -data-evaluate-expression @var{expr}
30313 Evaluate @var{expr} as an expression. The expression could contain an
30314 inferior function call. The function call will execute synchronously.
30315 If the expression contains spaces, it must be enclosed in double quotes.
30317 @subsubheading @value{GDBN} Command
30319 The corresponding @value{GDBN} commands are @samp{print}, @samp{output}, and
30320 @samp{call}. In @code{gdbtk} only, there's a corresponding
30321 @samp{gdb_eval} command.
30323 @subsubheading Example
30325 In the following example, the numbers that precede the commands are the
30326 @dfn{tokens} described in @ref{GDB/MI Command Syntax, ,@sc{gdb/mi}
30327 Command Syntax}. Notice how @sc{gdb/mi} returns the same tokens in its
30331 211-data-evaluate-expression A
30334 311-data-evaluate-expression &A
30335 311^done,value="0xefffeb7c"
30337 411-data-evaluate-expression A+3
30340 511-data-evaluate-expression "A + 3"
30346 @subheading The @code{-data-list-changed-registers} Command
30347 @findex -data-list-changed-registers
30349 @subsubheading Synopsis
30352 -data-list-changed-registers
30355 Display a list of the registers that have changed.
30357 @subsubheading @value{GDBN} Command
30359 @value{GDBN} doesn't have a direct analog for this command; @code{gdbtk}
30360 has the corresponding command @samp{gdb_changed_register_list}.
30362 @subsubheading Example
30364 On a PPC MBX board:
30372 *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",frame=@{
30373 func="main",args=[],file="try.c",fullname="/home/foo/bar/try.c",
30376 -data-list-changed-registers
30377 ^done,changed-registers=["0","1","2","4","5","6","7","8","9",
30378 "10","11","13","14","15","16","17","18","19","20","21","22","23",
30379 "24","25","26","27","28","30","31","64","65","66","67","69"]
30384 @subheading The @code{-data-list-register-names} Command
30385 @findex -data-list-register-names
30387 @subsubheading Synopsis
30390 -data-list-register-names [ ( @var{regno} )+ ]
30393 Show a list of register names for the current target. If no arguments
30394 are given, it shows a list of the names of all the registers. If
30395 integer numbers are given as arguments, it will print a list of the
30396 names of the registers corresponding to the arguments. To ensure
30397 consistency between a register name and its number, the output list may
30398 include empty register names.
30400 @subsubheading @value{GDBN} Command
30402 @value{GDBN} does not have a command which corresponds to
30403 @samp{-data-list-register-names}. In @code{gdbtk} there is a
30404 corresponding command @samp{gdb_regnames}.
30406 @subsubheading Example
30408 For the PPC MBX board:
30411 -data-list-register-names
30412 ^done,register-names=["r0","r1","r2","r3","r4","r5","r6","r7",
30413 "r8","r9","r10","r11","r12","r13","r14","r15","r16","r17","r18",
30414 "r19","r20","r21","r22","r23","r24","r25","r26","r27","r28","r29",
30415 "r30","r31","f0","f1","f2","f3","f4","f5","f6","f7","f8","f9",
30416 "f10","f11","f12","f13","f14","f15","f16","f17","f18","f19","f20",
30417 "f21","f22","f23","f24","f25","f26","f27","f28","f29","f30","f31",
30418 "", "pc","ps","cr","lr","ctr","xer"]
30420 -data-list-register-names 1 2 3
30421 ^done,register-names=["r1","r2","r3"]
30425 @subheading The @code{-data-list-register-values} Command
30426 @findex -data-list-register-values
30428 @subsubheading Synopsis
30431 -data-list-register-values
30432 [ @code{--skip-unavailable} ] @var{fmt} [ ( @var{regno} )*]
30435 Display the registers' contents. The format according to which the
30436 registers' contents are to be returned is given by @var{fmt}, followed
30437 by an optional list of numbers specifying the registers to display. A
30438 missing list of numbers indicates that the contents of all the
30439 registers must be returned. The @code{--skip-unavailable} option
30440 indicates that only the available registers are to be returned.
30442 Allowed formats for @var{fmt} are:
30459 @subsubheading @value{GDBN} Command
30461 The corresponding @value{GDBN} commands are @samp{info reg}, @samp{info
30462 all-reg}, and (in @code{gdbtk}) @samp{gdb_fetch_registers}.
30464 @subsubheading Example
30466 For a PPC MBX board (note: line breaks are for readability only, they
30467 don't appear in the actual output):
30471 -data-list-register-values r 64 65
30472 ^done,register-values=[@{number="64",value="0xfe00a300"@},
30473 @{number="65",value="0x00029002"@}]
30475 -data-list-register-values x
30476 ^done,register-values=[@{number="0",value="0xfe0043c8"@},
30477 @{number="1",value="0x3fff88"@},@{number="2",value="0xfffffffe"@},
30478 @{number="3",value="0x0"@},@{number="4",value="0xa"@},
30479 @{number="5",value="0x3fff68"@},@{number="6",value="0x3fff58"@},
30480 @{number="7",value="0xfe011e98"@},@{number="8",value="0x2"@},
30481 @{number="9",value="0xfa202820"@},@{number="10",value="0xfa202808"@},
30482 @{number="11",value="0x1"@},@{number="12",value="0x0"@},
30483 @{number="13",value="0x4544"@},@{number="14",value="0xffdfffff"@},
30484 @{number="15",value="0xffffffff"@},@{number="16",value="0xfffffeff"@},
30485 @{number="17",value="0xefffffed"@},@{number="18",value="0xfffffffe"@},
30486 @{number="19",value="0xffffffff"@},@{number="20",value="0xffffffff"@},
30487 @{number="21",value="0xffffffff"@},@{number="22",value="0xfffffff7"@},
30488 @{number="23",value="0xffffffff"@},@{number="24",value="0xffffffff"@},
30489 @{number="25",value="0xffffffff"@},@{number="26",value="0xfffffffb"@},
30490 @{number="27",value="0xffffffff"@},@{number="28",value="0xf7bfffff"@},
30491 @{number="29",value="0x0"@},@{number="30",value="0xfe010000"@},
30492 @{number="31",value="0x0"@},@{number="32",value="0x0"@},
30493 @{number="33",value="0x0"@},@{number="34",value="0x0"@},
30494 @{number="35",value="0x0"@},@{number="36",value="0x0"@},
30495 @{number="37",value="0x0"@},@{number="38",value="0x0"@},
30496 @{number="39",value="0x0"@},@{number="40",value="0x0"@},
30497 @{number="41",value="0x0"@},@{number="42",value="0x0"@},
30498 @{number="43",value="0x0"@},@{number="44",value="0x0"@},
30499 @{number="45",value="0x0"@},@{number="46",value="0x0"@},
30500 @{number="47",value="0x0"@},@{number="48",value="0x0"@},
30501 @{number="49",value="0x0"@},@{number="50",value="0x0"@},
30502 @{number="51",value="0x0"@},@{number="52",value="0x0"@},
30503 @{number="53",value="0x0"@},@{number="54",value="0x0"@},
30504 @{number="55",value="0x0"@},@{number="56",value="0x0"@},
30505 @{number="57",value="0x0"@},@{number="58",value="0x0"@},
30506 @{number="59",value="0x0"@},@{number="60",value="0x0"@},
30507 @{number="61",value="0x0"@},@{number="62",value="0x0"@},
30508 @{number="63",value="0x0"@},@{number="64",value="0xfe00a300"@},
30509 @{number="65",value="0x29002"@},@{number="66",value="0x202f04b5"@},
30510 @{number="67",value="0xfe0043b0"@},@{number="68",value="0xfe00b3e4"@},
30511 @{number="69",value="0x20002b03"@}]
30516 @subheading The @code{-data-read-memory} Command
30517 @findex -data-read-memory
30519 This command is deprecated, use @code{-data-read-memory-bytes} instead.
30521 @subsubheading Synopsis
30524 -data-read-memory [ -o @var{byte-offset} ]
30525 @var{address} @var{word-format} @var{word-size}
30526 @var{nr-rows} @var{nr-cols} [ @var{aschar} ]
30533 @item @var{address}
30534 An expression specifying the address of the first memory word to be
30535 read. Complex expressions containing embedded white space should be
30536 quoted using the C convention.
30538 @item @var{word-format}
30539 The format to be used to print the memory words. The notation is the
30540 same as for @value{GDBN}'s @code{print} command (@pxref{Output Formats,
30543 @item @var{word-size}
30544 The size of each memory word in bytes.
30546 @item @var{nr-rows}
30547 The number of rows in the output table.
30549 @item @var{nr-cols}
30550 The number of columns in the output table.
30553 If present, indicates that each row should include an @sc{ascii} dump. The
30554 value of @var{aschar} is used as a padding character when a byte is not a
30555 member of the printable @sc{ascii} character set (printable @sc{ascii}
30556 characters are those whose code is between 32 and 126, inclusively).
30558 @item @var{byte-offset}
30559 An offset to add to the @var{address} before fetching memory.
30562 This command displays memory contents as a table of @var{nr-rows} by
30563 @var{nr-cols} words, each word being @var{word-size} bytes. In total,
30564 @code{@var{nr-rows} * @var{nr-cols} * @var{word-size}} bytes are read
30565 (returned as @samp{total-bytes}). Should less than the requested number
30566 of bytes be returned by the target, the missing words are identified
30567 using @samp{N/A}. The number of bytes read from the target is returned
30568 in @samp{nr-bytes} and the starting address used to read memory in
30571 The address of the next/previous row or page is available in
30572 @samp{next-row} and @samp{prev-row}, @samp{next-page} and
30575 @subsubheading @value{GDBN} Command
30577 The corresponding @value{GDBN} command is @samp{x}. @code{gdbtk} has
30578 @samp{gdb_get_mem} memory read command.
30580 @subsubheading Example
30582 Read six bytes of memory starting at @code{bytes+6} but then offset by
30583 @code{-6} bytes. Format as three rows of two columns. One byte per
30584 word. Display each word in hex.
30588 9-data-read-memory -o -6 -- bytes+6 x 1 3 2
30589 9^done,addr="0x00001390",nr-bytes="6",total-bytes="6",
30590 next-row="0x00001396",prev-row="0x0000138e",next-page="0x00001396",
30591 prev-page="0x0000138a",memory=[
30592 @{addr="0x00001390",data=["0x00","0x01"]@},
30593 @{addr="0x00001392",data=["0x02","0x03"]@},
30594 @{addr="0x00001394",data=["0x04","0x05"]@}]
30598 Read two bytes of memory starting at address @code{shorts + 64} and
30599 display as a single word formatted in decimal.
30603 5-data-read-memory shorts+64 d 2 1 1
30604 5^done,addr="0x00001510",nr-bytes="2",total-bytes="2",
30605 next-row="0x00001512",prev-row="0x0000150e",
30606 next-page="0x00001512",prev-page="0x0000150e",memory=[
30607 @{addr="0x00001510",data=["128"]@}]
30611 Read thirty two bytes of memory starting at @code{bytes+16} and format
30612 as eight rows of four columns. Include a string encoding with @samp{x}
30613 used as the non-printable character.
30617 4-data-read-memory bytes+16 x 1 8 4 x
30618 4^done,addr="0x000013a0",nr-bytes="32",total-bytes="32",
30619 next-row="0x000013c0",prev-row="0x0000139c",
30620 next-page="0x000013c0",prev-page="0x00001380",memory=[
30621 @{addr="0x000013a0",data=["0x10","0x11","0x12","0x13"],ascii="xxxx"@},
30622 @{addr="0x000013a4",data=["0x14","0x15","0x16","0x17"],ascii="xxxx"@},
30623 @{addr="0x000013a8",data=["0x18","0x19","0x1a","0x1b"],ascii="xxxx"@},
30624 @{addr="0x000013ac",data=["0x1c","0x1d","0x1e","0x1f"],ascii="xxxx"@},
30625 @{addr="0x000013b0",data=["0x20","0x21","0x22","0x23"],ascii=" !\"#"@},
30626 @{addr="0x000013b4",data=["0x24","0x25","0x26","0x27"],ascii="$%&'"@},
30627 @{addr="0x000013b8",data=["0x28","0x29","0x2a","0x2b"],ascii="()*+"@},
30628 @{addr="0x000013bc",data=["0x2c","0x2d","0x2e","0x2f"],ascii=",-./"@}]
30632 @subheading The @code{-data-read-memory-bytes} Command
30633 @findex -data-read-memory-bytes
30635 @subsubheading Synopsis
30638 -data-read-memory-bytes [ -o @var{offset} ]
30639 @var{address} @var{count}
30646 @item @var{address}
30647 An expression specifying the address of the first addressable memory unit
30648 to be read. Complex expressions containing embedded white space should be
30649 quoted using the C convention.
30652 The number of addressable memory units to read. This should be an integer
30656 The offset relative to @var{address} at which to start reading. This
30657 should be an integer literal. This option is provided so that a frontend
30658 is not required to first evaluate address and then perform address
30659 arithmetics itself.
30663 This command attempts to read all accessible memory regions in the
30664 specified range. First, all regions marked as unreadable in the memory
30665 map (if one is defined) will be skipped. @xref{Memory Region
30666 Attributes}. Second, @value{GDBN} will attempt to read the remaining
30667 regions. For each one, if reading full region results in an errors,
30668 @value{GDBN} will try to read a subset of the region.
30670 In general, every single memory unit in the region may be readable or not,
30671 and the only way to read every readable unit is to try a read at
30672 every address, which is not practical. Therefore, @value{GDBN} will
30673 attempt to read all accessible memory units at either beginning or the end
30674 of the region, using a binary division scheme. This heuristic works
30675 well for reading accross a memory map boundary. Note that if a region
30676 has a readable range that is neither at the beginning or the end,
30677 @value{GDBN} will not read it.
30679 The result record (@pxref{GDB/MI Result Records}) that is output of
30680 the command includes a field named @samp{memory} whose content is a
30681 list of tuples. Each tuple represent a successfully read memory block
30682 and has the following fields:
30686 The start address of the memory block, as hexadecimal literal.
30689 The end address of the memory block, as hexadecimal literal.
30692 The offset of the memory block, as hexadecimal literal, relative to
30693 the start address passed to @code{-data-read-memory-bytes}.
30696 The contents of the memory block, in hex.
30702 @subsubheading @value{GDBN} Command
30704 The corresponding @value{GDBN} command is @samp{x}.
30706 @subsubheading Example
30710 -data-read-memory-bytes &a 10
30711 ^done,memory=[@{begin="0xbffff154",offset="0x00000000",
30713 contents="01000000020000000300"@}]
30718 @subheading The @code{-data-write-memory-bytes} Command
30719 @findex -data-write-memory-bytes
30721 @subsubheading Synopsis
30724 -data-write-memory-bytes @var{address} @var{contents}
30725 -data-write-memory-bytes @var{address} @var{contents} @r{[}@var{count}@r{]}
30732 @item @var{address}
30733 An expression specifying the address of the first addressable memory unit
30734 to be written. Complex expressions containing embedded white space should
30735 be quoted using the C convention.
30737 @item @var{contents}
30738 The hex-encoded data to write. It is an error if @var{contents} does
30739 not represent an integral number of addressable memory units.
30742 Optional argument indicating the number of addressable memory units to be
30743 written. If @var{count} is greater than @var{contents}' length,
30744 @value{GDBN} will repeatedly write @var{contents} until it fills
30745 @var{count} memory units.
30749 @subsubheading @value{GDBN} Command
30751 There's no corresponding @value{GDBN} command.
30753 @subsubheading Example
30757 -data-write-memory-bytes &a "aabbccdd"
30764 -data-write-memory-bytes &a "aabbccdd" 16e
30769 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
30770 @node GDB/MI Tracepoint Commands
30771 @section @sc{gdb/mi} Tracepoint Commands
30773 The commands defined in this section implement MI support for
30774 tracepoints. For detailed introduction, see @ref{Tracepoints}.
30776 @subheading The @code{-trace-find} Command
30777 @findex -trace-find
30779 @subsubheading Synopsis
30782 -trace-find @var{mode} [@var{parameters}@dots{}]
30785 Find a trace frame using criteria defined by @var{mode} and
30786 @var{parameters}. The following table lists permissible
30787 modes and their parameters. For details of operation, see @ref{tfind}.
30792 No parameters are required. Stops examining trace frames.
30795 An integer is required as parameter. Selects tracepoint frame with
30798 @item tracepoint-number
30799 An integer is required as parameter. Finds next
30800 trace frame that corresponds to tracepoint with the specified number.
30803 An address is required as parameter. Finds
30804 next trace frame that corresponds to any tracepoint at the specified
30807 @item pc-inside-range
30808 Two addresses are required as parameters. Finds next trace
30809 frame that corresponds to a tracepoint at an address inside the
30810 specified range. Both bounds are considered to be inside the range.
30812 @item pc-outside-range
30813 Two addresses are required as parameters. Finds
30814 next trace frame that corresponds to a tracepoint at an address outside
30815 the specified range. Both bounds are considered to be inside the range.
30818 Line specification is required as parameter. @xref{Specify Location}.
30819 Finds next trace frame that corresponds to a tracepoint at
30820 the specified location.
30824 If @samp{none} was passed as @var{mode}, the response does not
30825 have fields. Otherwise, the response may have the following fields:
30829 This field has either @samp{0} or @samp{1} as the value, depending
30830 on whether a matching tracepoint was found.
30833 The index of the found traceframe. This field is present iff
30834 the @samp{found} field has value of @samp{1}.
30837 The index of the found tracepoint. This field is present iff
30838 the @samp{found} field has value of @samp{1}.
30841 The information about the frame corresponding to the found trace
30842 frame. This field is present only if a trace frame was found.
30843 @xref{GDB/MI Frame Information}, for description of this field.
30847 @subsubheading @value{GDBN} Command
30849 The corresponding @value{GDBN} command is @samp{tfind}.
30851 @subheading -trace-define-variable
30852 @findex -trace-define-variable
30854 @subsubheading Synopsis
30857 -trace-define-variable @var{name} [ @var{value} ]
30860 Create trace variable @var{name} if it does not exist. If
30861 @var{value} is specified, sets the initial value of the specified
30862 trace variable to that value. Note that the @var{name} should start
30863 with the @samp{$} character.
30865 @subsubheading @value{GDBN} Command
30867 The corresponding @value{GDBN} command is @samp{tvariable}.
30869 @subheading The @code{-trace-frame-collected} Command
30870 @findex -trace-frame-collected
30872 @subsubheading Synopsis
30875 -trace-frame-collected
30876 [--var-print-values @var{var_pval}]
30877 [--comp-print-values @var{comp_pval}]
30878 [--registers-format @var{regformat}]
30879 [--memory-contents]
30882 This command returns the set of collected objects, register names,
30883 trace state variable names, memory ranges and computed expressions
30884 that have been collected at a particular trace frame. The optional
30885 parameters to the command affect the output format in different ways.
30886 See the output description table below for more details.
30888 The reported names can be used in the normal manner to create
30889 varobjs and inspect the objects themselves. The items returned by
30890 this command are categorized so that it is clear which is a variable,
30891 which is a register, which is a trace state variable, which is a
30892 memory range and which is a computed expression.
30894 For instance, if the actions were
30896 collect myVar, myArray[myIndex], myObj.field, myPtr->field, myCount + 2
30897 collect *(int*)0xaf02bef0@@40
30901 the object collected in its entirety would be @code{myVar}. The
30902 object @code{myArray} would be partially collected, because only the
30903 element at index @code{myIndex} would be collected. The remaining
30904 objects would be computed expressions.
30906 An example output would be:
30910 -trace-frame-collected
30912 explicit-variables=[@{name="myVar",value="1"@}],
30913 computed-expressions=[@{name="myArray[myIndex]",value="0"@},
30914 @{name="myObj.field",value="0"@},
30915 @{name="myPtr->field",value="1"@},
30916 @{name="myCount + 2",value="3"@},
30917 @{name="$tvar1 + 1",value="43970027"@}],
30918 registers=[@{number="0",value="0x7fe2c6e79ec8"@},
30919 @{number="1",value="0x0"@},
30920 @{number="2",value="0x4"@},
30922 @{number="125",value="0x0"@}],
30923 tvars=[@{name="$tvar1",current="43970026"@}],
30924 memory=[@{address="0x0000000000602264",length="4"@},
30925 @{address="0x0000000000615bc0",length="4"@}]
30932 @item explicit-variables
30933 The set of objects that have been collected in their entirety (as
30934 opposed to collecting just a few elements of an array or a few struct
30935 members). For each object, its name and value are printed.
30936 The @code{--var-print-values} option affects how or whether the value
30937 field is output. If @var{var_pval} is 0, then print only the names;
30938 if it is 1, print also their values; and if it is 2, print the name,
30939 type and value for simple data types, and the name and type for
30940 arrays, structures and unions.
30942 @item computed-expressions
30943 The set of computed expressions that have been collected at the
30944 current trace frame. The @code{--comp-print-values} option affects
30945 this set like the @code{--var-print-values} option affects the
30946 @code{explicit-variables} set. See above.
30949 The registers that have been collected at the current trace frame.
30950 For each register collected, the name and current value are returned.
30951 The value is formatted according to the @code{--registers-format}
30952 option. See the @command{-data-list-register-values} command for a
30953 list of the allowed formats. The default is @samp{x}.
30956 The trace state variables that have been collected at the current
30957 trace frame. For each trace state variable collected, the name and
30958 current value are returned.
30961 The set of memory ranges that have been collected at the current trace
30962 frame. Its content is a list of tuples. Each tuple represents a
30963 collected memory range and has the following fields:
30967 The start address of the memory range, as hexadecimal literal.
30970 The length of the memory range, as decimal literal.
30973 The contents of the memory block, in hex. This field is only present
30974 if the @code{--memory-contents} option is specified.
30980 @subsubheading @value{GDBN} Command
30982 There is no corresponding @value{GDBN} command.
30984 @subsubheading Example
30986 @subheading -trace-list-variables
30987 @findex -trace-list-variables
30989 @subsubheading Synopsis
30992 -trace-list-variables
30995 Return a table of all defined trace variables. Each element of the
30996 table has the following fields:
31000 The name of the trace variable. This field is always present.
31003 The initial value. This is a 64-bit signed integer. This
31004 field is always present.
31007 The value the trace variable has at the moment. This is a 64-bit
31008 signed integer. This field is absent iff current value is
31009 not defined, for example if the trace was never run, or is
31014 @subsubheading @value{GDBN} Command
31016 The corresponding @value{GDBN} command is @samp{tvariables}.
31018 @subsubheading Example
31022 -trace-list-variables
31023 ^done,trace-variables=@{nr_rows="1",nr_cols="3",
31024 hdr=[@{width="15",alignment="-1",col_name="name",colhdr="Name"@},
31025 @{width="11",alignment="-1",col_name="initial",colhdr="Initial"@},
31026 @{width="11",alignment="-1",col_name="current",colhdr="Current"@}],
31027 body=[variable=@{name="$trace_timestamp",initial="0"@}
31028 variable=@{name="$foo",initial="10",current="15"@}]@}
31032 @subheading -trace-save
31033 @findex -trace-save
31035 @subsubheading Synopsis
31038 -trace-save [ -r ] [ -ctf ] @var{filename}
31041 Saves the collected trace data to @var{filename}. Without the
31042 @samp{-r} option, the data is downloaded from the target and saved
31043 in a local file. With the @samp{-r} option the target is asked
31044 to perform the save.
31046 By default, this command will save the trace in the tfile format. You can
31047 supply the optional @samp{-ctf} argument to save it the CTF format. See
31048 @ref{Trace Files} for more information about CTF.
31050 @subsubheading @value{GDBN} Command
31052 The corresponding @value{GDBN} command is @samp{tsave}.
31055 @subheading -trace-start
31056 @findex -trace-start
31058 @subsubheading Synopsis
31064 Starts a tracing experiment. The result of this command does not
31067 @subsubheading @value{GDBN} Command
31069 The corresponding @value{GDBN} command is @samp{tstart}.
31071 @subheading -trace-status
31072 @findex -trace-status
31074 @subsubheading Synopsis
31080 Obtains the status of a tracing experiment. The result may include
31081 the following fields:
31086 May have a value of either @samp{0}, when no tracing operations are
31087 supported, @samp{1}, when all tracing operations are supported, or
31088 @samp{file} when examining trace file. In the latter case, examining
31089 of trace frame is possible but new tracing experiement cannot be
31090 started. This field is always present.
31093 May have a value of either @samp{0} or @samp{1} depending on whether
31094 tracing experiement is in progress on target. This field is present
31095 if @samp{supported} field is not @samp{0}.
31098 Report the reason why the tracing was stopped last time. This field
31099 may be absent iff tracing was never stopped on target yet. The
31100 value of @samp{request} means the tracing was stopped as result of
31101 the @code{-trace-stop} command. The value of @samp{overflow} means
31102 the tracing buffer is full. The value of @samp{disconnection} means
31103 tracing was automatically stopped when @value{GDBN} has disconnected.
31104 The value of @samp{passcount} means tracing was stopped when a
31105 tracepoint was passed a maximal number of times for that tracepoint.
31106 This field is present if @samp{supported} field is not @samp{0}.
31108 @item stopping-tracepoint
31109 The number of tracepoint whose passcount as exceeded. This field is
31110 present iff the @samp{stop-reason} field has the value of
31114 @itemx frames-created
31115 The @samp{frames} field is a count of the total number of trace frames
31116 in the trace buffer, while @samp{frames-created} is the total created
31117 during the run, including ones that were discarded, such as when a
31118 circular trace buffer filled up. Both fields are optional.
31122 These fields tell the current size of the tracing buffer and the
31123 remaining space. These fields are optional.
31126 The value of the circular trace buffer flag. @code{1} means that the
31127 trace buffer is circular and old trace frames will be discarded if
31128 necessary to make room, @code{0} means that the trace buffer is linear
31132 The value of the disconnected tracing flag. @code{1} means that
31133 tracing will continue after @value{GDBN} disconnects, @code{0} means
31134 that the trace run will stop.
31137 The filename of the trace file being examined. This field is
31138 optional, and only present when examining a trace file.
31142 @subsubheading @value{GDBN} Command
31144 The corresponding @value{GDBN} command is @samp{tstatus}.
31146 @subheading -trace-stop
31147 @findex -trace-stop
31149 @subsubheading Synopsis
31155 Stops a tracing experiment. The result of this command has the same
31156 fields as @code{-trace-status}, except that the @samp{supported} and
31157 @samp{running} fields are not output.
31159 @subsubheading @value{GDBN} Command
31161 The corresponding @value{GDBN} command is @samp{tstop}.
31164 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
31165 @node GDB/MI Symbol Query
31166 @section @sc{gdb/mi} Symbol Query Commands
31170 @subheading The @code{-symbol-info-address} Command
31171 @findex -symbol-info-address
31173 @subsubheading Synopsis
31176 -symbol-info-address @var{symbol}
31179 Describe where @var{symbol} is stored.
31181 @subsubheading @value{GDBN} Command
31183 The corresponding @value{GDBN} command is @samp{info address}.
31185 @subsubheading Example
31189 @subheading The @code{-symbol-info-file} Command
31190 @findex -symbol-info-file
31192 @subsubheading Synopsis
31198 Show the file for the symbol.
31200 @subsubheading @value{GDBN} Command
31202 There's no equivalent @value{GDBN} command. @code{gdbtk} has
31203 @samp{gdb_find_file}.
31205 @subsubheading Example
31209 @subheading The @code{-symbol-info-function} Command
31210 @findex -symbol-info-function
31212 @subsubheading Synopsis
31215 -symbol-info-function
31218 Show which function the symbol lives in.
31220 @subsubheading @value{GDBN} Command
31222 @samp{gdb_get_function} in @code{gdbtk}.
31224 @subsubheading Example
31228 @subheading The @code{-symbol-info-line} Command
31229 @findex -symbol-info-line
31231 @subsubheading Synopsis
31237 Show the core addresses of the code for a source line.
31239 @subsubheading @value{GDBN} Command
31241 The corresponding @value{GDBN} command is @samp{info line}.
31242 @code{gdbtk} has the @samp{gdb_get_line} and @samp{gdb_get_file} commands.
31244 @subsubheading Example
31248 @subheading The @code{-symbol-info-symbol} Command
31249 @findex -symbol-info-symbol
31251 @subsubheading Synopsis
31254 -symbol-info-symbol @var{addr}
31257 Describe what symbol is at location @var{addr}.
31259 @subsubheading @value{GDBN} Command
31261 The corresponding @value{GDBN} command is @samp{info symbol}.
31263 @subsubheading Example
31267 @subheading The @code{-symbol-list-functions} Command
31268 @findex -symbol-list-functions
31270 @subsubheading Synopsis
31273 -symbol-list-functions
31276 List the functions in the executable.
31278 @subsubheading @value{GDBN} Command
31280 @samp{info functions} in @value{GDBN}, @samp{gdb_listfunc} and
31281 @samp{gdb_search} in @code{gdbtk}.
31283 @subsubheading Example
31288 @subheading The @code{-symbol-list-lines} Command
31289 @findex -symbol-list-lines
31291 @subsubheading Synopsis
31294 -symbol-list-lines @var{filename}
31297 Print the list of lines that contain code and their associated program
31298 addresses for the given source filename. The entries are sorted in
31299 ascending PC order.
31301 @subsubheading @value{GDBN} Command
31303 There is no corresponding @value{GDBN} command.
31305 @subsubheading Example
31308 -symbol-list-lines basics.c
31309 ^done,lines=[@{pc="0x08048554",line="7"@},@{pc="0x0804855a",line="8"@}]
31315 @subheading The @code{-symbol-list-types} Command
31316 @findex -symbol-list-types
31318 @subsubheading Synopsis
31324 List all the type names.
31326 @subsubheading @value{GDBN} Command
31328 The corresponding commands are @samp{info types} in @value{GDBN},
31329 @samp{gdb_search} in @code{gdbtk}.
31331 @subsubheading Example
31335 @subheading The @code{-symbol-list-variables} Command
31336 @findex -symbol-list-variables
31338 @subsubheading Synopsis
31341 -symbol-list-variables
31344 List all the global and static variable names.
31346 @subsubheading @value{GDBN} Command
31348 @samp{info variables} in @value{GDBN}, @samp{gdb_search} in @code{gdbtk}.
31350 @subsubheading Example
31354 @subheading The @code{-symbol-locate} Command
31355 @findex -symbol-locate
31357 @subsubheading Synopsis
31363 @subsubheading @value{GDBN} Command
31365 @samp{gdb_loc} in @code{gdbtk}.
31367 @subsubheading Example
31371 @subheading The @code{-symbol-type} Command
31372 @findex -symbol-type
31374 @subsubheading Synopsis
31377 -symbol-type @var{variable}
31380 Show type of @var{variable}.
31382 @subsubheading @value{GDBN} Command
31384 The corresponding @value{GDBN} command is @samp{ptype}, @code{gdbtk} has
31385 @samp{gdb_obj_variable}.
31387 @subsubheading Example
31392 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
31393 @node GDB/MI File Commands
31394 @section @sc{gdb/mi} File Commands
31396 This section describes the GDB/MI commands to specify executable file names
31397 and to read in and obtain symbol table information.
31399 @subheading The @code{-file-exec-and-symbols} Command
31400 @findex -file-exec-and-symbols
31402 @subsubheading Synopsis
31405 -file-exec-and-symbols @var{file}
31408 Specify the executable file to be debugged. This file is the one from
31409 which the symbol table is also read. If no file is specified, the
31410 command clears the executable and symbol information. If breakpoints
31411 are set when using this command with no arguments, @value{GDBN} will produce
31412 error messages. Otherwise, no output is produced, except a completion
31415 @subsubheading @value{GDBN} Command
31417 The corresponding @value{GDBN} command is @samp{file}.
31419 @subsubheading Example
31423 -file-exec-and-symbols /kwikemart/marge/ezannoni/TRUNK/mbx/hello.mbx
31429 @subheading The @code{-file-exec-file} Command
31430 @findex -file-exec-file
31432 @subsubheading Synopsis
31435 -file-exec-file @var{file}
31438 Specify the executable file to be debugged. Unlike
31439 @samp{-file-exec-and-symbols}, the symbol table is @emph{not} read
31440 from this file. If used without argument, @value{GDBN} clears the information
31441 about the executable file. No output is produced, except a completion
31444 @subsubheading @value{GDBN} Command
31446 The corresponding @value{GDBN} command is @samp{exec-file}.
31448 @subsubheading Example
31452 -file-exec-file /kwikemart/marge/ezannoni/TRUNK/mbx/hello.mbx
31459 @subheading The @code{-file-list-exec-sections} Command
31460 @findex -file-list-exec-sections
31462 @subsubheading Synopsis
31465 -file-list-exec-sections
31468 List the sections of the current executable file.
31470 @subsubheading @value{GDBN} Command
31472 The @value{GDBN} command @samp{info file} shows, among the rest, the same
31473 information as this command. @code{gdbtk} has a corresponding command
31474 @samp{gdb_load_info}.
31476 @subsubheading Example
31481 @subheading The @code{-file-list-exec-source-file} Command
31482 @findex -file-list-exec-source-file
31484 @subsubheading Synopsis
31487 -file-list-exec-source-file
31490 List the line number, the current source file, and the absolute path
31491 to the current source file for the current executable. The macro
31492 information field has a value of @samp{1} or @samp{0} depending on
31493 whether or not the file includes preprocessor macro information.
31495 @subsubheading @value{GDBN} Command
31497 The @value{GDBN} equivalent is @samp{info source}
31499 @subsubheading Example
31503 123-file-list-exec-source-file
31504 123^done,line="1",file="foo.c",fullname="/home/bar/foo.c,macro-info="1"
31509 @subheading The @code{-file-list-exec-source-files} Command
31510 @findex -file-list-exec-source-files
31512 @subsubheading Synopsis
31515 -file-list-exec-source-files
31518 List the source files for the current executable.
31520 It will always output both the filename and fullname (absolute file
31521 name) of a source file.
31523 @subsubheading @value{GDBN} Command
31525 The @value{GDBN} equivalent is @samp{info sources}.
31526 @code{gdbtk} has an analogous command @samp{gdb_listfiles}.
31528 @subsubheading Example
31531 -file-list-exec-source-files
31533 @{file=foo.c,fullname=/home/foo.c@},
31534 @{file=/home/bar.c,fullname=/home/bar.c@},
31535 @{file=gdb_could_not_find_fullpath.c@}]
31539 @subheading The @code{-file-list-shared-libraries} Command
31540 @findex -file-list-shared-libraries
31542 @subsubheading Synopsis
31545 -file-list-shared-libraries [ @var{regexp} ]
31548 List the shared libraries in the program.
31549 With a regular expression @var{regexp}, only those libraries whose
31550 names match @var{regexp} are listed.
31552 @subsubheading @value{GDBN} Command
31554 The corresponding @value{GDBN} command is @samp{info shared}. The fields
31555 have a similar meaning to the @code{=library-loaded} notification.
31556 The @code{ranges} field specifies the multiple segments belonging to this
31557 library. Each range has the following fields:
31561 The address defining the inclusive lower bound of the segment.
31563 The address defining the exclusive upper bound of the segment.
31566 @subsubheading Example
31569 -file-list-exec-source-files
31570 ^done,shared-libraries=[
31571 @{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"@}]@},
31572 @{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"@}]@}]
31578 @subheading The @code{-file-list-symbol-files} Command
31579 @findex -file-list-symbol-files
31581 @subsubheading Synopsis
31584 -file-list-symbol-files
31589 @subsubheading @value{GDBN} Command
31591 The corresponding @value{GDBN} command is @samp{info file} (part of it).
31593 @subsubheading Example
31598 @subheading The @code{-file-symbol-file} Command
31599 @findex -file-symbol-file
31601 @subsubheading Synopsis
31604 -file-symbol-file @var{file}
31607 Read symbol table info from the specified @var{file} argument. When
31608 used without arguments, clears @value{GDBN}'s symbol table info. No output is
31609 produced, except for a completion notification.
31611 @subsubheading @value{GDBN} Command
31613 The corresponding @value{GDBN} command is @samp{symbol-file}.
31615 @subsubheading Example
31619 -file-symbol-file /kwikemart/marge/ezannoni/TRUNK/mbx/hello.mbx
31625 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
31626 @node GDB/MI Memory Overlay Commands
31627 @section @sc{gdb/mi} Memory Overlay Commands
31629 The memory overlay commands are not implemented.
31631 @c @subheading -overlay-auto
31633 @c @subheading -overlay-list-mapping-state
31635 @c @subheading -overlay-list-overlays
31637 @c @subheading -overlay-map
31639 @c @subheading -overlay-off
31641 @c @subheading -overlay-on
31643 @c @subheading -overlay-unmap
31645 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
31646 @node GDB/MI Signal Handling Commands
31647 @section @sc{gdb/mi} Signal Handling Commands
31649 Signal handling commands are not implemented.
31651 @c @subheading -signal-handle
31653 @c @subheading -signal-list-handle-actions
31655 @c @subheading -signal-list-signal-types
31659 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
31660 @node GDB/MI Target Manipulation
31661 @section @sc{gdb/mi} Target Manipulation Commands
31664 @subheading The @code{-target-attach} Command
31665 @findex -target-attach
31667 @subsubheading Synopsis
31670 -target-attach @var{pid} | @var{gid} | @var{file}
31673 Attach to a process @var{pid} or a file @var{file} outside of
31674 @value{GDBN}, or a thread group @var{gid}. If attaching to a thread
31675 group, the id previously returned by
31676 @samp{-list-thread-groups --available} must be used.
31678 @subsubheading @value{GDBN} Command
31680 The corresponding @value{GDBN} command is @samp{attach}.
31682 @subsubheading Example
31686 =thread-created,id="1"
31687 *stopped,thread-id="1",frame=@{addr="0xb7f7e410",func="bar",args=[]@}
31693 @subheading The @code{-target-compare-sections} Command
31694 @findex -target-compare-sections
31696 @subsubheading Synopsis
31699 -target-compare-sections [ @var{section} ]
31702 Compare data of section @var{section} on target to the exec file.
31703 Without the argument, all sections are compared.
31705 @subsubheading @value{GDBN} Command
31707 The @value{GDBN} equivalent is @samp{compare-sections}.
31709 @subsubheading Example
31714 @subheading The @code{-target-detach} Command
31715 @findex -target-detach
31717 @subsubheading Synopsis
31720 -target-detach [ @var{pid} | @var{gid} ]
31723 Detach from the remote target which normally resumes its execution.
31724 If either @var{pid} or @var{gid} is specified, detaches from either
31725 the specified process, or specified thread group. There's no output.
31727 @subsubheading @value{GDBN} Command
31729 The corresponding @value{GDBN} command is @samp{detach}.
31731 @subsubheading Example
31741 @subheading The @code{-target-disconnect} Command
31742 @findex -target-disconnect
31744 @subsubheading Synopsis
31750 Disconnect from the remote target. There's no output and the target is
31751 generally not resumed.
31753 @subsubheading @value{GDBN} Command
31755 The corresponding @value{GDBN} command is @samp{disconnect}.
31757 @subsubheading Example
31767 @subheading The @code{-target-download} Command
31768 @findex -target-download
31770 @subsubheading Synopsis
31776 Loads the executable onto the remote target.
31777 It prints out an update message every half second, which includes the fields:
31781 The name of the section.
31783 The size of what has been sent so far for that section.
31785 The size of the section.
31787 The total size of what was sent so far (the current and the previous sections).
31789 The size of the overall executable to download.
31793 Each message is sent as status record (@pxref{GDB/MI Output Syntax, ,
31794 @sc{gdb/mi} Output Syntax}).
31796 In addition, it prints the name and size of the sections, as they are
31797 downloaded. These messages include the following fields:
31801 The name of the section.
31803 The size of the section.
31805 The size of the overall executable to download.
31809 At the end, a summary is printed.
31811 @subsubheading @value{GDBN} Command
31813 The corresponding @value{GDBN} command is @samp{load}.
31815 @subsubheading Example
31817 Note: each status message appears on a single line. Here the messages
31818 have been broken down so that they can fit onto a page.
31823 +download,@{section=".text",section-size="6668",total-size="9880"@}
31824 +download,@{section=".text",section-sent="512",section-size="6668",
31825 total-sent="512",total-size="9880"@}
31826 +download,@{section=".text",section-sent="1024",section-size="6668",
31827 total-sent="1024",total-size="9880"@}
31828 +download,@{section=".text",section-sent="1536",section-size="6668",
31829 total-sent="1536",total-size="9880"@}
31830 +download,@{section=".text",section-sent="2048",section-size="6668",
31831 total-sent="2048",total-size="9880"@}
31832 +download,@{section=".text",section-sent="2560",section-size="6668",
31833 total-sent="2560",total-size="9880"@}
31834 +download,@{section=".text",section-sent="3072",section-size="6668",
31835 total-sent="3072",total-size="9880"@}
31836 +download,@{section=".text",section-sent="3584",section-size="6668",
31837 total-sent="3584",total-size="9880"@}
31838 +download,@{section=".text",section-sent="4096",section-size="6668",
31839 total-sent="4096",total-size="9880"@}
31840 +download,@{section=".text",section-sent="4608",section-size="6668",
31841 total-sent="4608",total-size="9880"@}
31842 +download,@{section=".text",section-sent="5120",section-size="6668",
31843 total-sent="5120",total-size="9880"@}
31844 +download,@{section=".text",section-sent="5632",section-size="6668",
31845 total-sent="5632",total-size="9880"@}
31846 +download,@{section=".text",section-sent="6144",section-size="6668",
31847 total-sent="6144",total-size="9880"@}
31848 +download,@{section=".text",section-sent="6656",section-size="6668",
31849 total-sent="6656",total-size="9880"@}
31850 +download,@{section=".init",section-size="28",total-size="9880"@}
31851 +download,@{section=".fini",section-size="28",total-size="9880"@}
31852 +download,@{section=".data",section-size="3156",total-size="9880"@}
31853 +download,@{section=".data",section-sent="512",section-size="3156",
31854 total-sent="7236",total-size="9880"@}
31855 +download,@{section=".data",section-sent="1024",section-size="3156",
31856 total-sent="7748",total-size="9880"@}
31857 +download,@{section=".data",section-sent="1536",section-size="3156",
31858 total-sent="8260",total-size="9880"@}
31859 +download,@{section=".data",section-sent="2048",section-size="3156",
31860 total-sent="8772",total-size="9880"@}
31861 +download,@{section=".data",section-sent="2560",section-size="3156",
31862 total-sent="9284",total-size="9880"@}
31863 +download,@{section=".data",section-sent="3072",section-size="3156",
31864 total-sent="9796",total-size="9880"@}
31865 ^done,address="0x10004",load-size="9880",transfer-rate="6586",
31872 @subheading The @code{-target-exec-status} Command
31873 @findex -target-exec-status
31875 @subsubheading Synopsis
31878 -target-exec-status
31881 Provide information on the state of the target (whether it is running or
31882 not, for instance).
31884 @subsubheading @value{GDBN} Command
31886 There's no equivalent @value{GDBN} command.
31888 @subsubheading Example
31892 @subheading The @code{-target-list-available-targets} Command
31893 @findex -target-list-available-targets
31895 @subsubheading Synopsis
31898 -target-list-available-targets
31901 List the possible targets to connect to.
31903 @subsubheading @value{GDBN} Command
31905 The corresponding @value{GDBN} command is @samp{help target}.
31907 @subsubheading Example
31911 @subheading The @code{-target-list-current-targets} Command
31912 @findex -target-list-current-targets
31914 @subsubheading Synopsis
31917 -target-list-current-targets
31920 Describe the current target.
31922 @subsubheading @value{GDBN} Command
31924 The corresponding information is printed by @samp{info file} (among
31927 @subsubheading Example
31931 @subheading The @code{-target-list-parameters} Command
31932 @findex -target-list-parameters
31934 @subsubheading Synopsis
31937 -target-list-parameters
31943 @subsubheading @value{GDBN} Command
31947 @subsubheading Example
31950 @subheading The @code{-target-flash-erase} Command
31951 @findex -target-flash-erase
31953 @subsubheading Synopsis
31956 -target-flash-erase
31959 Erases all known flash memory regions on the target.
31961 The corresponding @value{GDBN} command is @samp{flash-erase}.
31963 The output is a list of flash regions that have been erased, with starting
31964 addresses and memory region sizes.
31968 -target-flash-erase
31969 ^done,erased-regions=@{address="0x0",size="0x40000"@}
31973 @subheading The @code{-target-select} Command
31974 @findex -target-select
31976 @subsubheading Synopsis
31979 -target-select @var{type} @var{parameters @dots{}}
31982 Connect @value{GDBN} to the remote target. This command takes two args:
31986 The type of target, for instance @samp{remote}, etc.
31987 @item @var{parameters}
31988 Device names, host names and the like. @xref{Target Commands, ,
31989 Commands for Managing Targets}, for more details.
31992 The output is a connection notification, followed by the address at
31993 which the target program is, in the following form:
31996 ^connected,addr="@var{address}",func="@var{function name}",
31997 args=[@var{arg list}]
32000 @subsubheading @value{GDBN} Command
32002 The corresponding @value{GDBN} command is @samp{target}.
32004 @subsubheading Example
32008 -target-select remote /dev/ttya
32009 ^connected,addr="0xfe00a300",func="??",args=[]
32013 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
32014 @node GDB/MI File Transfer Commands
32015 @section @sc{gdb/mi} File Transfer Commands
32018 @subheading The @code{-target-file-put} Command
32019 @findex -target-file-put
32021 @subsubheading Synopsis
32024 -target-file-put @var{hostfile} @var{targetfile}
32027 Copy file @var{hostfile} from the host system (the machine running
32028 @value{GDBN}) to @var{targetfile} on the target system.
32030 @subsubheading @value{GDBN} Command
32032 The corresponding @value{GDBN} command is @samp{remote put}.
32034 @subsubheading Example
32038 -target-file-put localfile remotefile
32044 @subheading The @code{-target-file-get} Command
32045 @findex -target-file-get
32047 @subsubheading Synopsis
32050 -target-file-get @var{targetfile} @var{hostfile}
32053 Copy file @var{targetfile} from the target system to @var{hostfile}
32054 on the host system.
32056 @subsubheading @value{GDBN} Command
32058 The corresponding @value{GDBN} command is @samp{remote get}.
32060 @subsubheading Example
32064 -target-file-get remotefile localfile
32070 @subheading The @code{-target-file-delete} Command
32071 @findex -target-file-delete
32073 @subsubheading Synopsis
32076 -target-file-delete @var{targetfile}
32079 Delete @var{targetfile} from the target system.
32081 @subsubheading @value{GDBN} Command
32083 The corresponding @value{GDBN} command is @samp{remote delete}.
32085 @subsubheading Example
32089 -target-file-delete remotefile
32095 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
32096 @node GDB/MI Ada Exceptions Commands
32097 @section Ada Exceptions @sc{gdb/mi} Commands
32099 @subheading The @code{-info-ada-exceptions} Command
32100 @findex -info-ada-exceptions
32102 @subsubheading Synopsis
32105 -info-ada-exceptions [ @var{regexp}]
32108 List all Ada exceptions defined within the program being debugged.
32109 With a regular expression @var{regexp}, only those exceptions whose
32110 names match @var{regexp} are listed.
32112 @subsubheading @value{GDBN} Command
32114 The corresponding @value{GDBN} command is @samp{info exceptions}.
32116 @subsubheading Result
32118 The result is a table of Ada exceptions. The following columns are
32119 defined for each exception:
32123 The name of the exception.
32126 The address of the exception.
32130 @subsubheading Example
32133 -info-ada-exceptions aint
32134 ^done,ada-exceptions=@{nr_rows="2",nr_cols="2",
32135 hdr=[@{width="1",alignment="-1",col_name="name",colhdr="Name"@},
32136 @{width="1",alignment="-1",col_name="address",colhdr="Address"@}],
32137 body=[@{name="constraint_error",address="0x0000000000613da0"@},
32138 @{name="const.aint_global_e",address="0x0000000000613b00"@}]@}
32141 @subheading Catching Ada Exceptions
32143 The commands describing how to ask @value{GDBN} to stop when a program
32144 raises an exception are described at @ref{Ada Exception GDB/MI
32145 Catchpoint Commands}.
32148 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
32149 @node GDB/MI Support Commands
32150 @section @sc{gdb/mi} Support Commands
32152 Since new commands and features get regularly added to @sc{gdb/mi},
32153 some commands are available to help front-ends query the debugger
32154 about support for these capabilities. Similarly, it is also possible
32155 to query @value{GDBN} about target support of certain features.
32157 @subheading The @code{-info-gdb-mi-command} Command
32158 @cindex @code{-info-gdb-mi-command}
32159 @findex -info-gdb-mi-command
32161 @subsubheading Synopsis
32164 -info-gdb-mi-command @var{cmd_name}
32167 Query support for the @sc{gdb/mi} command named @var{cmd_name}.
32169 Note that the dash (@code{-}) starting all @sc{gdb/mi} commands
32170 is technically not part of the command name (@pxref{GDB/MI Input
32171 Syntax}), and thus should be omitted in @var{cmd_name}. However,
32172 for ease of use, this command also accepts the form with the leading
32175 @subsubheading @value{GDBN} Command
32177 There is no corresponding @value{GDBN} command.
32179 @subsubheading Result
32181 The result is a tuple. There is currently only one field:
32185 This field is equal to @code{"true"} if the @sc{gdb/mi} command exists,
32186 @code{"false"} otherwise.
32190 @subsubheading Example
32192 Here is an example where the @sc{gdb/mi} command does not exist:
32195 -info-gdb-mi-command unsupported-command
32196 ^done,command=@{exists="false"@}
32200 And here is an example where the @sc{gdb/mi} command is known
32204 -info-gdb-mi-command symbol-list-lines
32205 ^done,command=@{exists="true"@}
32208 @subheading The @code{-list-features} Command
32209 @findex -list-features
32210 @cindex supported @sc{gdb/mi} features, list
32212 Returns a list of particular features of the MI protocol that
32213 this version of gdb implements. A feature can be a command,
32214 or a new field in an output of some command, or even an
32215 important bugfix. While a frontend can sometimes detect presence
32216 of a feature at runtime, it is easier to perform detection at debugger
32219 The command returns a list of strings, with each string naming an
32220 available feature. Each returned string is just a name, it does not
32221 have any internal structure. The list of possible feature names
32227 (gdb) -list-features
32228 ^done,result=["feature1","feature2"]
32231 The current list of features is:
32234 @item frozen-varobjs
32235 Indicates support for the @code{-var-set-frozen} command, as well
32236 as possible presense of the @code{frozen} field in the output
32237 of @code{-varobj-create}.
32238 @item pending-breakpoints
32239 Indicates support for the @option{-f} option to the @code{-break-insert}
32242 Indicates Python scripting support, Python-based
32243 pretty-printing commands, and possible presence of the
32244 @samp{display_hint} field in the output of @code{-var-list-children}
32246 Indicates support for the @code{-thread-info} command.
32247 @item data-read-memory-bytes
32248 Indicates support for the @code{-data-read-memory-bytes} and the
32249 @code{-data-write-memory-bytes} commands.
32250 @item breakpoint-notifications
32251 Indicates that changes to breakpoints and breakpoints created via the
32252 CLI will be announced via async records.
32253 @item ada-task-info
32254 Indicates support for the @code{-ada-task-info} command.
32255 @item language-option
32256 Indicates that all @sc{gdb/mi} commands accept the @option{--language}
32257 option (@pxref{Context management}).
32258 @item info-gdb-mi-command
32259 Indicates support for the @code{-info-gdb-mi-command} command.
32260 @item undefined-command-error-code
32261 Indicates support for the "undefined-command" error code in error result
32262 records, produced when trying to execute an undefined @sc{gdb/mi} command
32263 (@pxref{GDB/MI Result Records}).
32264 @item exec-run-start-option
32265 Indicates that the @code{-exec-run} command supports the @option{--start}
32266 option (@pxref{GDB/MI Program Execution}).
32269 @subheading The @code{-list-target-features} Command
32270 @findex -list-target-features
32272 Returns a list of particular features that are supported by the
32273 target. Those features affect the permitted MI commands, but
32274 unlike the features reported by the @code{-list-features} command, the
32275 features depend on which target GDB is using at the moment. Whenever
32276 a target can change, due to commands such as @code{-target-select},
32277 @code{-target-attach} or @code{-exec-run}, the list of target features
32278 may change, and the frontend should obtain it again.
32282 (gdb) -list-target-features
32283 ^done,result=["async"]
32286 The current list of features is:
32290 Indicates that the target is capable of asynchronous command
32291 execution, which means that @value{GDBN} will accept further commands
32292 while the target is running.
32295 Indicates that the target is capable of reverse execution.
32296 @xref{Reverse Execution}, for more information.
32300 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
32301 @node GDB/MI Miscellaneous Commands
32302 @section Miscellaneous @sc{gdb/mi} Commands
32304 @c @subheading -gdb-complete
32306 @subheading The @code{-gdb-exit} Command
32309 @subsubheading Synopsis
32315 Exit @value{GDBN} immediately.
32317 @subsubheading @value{GDBN} Command
32319 Approximately corresponds to @samp{quit}.
32321 @subsubheading Example
32331 @subheading The @code{-exec-abort} Command
32332 @findex -exec-abort
32334 @subsubheading Synopsis
32340 Kill the inferior running program.
32342 @subsubheading @value{GDBN} Command
32344 The corresponding @value{GDBN} command is @samp{kill}.
32346 @subsubheading Example
32351 @subheading The @code{-gdb-set} Command
32354 @subsubheading Synopsis
32360 Set an internal @value{GDBN} variable.
32361 @c IS THIS A DOLLAR VARIABLE? OR SOMETHING LIKE ANNOTATE ?????
32363 @subsubheading @value{GDBN} Command
32365 The corresponding @value{GDBN} command is @samp{set}.
32367 @subsubheading Example
32377 @subheading The @code{-gdb-show} Command
32380 @subsubheading Synopsis
32386 Show the current value of a @value{GDBN} variable.
32388 @subsubheading @value{GDBN} Command
32390 The corresponding @value{GDBN} command is @samp{show}.
32392 @subsubheading Example
32401 @c @subheading -gdb-source
32404 @subheading The @code{-gdb-version} Command
32405 @findex -gdb-version
32407 @subsubheading Synopsis
32413 Show version information for @value{GDBN}. Used mostly in testing.
32415 @subsubheading @value{GDBN} Command
32417 The @value{GDBN} equivalent is @samp{show version}. @value{GDBN} by
32418 default shows this information when you start an interactive session.
32420 @subsubheading Example
32422 @c This example modifies the actual output from GDB to avoid overfull
32428 ~Copyright 2000 Free Software Foundation, Inc.
32429 ~GDB is free software, covered by the GNU General Public License, and
32430 ~you are welcome to change it and/or distribute copies of it under
32431 ~ certain conditions.
32432 ~Type "show copying" to see the conditions.
32433 ~There is absolutely no warranty for GDB. Type "show warranty" for
32435 ~This GDB was configured as
32436 "--host=sparc-sun-solaris2.5.1 --target=ppc-eabi".
32441 @subheading The @code{-list-thread-groups} Command
32442 @findex -list-thread-groups
32444 @subheading Synopsis
32447 -list-thread-groups [ --available ] [ --recurse 1 ] [ @var{group} ... ]
32450 Lists thread groups (@pxref{Thread groups}). When a single thread
32451 group is passed as the argument, lists the children of that group.
32452 When several thread group are passed, lists information about those
32453 thread groups. Without any parameters, lists information about all
32454 top-level thread groups.
32456 Normally, thread groups that are being debugged are reported.
32457 With the @samp{--available} option, @value{GDBN} reports thread groups
32458 available on the target.
32460 The output of this command may have either a @samp{threads} result or
32461 a @samp{groups} result. The @samp{thread} result has a list of tuples
32462 as value, with each tuple describing a thread (@pxref{GDB/MI Thread
32463 Information}). The @samp{groups} result has a list of tuples as value,
32464 each tuple describing a thread group. If top-level groups are
32465 requested (that is, no parameter is passed), or when several groups
32466 are passed, the output always has a @samp{groups} result. The format
32467 of the @samp{group} result is described below.
32469 To reduce the number of roundtrips it's possible to list thread groups
32470 together with their children, by passing the @samp{--recurse} option
32471 and the recursion depth. Presently, only recursion depth of 1 is
32472 permitted. If this option is present, then every reported thread group
32473 will also include its children, either as @samp{group} or
32474 @samp{threads} field.
32476 In general, any combination of option and parameters is permitted, with
32477 the following caveats:
32481 When a single thread group is passed, the output will typically
32482 be the @samp{threads} result. Because threads may not contain
32483 anything, the @samp{recurse} option will be ignored.
32486 When the @samp{--available} option is passed, limited information may
32487 be available. In particular, the list of threads of a process might
32488 be inaccessible. Further, specifying specific thread groups might
32489 not give any performance advantage over listing all thread groups.
32490 The frontend should assume that @samp{-list-thread-groups --available}
32491 is always an expensive operation and cache the results.
32495 The @samp{groups} result is a list of tuples, where each tuple may
32496 have the following fields:
32500 Identifier of the thread group. This field is always present.
32501 The identifier is an opaque string; frontends should not try to
32502 convert it to an integer, even though it might look like one.
32505 The type of the thread group. At present, only @samp{process} is a
32509 The target-specific process identifier. This field is only present
32510 for thread groups of type @samp{process} and only if the process exists.
32513 The exit code of this group's last exited thread, formatted in octal.
32514 This field is only present for thread groups of type @samp{process} and
32515 only if the process is not running.
32518 The number of children this thread group has. This field may be
32519 absent for an available thread group.
32522 This field has a list of tuples as value, each tuple describing a
32523 thread. It may be present if the @samp{--recurse} option is
32524 specified, and it's actually possible to obtain the threads.
32527 This field is a list of integers, each identifying a core that one
32528 thread of the group is running on. This field may be absent if
32529 such information is not available.
32532 The name of the executable file that corresponds to this thread group.
32533 The field is only present for thread groups of type @samp{process},
32534 and only if there is a corresponding executable file.
32538 @subheading Example
32542 -list-thread-groups
32543 ^done,groups=[@{id="17",type="process",pid="yyy",num_children="2"@}]
32544 -list-thread-groups 17
32545 ^done,threads=[@{id="2",target-id="Thread 0xb7e14b90 (LWP 21257)",
32546 frame=@{level="0",addr="0xffffe410",func="__kernel_vsyscall",args=[]@},state="running"@},
32547 @{id="1",target-id="Thread 0xb7e156b0 (LWP 21254)",
32548 frame=@{level="0",addr="0x0804891f",func="foo",args=[@{name="i",value="10"@}],
32549 file="/tmp/a.c",fullname="/tmp/a.c",line="158"@},state="running"@}]]
32550 -list-thread-groups --available
32551 ^done,groups=[@{id="17",type="process",pid="yyy",num_children="2",cores=[1,2]@}]
32552 -list-thread-groups --available --recurse 1
32553 ^done,groups=[@{id="17", types="process",pid="yyy",num_children="2",cores=[1,2],
32554 threads=[@{id="1",target-id="Thread 0xb7e14b90",cores=[1]@},
32555 @{id="2",target-id="Thread 0xb7e14b90",cores=[2]@}]@},..]
32556 -list-thread-groups --available --recurse 1 17 18
32557 ^done,groups=[@{id="17", types="process",pid="yyy",num_children="2",cores=[1,2],
32558 threads=[@{id="1",target-id="Thread 0xb7e14b90",cores=[1]@},
32559 @{id="2",target-id="Thread 0xb7e14b90",cores=[2]@}]@},...]
32562 @subheading The @code{-info-os} Command
32565 @subsubheading Synopsis
32568 -info-os [ @var{type} ]
32571 If no argument is supplied, the command returns a table of available
32572 operating-system-specific information types. If one of these types is
32573 supplied as an argument @var{type}, then the command returns a table
32574 of data of that type.
32576 The types of information available depend on the target operating
32579 @subsubheading @value{GDBN} Command
32581 The corresponding @value{GDBN} command is @samp{info os}.
32583 @subsubheading Example
32585 When run on a @sc{gnu}/Linux system, the output will look something
32591 ^done,OSDataTable=@{nr_rows="10",nr_cols="3",
32592 hdr=[@{width="10",alignment="-1",col_name="col0",colhdr="Type"@},
32593 @{width="10",alignment="-1",col_name="col1",colhdr="Description"@},
32594 @{width="10",alignment="-1",col_name="col2",colhdr="Title"@}],
32595 body=[item=@{col0="cpus",col1="Listing of all cpus/cores on the system",
32597 item=@{col0="files",col1="Listing of all file descriptors",
32598 col2="File descriptors"@},
32599 item=@{col0="modules",col1="Listing of all loaded kernel modules",
32600 col2="Kernel modules"@},
32601 item=@{col0="msg",col1="Listing of all message queues",
32602 col2="Message queues"@},
32603 item=@{col0="processes",col1="Listing of all processes",
32604 col2="Processes"@},
32605 item=@{col0="procgroups",col1="Listing of all process groups",
32606 col2="Process groups"@},
32607 item=@{col0="semaphores",col1="Listing of all semaphores",
32608 col2="Semaphores"@},
32609 item=@{col0="shm",col1="Listing of all shared-memory regions",
32610 col2="Shared-memory regions"@},
32611 item=@{col0="sockets",col1="Listing of all internet-domain sockets",
32613 item=@{col0="threads",col1="Listing of all threads",
32617 ^done,OSDataTable=@{nr_rows="190",nr_cols="4",
32618 hdr=[@{width="10",alignment="-1",col_name="col0",colhdr="pid"@},
32619 @{width="10",alignment="-1",col_name="col1",colhdr="user"@},
32620 @{width="10",alignment="-1",col_name="col2",colhdr="command"@},
32621 @{width="10",alignment="-1",col_name="col3",colhdr="cores"@}],
32622 body=[item=@{col0="1",col1="root",col2="/sbin/init",col3="0"@},
32623 item=@{col0="2",col1="root",col2="[kthreadd]",col3="1"@},
32624 item=@{col0="3",col1="root",col2="[ksoftirqd/0]",col3="0"@},
32626 item=@{col0="26446",col1="stan",col2="bash",col3="0"@},
32627 item=@{col0="28152",col1="stan",col2="bash",col3="1"@}]@}
32631 (Note that the MI output here includes a @code{"Title"} column that
32632 does not appear in command-line @code{info os}; this column is useful
32633 for MI clients that want to enumerate the types of data, such as in a
32634 popup menu, but is needless clutter on the command line, and
32635 @code{info os} omits it.)
32637 @subheading The @code{-add-inferior} Command
32638 @findex -add-inferior
32640 @subheading Synopsis
32646 Creates a new inferior (@pxref{Inferiors and Programs}). The created
32647 inferior is not associated with any executable. Such association may
32648 be established with the @samp{-file-exec-and-symbols} command
32649 (@pxref{GDB/MI File Commands}). The command response has a single
32650 field, @samp{inferior}, whose value is the identifier of the
32651 thread group corresponding to the new inferior.
32653 @subheading Example
32658 ^done,inferior="i3"
32661 @subheading The @code{-interpreter-exec} Command
32662 @findex -interpreter-exec
32664 @subheading Synopsis
32667 -interpreter-exec @var{interpreter} @var{command}
32669 @anchor{-interpreter-exec}
32671 Execute the specified @var{command} in the given @var{interpreter}.
32673 @subheading @value{GDBN} Command
32675 The corresponding @value{GDBN} command is @samp{interpreter-exec}.
32677 @subheading Example
32681 -interpreter-exec console "break main"
32682 &"During symbol reading, couldn't parse type; debugger out of date?.\n"
32683 &"During symbol reading, bad structure-type format.\n"
32684 ~"Breakpoint 1 at 0x8074fc6: file ../../src/gdb/main.c, line 743.\n"
32689 @subheading The @code{-inferior-tty-set} Command
32690 @findex -inferior-tty-set
32692 @subheading Synopsis
32695 -inferior-tty-set /dev/pts/1
32698 Set terminal for future runs of the program being debugged.
32700 @subheading @value{GDBN} Command
32702 The corresponding @value{GDBN} command is @samp{set inferior-tty} /dev/pts/1.
32704 @subheading Example
32708 -inferior-tty-set /dev/pts/1
32713 @subheading The @code{-inferior-tty-show} Command
32714 @findex -inferior-tty-show
32716 @subheading Synopsis
32722 Show terminal for future runs of program being debugged.
32724 @subheading @value{GDBN} Command
32726 The corresponding @value{GDBN} command is @samp{show inferior-tty}.
32728 @subheading Example
32732 -inferior-tty-set /dev/pts/1
32736 ^done,inferior_tty_terminal="/dev/pts/1"
32740 @subheading The @code{-enable-timings} Command
32741 @findex -enable-timings
32743 @subheading Synopsis
32746 -enable-timings [yes | no]
32749 Toggle the printing of the wallclock, user and system times for an MI
32750 command as a field in its output. This command is to help frontend
32751 developers optimize the performance of their code. No argument is
32752 equivalent to @samp{yes}.
32754 @subheading @value{GDBN} Command
32758 @subheading Example
32766 ^done,bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
32767 addr="0x080484ed",func="main",file="myprog.c",
32768 fullname="/home/nickrob/myprog.c",line="73",thread-groups=["i1"],
32770 time=@{wallclock="0.05185",user="0.00800",system="0.00000"@}
32778 *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",thread-id="0",
32779 frame=@{addr="0x080484ed",func="main",args=[@{name="argc",value="1"@},
32780 @{name="argv",value="0xbfb60364"@}],file="myprog.c",
32781 fullname="/home/nickrob/myprog.c",line="73"@}
32786 @chapter @value{GDBN} Annotations
32788 This chapter describes annotations in @value{GDBN}. Annotations were
32789 designed to interface @value{GDBN} to graphical user interfaces or other
32790 similar programs which want to interact with @value{GDBN} at a
32791 relatively high level.
32793 The annotation mechanism has largely been superseded by @sc{gdb/mi}
32797 This is Edition @value{EDITION}, @value{DATE}.
32801 * Annotations Overview:: What annotations are; the general syntax.
32802 * Server Prefix:: Issuing a command without affecting user state.
32803 * Prompting:: Annotations marking @value{GDBN}'s need for input.
32804 * Errors:: Annotations for error messages.
32805 * Invalidation:: Some annotations describe things now invalid.
32806 * Annotations for Running::
32807 Whether the program is running, how it stopped, etc.
32808 * Source Annotations:: Annotations describing source code.
32811 @node Annotations Overview
32812 @section What is an Annotation?
32813 @cindex annotations
32815 Annotations start with a newline character, two @samp{control-z}
32816 characters, and the name of the annotation. If there is no additional
32817 information associated with this annotation, the name of the annotation
32818 is followed immediately by a newline. If there is additional
32819 information, the name of the annotation is followed by a space, the
32820 additional information, and a newline. The additional information
32821 cannot contain newline characters.
32823 Any output not beginning with a newline and two @samp{control-z}
32824 characters denotes literal output from @value{GDBN}. Currently there is
32825 no need for @value{GDBN} to output a newline followed by two
32826 @samp{control-z} characters, but if there was such a need, the
32827 annotations could be extended with an @samp{escape} annotation which
32828 means those three characters as output.
32830 The annotation @var{level}, which is specified using the
32831 @option{--annotate} command line option (@pxref{Mode Options}), controls
32832 how much information @value{GDBN} prints together with its prompt,
32833 values of expressions, source lines, and other types of output. Level 0
32834 is for no annotations, level 1 is for use when @value{GDBN} is run as a
32835 subprocess of @sc{gnu} Emacs, level 3 is the maximum annotation suitable
32836 for programs that control @value{GDBN}, and level 2 annotations have
32837 been made obsolete (@pxref{Limitations, , Limitations of the Annotation
32838 Interface, annotate, GDB's Obsolete Annotations}).
32841 @kindex set annotate
32842 @item set annotate @var{level}
32843 The @value{GDBN} command @code{set annotate} sets the level of
32844 annotations to the specified @var{level}.
32846 @item show annotate
32847 @kindex show annotate
32848 Show the current annotation level.
32851 This chapter describes level 3 annotations.
32853 A simple example of starting up @value{GDBN} with annotations is:
32856 $ @kbd{gdb --annotate=3}
32858 Copyright 2003 Free Software Foundation, Inc.
32859 GDB is free software, covered by the GNU General Public License,
32860 and you are welcome to change it and/or distribute copies of it
32861 under certain conditions.
32862 Type "show copying" to see the conditions.
32863 There is absolutely no warranty for GDB. Type "show warranty"
32865 This GDB was configured as "i386-pc-linux-gnu"
32876 Here @samp{quit} is input to @value{GDBN}; the rest is output from
32877 @value{GDBN}. The three lines beginning @samp{^Z^Z} (where @samp{^Z}
32878 denotes a @samp{control-z} character) are annotations; the rest is
32879 output from @value{GDBN}.
32881 @node Server Prefix
32882 @section The Server Prefix
32883 @cindex server prefix
32885 If you prefix a command with @samp{server } then it will not affect
32886 the command history, nor will it affect @value{GDBN}'s notion of which
32887 command to repeat if @key{RET} is pressed on a line by itself. This
32888 means that commands can be run behind a user's back by a front-end in
32889 a transparent manner.
32891 The @code{server } prefix does not affect the recording of values into
32892 the value history; to print a value without recording it into the
32893 value history, use the @code{output} command instead of the
32894 @code{print} command.
32896 Using this prefix also disables confirmation requests
32897 (@pxref{confirmation requests}).
32900 @section Annotation for @value{GDBN} Input
32902 @cindex annotations for prompts
32903 When @value{GDBN} prompts for input, it annotates this fact so it is possible
32904 to know when to send output, when the output from a given command is
32907 Different kinds of input each have a different @dfn{input type}. Each
32908 input type has three annotations: a @code{pre-} annotation, which
32909 denotes the beginning of any prompt which is being output, a plain
32910 annotation, which denotes the end of the prompt, and then a @code{post-}
32911 annotation which denotes the end of any echo which may (or may not) be
32912 associated with the input. For example, the @code{prompt} input type
32913 features the following annotations:
32921 The input types are
32924 @findex pre-prompt annotation
32925 @findex prompt annotation
32926 @findex post-prompt annotation
32928 When @value{GDBN} is prompting for a command (the main @value{GDBN} prompt).
32930 @findex pre-commands annotation
32931 @findex commands annotation
32932 @findex post-commands annotation
32934 When @value{GDBN} prompts for a set of commands, like in the @code{commands}
32935 command. The annotations are repeated for each command which is input.
32937 @findex pre-overload-choice annotation
32938 @findex overload-choice annotation
32939 @findex post-overload-choice annotation
32940 @item overload-choice
32941 When @value{GDBN} wants the user to select between various overloaded functions.
32943 @findex pre-query annotation
32944 @findex query annotation
32945 @findex post-query annotation
32947 When @value{GDBN} wants the user to confirm a potentially dangerous operation.
32949 @findex pre-prompt-for-continue annotation
32950 @findex prompt-for-continue annotation
32951 @findex post-prompt-for-continue annotation
32952 @item prompt-for-continue
32953 When @value{GDBN} is asking the user to press return to continue. Note: Don't
32954 expect this to work well; instead use @code{set height 0} to disable
32955 prompting. This is because the counting of lines is buggy in the
32956 presence of annotations.
32961 @cindex annotations for errors, warnings and interrupts
32963 @findex quit annotation
32968 This annotation occurs right before @value{GDBN} responds to an interrupt.
32970 @findex error annotation
32975 This annotation occurs right before @value{GDBN} responds to an error.
32977 Quit and error annotations indicate that any annotations which @value{GDBN} was
32978 in the middle of may end abruptly. For example, if a
32979 @code{value-history-begin} annotation is followed by a @code{error}, one
32980 cannot expect to receive the matching @code{value-history-end}. One
32981 cannot expect not to receive it either, however; an error annotation
32982 does not necessarily mean that @value{GDBN} is immediately returning all the way
32985 @findex error-begin annotation
32986 A quit or error annotation may be preceded by
32992 Any output between that and the quit or error annotation is the error
32995 Warning messages are not yet annotated.
32996 @c If we want to change that, need to fix warning(), type_error(),
32997 @c range_error(), and possibly other places.
33000 @section Invalidation Notices
33002 @cindex annotations for invalidation messages
33003 The following annotations say that certain pieces of state may have
33007 @findex frames-invalid annotation
33008 @item ^Z^Zframes-invalid
33010 The frames (for example, output from the @code{backtrace} command) may
33013 @findex breakpoints-invalid annotation
33014 @item ^Z^Zbreakpoints-invalid
33016 The breakpoints may have changed. For example, the user just added or
33017 deleted a breakpoint.
33020 @node Annotations for Running
33021 @section Running the Program
33022 @cindex annotations for running programs
33024 @findex starting annotation
33025 @findex stopping annotation
33026 When the program starts executing due to a @value{GDBN} command such as
33027 @code{step} or @code{continue},
33033 is output. When the program stops,
33039 is output. Before the @code{stopped} annotation, a variety of
33040 annotations describe how the program stopped.
33043 @findex exited annotation
33044 @item ^Z^Zexited @var{exit-status}
33045 The program exited, and @var{exit-status} is the exit status (zero for
33046 successful exit, otherwise nonzero).
33048 @findex signalled annotation
33049 @findex signal-name annotation
33050 @findex signal-name-end annotation
33051 @findex signal-string annotation
33052 @findex signal-string-end annotation
33053 @item ^Z^Zsignalled
33054 The program exited with a signal. After the @code{^Z^Zsignalled}, the
33055 annotation continues:
33061 ^Z^Zsignal-name-end
33065 ^Z^Zsignal-string-end
33070 where @var{name} is the name of the signal, such as @code{SIGILL} or
33071 @code{SIGSEGV}, and @var{string} is the explanation of the signal, such
33072 as @code{Illegal Instruction} or @code{Segmentation fault}. The arguments
33073 @var{intro-text}, @var{middle-text}, and @var{end-text} are for the
33074 user's benefit and have no particular format.
33076 @findex signal annotation
33078 The syntax of this annotation is just like @code{signalled}, but @value{GDBN} is
33079 just saying that the program received the signal, not that it was
33080 terminated with it.
33082 @findex breakpoint annotation
33083 @item ^Z^Zbreakpoint @var{number}
33084 The program hit breakpoint number @var{number}.
33086 @findex watchpoint annotation
33087 @item ^Z^Zwatchpoint @var{number}
33088 The program hit watchpoint number @var{number}.
33091 @node Source Annotations
33092 @section Displaying Source
33093 @cindex annotations for source display
33095 @findex source annotation
33096 The following annotation is used instead of displaying source code:
33099 ^Z^Zsource @var{filename}:@var{line}:@var{character}:@var{middle}:@var{addr}
33102 where @var{filename} is an absolute file name indicating which source
33103 file, @var{line} is the line number within that file (where 1 is the
33104 first line in the file), @var{character} is the character position
33105 within the file (where 0 is the first character in the file) (for most
33106 debug formats this will necessarily point to the beginning of a line),
33107 @var{middle} is @samp{middle} if @var{addr} is in the middle of the
33108 line, or @samp{beg} if @var{addr} is at the beginning of the line, and
33109 @var{addr} is the address in the target program associated with the
33110 source which is being displayed. The @var{addr} is in the form @samp{0x}
33111 followed by one or more lowercase hex digits (note that this does not
33112 depend on the language).
33114 @node JIT Interface
33115 @chapter JIT Compilation Interface
33116 @cindex just-in-time compilation
33117 @cindex JIT compilation interface
33119 This chapter documents @value{GDBN}'s @dfn{just-in-time} (JIT) compilation
33120 interface. A JIT compiler is a program or library that generates native
33121 executable code at runtime and executes it, usually in order to achieve good
33122 performance while maintaining platform independence.
33124 Programs that use JIT compilation are normally difficult to debug because
33125 portions of their code are generated at runtime, instead of being loaded from
33126 object files, which is where @value{GDBN} normally finds the program's symbols
33127 and debug information. In order to debug programs that use JIT compilation,
33128 @value{GDBN} has an interface that allows the program to register in-memory
33129 symbol files with @value{GDBN} at runtime.
33131 If you are using @value{GDBN} to debug a program that uses this interface, then
33132 it should work transparently so long as you have not stripped the binary. If
33133 you are developing a JIT compiler, then the interface is documented in the rest
33134 of this chapter. At this time, the only known client of this interface is the
33137 Broadly speaking, the JIT interface mirrors the dynamic loader interface. The
33138 JIT compiler communicates with @value{GDBN} by writing data into a global
33139 variable and calling a fuction at a well-known symbol. When @value{GDBN}
33140 attaches, it reads a linked list of symbol files from the global variable to
33141 find existing code, and puts a breakpoint in the function so that it can find
33142 out about additional code.
33145 * Declarations:: Relevant C struct declarations
33146 * Registering Code:: Steps to register code
33147 * Unregistering Code:: Steps to unregister code
33148 * Custom Debug Info:: Emit debug information in a custom format
33152 @section JIT Declarations
33154 These are the relevant struct declarations that a C program should include to
33155 implement the interface:
33165 struct jit_code_entry
33167 struct jit_code_entry *next_entry;
33168 struct jit_code_entry *prev_entry;
33169 const char *symfile_addr;
33170 uint64_t symfile_size;
33173 struct jit_descriptor
33176 /* This type should be jit_actions_t, but we use uint32_t
33177 to be explicit about the bitwidth. */
33178 uint32_t action_flag;
33179 struct jit_code_entry *relevant_entry;
33180 struct jit_code_entry *first_entry;
33183 /* GDB puts a breakpoint in this function. */
33184 void __attribute__((noinline)) __jit_debug_register_code() @{ @};
33186 /* Make sure to specify the version statically, because the
33187 debugger may check the version before we can set it. */
33188 struct jit_descriptor __jit_debug_descriptor = @{ 1, 0, 0, 0 @};
33191 If the JIT is multi-threaded, then it is important that the JIT synchronize any
33192 modifications to this global data properly, which can easily be done by putting
33193 a global mutex around modifications to these structures.
33195 @node Registering Code
33196 @section Registering Code
33198 To register code with @value{GDBN}, the JIT should follow this protocol:
33202 Generate an object file in memory with symbols and other desired debug
33203 information. The file must include the virtual addresses of the sections.
33206 Create a code entry for the file, which gives the start and size of the symbol
33210 Add it to the linked list in the JIT descriptor.
33213 Point the relevant_entry field of the descriptor at the entry.
33216 Set @code{action_flag} to @code{JIT_REGISTER} and call
33217 @code{__jit_debug_register_code}.
33220 When @value{GDBN} is attached and the breakpoint fires, @value{GDBN} uses the
33221 @code{relevant_entry} pointer so it doesn't have to walk the list looking for
33222 new code. However, the linked list must still be maintained in order to allow
33223 @value{GDBN} to attach to a running process and still find the symbol files.
33225 @node Unregistering Code
33226 @section Unregistering Code
33228 If code is freed, then the JIT should use the following protocol:
33232 Remove the code entry corresponding to the code from the linked list.
33235 Point the @code{relevant_entry} field of the descriptor at the code entry.
33238 Set @code{action_flag} to @code{JIT_UNREGISTER} and call
33239 @code{__jit_debug_register_code}.
33242 If the JIT frees or recompiles code without unregistering it, then @value{GDBN}
33243 and the JIT will leak the memory used for the associated symbol files.
33245 @node Custom Debug Info
33246 @section Custom Debug Info
33247 @cindex custom JIT debug info
33248 @cindex JIT debug info reader
33250 Generating debug information in platform-native file formats (like ELF
33251 or COFF) may be an overkill for JIT compilers; especially if all the
33252 debug info is used for is displaying a meaningful backtrace. The
33253 issue can be resolved by having the JIT writers decide on a debug info
33254 format and also provide a reader that parses the debug info generated
33255 by the JIT compiler. This section gives a brief overview on writing
33256 such a parser. More specific details can be found in the source file
33257 @file{gdb/jit-reader.in}, which is also installed as a header at
33258 @file{@var{includedir}/gdb/jit-reader.h} for easy inclusion.
33260 The reader is implemented as a shared object (so this functionality is
33261 not available on platforms which don't allow loading shared objects at
33262 runtime). Two @value{GDBN} commands, @code{jit-reader-load} and
33263 @code{jit-reader-unload} are provided, to be used to load and unload
33264 the readers from a preconfigured directory. Once loaded, the shared
33265 object is used the parse the debug information emitted by the JIT
33269 * Using JIT Debug Info Readers:: How to use supplied readers correctly
33270 * Writing JIT Debug Info Readers:: Creating a debug-info reader
33273 @node Using JIT Debug Info Readers
33274 @subsection Using JIT Debug Info Readers
33275 @kindex jit-reader-load
33276 @kindex jit-reader-unload
33278 Readers can be loaded and unloaded using the @code{jit-reader-load}
33279 and @code{jit-reader-unload} commands.
33282 @item jit-reader-load @var{reader}
33283 Load the JIT reader named @var{reader}, which is a shared
33284 object specified as either an absolute or a relative file name. In
33285 the latter case, @value{GDBN} will try to load the reader from a
33286 pre-configured directory, usually @file{@var{libdir}/gdb/} on a UNIX
33287 system (here @var{libdir} is the system library directory, often
33288 @file{/usr/local/lib}).
33290 Only one reader can be active at a time; trying to load a second
33291 reader when one is already loaded will result in @value{GDBN}
33292 reporting an error. A new JIT reader can be loaded by first unloading
33293 the current one using @code{jit-reader-unload} and then invoking
33294 @code{jit-reader-load}.
33296 @item jit-reader-unload
33297 Unload the currently loaded JIT reader.
33301 @node Writing JIT Debug Info Readers
33302 @subsection Writing JIT Debug Info Readers
33303 @cindex writing JIT debug info readers
33305 As mentioned, a reader is essentially a shared object conforming to a
33306 certain ABI. This ABI is described in @file{jit-reader.h}.
33308 @file{jit-reader.h} defines the structures, macros and functions
33309 required to write a reader. It is installed (along with
33310 @value{GDBN}), in @file{@var{includedir}/gdb} where @var{includedir} is
33311 the system include directory.
33313 Readers need to be released under a GPL compatible license. A reader
33314 can be declared as released under such a license by placing the macro
33315 @code{GDB_DECLARE_GPL_COMPATIBLE_READER} in a source file.
33317 The entry point for readers is the symbol @code{gdb_init_reader},
33318 which is expected to be a function with the prototype
33320 @findex gdb_init_reader
33322 extern struct gdb_reader_funcs *gdb_init_reader (void);
33325 @cindex @code{struct gdb_reader_funcs}
33327 @code{struct gdb_reader_funcs} contains a set of pointers to callback
33328 functions. These functions are executed to read the debug info
33329 generated by the JIT compiler (@code{read}), to unwind stack frames
33330 (@code{unwind}) and to create canonical frame IDs
33331 (@code{get_Frame_id}). It also has a callback that is called when the
33332 reader is being unloaded (@code{destroy}). The struct looks like this
33335 struct gdb_reader_funcs
33337 /* Must be set to GDB_READER_INTERFACE_VERSION. */
33338 int reader_version;
33340 /* For use by the reader. */
33343 gdb_read_debug_info *read;
33344 gdb_unwind_frame *unwind;
33345 gdb_get_frame_id *get_frame_id;
33346 gdb_destroy_reader *destroy;
33350 @cindex @code{struct gdb_symbol_callbacks}
33351 @cindex @code{struct gdb_unwind_callbacks}
33353 The callbacks are provided with another set of callbacks by
33354 @value{GDBN} to do their job. For @code{read}, these callbacks are
33355 passed in a @code{struct gdb_symbol_callbacks} and for @code{unwind}
33356 and @code{get_frame_id}, in a @code{struct gdb_unwind_callbacks}.
33357 @code{struct gdb_symbol_callbacks} has callbacks to create new object
33358 files and new symbol tables inside those object files. @code{struct
33359 gdb_unwind_callbacks} has callbacks to read registers off the current
33360 frame and to write out the values of the registers in the previous
33361 frame. Both have a callback (@code{target_read}) to read bytes off the
33362 target's address space.
33364 @node In-Process Agent
33365 @chapter In-Process Agent
33366 @cindex debugging agent
33367 The traditional debugging model is conceptually low-speed, but works fine,
33368 because most bugs can be reproduced in debugging-mode execution. However,
33369 as multi-core or many-core processors are becoming mainstream, and
33370 multi-threaded programs become more and more popular, there should be more
33371 and more bugs that only manifest themselves at normal-mode execution, for
33372 example, thread races, because debugger's interference with the program's
33373 timing may conceal the bugs. On the other hand, in some applications,
33374 it is not feasible for the debugger to interrupt the program's execution
33375 long enough for the developer to learn anything helpful about its behavior.
33376 If the program's correctness depends on its real-time behavior, delays
33377 introduced by a debugger might cause the program to fail, even when the
33378 code itself is correct. It is useful to be able to observe the program's
33379 behavior without interrupting it.
33381 Therefore, traditional debugging model is too intrusive to reproduce
33382 some bugs. In order to reduce the interference with the program, we can
33383 reduce the number of operations performed by debugger. The
33384 @dfn{In-Process Agent}, a shared library, is running within the same
33385 process with inferior, and is able to perform some debugging operations
33386 itself. As a result, debugger is only involved when necessary, and
33387 performance of debugging can be improved accordingly. Note that
33388 interference with program can be reduced but can't be removed completely,
33389 because the in-process agent will still stop or slow down the program.
33391 The in-process agent can interpret and execute Agent Expressions
33392 (@pxref{Agent Expressions}) during performing debugging operations. The
33393 agent expressions can be used for different purposes, such as collecting
33394 data in tracepoints, and condition evaluation in breakpoints.
33396 @anchor{Control Agent}
33397 You can control whether the in-process agent is used as an aid for
33398 debugging with the following commands:
33401 @kindex set agent on
33403 Causes the in-process agent to perform some operations on behalf of the
33404 debugger. Just which operations requested by the user will be done
33405 by the in-process agent depends on the its capabilities. For example,
33406 if you request to evaluate breakpoint conditions in the in-process agent,
33407 and the in-process agent has such capability as well, then breakpoint
33408 conditions will be evaluated in the in-process agent.
33410 @kindex set agent off
33411 @item set agent off
33412 Disables execution of debugging operations by the in-process agent. All
33413 of the operations will be performed by @value{GDBN}.
33417 Display the current setting of execution of debugging operations by
33418 the in-process agent.
33422 * In-Process Agent Protocol::
33425 @node In-Process Agent Protocol
33426 @section In-Process Agent Protocol
33427 @cindex in-process agent protocol
33429 The in-process agent is able to communicate with both @value{GDBN} and
33430 GDBserver (@pxref{In-Process Agent}). This section documents the protocol
33431 used for communications between @value{GDBN} or GDBserver and the IPA.
33432 In general, @value{GDBN} or GDBserver sends commands
33433 (@pxref{IPA Protocol Commands}) and data to in-process agent, and then
33434 in-process agent replies back with the return result of the command, or
33435 some other information. The data sent to in-process agent is composed
33436 of primitive data types, such as 4-byte or 8-byte type, and composite
33437 types, which are called objects (@pxref{IPA Protocol Objects}).
33440 * IPA Protocol Objects::
33441 * IPA Protocol Commands::
33444 @node IPA Protocol Objects
33445 @subsection IPA Protocol Objects
33446 @cindex ipa protocol objects
33448 The commands sent to and results received from agent may contain some
33449 complex data types called @dfn{objects}.
33451 The in-process agent is running on the same machine with @value{GDBN}
33452 or GDBserver, so it doesn't have to handle as much differences between
33453 two ends as remote protocol (@pxref{Remote Protocol}) tries to handle.
33454 However, there are still some differences of two ends in two processes:
33458 word size. On some 64-bit machines, @value{GDBN} or GDBserver can be
33459 compiled as a 64-bit executable, while in-process agent is a 32-bit one.
33461 ABI. Some machines may have multiple types of ABI, @value{GDBN} or
33462 GDBserver is compiled with one, and in-process agent is compiled with
33466 Here are the IPA Protocol Objects:
33470 agent expression object. It represents an agent expression
33471 (@pxref{Agent Expressions}).
33472 @anchor{agent expression object}
33474 tracepoint action object. It represents a tracepoint action
33475 (@pxref{Tracepoint Actions,,Tracepoint Action Lists}) to collect registers,
33476 memory, static trace data and to evaluate expression.
33477 @anchor{tracepoint action object}
33479 tracepoint object. It represents a tracepoint (@pxref{Tracepoints}).
33480 @anchor{tracepoint object}
33484 The following table describes important attributes of each IPA protocol
33487 @multitable @columnfractions .30 .20 .50
33488 @headitem Name @tab Size @tab Description
33489 @item @emph{agent expression object} @tab @tab
33490 @item length @tab 4 @tab length of bytes code
33491 @item byte code @tab @var{length} @tab contents of byte code
33492 @item @emph{tracepoint action for collecting memory} @tab @tab
33493 @item 'M' @tab 1 @tab type of tracepoint action
33494 @item addr @tab 8 @tab if @var{basereg} is @samp{-1}, @var{addr} is the
33495 address of the lowest byte to collect, otherwise @var{addr} is the offset
33496 of @var{basereg} for memory collecting.
33497 @item len @tab 8 @tab length of memory for collecting
33498 @item basereg @tab 4 @tab the register number containing the starting
33499 memory address for collecting.
33500 @item @emph{tracepoint action for collecting registers} @tab @tab
33501 @item 'R' @tab 1 @tab type of tracepoint action
33502 @item @emph{tracepoint action for collecting static trace data} @tab @tab
33503 @item 'L' @tab 1 @tab type of tracepoint action
33504 @item @emph{tracepoint action for expression evaluation} @tab @tab
33505 @item 'X' @tab 1 @tab type of tracepoint action
33506 @item agent expression @tab length of @tab @ref{agent expression object}
33507 @item @emph{tracepoint object} @tab @tab
33508 @item number @tab 4 @tab number of tracepoint
33509 @item address @tab 8 @tab address of tracepoint inserted on
33510 @item type @tab 4 @tab type of tracepoint
33511 @item enabled @tab 1 @tab enable or disable of tracepoint
33512 @item step_count @tab 8 @tab step
33513 @item pass_count @tab 8 @tab pass
33514 @item numactions @tab 4 @tab number of tracepoint actions
33515 @item hit count @tab 8 @tab hit count
33516 @item trace frame usage @tab 8 @tab trace frame usage
33517 @item compiled_cond @tab 8 @tab compiled condition
33518 @item orig_size @tab 8 @tab orig size
33519 @item condition @tab 4 if condition is NULL otherwise length of
33520 @ref{agent expression object}
33521 @tab zero if condition is NULL, otherwise is
33522 @ref{agent expression object}
33523 @item actions @tab variable
33524 @tab numactions number of @ref{tracepoint action object}
33527 @node IPA Protocol Commands
33528 @subsection IPA Protocol Commands
33529 @cindex ipa protocol commands
33531 The spaces in each command are delimiters to ease reading this commands
33532 specification. They don't exist in real commands.
33536 @item FastTrace:@var{tracepoint_object} @var{gdb_jump_pad_head}
33537 Installs a new fast tracepoint described by @var{tracepoint_object}
33538 (@pxref{tracepoint object}). The @var{gdb_jump_pad_head}, 8-byte long, is the
33539 head of @dfn{jumppad}, which is used to jump to data collection routine
33544 @item OK @var{target_address} @var{gdb_jump_pad_head} @var{fjump_size} @var{fjump}
33545 @var{target_address} is address of tracepoint in the inferior.
33546 The @var{gdb_jump_pad_head} is updated head of jumppad. Both of
33547 @var{target_address} and @var{gdb_jump_pad_head} are 8-byte long.
33548 The @var{fjump} contains a sequence of instructions jump to jumppad entry.
33549 The @var{fjump_size}, 4-byte long, is the size of @var{fjump}.
33556 Closes the in-process agent. This command is sent when @value{GDBN} or GDBserver
33557 is about to kill inferiors.
33565 @item probe_marker_at:@var{address}
33566 Asks in-process agent to probe the marker at @var{address}.
33573 @item unprobe_marker_at:@var{address}
33574 Asks in-process agent to unprobe the marker at @var{address}.
33578 @chapter Reporting Bugs in @value{GDBN}
33579 @cindex bugs in @value{GDBN}
33580 @cindex reporting bugs in @value{GDBN}
33582 Your bug reports play an essential role in making @value{GDBN} reliable.
33584 Reporting a bug may help you by bringing a solution to your problem, or it
33585 may not. But in any case the principal function of a bug report is to help
33586 the entire community by making the next version of @value{GDBN} work better. Bug
33587 reports are your contribution to the maintenance of @value{GDBN}.
33589 In order for a bug report to serve its purpose, you must include the
33590 information that enables us to fix the bug.
33593 * Bug Criteria:: Have you found a bug?
33594 * Bug Reporting:: How to report bugs
33598 @section Have You Found a Bug?
33599 @cindex bug criteria
33601 If you are not sure whether you have found a bug, here are some guidelines:
33604 @cindex fatal signal
33605 @cindex debugger crash
33606 @cindex crash of debugger
33608 If the debugger gets a fatal signal, for any input whatever, that is a
33609 @value{GDBN} bug. Reliable debuggers never crash.
33611 @cindex error on valid input
33613 If @value{GDBN} produces an error message for valid input, that is a
33614 bug. (Note that if you're cross debugging, the problem may also be
33615 somewhere in the connection to the target.)
33617 @cindex invalid input
33619 If @value{GDBN} does not produce an error message for invalid input,
33620 that is a bug. However, you should note that your idea of
33621 ``invalid input'' might be our idea of ``an extension'' or ``support
33622 for traditional practice''.
33625 If you are an experienced user of debugging tools, your suggestions
33626 for improvement of @value{GDBN} are welcome in any case.
33629 @node Bug Reporting
33630 @section How to Report Bugs
33631 @cindex bug reports
33632 @cindex @value{GDBN} bugs, reporting
33634 A number of companies and individuals offer support for @sc{gnu} products.
33635 If you obtained @value{GDBN} from a support organization, we recommend you
33636 contact that organization first.
33638 You can find contact information for many support companies and
33639 individuals in the file @file{etc/SERVICE} in the @sc{gnu} Emacs
33641 @c should add a web page ref...
33644 @ifset BUGURL_DEFAULT
33645 In any event, we also recommend that you submit bug reports for
33646 @value{GDBN}. The preferred method is to submit them directly using
33647 @uref{http://www.gnu.org/software/gdb/bugs/, @value{GDBN}'s Bugs web
33648 page}. Alternatively, the @email{bug-gdb@@gnu.org, e-mail gateway} can
33651 @strong{Do not send bug reports to @samp{info-gdb}, or to
33652 @samp{help-gdb}, or to any newsgroups.} Most users of @value{GDBN} do
33653 not want to receive bug reports. Those that do have arranged to receive
33656 The mailing list @samp{bug-gdb} has a newsgroup @samp{gnu.gdb.bug} which
33657 serves as a repeater. The mailing list and the newsgroup carry exactly
33658 the same messages. Often people think of posting bug reports to the
33659 newsgroup instead of mailing them. This appears to work, but it has one
33660 problem which can be crucial: a newsgroup posting often lacks a mail
33661 path back to the sender. Thus, if we need to ask for more information,
33662 we may be unable to reach you. For this reason, it is better to send
33663 bug reports to the mailing list.
33665 @ifclear BUGURL_DEFAULT
33666 In any event, we also recommend that you submit bug reports for
33667 @value{GDBN} to @value{BUGURL}.
33671 The fundamental principle of reporting bugs usefully is this:
33672 @strong{report all the facts}. If you are not sure whether to state a
33673 fact or leave it out, state it!
33675 Often people omit facts because they think they know what causes the
33676 problem and assume that some details do not matter. Thus, you might
33677 assume that the name of the variable you use in an example does not matter.
33678 Well, probably it does not, but one cannot be sure. Perhaps the bug is a
33679 stray memory reference which happens to fetch from the location where that
33680 name is stored in memory; perhaps, if the name were different, the contents
33681 of that location would fool the debugger into doing the right thing despite
33682 the bug. Play it safe and give a specific, complete example. That is the
33683 easiest thing for you to do, and the most helpful.
33685 Keep in mind that the purpose of a bug report is to enable us to fix the
33686 bug. It may be that the bug has been reported previously, but neither
33687 you nor we can know that unless your bug report is complete and
33690 Sometimes people give a few sketchy facts and ask, ``Does this ring a
33691 bell?'' Those bug reports are useless, and we urge everyone to
33692 @emph{refuse to respond to them} except to chide the sender to report
33695 To enable us to fix the bug, you should include all these things:
33699 The version of @value{GDBN}. @value{GDBN} announces it if you start
33700 with no arguments; you can also print it at any time using @code{show
33703 Without this, we will not know whether there is any point in looking for
33704 the bug in the current version of @value{GDBN}.
33707 The type of machine you are using, and the operating system name and
33711 The details of the @value{GDBN} build-time configuration.
33712 @value{GDBN} shows these details if you invoke it with the
33713 @option{--configuration} command-line option, or if you type
33714 @code{show configuration} at @value{GDBN}'s prompt.
33717 What compiler (and its version) was used to compile @value{GDBN}---e.g.@:
33718 ``@value{GCC}--2.8.1''.
33721 What compiler (and its version) was used to compile the program you are
33722 debugging---e.g.@: ``@value{GCC}--2.8.1'', or ``HP92453-01 A.10.32.03 HP
33723 C Compiler''. For @value{NGCC}, you can say @kbd{@value{GCC} --version}
33724 to get this information; for other compilers, see the documentation for
33728 The command arguments you gave the compiler to compile your example and
33729 observe the bug. For example, did you use @samp{-O}? To guarantee
33730 you will not omit something important, list them all. A copy of the
33731 Makefile (or the output from make) is sufficient.
33733 If we were to try to guess the arguments, we would probably guess wrong
33734 and then we might not encounter the bug.
33737 A complete input script, and all necessary source files, that will
33741 A description of what behavior you observe that you believe is
33742 incorrect. For example, ``It gets a fatal signal.''
33744 Of course, if the bug is that @value{GDBN} gets a fatal signal, then we
33745 will certainly notice it. But if the bug is incorrect output, we might
33746 not notice unless it is glaringly wrong. You might as well not give us
33747 a chance to make a mistake.
33749 Even if the problem you experience is a fatal signal, you should still
33750 say so explicitly. Suppose something strange is going on, such as, your
33751 copy of @value{GDBN} is out of synch, or you have encountered a bug in
33752 the C library on your system. (This has happened!) Your copy might
33753 crash and ours would not. If you told us to expect a crash, then when
33754 ours fails to crash, we would know that the bug was not happening for
33755 us. If you had not told us to expect a crash, then we would not be able
33756 to draw any conclusion from our observations.
33759 @cindex recording a session script
33760 To collect all this information, you can use a session recording program
33761 such as @command{script}, which is available on many Unix systems.
33762 Just run your @value{GDBN} session inside @command{script} and then
33763 include the @file{typescript} file with your bug report.
33765 Another way to record a @value{GDBN} session is to run @value{GDBN}
33766 inside Emacs and then save the entire buffer to a file.
33769 If you wish to suggest changes to the @value{GDBN} source, send us context
33770 diffs. If you even discuss something in the @value{GDBN} source, refer to
33771 it by context, not by line number.
33773 The line numbers in our development sources will not match those in your
33774 sources. Your line numbers would convey no useful information to us.
33778 Here are some things that are not necessary:
33782 A description of the envelope of the bug.
33784 Often people who encounter a bug spend a lot of time investigating
33785 which changes to the input file will make the bug go away and which
33786 changes will not affect it.
33788 This is often time consuming and not very useful, because the way we
33789 will find the bug is by running a single example under the debugger
33790 with breakpoints, not by pure deduction from a series of examples.
33791 We recommend that you save your time for something else.
33793 Of course, if you can find a simpler example to report @emph{instead}
33794 of the original one, that is a convenience for us. Errors in the
33795 output will be easier to spot, running under the debugger will take
33796 less time, and so on.
33798 However, simplification is not vital; if you do not want to do this,
33799 report the bug anyway and send us the entire test case you used.
33802 A patch for the bug.
33804 A patch for the bug does help us if it is a good one. But do not omit
33805 the necessary information, such as the test case, on the assumption that
33806 a patch is all we need. We might see problems with your patch and decide
33807 to fix the problem another way, or we might not understand it at all.
33809 Sometimes with a program as complicated as @value{GDBN} it is very hard to
33810 construct an example that will make the program follow a certain path
33811 through the code. If you do not send us the example, we will not be able
33812 to construct one, so we will not be able to verify that the bug is fixed.
33814 And if we cannot understand what bug you are trying to fix, or why your
33815 patch should be an improvement, we will not install it. A test case will
33816 help us to understand.
33819 A guess about what the bug is or what it depends on.
33821 Such guesses are usually wrong. Even we cannot guess right about such
33822 things without first using the debugger to find the facts.
33825 @c The readline documentation is distributed with the readline code
33826 @c and consists of the two following files:
33829 @c Use -I with makeinfo to point to the appropriate directory,
33830 @c environment var TEXINPUTS with TeX.
33831 @ifclear SYSTEM_READLINE
33832 @include rluser.texi
33833 @include hsuser.texi
33837 @appendix In Memoriam
33839 The @value{GDBN} project mourns the loss of the following long-time
33844 Fred was a long-standing contributor to @value{GDBN} (1991-2006), and
33845 to Free Software in general. Outside of @value{GDBN}, he was known in
33846 the Amiga world for his series of Fish Disks, and the GeekGadget project.
33848 @item Michael Snyder
33849 Michael was one of the Global Maintainers of the @value{GDBN} project,
33850 with contributions recorded as early as 1996, until 2011. In addition
33851 to his day to day participation, he was a large driving force behind
33852 adding Reverse Debugging to @value{GDBN}.
33855 Beyond their technical contributions to the project, they were also
33856 enjoyable members of the Free Software Community. We will miss them.
33858 @node Formatting Documentation
33859 @appendix Formatting Documentation
33861 @cindex @value{GDBN} reference card
33862 @cindex reference card
33863 The @value{GDBN} 4 release includes an already-formatted reference card, ready
33864 for printing with PostScript or Ghostscript, in the @file{gdb}
33865 subdirectory of the main source directory@footnote{In
33866 @file{gdb-@value{GDBVN}/gdb/refcard.ps} of the version @value{GDBVN}
33867 release.}. If you can use PostScript or Ghostscript with your printer,
33868 you can print the reference card immediately with @file{refcard.ps}.
33870 The release also includes the source for the reference card. You
33871 can format it, using @TeX{}, by typing:
33877 The @value{GDBN} reference card is designed to print in @dfn{landscape}
33878 mode on US ``letter'' size paper;
33879 that is, on a sheet 11 inches wide by 8.5 inches
33880 high. You will need to specify this form of printing as an option to
33881 your @sc{dvi} output program.
33883 @cindex documentation
33885 All the documentation for @value{GDBN} comes as part of the machine-readable
33886 distribution. The documentation is written in Texinfo format, which is
33887 a documentation system that uses a single source file to produce both
33888 on-line information and a printed manual. You can use one of the Info
33889 formatting commands to create the on-line version of the documentation
33890 and @TeX{} (or @code{texi2roff}) to typeset the printed version.
33892 @value{GDBN} includes an already formatted copy of the on-line Info
33893 version of this manual in the @file{gdb} subdirectory. The main Info
33894 file is @file{gdb-@value{GDBVN}/gdb/gdb.info}, and it refers to
33895 subordinate files matching @samp{gdb.info*} in the same directory. If
33896 necessary, you can print out these files, or read them with any editor;
33897 but they are easier to read using the @code{info} subsystem in @sc{gnu}
33898 Emacs or the standalone @code{info} program, available as part of the
33899 @sc{gnu} Texinfo distribution.
33901 If you want to format these Info files yourself, you need one of the
33902 Info formatting programs, such as @code{texinfo-format-buffer} or
33905 If you have @code{makeinfo} installed, and are in the top level
33906 @value{GDBN} source directory (@file{gdb-@value{GDBVN}}, in the case of
33907 version @value{GDBVN}), you can make the Info file by typing:
33914 If you want to typeset and print copies of this manual, you need @TeX{},
33915 a program to print its @sc{dvi} output files, and @file{texinfo.tex}, the
33916 Texinfo definitions file.
33918 @TeX{} is a typesetting program; it does not print files directly, but
33919 produces output files called @sc{dvi} files. To print a typeset
33920 document, you need a program to print @sc{dvi} files. If your system
33921 has @TeX{} installed, chances are it has such a program. The precise
33922 command to use depends on your system; @kbd{lpr -d} is common; another
33923 (for PostScript devices) is @kbd{dvips}. The @sc{dvi} print command may
33924 require a file name without any extension or a @samp{.dvi} extension.
33926 @TeX{} also requires a macro definitions file called
33927 @file{texinfo.tex}. This file tells @TeX{} how to typeset a document
33928 written in Texinfo format. On its own, @TeX{} cannot either read or
33929 typeset a Texinfo file. @file{texinfo.tex} is distributed with GDB
33930 and is located in the @file{gdb-@var{version-number}/texinfo}
33933 If you have @TeX{} and a @sc{dvi} printer program installed, you can
33934 typeset and print this manual. First switch to the @file{gdb}
33935 subdirectory of the main source directory (for example, to
33936 @file{gdb-@value{GDBVN}/gdb}) and type:
33942 Then give @file{gdb.dvi} to your @sc{dvi} printing program.
33944 @node Installing GDB
33945 @appendix Installing @value{GDBN}
33946 @cindex installation
33949 * Requirements:: Requirements for building @value{GDBN}
33950 * Running Configure:: Invoking the @value{GDBN} @file{configure} script
33951 * Separate Objdir:: Compiling @value{GDBN} in another directory
33952 * Config Names:: Specifying names for hosts and targets
33953 * Configure Options:: Summary of options for configure
33954 * System-wide configuration:: Having a system-wide init file
33958 @section Requirements for Building @value{GDBN}
33959 @cindex building @value{GDBN}, requirements for
33961 Building @value{GDBN} requires various tools and packages to be available.
33962 Other packages will be used only if they are found.
33964 @heading Tools/Packages Necessary for Building @value{GDBN}
33966 @item ISO C90 compiler
33967 @value{GDBN} is written in ISO C90. It should be buildable with any
33968 working C90 compiler, e.g.@: GCC.
33972 @heading Tools/Packages Optional for Building @value{GDBN}
33976 @value{GDBN} can use the Expat XML parsing library. This library may be
33977 included with your operating system distribution; if it is not, you
33978 can get the latest version from @url{http://expat.sourceforge.net}.
33979 The @file{configure} script will search for this library in several
33980 standard locations; if it is installed in an unusual path, you can
33981 use the @option{--with-libexpat-prefix} option to specify its location.
33987 Remote protocol memory maps (@pxref{Memory Map Format})
33989 Target descriptions (@pxref{Target Descriptions})
33991 Remote shared library lists (@xref{Library List Format},
33992 or alternatively @pxref{Library List Format for SVR4 Targets})
33994 MS-Windows shared libraries (@pxref{Shared Libraries})
33996 Traceframe info (@pxref{Traceframe Info Format})
33998 Branch trace (@pxref{Branch Trace Format},
33999 @pxref{Branch Trace Configuration Format})
34003 @cindex compressed debug sections
34004 @value{GDBN} will use the @samp{zlib} library, if available, to read
34005 compressed debug sections. Some linkers, such as GNU gold, are capable
34006 of producing binaries with compressed debug sections. If @value{GDBN}
34007 is compiled with @samp{zlib}, it will be able to read the debug
34008 information in such binaries.
34010 The @samp{zlib} library is likely included with your operating system
34011 distribution; if it is not, you can get the latest version from
34012 @url{http://zlib.net}.
34015 @value{GDBN}'s features related to character sets (@pxref{Character
34016 Sets}) require a functioning @code{iconv} implementation. If you are
34017 on a GNU system, then this is provided by the GNU C Library. Some
34018 other systems also provide a working @code{iconv}.
34020 If @value{GDBN} is using the @code{iconv} program which is installed
34021 in a non-standard place, you will need to tell @value{GDBN} where to find it.
34022 This is done with @option{--with-iconv-bin} which specifies the
34023 directory that contains the @code{iconv} program.
34025 On systems without @code{iconv}, you can install GNU Libiconv. If you
34026 have previously installed Libiconv, you can use the
34027 @option{--with-libiconv-prefix} option to configure.
34029 @value{GDBN}'s top-level @file{configure} and @file{Makefile} will
34030 arrange to build Libiconv if a directory named @file{libiconv} appears
34031 in the top-most source directory. If Libiconv is built this way, and
34032 if the operating system does not provide a suitable @code{iconv}
34033 implementation, then the just-built library will automatically be used
34034 by @value{GDBN}. One easy way to set this up is to download GNU
34035 Libiconv, unpack it, and then rename the directory holding the
34036 Libiconv source code to @samp{libiconv}.
34039 @node Running Configure
34040 @section Invoking the @value{GDBN} @file{configure} Script
34041 @cindex configuring @value{GDBN}
34042 @value{GDBN} comes with a @file{configure} script that automates the process
34043 of preparing @value{GDBN} for installation; you can then use @code{make} to
34044 build the @code{gdb} program.
34046 @c irrelevant in info file; it's as current as the code it lives with.
34047 @footnote{If you have a more recent version of @value{GDBN} than @value{GDBVN},
34048 look at the @file{README} file in the sources; we may have improved the
34049 installation procedures since publishing this manual.}
34052 The @value{GDBN} distribution includes all the source code you need for
34053 @value{GDBN} in a single directory, whose name is usually composed by
34054 appending the version number to @samp{gdb}.
34056 For example, the @value{GDBN} version @value{GDBVN} distribution is in the
34057 @file{gdb-@value{GDBVN}} directory. That directory contains:
34060 @item gdb-@value{GDBVN}/configure @r{(and supporting files)}
34061 script for configuring @value{GDBN} and all its supporting libraries
34063 @item gdb-@value{GDBVN}/gdb
34064 the source specific to @value{GDBN} itself
34066 @item gdb-@value{GDBVN}/bfd
34067 source for the Binary File Descriptor library
34069 @item gdb-@value{GDBVN}/include
34070 @sc{gnu} include files
34072 @item gdb-@value{GDBVN}/libiberty
34073 source for the @samp{-liberty} free software library
34075 @item gdb-@value{GDBVN}/opcodes
34076 source for the library of opcode tables and disassemblers
34078 @item gdb-@value{GDBVN}/readline
34079 source for the @sc{gnu} command-line interface
34081 @item gdb-@value{GDBVN}/glob
34082 source for the @sc{gnu} filename pattern-matching subroutine
34084 @item gdb-@value{GDBVN}/mmalloc
34085 source for the @sc{gnu} memory-mapped malloc package
34088 The simplest way to configure and build @value{GDBN} is to run @file{configure}
34089 from the @file{gdb-@var{version-number}} source directory, which in
34090 this example is the @file{gdb-@value{GDBVN}} directory.
34092 First switch to the @file{gdb-@var{version-number}} source directory
34093 if you are not already in it; then run @file{configure}. Pass the
34094 identifier for the platform on which @value{GDBN} will run as an
34100 cd gdb-@value{GDBVN}
34101 ./configure @var{host}
34106 where @var{host} is an identifier such as @samp{sun4} or
34107 @samp{decstation}, that identifies the platform where @value{GDBN} will run.
34108 (You can often leave off @var{host}; @file{configure} tries to guess the
34109 correct value by examining your system.)
34111 Running @samp{configure @var{host}} and then running @code{make} builds the
34112 @file{bfd}, @file{readline}, @file{mmalloc}, and @file{libiberty}
34113 libraries, then @code{gdb} itself. The configured source files, and the
34114 binaries, are left in the corresponding source directories.
34117 @file{configure} is a Bourne-shell (@code{/bin/sh}) script; if your
34118 system does not recognize this automatically when you run a different
34119 shell, you may need to run @code{sh} on it explicitly:
34122 sh configure @var{host}
34125 If you run @file{configure} from a directory that contains source
34126 directories for multiple libraries or programs, such as the
34127 @file{gdb-@value{GDBVN}} source directory for version @value{GDBVN},
34129 creates configuration files for every directory level underneath (unless
34130 you tell it not to, with the @samp{--norecursion} option).
34132 You should run the @file{configure} script from the top directory in the
34133 source tree, the @file{gdb-@var{version-number}} directory. If you run
34134 @file{configure} from one of the subdirectories, you will configure only
34135 that subdirectory. That is usually not what you want. In particular,
34136 if you run the first @file{configure} from the @file{gdb} subdirectory
34137 of the @file{gdb-@var{version-number}} directory, you will omit the
34138 configuration of @file{bfd}, @file{readline}, and other sibling
34139 directories of the @file{gdb} subdirectory. This leads to build errors
34140 about missing include files such as @file{bfd/bfd.h}.
34142 You can install @code{@value{GDBP}} anywhere; it has no hardwired paths.
34143 However, you should make sure that the shell on your path (named by
34144 the @samp{SHELL} environment variable) is publicly readable. Remember
34145 that @value{GDBN} uses the shell to start your program---some systems refuse to
34146 let @value{GDBN} debug child processes whose programs are not readable.
34148 @node Separate Objdir
34149 @section Compiling @value{GDBN} in Another Directory
34151 If you want to run @value{GDBN} versions for several host or target machines,
34152 you need a different @code{gdb} compiled for each combination of
34153 host and target. @file{configure} is designed to make this easy by
34154 allowing you to generate each configuration in a separate subdirectory,
34155 rather than in the source directory. If your @code{make} program
34156 handles the @samp{VPATH} feature (@sc{gnu} @code{make} does), running
34157 @code{make} in each of these directories builds the @code{gdb}
34158 program specified there.
34160 To build @code{gdb} in a separate directory, run @file{configure}
34161 with the @samp{--srcdir} option to specify where to find the source.
34162 (You also need to specify a path to find @file{configure}
34163 itself from your working directory. If the path to @file{configure}
34164 would be the same as the argument to @samp{--srcdir}, you can leave out
34165 the @samp{--srcdir} option; it is assumed.)
34167 For example, with version @value{GDBVN}, you can build @value{GDBN} in a
34168 separate directory for a Sun 4 like this:
34172 cd gdb-@value{GDBVN}
34175 ../gdb-@value{GDBVN}/configure sun4
34180 When @file{configure} builds a configuration using a remote source
34181 directory, it creates a tree for the binaries with the same structure
34182 (and using the same names) as the tree under the source directory. In
34183 the example, you'd find the Sun 4 library @file{libiberty.a} in the
34184 directory @file{gdb-sun4/libiberty}, and @value{GDBN} itself in
34185 @file{gdb-sun4/gdb}.
34187 Make sure that your path to the @file{configure} script has just one
34188 instance of @file{gdb} in it. If your path to @file{configure} looks
34189 like @file{../gdb-@value{GDBVN}/gdb/configure}, you are configuring only
34190 one subdirectory of @value{GDBN}, not the whole package. This leads to
34191 build errors about missing include files such as @file{bfd/bfd.h}.
34193 One popular reason to build several @value{GDBN} configurations in separate
34194 directories is to configure @value{GDBN} for cross-compiling (where
34195 @value{GDBN} runs on one machine---the @dfn{host}---while debugging
34196 programs that run on another machine---the @dfn{target}).
34197 You specify a cross-debugging target by
34198 giving the @samp{--target=@var{target}} option to @file{configure}.
34200 When you run @code{make} to build a program or library, you must run
34201 it in a configured directory---whatever directory you were in when you
34202 called @file{configure} (or one of its subdirectories).
34204 The @code{Makefile} that @file{configure} generates in each source
34205 directory also runs recursively. If you type @code{make} in a source
34206 directory such as @file{gdb-@value{GDBVN}} (or in a separate configured
34207 directory configured with @samp{--srcdir=@var{dirname}/gdb-@value{GDBVN}}), you
34208 will build all the required libraries, and then build GDB.
34210 When you have multiple hosts or targets configured in separate
34211 directories, you can run @code{make} on them in parallel (for example,
34212 if they are NFS-mounted on each of the hosts); they will not interfere
34216 @section Specifying Names for Hosts and Targets
34218 The specifications used for hosts and targets in the @file{configure}
34219 script are based on a three-part naming scheme, but some short predefined
34220 aliases are also supported. The full naming scheme encodes three pieces
34221 of information in the following pattern:
34224 @var{architecture}-@var{vendor}-@var{os}
34227 For example, you can use the alias @code{sun4} as a @var{host} argument,
34228 or as the value for @var{target} in a @code{--target=@var{target}}
34229 option. The equivalent full name is @samp{sparc-sun-sunos4}.
34231 The @file{configure} script accompanying @value{GDBN} does not provide
34232 any query facility to list all supported host and target names or
34233 aliases. @file{configure} calls the Bourne shell script
34234 @code{config.sub} to map abbreviations to full names; you can read the
34235 script, if you wish, or you can use it to test your guesses on
34236 abbreviations---for example:
34239 % sh config.sub i386-linux
34241 % sh config.sub alpha-linux
34242 alpha-unknown-linux-gnu
34243 % sh config.sub hp9k700
34245 % sh config.sub sun4
34246 sparc-sun-sunos4.1.1
34247 % sh config.sub sun3
34248 m68k-sun-sunos4.1.1
34249 % sh config.sub i986v
34250 Invalid configuration `i986v': machine `i986v' not recognized
34254 @code{config.sub} is also distributed in the @value{GDBN} source
34255 directory (@file{gdb-@value{GDBVN}}, for version @value{GDBVN}).
34257 @node Configure Options
34258 @section @file{configure} Options
34260 Here is a summary of the @file{configure} options and arguments that
34261 are most often useful for building @value{GDBN}. @file{configure} also has
34262 several other options not listed here. @inforef{What Configure
34263 Does,,configure.info}, for a full explanation of @file{configure}.
34266 configure @r{[}--help@r{]}
34267 @r{[}--prefix=@var{dir}@r{]}
34268 @r{[}--exec-prefix=@var{dir}@r{]}
34269 @r{[}--srcdir=@var{dirname}@r{]}
34270 @r{[}--norecursion@r{]} @r{[}--rm@r{]}
34271 @r{[}--target=@var{target}@r{]}
34276 You may introduce options with a single @samp{-} rather than
34277 @samp{--} if you prefer; but you may abbreviate option names if you use
34282 Display a quick summary of how to invoke @file{configure}.
34284 @item --prefix=@var{dir}
34285 Configure the source to install programs and files under directory
34288 @item --exec-prefix=@var{dir}
34289 Configure the source to install programs under directory
34292 @c avoid splitting the warning from the explanation:
34294 @item --srcdir=@var{dirname}
34295 @strong{Warning: using this option requires @sc{gnu} @code{make}, or another
34296 @code{make} that implements the @code{VPATH} feature.}@*
34297 Use this option to make configurations in directories separate from the
34298 @value{GDBN} source directories. Among other things, you can use this to
34299 build (or maintain) several configurations simultaneously, in separate
34300 directories. @file{configure} writes configuration-specific files in
34301 the current directory, but arranges for them to use the source in the
34302 directory @var{dirname}. @file{configure} creates directories under
34303 the working directory in parallel to the source directories below
34306 @item --norecursion
34307 Configure only the directory level where @file{configure} is executed; do not
34308 propagate configuration to subdirectories.
34310 @item --target=@var{target}
34311 Configure @value{GDBN} for cross-debugging programs running on the specified
34312 @var{target}. Without this option, @value{GDBN} is configured to debug
34313 programs that run on the same machine (@var{host}) as @value{GDBN} itself.
34315 There is no convenient way to generate a list of all available targets.
34317 @item @var{host} @dots{}
34318 Configure @value{GDBN} to run on the specified @var{host}.
34320 There is no convenient way to generate a list of all available hosts.
34323 There are many other options available as well, but they are generally
34324 needed for special purposes only.
34326 @node System-wide configuration
34327 @section System-wide configuration and settings
34328 @cindex system-wide init file
34330 @value{GDBN} can be configured to have a system-wide init file;
34331 this file will be read and executed at startup (@pxref{Startup, , What
34332 @value{GDBN} does during startup}).
34334 Here is the corresponding configure option:
34337 @item --with-system-gdbinit=@var{file}
34338 Specify that the default location of the system-wide init file is
34342 If @value{GDBN} has been configured with the option @option{--prefix=$prefix},
34343 it may be subject to relocation. Two possible cases:
34347 If the default location of this init file contains @file{$prefix},
34348 it will be subject to relocation. Suppose that the configure options
34349 are @option{--prefix=$prefix --with-system-gdbinit=$prefix/etc/gdbinit};
34350 if @value{GDBN} is moved from @file{$prefix} to @file{$install}, the system
34351 init file is looked for as @file{$install/etc/gdbinit} instead of
34352 @file{$prefix/etc/gdbinit}.
34355 By contrast, if the default location does not contain the prefix,
34356 it will not be relocated. E.g.@: if @value{GDBN} has been configured with
34357 @option{--prefix=/usr/local --with-system-gdbinit=/usr/share/gdb/gdbinit},
34358 then @value{GDBN} will always look for @file{/usr/share/gdb/gdbinit},
34359 wherever @value{GDBN} is installed.
34362 If the configured location of the system-wide init file (as given by the
34363 @option{--with-system-gdbinit} option at configure time) is in the
34364 data-directory (as specified by @option{--with-gdb-datadir} at configure
34365 time) or in one of its subdirectories, then @value{GDBN} will look for the
34366 system-wide init file in the directory specified by the
34367 @option{--data-directory} command-line option.
34368 Note that the system-wide init file is only read once, during @value{GDBN}
34369 initialization. If the data-directory is changed after @value{GDBN} has
34370 started with the @code{set data-directory} command, the file will not be
34374 * System-wide Configuration Scripts:: Installed System-wide Configuration Scripts
34377 @node System-wide Configuration Scripts
34378 @subsection Installed System-wide Configuration Scripts
34379 @cindex system-wide configuration scripts
34381 The @file{system-gdbinit} directory, located inside the data-directory
34382 (as specified by @option{--with-gdb-datadir} at configure time) contains
34383 a number of scripts which can be used as system-wide init files. To
34384 automatically source those scripts at startup, @value{GDBN} should be
34385 configured with @option{--with-system-gdbinit}. Otherwise, any user
34386 should be able to source them by hand as needed.
34388 The following scripts are currently available:
34391 @item @file{elinos.py}
34393 @cindex ELinOS system-wide configuration script
34394 This script is useful when debugging a program on an ELinOS target.
34395 It takes advantage of the environment variables defined in a standard
34396 ELinOS environment in order to determine the location of the system
34397 shared libraries, and then sets the @samp{solib-absolute-prefix}
34398 and @samp{solib-search-path} variables appropriately.
34400 @item @file{wrs-linux.py}
34401 @pindex wrs-linux.py
34402 @cindex Wind River Linux system-wide configuration script
34403 This script is useful when debugging a program on a target running
34404 Wind River Linux. It expects the @env{ENV_PREFIX} to be set to
34405 the host-side sysroot used by the target system.
34409 @node Maintenance Commands
34410 @appendix Maintenance Commands
34411 @cindex maintenance commands
34412 @cindex internal commands
34414 In addition to commands intended for @value{GDBN} users, @value{GDBN}
34415 includes a number of commands intended for @value{GDBN} developers,
34416 that are not documented elsewhere in this manual. These commands are
34417 provided here for reference. (For commands that turn on debugging
34418 messages, see @ref{Debugging Output}.)
34421 @kindex maint agent
34422 @kindex maint agent-eval
34423 @item maint agent @r{[}-at @var{location}@r{,}@r{]} @var{expression}
34424 @itemx maint agent-eval @r{[}-at @var{location}@r{,}@r{]} @var{expression}
34425 Translate the given @var{expression} into remote agent bytecodes.
34426 This command is useful for debugging the Agent Expression mechanism
34427 (@pxref{Agent Expressions}). The @samp{agent} version produces an
34428 expression useful for data collection, such as by tracepoints, while
34429 @samp{maint agent-eval} produces an expression that evaluates directly
34430 to a result. For instance, a collection expression for @code{globa +
34431 globb} will include bytecodes to record four bytes of memory at each
34432 of the addresses of @code{globa} and @code{globb}, while discarding
34433 the result of the addition, while an evaluation expression will do the
34434 addition and return the sum.
34435 If @code{-at} is given, generate remote agent bytecode for @var{location}.
34436 If not, generate remote agent bytecode for current frame PC address.
34438 @kindex maint agent-printf
34439 @item maint agent-printf @var{format},@var{expr},...
34440 Translate the given format string and list of argument expressions
34441 into remote agent bytecodes and display them as a disassembled list.
34442 This command is useful for debugging the agent version of dynamic
34443 printf (@pxref{Dynamic Printf}).
34445 @kindex maint info breakpoints
34446 @item @anchor{maint info breakpoints}maint info breakpoints
34447 Using the same format as @samp{info breakpoints}, display both the
34448 breakpoints you've set explicitly, and those @value{GDBN} is using for
34449 internal purposes. Internal breakpoints are shown with negative
34450 breakpoint numbers. The type column identifies what kind of breakpoint
34455 Normal, explicitly set breakpoint.
34458 Normal, explicitly set watchpoint.
34461 Internal breakpoint, used to handle correctly stepping through
34462 @code{longjmp} calls.
34464 @item longjmp resume
34465 Internal breakpoint at the target of a @code{longjmp}.
34468 Temporary internal breakpoint used by the @value{GDBN} @code{until} command.
34471 Temporary internal breakpoint used by the @value{GDBN} @code{finish} command.
34474 Shared library events.
34478 @kindex maint info btrace
34479 @item maint info btrace
34480 Pint information about raw branch tracing data.
34482 @kindex maint btrace packet-history
34483 @item maint btrace packet-history
34484 Print the raw branch trace packets that are used to compute the
34485 execution history for the @samp{record btrace} command. Both the
34486 information and the format in which it is printed depend on the btrace
34491 For the BTS recording format, print a list of blocks of sequential
34492 code. For each block, the following information is printed:
34496 Newer blocks have higher numbers. The oldest block has number zero.
34497 @item Lowest @samp{PC}
34498 @item Highest @samp{PC}
34502 For the Intel Processor Trace recording format, print a list of
34503 Intel Processor Trace packets. For each packet, the following
34504 information is printed:
34507 @item Packet number
34508 Newer packets have higher numbers. The oldest packet has number zero.
34510 The packet's offset in the trace stream.
34511 @item Packet opcode and payload
34515 @kindex maint btrace clear-packet-history
34516 @item maint btrace clear-packet-history
34517 Discards the cached packet history printed by the @samp{maint btrace
34518 packet-history} command. The history will be computed again when
34521 @kindex maint btrace clear
34522 @item maint btrace clear
34523 Discard the branch trace data. The data will be fetched anew and the
34524 branch trace will be recomputed when needed.
34526 This implicitly truncates the branch trace to a single branch trace
34527 buffer. When updating branch trace incrementally, the branch trace
34528 available to @value{GDBN} may be bigger than a single branch trace
34531 @kindex maint set btrace pt skip-pad
34532 @item maint set btrace pt skip-pad
34533 @kindex maint show btrace pt skip-pad
34534 @item maint show btrace pt skip-pad
34535 Control whether @value{GDBN} will skip PAD packets when computing the
34538 @kindex set displaced-stepping
34539 @kindex show displaced-stepping
34540 @cindex displaced stepping support
34541 @cindex out-of-line single-stepping
34542 @item set displaced-stepping
34543 @itemx show displaced-stepping
34544 Control whether or not @value{GDBN} will do @dfn{displaced stepping}
34545 if the target supports it. Displaced stepping is a way to single-step
34546 over breakpoints without removing them from the inferior, by executing
34547 an out-of-line copy of the instruction that was originally at the
34548 breakpoint location. It is also known as out-of-line single-stepping.
34551 @item set displaced-stepping on
34552 If the target architecture supports it, @value{GDBN} will use
34553 displaced stepping to step over breakpoints.
34555 @item set displaced-stepping off
34556 @value{GDBN} will not use displaced stepping to step over breakpoints,
34557 even if such is supported by the target architecture.
34559 @cindex non-stop mode, and @samp{set displaced-stepping}
34560 @item set displaced-stepping auto
34561 This is the default mode. @value{GDBN} will use displaced stepping
34562 only if non-stop mode is active (@pxref{Non-Stop Mode}) and the target
34563 architecture supports displaced stepping.
34566 @kindex maint check-psymtabs
34567 @item maint check-psymtabs
34568 Check the consistency of currently expanded psymtabs versus symtabs.
34569 Use this to check, for example, whether a symbol is in one but not the other.
34571 @kindex maint check-symtabs
34572 @item maint check-symtabs
34573 Check the consistency of currently expanded symtabs.
34575 @kindex maint expand-symtabs
34576 @item maint expand-symtabs [@var{regexp}]
34577 Expand symbol tables.
34578 If @var{regexp} is specified, only expand symbol tables for file
34579 names matching @var{regexp}.
34581 @kindex maint set catch-demangler-crashes
34582 @kindex maint show catch-demangler-crashes
34583 @cindex demangler crashes
34584 @item maint set catch-demangler-crashes [on|off]
34585 @itemx maint show catch-demangler-crashes
34586 Control whether @value{GDBN} should attempt to catch crashes in the
34587 symbol name demangler. The default is to attempt to catch crashes.
34588 If enabled, the first time a crash is caught, a core file is created,
34589 the offending symbol is displayed and the user is presented with the
34590 option to terminate the current session.
34592 @kindex maint cplus first_component
34593 @item maint cplus first_component @var{name}
34594 Print the first C@t{++} class/namespace component of @var{name}.
34596 @kindex maint cplus namespace
34597 @item maint cplus namespace
34598 Print the list of possible C@t{++} namespaces.
34600 @kindex maint deprecate
34601 @kindex maint undeprecate
34602 @cindex deprecated commands
34603 @item maint deprecate @var{command} @r{[}@var{replacement}@r{]}
34604 @itemx maint undeprecate @var{command}
34605 Deprecate or undeprecate the named @var{command}. Deprecated commands
34606 cause @value{GDBN} to issue a warning when you use them. The optional
34607 argument @var{replacement} says which newer command should be used in
34608 favor of the deprecated one; if it is given, @value{GDBN} will mention
34609 the replacement as part of the warning.
34611 @kindex maint dump-me
34612 @item maint dump-me
34613 @cindex @code{SIGQUIT} signal, dump core of @value{GDBN}
34614 Cause a fatal signal in the debugger and force it to dump its core.
34615 This is supported only on systems which support aborting a program
34616 with the @code{SIGQUIT} signal.
34618 @kindex maint internal-error
34619 @kindex maint internal-warning
34620 @kindex maint demangler-warning
34621 @cindex demangler crashes
34622 @item maint internal-error @r{[}@var{message-text}@r{]}
34623 @itemx maint internal-warning @r{[}@var{message-text}@r{]}
34624 @itemx maint demangler-warning @r{[}@var{message-text}@r{]}
34626 Cause @value{GDBN} to call the internal function @code{internal_error},
34627 @code{internal_warning} or @code{demangler_warning} and hence behave
34628 as though an internal problem has been detected. In addition to
34629 reporting the internal problem, these functions give the user the
34630 opportunity to either quit @value{GDBN} or (for @code{internal_error}
34631 and @code{internal_warning}) create a core file of the current
34632 @value{GDBN} session.
34634 These commands take an optional parameter @var{message-text} that is
34635 used as the text of the error or warning message.
34637 Here's an example of using @code{internal-error}:
34640 (@value{GDBP}) @kbd{maint internal-error testing, 1, 2}
34641 @dots{}/maint.c:121: internal-error: testing, 1, 2
34642 A problem internal to GDB has been detected. Further
34643 debugging may prove unreliable.
34644 Quit this debugging session? (y or n) @kbd{n}
34645 Create a core file? (y or n) @kbd{n}
34649 @cindex @value{GDBN} internal error
34650 @cindex internal errors, control of @value{GDBN} behavior
34651 @cindex demangler crashes
34653 @kindex maint set internal-error
34654 @kindex maint show internal-error
34655 @kindex maint set internal-warning
34656 @kindex maint show internal-warning
34657 @kindex maint set demangler-warning
34658 @kindex maint show demangler-warning
34659 @item maint set internal-error @var{action} [ask|yes|no]
34660 @itemx maint show internal-error @var{action}
34661 @itemx maint set internal-warning @var{action} [ask|yes|no]
34662 @itemx maint show internal-warning @var{action}
34663 @itemx maint set demangler-warning @var{action} [ask|yes|no]
34664 @itemx maint show demangler-warning @var{action}
34665 When @value{GDBN} reports an internal problem (error or warning) it
34666 gives the user the opportunity to both quit @value{GDBN} and create a
34667 core file of the current @value{GDBN} session. These commands let you
34668 override the default behaviour for each particular @var{action},
34669 described in the table below.
34673 You can specify that @value{GDBN} should always (yes) or never (no)
34674 quit. The default is to ask the user what to do.
34677 You can specify that @value{GDBN} should always (yes) or never (no)
34678 create a core file. The default is to ask the user what to do. Note
34679 that there is no @code{corefile} option for @code{demangler-warning}:
34680 demangler warnings always create a core file and this cannot be
34684 @kindex maint packet
34685 @item maint packet @var{text}
34686 If @value{GDBN} is talking to an inferior via the serial protocol,
34687 then this command sends the string @var{text} to the inferior, and
34688 displays the response packet. @value{GDBN} supplies the initial
34689 @samp{$} character, the terminating @samp{#} character, and the
34692 @kindex maint print architecture
34693 @item maint print architecture @r{[}@var{file}@r{]}
34694 Print the entire architecture configuration. The optional argument
34695 @var{file} names the file where the output goes.
34697 @kindex maint print c-tdesc
34698 @item maint print c-tdesc
34699 Print the current target description (@pxref{Target Descriptions}) as
34700 a C source file. The created source file can be used in @value{GDBN}
34701 when an XML parser is not available to parse the description.
34703 @kindex maint print dummy-frames
34704 @item maint print dummy-frames
34705 Prints the contents of @value{GDBN}'s internal dummy-frame stack.
34708 (@value{GDBP}) @kbd{b add}
34710 (@value{GDBP}) @kbd{print add(2,3)}
34711 Breakpoint 2, add (a=2, b=3) at @dots{}
34713 The program being debugged stopped while in a function called from GDB.
34715 (@value{GDBP}) @kbd{maint print dummy-frames}
34716 0xa8206d8: id=@{stack=0xbfffe734,code=0xbfffe73f,!special@}, ptid=process 9353
34720 Takes an optional file parameter.
34722 @kindex maint print registers
34723 @kindex maint print raw-registers
34724 @kindex maint print cooked-registers
34725 @kindex maint print register-groups
34726 @kindex maint print remote-registers
34727 @item maint print registers @r{[}@var{file}@r{]}
34728 @itemx maint print raw-registers @r{[}@var{file}@r{]}
34729 @itemx maint print cooked-registers @r{[}@var{file}@r{]}
34730 @itemx maint print register-groups @r{[}@var{file}@r{]}
34731 @itemx maint print remote-registers @r{[}@var{file}@r{]}
34732 Print @value{GDBN}'s internal register data structures.
34734 The command @code{maint print raw-registers} includes the contents of
34735 the raw register cache; the command @code{maint print
34736 cooked-registers} includes the (cooked) value of all registers,
34737 including registers which aren't available on the target nor visible
34738 to user; the command @code{maint print register-groups} includes the
34739 groups that each register is a member of; and the command @code{maint
34740 print remote-registers} includes the remote target's register numbers
34741 and offsets in the `G' packets.
34743 These commands take an optional parameter, a file name to which to
34744 write the information.
34746 @kindex maint print reggroups
34747 @item maint print reggroups @r{[}@var{file}@r{]}
34748 Print @value{GDBN}'s internal register group data structures. The
34749 optional argument @var{file} tells to what file to write the
34752 The register groups info looks like this:
34755 (@value{GDBP}) @kbd{maint print reggroups}
34768 This command forces @value{GDBN} to flush its internal register cache.
34770 @kindex maint print objfiles
34771 @cindex info for known object files
34772 @item maint print objfiles @r{[}@var{regexp}@r{]}
34773 Print a dump of all known object files.
34774 If @var{regexp} is specified, only print object files whose names
34775 match @var{regexp}. For each object file, this command prints its name,
34776 address in memory, and all of its psymtabs and symtabs.
34778 @kindex maint print user-registers
34779 @cindex user registers
34780 @item maint print user-registers
34781 List all currently available @dfn{user registers}. User registers
34782 typically provide alternate names for actual hardware registers. They
34783 include the four ``standard'' registers @code{$fp}, @code{$pc},
34784 @code{$sp}, and @code{$ps}. @xref{standard registers}. User
34785 registers can be used in expressions in the same way as the canonical
34786 register names, but only the latter are listed by the @code{info
34787 registers} and @code{maint print registers} commands.
34789 @kindex maint print section-scripts
34790 @cindex info for known .debug_gdb_scripts-loaded scripts
34791 @item maint print section-scripts [@var{regexp}]
34792 Print a dump of scripts specified in the @code{.debug_gdb_section} section.
34793 If @var{regexp} is specified, only print scripts loaded by object files
34794 matching @var{regexp}.
34795 For each script, this command prints its name as specified in the objfile,
34796 and the full path if known.
34797 @xref{dotdebug_gdb_scripts section}.
34799 @kindex maint print statistics
34800 @cindex bcache statistics
34801 @item maint print statistics
34802 This command prints, for each object file in the program, various data
34803 about that object file followed by the byte cache (@dfn{bcache})
34804 statistics for the object file. The objfile data includes the number
34805 of minimal, partial, full, and stabs symbols, the number of types
34806 defined by the objfile, the number of as yet unexpanded psym tables,
34807 the number of line tables and string tables, and the amount of memory
34808 used by the various tables. The bcache statistics include the counts,
34809 sizes, and counts of duplicates of all and unique objects, max,
34810 average, and median entry size, total memory used and its overhead and
34811 savings, and various measures of the hash table size and chain
34814 @kindex maint print target-stack
34815 @cindex target stack description
34816 @item maint print target-stack
34817 A @dfn{target} is an interface between the debugger and a particular
34818 kind of file or process. Targets can be stacked in @dfn{strata},
34819 so that more than one target can potentially respond to a request.
34820 In particular, memory accesses will walk down the stack of targets
34821 until they find a target that is interested in handling that particular
34824 This command prints a short description of each layer that was pushed on
34825 the @dfn{target stack}, starting from the top layer down to the bottom one.
34827 @kindex maint print type
34828 @cindex type chain of a data type
34829 @item maint print type @var{expr}
34830 Print the type chain for a type specified by @var{expr}. The argument
34831 can be either a type name or a symbol. If it is a symbol, the type of
34832 that symbol is described. The type chain produced by this command is
34833 a recursive definition of the data type as stored in @value{GDBN}'s
34834 data structures, including its flags and contained types.
34836 @kindex maint selftest
34838 Run any self tests that were compiled in to @value{GDBN}. This will
34839 print a message showing how many tests were run, and how many failed.
34841 @kindex maint set dwarf always-disassemble
34842 @kindex maint show dwarf always-disassemble
34843 @item maint set dwarf always-disassemble
34844 @item maint show dwarf always-disassemble
34845 Control the behavior of @code{info address} when using DWARF debugging
34848 The default is @code{off}, which means that @value{GDBN} should try to
34849 describe a variable's location in an easily readable format. When
34850 @code{on}, @value{GDBN} will instead display the DWARF location
34851 expression in an assembly-like format. Note that some locations are
34852 too complex for @value{GDBN} to describe simply; in this case you will
34853 always see the disassembly form.
34855 Here is an example of the resulting disassembly:
34858 (gdb) info addr argc
34859 Symbol "argc" is a complex DWARF expression:
34863 For more information on these expressions, see
34864 @uref{http://www.dwarfstd.org/, the DWARF standard}.
34866 @kindex maint set dwarf max-cache-age
34867 @kindex maint show dwarf max-cache-age
34868 @item maint set dwarf max-cache-age
34869 @itemx maint show dwarf max-cache-age
34870 Control the DWARF compilation unit cache.
34872 @cindex DWARF compilation units cache
34873 In object files with inter-compilation-unit references, such as those
34874 produced by the GCC option @samp{-feliminate-dwarf2-dups}, the DWARF
34875 reader needs to frequently refer to previously read compilation units.
34876 This setting controls how long a compilation unit will remain in the
34877 cache if it is not referenced. A higher limit means that cached
34878 compilation units will be stored in memory longer, and more total
34879 memory will be used. Setting it to zero disables caching, which will
34880 slow down @value{GDBN} startup, but reduce memory consumption.
34882 @kindex maint set profile
34883 @kindex maint show profile
34884 @cindex profiling GDB
34885 @item maint set profile
34886 @itemx maint show profile
34887 Control profiling of @value{GDBN}.
34889 Profiling will be disabled until you use the @samp{maint set profile}
34890 command to enable it. When you enable profiling, the system will begin
34891 collecting timing and execution count data; when you disable profiling or
34892 exit @value{GDBN}, the results will be written to a log file. Remember that
34893 if you use profiling, @value{GDBN} will overwrite the profiling log file
34894 (often called @file{gmon.out}). If you have a record of important profiling
34895 data in a @file{gmon.out} file, be sure to move it to a safe location.
34897 Configuring with @samp{--enable-profiling} arranges for @value{GDBN} to be
34898 compiled with the @samp{-pg} compiler option.
34900 @kindex maint set show-debug-regs
34901 @kindex maint show show-debug-regs
34902 @cindex hardware debug registers
34903 @item maint set show-debug-regs
34904 @itemx maint show show-debug-regs
34905 Control whether to show variables that mirror the hardware debug
34906 registers. Use @code{on} to enable, @code{off} to disable. If
34907 enabled, the debug registers values are shown when @value{GDBN} inserts or
34908 removes a hardware breakpoint or watchpoint, and when the inferior
34909 triggers a hardware-assisted breakpoint or watchpoint.
34911 @kindex maint set show-all-tib
34912 @kindex maint show show-all-tib
34913 @item maint set show-all-tib
34914 @itemx maint show show-all-tib
34915 Control whether to show all non zero areas within a 1k block starting
34916 at thread local base, when using the @samp{info w32 thread-information-block}
34919 @kindex maint set target-async
34920 @kindex maint show target-async
34921 @item maint set target-async
34922 @itemx maint show target-async
34923 This controls whether @value{GDBN} targets operate in synchronous or
34924 asynchronous mode (@pxref{Background Execution}). Normally the
34925 default is asynchronous, if it is available; but this can be changed
34926 to more easily debug problems occurring only in synchronous mode.
34928 @kindex maint set target-non-stop @var{mode} [on|off|auto]
34929 @kindex maint show target-non-stop
34930 @item maint set target-non-stop
34931 @itemx maint show target-non-stop
34933 This controls whether @value{GDBN} targets always operate in non-stop
34934 mode even if @code{set non-stop} is @code{off} (@pxref{Non-Stop
34935 Mode}). The default is @code{auto}, meaning non-stop mode is enabled
34936 if supported by the target.
34939 @item maint set target-non-stop auto
34940 This is the default mode. @value{GDBN} controls the target in
34941 non-stop mode if the target supports it.
34943 @item maint set target-non-stop on
34944 @value{GDBN} controls the target in non-stop mode even if the target
34945 does not indicate support.
34947 @item maint set target-non-stop off
34948 @value{GDBN} does not control the target in non-stop mode even if the
34949 target supports it.
34952 @kindex maint set per-command
34953 @kindex maint show per-command
34954 @item maint set per-command
34955 @itemx maint show per-command
34956 @cindex resources used by commands
34958 @value{GDBN} can display the resources used by each command.
34959 This is useful in debugging performance problems.
34962 @item maint set per-command space [on|off]
34963 @itemx maint show per-command space
34964 Enable or disable the printing of the memory used by GDB for each command.
34965 If enabled, @value{GDBN} will display how much memory each command
34966 took, following the command's own output.
34967 This can also be requested by invoking @value{GDBN} with the
34968 @option{--statistics} command-line switch (@pxref{Mode Options}).
34970 @item maint set per-command time [on|off]
34971 @itemx maint show per-command time
34972 Enable or disable the printing of the execution time of @value{GDBN}
34974 If enabled, @value{GDBN} will display how much time it
34975 took to execute each command, following the command's own output.
34976 Both CPU time and wallclock time are printed.
34977 Printing both is useful when trying to determine whether the cost is
34978 CPU or, e.g., disk/network latency.
34979 Note that the CPU time printed is for @value{GDBN} only, it does not include
34980 the execution time of the inferior because there's no mechanism currently
34981 to compute how much time was spent by @value{GDBN} and how much time was
34982 spent by the program been debugged.
34983 This can also be requested by invoking @value{GDBN} with the
34984 @option{--statistics} command-line switch (@pxref{Mode Options}).
34986 @item maint set per-command symtab [on|off]
34987 @itemx maint show per-command symtab
34988 Enable or disable the printing of basic symbol table statistics
34990 If enabled, @value{GDBN} will display the following information:
34994 number of symbol tables
34996 number of primary symbol tables
34998 number of blocks in the blockvector
35002 @kindex maint space
35003 @cindex memory used by commands
35004 @item maint space @var{value}
35005 An alias for @code{maint set per-command space}.
35006 A non-zero value enables it, zero disables it.
35009 @cindex time of command execution
35010 @item maint time @var{value}
35011 An alias for @code{maint set per-command time}.
35012 A non-zero value enables it, zero disables it.
35014 @kindex maint translate-address
35015 @item maint translate-address @r{[}@var{section}@r{]} @var{addr}
35016 Find the symbol stored at the location specified by the address
35017 @var{addr} and an optional section name @var{section}. If found,
35018 @value{GDBN} prints the name of the closest symbol and an offset from
35019 the symbol's location to the specified address. This is similar to
35020 the @code{info address} command (@pxref{Symbols}), except that this
35021 command also allows to find symbols in other sections.
35023 If section was not specified, the section in which the symbol was found
35024 is also printed. For dynamically linked executables, the name of
35025 executable or shared library containing the symbol is printed as well.
35029 The following command is useful for non-interactive invocations of
35030 @value{GDBN}, such as in the test suite.
35033 @item set watchdog @var{nsec}
35034 @kindex set watchdog
35035 @cindex watchdog timer
35036 @cindex timeout for commands
35037 Set the maximum number of seconds @value{GDBN} will wait for the
35038 target operation to finish. If this time expires, @value{GDBN}
35039 reports and error and the command is aborted.
35041 @item show watchdog
35042 Show the current setting of the target wait timeout.
35045 @node Remote Protocol
35046 @appendix @value{GDBN} Remote Serial Protocol
35051 * Stop Reply Packets::
35052 * General Query Packets::
35053 * Architecture-Specific Protocol Details::
35054 * Tracepoint Packets::
35055 * Host I/O Packets::
35057 * Notification Packets::
35058 * Remote Non-Stop::
35059 * Packet Acknowledgment::
35061 * File-I/O Remote Protocol Extension::
35062 * Library List Format::
35063 * Library List Format for SVR4 Targets::
35064 * Memory Map Format::
35065 * Thread List Format::
35066 * Traceframe Info Format::
35067 * Branch Trace Format::
35068 * Branch Trace Configuration Format::
35074 There may be occasions when you need to know something about the
35075 protocol---for example, if there is only one serial port to your target
35076 machine, you might want your program to do something special if it
35077 recognizes a packet meant for @value{GDBN}.
35079 In the examples below, @samp{->} and @samp{<-} are used to indicate
35080 transmitted and received data, respectively.
35082 @cindex protocol, @value{GDBN} remote serial
35083 @cindex serial protocol, @value{GDBN} remote
35084 @cindex remote serial protocol
35085 All @value{GDBN} commands and responses (other than acknowledgments
35086 and notifications, see @ref{Notification Packets}) are sent as a
35087 @var{packet}. A @var{packet} is introduced with the character
35088 @samp{$}, the actual @var{packet-data}, and the terminating character
35089 @samp{#} followed by a two-digit @var{checksum}:
35092 @code{$}@var{packet-data}@code{#}@var{checksum}
35096 @cindex checksum, for @value{GDBN} remote
35098 The two-digit @var{checksum} is computed as the modulo 256 sum of all
35099 characters between the leading @samp{$} and the trailing @samp{#} (an
35100 eight bit unsigned checksum).
35102 Implementors should note that prior to @value{GDBN} 5.0 the protocol
35103 specification also included an optional two-digit @var{sequence-id}:
35106 @code{$}@var{sequence-id}@code{:}@var{packet-data}@code{#}@var{checksum}
35109 @cindex sequence-id, for @value{GDBN} remote
35111 That @var{sequence-id} was appended to the acknowledgment. @value{GDBN}
35112 has never output @var{sequence-id}s. Stubs that handle packets added
35113 since @value{GDBN} 5.0 must not accept @var{sequence-id}.
35115 When either the host or the target machine receives a packet, the first
35116 response expected is an acknowledgment: either @samp{+} (to indicate
35117 the package was received correctly) or @samp{-} (to request
35121 -> @code{$}@var{packet-data}@code{#}@var{checksum}
35126 The @samp{+}/@samp{-} acknowledgments can be disabled
35127 once a connection is established.
35128 @xref{Packet Acknowledgment}, for details.
35130 The host (@value{GDBN}) sends @var{command}s, and the target (the
35131 debugging stub incorporated in your program) sends a @var{response}. In
35132 the case of step and continue @var{command}s, the response is only sent
35133 when the operation has completed, and the target has again stopped all
35134 threads in all attached processes. This is the default all-stop mode
35135 behavior, but the remote protocol also supports @value{GDBN}'s non-stop
35136 execution mode; see @ref{Remote Non-Stop}, for details.
35138 @var{packet-data} consists of a sequence of characters with the
35139 exception of @samp{#} and @samp{$} (see @samp{X} packet for additional
35142 @cindex remote protocol, field separator
35143 Fields within the packet should be separated using @samp{,} @samp{;} or
35144 @samp{:}. Except where otherwise noted all numbers are represented in
35145 @sc{hex} with leading zeros suppressed.
35147 Implementors should note that prior to @value{GDBN} 5.0, the character
35148 @samp{:} could not appear as the third character in a packet (as it
35149 would potentially conflict with the @var{sequence-id}).
35151 @cindex remote protocol, binary data
35152 @anchor{Binary Data}
35153 Binary data in most packets is encoded either as two hexadecimal
35154 digits per byte of binary data. This allowed the traditional remote
35155 protocol to work over connections which were only seven-bit clean.
35156 Some packets designed more recently assume an eight-bit clean
35157 connection, and use a more efficient encoding to send and receive
35160 The binary data representation uses @code{7d} (@sc{ascii} @samp{@}})
35161 as an escape character. Any escaped byte is transmitted as the escape
35162 character followed by the original character XORed with @code{0x20}.
35163 For example, the byte @code{0x7d} would be transmitted as the two
35164 bytes @code{0x7d 0x5d}. The bytes @code{0x23} (@sc{ascii} @samp{#}),
35165 @code{0x24} (@sc{ascii} @samp{$}), and @code{0x7d} (@sc{ascii}
35166 @samp{@}}) must always be escaped. Responses sent by the stub
35167 must also escape @code{0x2a} (@sc{ascii} @samp{*}), so that it
35168 is not interpreted as the start of a run-length encoded sequence
35171 Response @var{data} can be run-length encoded to save space.
35172 Run-length encoding replaces runs of identical characters with one
35173 instance of the repeated character, followed by a @samp{*} and a
35174 repeat count. The repeat count is itself sent encoded, to avoid
35175 binary characters in @var{data}: a value of @var{n} is sent as
35176 @code{@var{n}+29}. For a repeat count greater or equal to 3, this
35177 produces a printable @sc{ascii} character, e.g.@: a space (@sc{ascii}
35178 code 32) for a repeat count of 3. (This is because run-length
35179 encoding starts to win for counts 3 or more.) Thus, for example,
35180 @samp{0* } is a run-length encoding of ``0000'': the space character
35181 after @samp{*} means repeat the leading @code{0} @w{@code{32 - 29 =
35184 The printable characters @samp{#} and @samp{$} or with a numeric value
35185 greater than 126 must not be used. Runs of six repeats (@samp{#}) or
35186 seven repeats (@samp{$}) can be expanded using a repeat count of only
35187 five (@samp{"}). For example, @samp{00000000} can be encoded as
35190 The error response returned for some packets includes a two character
35191 error number. That number is not well defined.
35193 @cindex empty response, for unsupported packets
35194 For any @var{command} not supported by the stub, an empty response
35195 (@samp{$#00}) should be returned. That way it is possible to extend the
35196 protocol. A newer @value{GDBN} can tell if a packet is supported based
35199 At a minimum, a stub is required to support the @samp{g} and @samp{G}
35200 commands for register access, and the @samp{m} and @samp{M} commands
35201 for memory access. Stubs that only control single-threaded targets
35202 can implement run control with the @samp{c} (continue), and @samp{s}
35203 (step) commands. Stubs that support multi-threading targets should
35204 support the @samp{vCont} command. All other commands are optional.
35209 The following table provides a complete list of all currently defined
35210 @var{command}s and their corresponding response @var{data}.
35211 @xref{File-I/O Remote Protocol Extension}, for details about the File
35212 I/O extension of the remote protocol.
35214 Each packet's description has a template showing the packet's overall
35215 syntax, followed by an explanation of the packet's meaning. We
35216 include spaces in some of the templates for clarity; these are not
35217 part of the packet's syntax. No @value{GDBN} packet uses spaces to
35218 separate its components. For example, a template like @samp{foo
35219 @var{bar} @var{baz}} describes a packet beginning with the three ASCII
35220 bytes @samp{foo}, followed by a @var{bar}, followed directly by a
35221 @var{baz}. @value{GDBN} does not transmit a space character between the
35222 @samp{foo} and the @var{bar}, or between the @var{bar} and the
35225 @cindex @var{thread-id}, in remote protocol
35226 @anchor{thread-id syntax}
35227 Several packets and replies include a @var{thread-id} field to identify
35228 a thread. Normally these are positive numbers with a target-specific
35229 interpretation, formatted as big-endian hex strings. A @var{thread-id}
35230 can also be a literal @samp{-1} to indicate all threads, or @samp{0} to
35233 In addition, the remote protocol supports a multiprocess feature in
35234 which the @var{thread-id} syntax is extended to optionally include both
35235 process and thread ID fields, as @samp{p@var{pid}.@var{tid}}.
35236 The @var{pid} (process) and @var{tid} (thread) components each have the
35237 format described above: a positive number with target-specific
35238 interpretation formatted as a big-endian hex string, literal @samp{-1}
35239 to indicate all processes or threads (respectively), or @samp{0} to
35240 indicate an arbitrary process or thread. Specifying just a process, as
35241 @samp{p@var{pid}}, is equivalent to @samp{p@var{pid}.-1}. It is an
35242 error to specify all processes but a specific thread, such as
35243 @samp{p-1.@var{tid}}. Note that the @samp{p} prefix is @emph{not} used
35244 for those packets and replies explicitly documented to include a process
35245 ID, rather than a @var{thread-id}.
35247 The multiprocess @var{thread-id} syntax extensions are only used if both
35248 @value{GDBN} and the stub report support for the @samp{multiprocess}
35249 feature using @samp{qSupported}. @xref{multiprocess extensions}, for
35252 Note that all packet forms beginning with an upper- or lower-case
35253 letter, other than those described here, are reserved for future use.
35255 Here are the packet descriptions.
35260 @cindex @samp{!} packet
35261 @anchor{extended mode}
35262 Enable extended mode. In extended mode, the remote server is made
35263 persistent. The @samp{R} packet is used to restart the program being
35269 The remote target both supports and has enabled extended mode.
35273 @cindex @samp{?} packet
35275 Indicate the reason the target halted. The reply is the same as for
35276 step and continue. This packet has a special interpretation when the
35277 target is in non-stop mode; see @ref{Remote Non-Stop}.
35280 @xref{Stop Reply Packets}, for the reply specifications.
35282 @item A @var{arglen},@var{argnum},@var{arg},@dots{}
35283 @cindex @samp{A} packet
35284 Initialized @code{argv[]} array passed into program. @var{arglen}
35285 specifies the number of bytes in the hex encoded byte stream
35286 @var{arg}. See @code{gdbserver} for more details.
35291 The arguments were set.
35297 @cindex @samp{b} packet
35298 (Don't use this packet; its behavior is not well-defined.)
35299 Change the serial line speed to @var{baud}.
35301 JTC: @emph{When does the transport layer state change? When it's
35302 received, or after the ACK is transmitted. In either case, there are
35303 problems if the command or the acknowledgment packet is dropped.}
35305 Stan: @emph{If people really wanted to add something like this, and get
35306 it working for the first time, they ought to modify ser-unix.c to send
35307 some kind of out-of-band message to a specially-setup stub and have the
35308 switch happen "in between" packets, so that from remote protocol's point
35309 of view, nothing actually happened.}
35311 @item B @var{addr},@var{mode}
35312 @cindex @samp{B} packet
35313 Set (@var{mode} is @samp{S}) or clear (@var{mode} is @samp{C}) a
35314 breakpoint at @var{addr}.
35316 Don't use this packet. Use the @samp{Z} and @samp{z} packets instead
35317 (@pxref{insert breakpoint or watchpoint packet}).
35319 @cindex @samp{bc} packet
35322 Backward continue. Execute the target system in reverse. No parameter.
35323 @xref{Reverse Execution}, for more information.
35326 @xref{Stop Reply Packets}, for the reply specifications.
35328 @cindex @samp{bs} packet
35331 Backward single step. Execute one instruction in reverse. No parameter.
35332 @xref{Reverse Execution}, for more information.
35335 @xref{Stop Reply Packets}, for the reply specifications.
35337 @item c @r{[}@var{addr}@r{]}
35338 @cindex @samp{c} packet
35339 Continue at @var{addr}, which is the address to resume. If @var{addr}
35340 is omitted, resume at current address.
35342 This packet is deprecated for multi-threading support. @xref{vCont
35346 @xref{Stop Reply Packets}, for the reply specifications.
35348 @item C @var{sig}@r{[};@var{addr}@r{]}
35349 @cindex @samp{C} packet
35350 Continue with signal @var{sig} (hex signal number). If
35351 @samp{;@var{addr}} is omitted, resume at same address.
35353 This packet is deprecated for multi-threading support. @xref{vCont
35357 @xref{Stop Reply Packets}, for the reply specifications.
35360 @cindex @samp{d} packet
35363 Don't use this packet; instead, define a general set packet
35364 (@pxref{General Query Packets}).
35368 @cindex @samp{D} packet
35369 The first form of the packet is used to detach @value{GDBN} from the
35370 remote system. It is sent to the remote target
35371 before @value{GDBN} disconnects via the @code{detach} command.
35373 The second form, including a process ID, is used when multiprocess
35374 protocol extensions are enabled (@pxref{multiprocess extensions}), to
35375 detach only a specific process. The @var{pid} is specified as a
35376 big-endian hex string.
35386 @item F @var{RC},@var{EE},@var{CF};@var{XX}
35387 @cindex @samp{F} packet
35388 A reply from @value{GDBN} to an @samp{F} packet sent by the target.
35389 This is part of the File-I/O protocol extension. @xref{File-I/O
35390 Remote Protocol Extension}, for the specification.
35393 @anchor{read registers packet}
35394 @cindex @samp{g} packet
35395 Read general registers.
35399 @item @var{XX@dots{}}
35400 Each byte of register data is described by two hex digits. The bytes
35401 with the register are transmitted in target byte order. The size of
35402 each register and their position within the @samp{g} packet are
35403 determined by the @value{GDBN} internal gdbarch functions
35404 @code{DEPRECATED_REGISTER_RAW_SIZE} and @code{gdbarch_register_name}.
35406 When reading registers from a trace frame (@pxref{Analyze Collected
35407 Data,,Using the Collected Data}), the stub may also return a string of
35408 literal @samp{x}'s in place of the register data digits, to indicate
35409 that the corresponding register has not been collected, thus its value
35410 is unavailable. For example, for an architecture with 4 registers of
35411 4 bytes each, the following reply indicates to @value{GDBN} that
35412 registers 0 and 2 have not been collected, while registers 1 and 3
35413 have been collected, and both have zero value:
35417 <- @code{xxxxxxxx00000000xxxxxxxx00000000}
35424 @item G @var{XX@dots{}}
35425 @cindex @samp{G} packet
35426 Write general registers. @xref{read registers packet}, for a
35427 description of the @var{XX@dots{}} data.
35437 @item H @var{op} @var{thread-id}
35438 @cindex @samp{H} packet
35439 Set thread for subsequent operations (@samp{m}, @samp{M}, @samp{g},
35440 @samp{G}, et.al.). Depending on the operation to be performed, @var{op}
35441 should be @samp{c} for step and continue operations (note that this
35442 is deprecated, supporting the @samp{vCont} command is a better
35443 option), and @samp{g} for other operations. The thread designator
35444 @var{thread-id} has the format and interpretation described in
35445 @ref{thread-id syntax}.
35456 @c 'H': How restrictive (or permissive) is the thread model. If a
35457 @c thread is selected and stopped, are other threads allowed
35458 @c to continue to execute? As I mentioned above, I think the
35459 @c semantics of each command when a thread is selected must be
35460 @c described. For example:
35462 @c 'g': If the stub supports threads and a specific thread is
35463 @c selected, returns the register block from that thread;
35464 @c otherwise returns current registers.
35466 @c 'G' If the stub supports threads and a specific thread is
35467 @c selected, sets the registers of the register block of
35468 @c that thread; otherwise sets current registers.
35470 @item i @r{[}@var{addr}@r{[},@var{nnn}@r{]]}
35471 @anchor{cycle step packet}
35472 @cindex @samp{i} packet
35473 Step the remote target by a single clock cycle. If @samp{,@var{nnn}} is
35474 present, cycle step @var{nnn} cycles. If @var{addr} is present, cycle
35475 step starting at that address.
35478 @cindex @samp{I} packet
35479 Signal, then cycle step. @xref{step with signal packet}. @xref{cycle
35483 @cindex @samp{k} packet
35486 The exact effect of this packet is not specified.
35488 For a bare-metal target, it may power cycle or reset the target
35489 system. For that reason, the @samp{k} packet has no reply.
35491 For a single-process target, it may kill that process if possible.
35493 A multiple-process target may choose to kill just one process, or all
35494 that are under @value{GDBN}'s control. For more precise control, use
35495 the vKill packet (@pxref{vKill packet}).
35497 If the target system immediately closes the connection in response to
35498 @samp{k}, @value{GDBN} does not consider the lack of packet
35499 acknowledgment to be an error, and assumes the kill was successful.
35501 If connected using @kbd{target extended-remote}, and the target does
35502 not close the connection in response to a kill request, @value{GDBN}
35503 probes the target state as if a new connection was opened
35504 (@pxref{? packet}).
35506 @item m @var{addr},@var{length}
35507 @cindex @samp{m} packet
35508 Read @var{length} addressable memory units starting at address @var{addr}
35509 (@pxref{addressable memory unit}). Note that @var{addr} may not be aligned to
35510 any particular boundary.
35512 The stub need not use any particular size or alignment when gathering
35513 data from memory for the response; even if @var{addr} is word-aligned
35514 and @var{length} is a multiple of the word size, the stub is free to
35515 use byte accesses, or not. For this reason, this packet may not be
35516 suitable for accessing memory-mapped I/O devices.
35517 @cindex alignment of remote memory accesses
35518 @cindex size of remote memory accesses
35519 @cindex memory, alignment and size of remote accesses
35523 @item @var{XX@dots{}}
35524 Memory contents; each byte is transmitted as a two-digit hexadecimal number.
35525 The reply may contain fewer addressable memory units than requested if the
35526 server was able to read only part of the region of memory.
35531 @item M @var{addr},@var{length}:@var{XX@dots{}}
35532 @cindex @samp{M} packet
35533 Write @var{length} addressable memory units starting at address @var{addr}
35534 (@pxref{addressable memory unit}). The data is given by @var{XX@dots{}}; each
35535 byte is transmitted as a two-digit hexadecimal number.
35542 for an error (this includes the case where only part of the data was
35547 @cindex @samp{p} packet
35548 Read the value of register @var{n}; @var{n} is in hex.
35549 @xref{read registers packet}, for a description of how the returned
35550 register value is encoded.
35554 @item @var{XX@dots{}}
35555 the register's value
35559 Indicating an unrecognized @var{query}.
35562 @item P @var{n@dots{}}=@var{r@dots{}}
35563 @anchor{write register packet}
35564 @cindex @samp{P} packet
35565 Write register @var{n@dots{}} with value @var{r@dots{}}. The register
35566 number @var{n} is in hexadecimal, and @var{r@dots{}} contains two hex
35567 digits for each byte in the register (target byte order).
35577 @item q @var{name} @var{params}@dots{}
35578 @itemx Q @var{name} @var{params}@dots{}
35579 @cindex @samp{q} packet
35580 @cindex @samp{Q} packet
35581 General query (@samp{q}) and set (@samp{Q}). These packets are
35582 described fully in @ref{General Query Packets}.
35585 @cindex @samp{r} packet
35586 Reset the entire system.
35588 Don't use this packet; use the @samp{R} packet instead.
35591 @cindex @samp{R} packet
35592 Restart the program being debugged. The @var{XX}, while needed, is ignored.
35593 This packet is only available in extended mode (@pxref{extended mode}).
35595 The @samp{R} packet has no reply.
35597 @item s @r{[}@var{addr}@r{]}
35598 @cindex @samp{s} packet
35599 Single step, resuming at @var{addr}. If
35600 @var{addr} is omitted, resume at same address.
35602 This packet is deprecated for multi-threading support. @xref{vCont
35606 @xref{Stop Reply Packets}, for the reply specifications.
35608 @item S @var{sig}@r{[};@var{addr}@r{]}
35609 @anchor{step with signal packet}
35610 @cindex @samp{S} packet
35611 Step with signal. This is analogous to the @samp{C} packet, but
35612 requests a single-step, rather than a normal resumption of execution.
35614 This packet is deprecated for multi-threading support. @xref{vCont
35618 @xref{Stop Reply Packets}, for the reply specifications.
35620 @item t @var{addr}:@var{PP},@var{MM}
35621 @cindex @samp{t} packet
35622 Search backwards starting at address @var{addr} for a match with pattern
35623 @var{PP} and mask @var{MM}, both of which are are 4 byte long.
35624 There must be at least 3 digits in @var{addr}.
35626 @item T @var{thread-id}
35627 @cindex @samp{T} packet
35628 Find out if the thread @var{thread-id} is alive. @xref{thread-id syntax}.
35633 thread is still alive
35639 Packets starting with @samp{v} are identified by a multi-letter name,
35640 up to the first @samp{;} or @samp{?} (or the end of the packet).
35642 @item vAttach;@var{pid}
35643 @cindex @samp{vAttach} packet
35644 Attach to a new process with the specified process ID @var{pid}.
35645 The process ID is a
35646 hexadecimal integer identifying the process. In all-stop mode, all
35647 threads in the attached process are stopped; in non-stop mode, it may be
35648 attached without being stopped if that is supported by the target.
35650 @c In non-stop mode, on a successful vAttach, the stub should set the
35651 @c current thread to a thread of the newly-attached process. After
35652 @c attaching, GDB queries for the attached process's thread ID with qC.
35653 @c Also note that, from a user perspective, whether or not the
35654 @c target is stopped on attach in non-stop mode depends on whether you
35655 @c use the foreground or background version of the attach command, not
35656 @c on what vAttach does; GDB does the right thing with respect to either
35657 @c stopping or restarting threads.
35659 This packet is only available in extended mode (@pxref{extended mode}).
35665 @item @r{Any stop packet}
35666 for success in all-stop mode (@pxref{Stop Reply Packets})
35668 for success in non-stop mode (@pxref{Remote Non-Stop})
35671 @item vCont@r{[};@var{action}@r{[}:@var{thread-id}@r{]]}@dots{}
35672 @cindex @samp{vCont} packet
35673 @anchor{vCont packet}
35674 Resume the inferior, specifying different actions for each thread.
35676 For each inferior thread, the leftmost action with a matching
35677 @var{thread-id} is applied. Threads that don't match any action
35678 remain in their current state. Thread IDs are specified using the
35679 syntax described in @ref{thread-id syntax}. If multiprocess
35680 extensions (@pxref{multiprocess extensions}) are supported, actions
35681 can be specified to match all threads in a process by using the
35682 @samp{p@var{pid}.-1} form of the @var{thread-id}. An action with no
35683 @var{thread-id} matches all threads. Specifying no actions is an
35686 Currently supported actions are:
35692 Continue with signal @var{sig}. The signal @var{sig} should be two hex digits.
35696 Step with signal @var{sig}. The signal @var{sig} should be two hex digits.
35699 @item r @var{start},@var{end}
35700 Step once, and then keep stepping as long as the thread stops at
35701 addresses between @var{start} (inclusive) and @var{end} (exclusive).
35702 The remote stub reports a stop reply when either the thread goes out
35703 of the range or is stopped due to an unrelated reason, such as hitting
35704 a breakpoint. @xref{range stepping}.
35706 If the range is empty (@var{start} == @var{end}), then the action
35707 becomes equivalent to the @samp{s} action. In other words,
35708 single-step once, and report the stop (even if the stepped instruction
35709 jumps to @var{start}).
35711 (A stop reply may be sent at any point even if the PC is still within
35712 the stepping range; for example, it is valid to implement this packet
35713 in a degenerate way as a single instruction step operation.)
35717 The optional argument @var{addr} normally associated with the
35718 @samp{c}, @samp{C}, @samp{s}, and @samp{S} packets is
35719 not supported in @samp{vCont}.
35721 The @samp{t} action is only relevant in non-stop mode
35722 (@pxref{Remote Non-Stop}) and may be ignored by the stub otherwise.
35723 A stop reply should be generated for any affected thread not already stopped.
35724 When a thread is stopped by means of a @samp{t} action,
35725 the corresponding stop reply should indicate that the thread has stopped with
35726 signal @samp{0}, regardless of whether the target uses some other signal
35727 as an implementation detail.
35729 The server must ignore @samp{c}, @samp{C}, @samp{s}, @samp{S}, and
35730 @samp{r} actions for threads that are already running. Conversely,
35731 the server must ignore @samp{t} actions for threads that are already
35734 @emph{Note:} In non-stop mode, a thread is considered running until
35735 @value{GDBN} acknowleges an asynchronous stop notification for it with
35736 the @samp{vStopped} packet (@pxref{Remote Non-Stop}).
35738 The stub must support @samp{vCont} if it reports support for
35739 multiprocess extensions (@pxref{multiprocess extensions}).
35742 @xref{Stop Reply Packets}, for the reply specifications.
35745 @cindex @samp{vCont?} packet
35746 Request a list of actions supported by the @samp{vCont} packet.
35750 @item vCont@r{[};@var{action}@dots{}@r{]}
35751 The @samp{vCont} packet is supported. Each @var{action} is a supported
35752 command in the @samp{vCont} packet.
35754 The @samp{vCont} packet is not supported.
35757 @anchor{vCtrlC packet}
35759 @cindex @samp{vCtrlC} packet
35760 Interrupt remote target as if a control-C was pressed on the remote
35761 terminal. This is the equivalent to reacting to the @code{^C}
35762 (@samp{\003}, the control-C character) character in all-stop mode
35763 while the target is running, except this works in non-stop mode.
35764 @xref{interrupting remote targets}, for more info on the all-stop
35775 @item vFile:@var{operation}:@var{parameter}@dots{}
35776 @cindex @samp{vFile} packet
35777 Perform a file operation on the target system. For details,
35778 see @ref{Host I/O Packets}.
35780 @item vFlashErase:@var{addr},@var{length}
35781 @cindex @samp{vFlashErase} packet
35782 Direct the stub to erase @var{length} bytes of flash starting at
35783 @var{addr}. The region may enclose any number of flash blocks, but
35784 its start and end must fall on block boundaries, as indicated by the
35785 flash block size appearing in the memory map (@pxref{Memory Map
35786 Format}). @value{GDBN} groups flash memory programming operations
35787 together, and sends a @samp{vFlashDone} request after each group; the
35788 stub is allowed to delay erase operation until the @samp{vFlashDone}
35789 packet is received.
35799 @item vFlashWrite:@var{addr}:@var{XX@dots{}}
35800 @cindex @samp{vFlashWrite} packet
35801 Direct the stub to write data to flash address @var{addr}. The data
35802 is passed in binary form using the same encoding as for the @samp{X}
35803 packet (@pxref{Binary Data}). The memory ranges specified by
35804 @samp{vFlashWrite} packets preceding a @samp{vFlashDone} packet must
35805 not overlap, and must appear in order of increasing addresses
35806 (although @samp{vFlashErase} packets for higher addresses may already
35807 have been received; the ordering is guaranteed only between
35808 @samp{vFlashWrite} packets). If a packet writes to an address that was
35809 neither erased by a preceding @samp{vFlashErase} packet nor by some other
35810 target-specific method, the results are unpredictable.
35818 for vFlashWrite addressing non-flash memory
35824 @cindex @samp{vFlashDone} packet
35825 Indicate to the stub that flash programming operation is finished.
35826 The stub is permitted to delay or batch the effects of a group of
35827 @samp{vFlashErase} and @samp{vFlashWrite} packets until a
35828 @samp{vFlashDone} packet is received. The contents of the affected
35829 regions of flash memory are unpredictable until the @samp{vFlashDone}
35830 request is completed.
35832 @item vKill;@var{pid}
35833 @cindex @samp{vKill} packet
35834 @anchor{vKill packet}
35835 Kill the process with the specified process ID @var{pid}, which is a
35836 hexadecimal integer identifying the process. This packet is used in
35837 preference to @samp{k} when multiprocess protocol extensions are
35838 supported; see @ref{multiprocess extensions}.
35848 @item vRun;@var{filename}@r{[};@var{argument}@r{]}@dots{}
35849 @cindex @samp{vRun} packet
35850 Run the program @var{filename}, passing it each @var{argument} on its
35851 command line. The file and arguments are hex-encoded strings. If
35852 @var{filename} is an empty string, the stub may use a default program
35853 (e.g.@: the last program run). The program is created in the stopped
35856 @c FIXME: What about non-stop mode?
35858 This packet is only available in extended mode (@pxref{extended mode}).
35864 @item @r{Any stop packet}
35865 for success (@pxref{Stop Reply Packets})
35869 @cindex @samp{vStopped} packet
35870 @xref{Notification Packets}.
35872 @item X @var{addr},@var{length}:@var{XX@dots{}}
35874 @cindex @samp{X} packet
35875 Write data to memory, where the data is transmitted in binary.
35876 Memory is specified by its address @var{addr} and number of addressable memory
35877 units @var{length} (@pxref{addressable memory unit});
35878 @samp{@var{XX}@dots{}} is binary data (@pxref{Binary Data}).
35888 @item z @var{type},@var{addr},@var{kind}
35889 @itemx Z @var{type},@var{addr},@var{kind}
35890 @anchor{insert breakpoint or watchpoint packet}
35891 @cindex @samp{z} packet
35892 @cindex @samp{Z} packets
35893 Insert (@samp{Z}) or remove (@samp{z}) a @var{type} breakpoint or
35894 watchpoint starting at address @var{address} of kind @var{kind}.
35896 Each breakpoint and watchpoint packet @var{type} is documented
35899 @emph{Implementation notes: A remote target shall return an empty string
35900 for an unrecognized breakpoint or watchpoint packet @var{type}. A
35901 remote target shall support either both or neither of a given
35902 @samp{Z@var{type}@dots{}} and @samp{z@var{type}@dots{}} packet pair. To
35903 avoid potential problems with duplicate packets, the operations should
35904 be implemented in an idempotent way.}
35906 @item z0,@var{addr},@var{kind}
35907 @itemx Z0,@var{addr},@var{kind}@r{[};@var{cond_list}@dots{}@r{]}@r{[};cmds:@var{persist},@var{cmd_list}@dots{}@r{]}
35908 @cindex @samp{z0} packet
35909 @cindex @samp{Z0} packet
35910 Insert (@samp{Z0}) or remove (@samp{z0}) a software breakpoint at address
35911 @var{addr} of type @var{kind}.
35913 A software breakpoint is implemented by replacing the instruction at
35914 @var{addr} with a software breakpoint or trap instruction. The
35915 @var{kind} is target-specific and typically indicates the size of the
35916 breakpoint in bytes that should be inserted. E.g., the @sc{arm} and
35917 @sc{mips} can insert either a 2 or 4 byte breakpoint. Some
35918 architectures have additional meanings for @var{kind}
35919 (@pxref{Architecture-Specific Protocol Details}); if no
35920 architecture-specific value is being used, it should be @samp{0}.
35921 @var{kind} is hex-encoded. @var{cond_list} is an optional list of
35922 conditional expressions in bytecode form that should be evaluated on
35923 the target's side. These are the conditions that should be taken into
35924 consideration when deciding if the breakpoint trigger should be
35925 reported back to @value{GDBN}.
35927 See also the @samp{swbreak} stop reason (@pxref{swbreak stop reason})
35928 for how to best report a software breakpoint event to @value{GDBN}.
35930 The @var{cond_list} parameter is comprised of a series of expressions,
35931 concatenated without separators. Each expression has the following form:
35935 @item X @var{len},@var{expr}
35936 @var{len} is the length of the bytecode expression and @var{expr} is the
35937 actual conditional expression in bytecode form.
35941 The optional @var{cmd_list} parameter introduces commands that may be
35942 run on the target, rather than being reported back to @value{GDBN}.
35943 The parameter starts with a numeric flag @var{persist}; if the flag is
35944 nonzero, then the breakpoint may remain active and the commands
35945 continue to be run even when @value{GDBN} disconnects from the target.
35946 Following this flag is a series of expressions concatenated with no
35947 separators. Each expression has the following form:
35951 @item X @var{len},@var{expr}
35952 @var{len} is the length of the bytecode expression and @var{expr} is the
35953 actual conditional expression in bytecode form.
35957 @emph{Implementation note: It is possible for a target to copy or move
35958 code that contains software breakpoints (e.g., when implementing
35959 overlays). The behavior of this packet, in the presence of such a
35960 target, is not defined.}
35972 @item z1,@var{addr},@var{kind}
35973 @itemx Z1,@var{addr},@var{kind}@r{[};@var{cond_list}@dots{}@r{]}@r{[};cmds:@var{persist},@var{cmd_list}@dots{}@r{]}
35974 @cindex @samp{z1} packet
35975 @cindex @samp{Z1} packet
35976 Insert (@samp{Z1}) or remove (@samp{z1}) a hardware breakpoint at
35977 address @var{addr}.
35979 A hardware breakpoint is implemented using a mechanism that is not
35980 dependent on being able to modify the target's memory. The
35981 @var{kind}, @var{cond_list}, and @var{cmd_list} arguments have the
35982 same meaning as in @samp{Z0} packets.
35984 @emph{Implementation note: A hardware breakpoint is not affected by code
35997 @item z2,@var{addr},@var{kind}
35998 @itemx Z2,@var{addr},@var{kind}
35999 @cindex @samp{z2} packet
36000 @cindex @samp{Z2} packet
36001 Insert (@samp{Z2}) or remove (@samp{z2}) a write watchpoint at @var{addr}.
36002 The number of bytes to watch is specified by @var{kind}.
36014 @item z3,@var{addr},@var{kind}
36015 @itemx Z3,@var{addr},@var{kind}
36016 @cindex @samp{z3} packet
36017 @cindex @samp{Z3} packet
36018 Insert (@samp{Z3}) or remove (@samp{z3}) a read watchpoint at @var{addr}.
36019 The number of bytes to watch is specified by @var{kind}.
36031 @item z4,@var{addr},@var{kind}
36032 @itemx Z4,@var{addr},@var{kind}
36033 @cindex @samp{z4} packet
36034 @cindex @samp{Z4} packet
36035 Insert (@samp{Z4}) or remove (@samp{z4}) an access watchpoint at @var{addr}.
36036 The number of bytes to watch is specified by @var{kind}.
36050 @node Stop Reply Packets
36051 @section Stop Reply Packets
36052 @cindex stop reply packets
36054 The @samp{C}, @samp{c}, @samp{S}, @samp{s}, @samp{vCont},
36055 @samp{vAttach}, @samp{vRun}, @samp{vStopped}, and @samp{?} packets can
36056 receive any of the below as a reply. Except for @samp{?}
36057 and @samp{vStopped}, that reply is only returned
36058 when the target halts. In the below the exact meaning of @dfn{signal
36059 number} is defined by the header @file{include/gdb/signals.h} in the
36060 @value{GDBN} source code.
36062 In non-stop mode, the server will simply reply @samp{OK} to commands
36063 such as @samp{vCont}; any stop will be the subject of a future
36064 notification. @xref{Remote Non-Stop}.
36066 As in the description of request packets, we include spaces in the
36067 reply templates for clarity; these are not part of the reply packet's
36068 syntax. No @value{GDBN} stop reply packet uses spaces to separate its
36074 The program received signal number @var{AA} (a two-digit hexadecimal
36075 number). This is equivalent to a @samp{T} response with no
36076 @var{n}:@var{r} pairs.
36078 @item T @var{AA} @var{n1}:@var{r1};@var{n2}:@var{r2};@dots{}
36079 @cindex @samp{T} packet reply
36080 The program received signal number @var{AA} (a two-digit hexadecimal
36081 number). This is equivalent to an @samp{S} response, except that the
36082 @samp{@var{n}:@var{r}} pairs can carry values of important registers
36083 and other information directly in the stop reply packet, reducing
36084 round-trip latency. Single-step and breakpoint traps are reported
36085 this way. Each @samp{@var{n}:@var{r}} pair is interpreted as follows:
36089 If @var{n} is a hexadecimal number, it is a register number, and the
36090 corresponding @var{r} gives that register's value. The data @var{r} is a
36091 series of bytes in target byte order, with each byte given by a
36092 two-digit hex number.
36095 If @var{n} is @samp{thread}, then @var{r} is the @var{thread-id} of
36096 the stopped thread, as specified in @ref{thread-id syntax}.
36099 If @var{n} is @samp{core}, then @var{r} is the hexadecimal number of
36100 the core on which the stop event was detected.
36103 If @var{n} is a recognized @dfn{stop reason}, it describes a more
36104 specific event that stopped the target. The currently defined stop
36105 reasons are listed below. The @var{aa} should be @samp{05}, the trap
36106 signal. At most one stop reason should be present.
36109 Otherwise, @value{GDBN} should ignore this @samp{@var{n}:@var{r}} pair
36110 and go on to the next; this allows us to extend the protocol in the
36114 The currently defined stop reasons are:
36120 The packet indicates a watchpoint hit, and @var{r} is the data address, in
36123 @item syscall_entry
36124 @itemx syscall_return
36125 The packet indicates a syscall entry or return, and @var{r} is the
36126 syscall number, in hex.
36128 @cindex shared library events, remote reply
36130 The packet indicates that the loaded libraries have changed.
36131 @value{GDBN} should use @samp{qXfer:libraries:read} to fetch a new
36132 list of loaded libraries. The @var{r} part is ignored.
36134 @cindex replay log events, remote reply
36136 The packet indicates that the target cannot continue replaying
36137 logged execution events, because it has reached the end (or the
36138 beginning when executing backward) of the log. The value of @var{r}
36139 will be either @samp{begin} or @samp{end}. @xref{Reverse Execution},
36140 for more information.
36143 @anchor{swbreak stop reason}
36144 The packet indicates a software breakpoint instruction was executed,
36145 irrespective of whether it was @value{GDBN} that planted the
36146 breakpoint or the breakpoint is hardcoded in the program. The @var{r}
36147 part must be left empty.
36149 On some architectures, such as x86, at the architecture level, when a
36150 breakpoint instruction executes the program counter points at the
36151 breakpoint address plus an offset. On such targets, the stub is
36152 responsible for adjusting the PC to point back at the breakpoint
36155 This packet should not be sent by default; older @value{GDBN} versions
36156 did not support it. @value{GDBN} requests it, by supplying an
36157 appropriate @samp{qSupported} feature (@pxref{qSupported}). The
36158 remote stub must also supply the appropriate @samp{qSupported} feature
36159 indicating support.
36161 This packet is required for correct non-stop mode operation.
36164 The packet indicates the target stopped for a hardware breakpoint.
36165 The @var{r} part must be left empty.
36167 The same remarks about @samp{qSupported} and non-stop mode above
36170 @cindex fork events, remote reply
36172 The packet indicates that @code{fork} was called, and @var{r}
36173 is the thread ID of the new child process. Refer to
36174 @ref{thread-id syntax} for the format of the @var{thread-id}
36175 field. This packet is only applicable to targets that support
36178 This packet should not be sent by default; older @value{GDBN} versions
36179 did not support it. @value{GDBN} requests it, by supplying an
36180 appropriate @samp{qSupported} feature (@pxref{qSupported}). The
36181 remote stub must also supply the appropriate @samp{qSupported} feature
36182 indicating support.
36184 @cindex vfork events, remote reply
36186 The packet indicates that @code{vfork} was called, and @var{r}
36187 is the thread ID of the new child process. Refer to
36188 @ref{thread-id syntax} for the format of the @var{thread-id}
36189 field. This packet is only applicable to targets that support
36192 This packet should not be sent by default; older @value{GDBN} versions
36193 did not support it. @value{GDBN} requests it, by supplying an
36194 appropriate @samp{qSupported} feature (@pxref{qSupported}). The
36195 remote stub must also supply the appropriate @samp{qSupported} feature
36196 indicating support.
36198 @cindex vforkdone events, remote reply
36200 The packet indicates that a child process created by a vfork
36201 has either called @code{exec} or terminated, so that the
36202 address spaces of the parent and child process are no longer
36203 shared. The @var{r} part is ignored. This packet is only
36204 applicable to targets that support vforkdone events.
36206 This packet should not be sent by default; older @value{GDBN} versions
36207 did not support it. @value{GDBN} requests it, by supplying an
36208 appropriate @samp{qSupported} feature (@pxref{qSupported}). The
36209 remote stub must also supply the appropriate @samp{qSupported} feature
36210 indicating support.
36212 @cindex exec events, remote reply
36214 The packet indicates that @code{execve} was called, and @var{r}
36215 is the absolute pathname of the file that was executed, in hex.
36216 This packet is only applicable to targets that support exec events.
36218 This packet should not be sent by default; older @value{GDBN} versions
36219 did not support it. @value{GDBN} requests it, by supplying an
36220 appropriate @samp{qSupported} feature (@pxref{qSupported}). The
36221 remote stub must also supply the appropriate @samp{qSupported} feature
36222 indicating support.
36224 @cindex thread create event, remote reply
36225 @anchor{thread create event}
36227 The packet indicates that the thread was just created. The new thread
36228 is stopped until @value{GDBN} sets it running with a resumption packet
36229 (@pxref{vCont packet}). This packet should not be sent by default;
36230 @value{GDBN} requests it with the @ref{QThreadEvents} packet. See
36231 also the @samp{w} (@pxref{thread exit event}) remote reply below. The
36232 @var{r} part is ignored.
36237 @itemx W @var{AA} ; process:@var{pid}
36238 The process exited, and @var{AA} is the exit status. This is only
36239 applicable to certain targets.
36241 The second form of the response, including the process ID of the
36242 exited process, can be used only when @value{GDBN} has reported
36243 support for multiprocess protocol extensions; see @ref{multiprocess
36244 extensions}. Both @var{AA} and @var{pid} are formatted as big-endian
36248 @itemx X @var{AA} ; process:@var{pid}
36249 The process terminated with signal @var{AA}.
36251 The second form of the response, including the process ID of the
36252 terminated process, can be used only when @value{GDBN} has reported
36253 support for multiprocess protocol extensions; see @ref{multiprocess
36254 extensions}. Both @var{AA} and @var{pid} are formatted as big-endian
36257 @anchor{thread exit event}
36258 @cindex thread exit event, remote reply
36259 @item w @var{AA} ; @var{tid}
36261 The thread exited, and @var{AA} is the exit status. This response
36262 should not be sent by default; @value{GDBN} requests it with the
36263 @ref{QThreadEvents} packet. See also @ref{thread create event} above.
36264 @var{AA} is formatted as a big-endian hex string.
36267 There are no resumed threads left in the target. In other words, even
36268 though the process is alive, the last resumed thread has exited. For
36269 example, say the target process has two threads: thread 1 and thread
36270 2. The client leaves thread 1 stopped, and resumes thread 2, which
36271 subsequently exits. At this point, even though the process is still
36272 alive, and thus no @samp{W} stop reply is sent, no thread is actually
36273 executing either. The @samp{N} stop reply thus informs the client
36274 that it can stop waiting for stop replies. This packet should not be
36275 sent by default; older @value{GDBN} versions did not support it.
36276 @value{GDBN} requests it, by supplying an appropriate
36277 @samp{qSupported} feature (@pxref{qSupported}). The remote stub must
36278 also supply the appropriate @samp{qSupported} feature indicating
36281 @item O @var{XX}@dots{}
36282 @samp{@var{XX}@dots{}} is hex encoding of @sc{ascii} data, to be
36283 written as the program's console output. This can happen at any time
36284 while the program is running and the debugger should continue to wait
36285 for @samp{W}, @samp{T}, etc. This reply is not permitted in non-stop mode.
36287 @item F @var{call-id},@var{parameter}@dots{}
36288 @var{call-id} is the identifier which says which host system call should
36289 be called. This is just the name of the function. Translation into the
36290 correct system call is only applicable as it's defined in @value{GDBN}.
36291 @xref{File-I/O Remote Protocol Extension}, for a list of implemented
36294 @samp{@var{parameter}@dots{}} is a list of parameters as defined for
36295 this very system call.
36297 The target replies with this packet when it expects @value{GDBN} to
36298 call a host system call on behalf of the target. @value{GDBN} replies
36299 with an appropriate @samp{F} packet and keeps up waiting for the next
36300 reply packet from the target. The latest @samp{C}, @samp{c}, @samp{S}
36301 or @samp{s} action is expected to be continued. @xref{File-I/O Remote
36302 Protocol Extension}, for more details.
36306 @node General Query Packets
36307 @section General Query Packets
36308 @cindex remote query requests
36310 Packets starting with @samp{q} are @dfn{general query packets};
36311 packets starting with @samp{Q} are @dfn{general set packets}. General
36312 query and set packets are a semi-unified form for retrieving and
36313 sending information to and from the stub.
36315 The initial letter of a query or set packet is followed by a name
36316 indicating what sort of thing the packet applies to. For example,
36317 @value{GDBN} may use a @samp{qSymbol} packet to exchange symbol
36318 definitions with the stub. These packet names follow some
36323 The name must not contain commas, colons or semicolons.
36325 Most @value{GDBN} query and set packets have a leading upper case
36328 The names of custom vendor packets should use a company prefix, in
36329 lower case, followed by a period. For example, packets designed at
36330 the Acme Corporation might begin with @samp{qacme.foo} (for querying
36331 foos) or @samp{Qacme.bar} (for setting bars).
36334 The name of a query or set packet should be separated from any
36335 parameters by a @samp{:}; the parameters themselves should be
36336 separated by @samp{,} or @samp{;}. Stubs must be careful to match the
36337 full packet name, and check for a separator or the end of the packet,
36338 in case two packet names share a common prefix. New packets should not begin
36339 with @samp{qC}, @samp{qP}, or @samp{qL}@footnote{The @samp{qP} and @samp{qL}
36340 packets predate these conventions, and have arguments without any terminator
36341 for the packet name; we suspect they are in widespread use in places that
36342 are difficult to upgrade. The @samp{qC} packet has no arguments, but some
36343 existing stubs (e.g.@: RedBoot) are known to not check for the end of the
36346 Like the descriptions of the other packets, each description here
36347 has a template showing the packet's overall syntax, followed by an
36348 explanation of the packet's meaning. We include spaces in some of the
36349 templates for clarity; these are not part of the packet's syntax. No
36350 @value{GDBN} packet uses spaces to separate its components.
36352 Here are the currently defined query and set packets:
36358 Turn on or off the agent as a helper to perform some debugging operations
36359 delegated from @value{GDBN} (@pxref{Control Agent}).
36361 @item QAllow:@var{op}:@var{val}@dots{}
36362 @cindex @samp{QAllow} packet
36363 Specify which operations @value{GDBN} expects to request of the
36364 target, as a semicolon-separated list of operation name and value
36365 pairs. Possible values for @var{op} include @samp{WriteReg},
36366 @samp{WriteMem}, @samp{InsertBreak}, @samp{InsertTrace},
36367 @samp{InsertFastTrace}, and @samp{Stop}. @var{val} is either 0,
36368 indicating that @value{GDBN} will not request the operation, or 1,
36369 indicating that it may. (The target can then use this to set up its
36370 own internals optimally, for instance if the debugger never expects to
36371 insert breakpoints, it may not need to install its own trap handler.)
36374 @cindex current thread, remote request
36375 @cindex @samp{qC} packet
36376 Return the current thread ID.
36380 @item QC @var{thread-id}
36381 Where @var{thread-id} is a thread ID as documented in
36382 @ref{thread-id syntax}.
36383 @item @r{(anything else)}
36384 Any other reply implies the old thread ID.
36387 @item qCRC:@var{addr},@var{length}
36388 @cindex CRC of memory block, remote request
36389 @cindex @samp{qCRC} packet
36390 @anchor{qCRC packet}
36391 Compute the CRC checksum of a block of memory using CRC-32 defined in
36392 IEEE 802.3. The CRC is computed byte at a time, taking the most
36393 significant bit of each byte first. The initial pattern code
36394 @code{0xffffffff} is used to ensure leading zeros affect the CRC.
36396 @emph{Note:} This is the same CRC used in validating separate debug
36397 files (@pxref{Separate Debug Files, , Debugging Information in Separate
36398 Files}). However the algorithm is slightly different. When validating
36399 separate debug files, the CRC is computed taking the @emph{least}
36400 significant bit of each byte first, and the final result is inverted to
36401 detect trailing zeros.
36406 An error (such as memory fault)
36407 @item C @var{crc32}
36408 The specified memory region's checksum is @var{crc32}.
36411 @item QDisableRandomization:@var{value}
36412 @cindex disable address space randomization, remote request
36413 @cindex @samp{QDisableRandomization} packet
36414 Some target operating systems will randomize the virtual address space
36415 of the inferior process as a security feature, but provide a feature
36416 to disable such randomization, e.g.@: to allow for a more deterministic
36417 debugging experience. On such systems, this packet with a @var{value}
36418 of 1 directs the target to disable address space randomization for
36419 processes subsequently started via @samp{vRun} packets, while a packet
36420 with a @var{value} of 0 tells the target to enable address space
36423 This packet is only available in extended mode (@pxref{extended mode}).
36428 The request succeeded.
36431 An error occurred. The error number @var{nn} is given as hex digits.
36434 An empty reply indicates that @samp{QDisableRandomization} is not supported
36438 This packet is not probed by default; the remote stub must request it,
36439 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
36440 This should only be done on targets that actually support disabling
36441 address space randomization.
36444 @itemx qsThreadInfo
36445 @cindex list active threads, remote request
36446 @cindex @samp{qfThreadInfo} packet
36447 @cindex @samp{qsThreadInfo} packet
36448 Obtain a list of all active thread IDs from the target (OS). Since there
36449 may be too many active threads to fit into one reply packet, this query
36450 works iteratively: it may require more than one query/reply sequence to
36451 obtain the entire list of threads. The first query of the sequence will
36452 be the @samp{qfThreadInfo} query; subsequent queries in the
36453 sequence will be the @samp{qsThreadInfo} query.
36455 NOTE: This packet replaces the @samp{qL} query (see below).
36459 @item m @var{thread-id}
36461 @item m @var{thread-id},@var{thread-id}@dots{}
36462 a comma-separated list of thread IDs
36464 (lower case letter @samp{L}) denotes end of list.
36467 In response to each query, the target will reply with a list of one or
36468 more thread IDs, separated by commas.
36469 @value{GDBN} will respond to each reply with a request for more thread
36470 ids (using the @samp{qs} form of the query), until the target responds
36471 with @samp{l} (lower-case ell, for @dfn{last}).
36472 Refer to @ref{thread-id syntax}, for the format of the @var{thread-id}
36475 @emph{Note: @value{GDBN} will send the @code{qfThreadInfo} query during the
36476 initial connection with the remote target, and the very first thread ID
36477 mentioned in the reply will be stopped by @value{GDBN} in a subsequent
36478 message. Therefore, the stub should ensure that the first thread ID in
36479 the @code{qfThreadInfo} reply is suitable for being stopped by @value{GDBN}.}
36481 @item qGetTLSAddr:@var{thread-id},@var{offset},@var{lm}
36482 @cindex get thread-local storage address, remote request
36483 @cindex @samp{qGetTLSAddr} packet
36484 Fetch the address associated with thread local storage specified
36485 by @var{thread-id}, @var{offset}, and @var{lm}.
36487 @var{thread-id} is the thread ID associated with the
36488 thread for which to fetch the TLS address. @xref{thread-id syntax}.
36490 @var{offset} is the (big endian, hex encoded) offset associated with the
36491 thread local variable. (This offset is obtained from the debug
36492 information associated with the variable.)
36494 @var{lm} is the (big endian, hex encoded) OS/ABI-specific encoding of the
36495 load module associated with the thread local storage. For example,
36496 a @sc{gnu}/Linux system will pass the link map address of the shared
36497 object associated with the thread local storage under consideration.
36498 Other operating environments may choose to represent the load module
36499 differently, so the precise meaning of this parameter will vary.
36503 @item @var{XX}@dots{}
36504 Hex encoded (big endian) bytes representing the address of the thread
36505 local storage requested.
36508 An error occurred. The error number @var{nn} is given as hex digits.
36511 An empty reply indicates that @samp{qGetTLSAddr} is not supported by the stub.
36514 @item qGetTIBAddr:@var{thread-id}
36515 @cindex get thread information block address
36516 @cindex @samp{qGetTIBAddr} packet
36517 Fetch address of the Windows OS specific Thread Information Block.
36519 @var{thread-id} is the thread ID associated with the thread.
36523 @item @var{XX}@dots{}
36524 Hex encoded (big endian) bytes representing the linear address of the
36525 thread information block.
36528 An error occured. This means that either the thread was not found, or the
36529 address could not be retrieved.
36532 An empty reply indicates that @samp{qGetTIBAddr} is not supported by the stub.
36535 @item qL @var{startflag} @var{threadcount} @var{nextthread}
36536 Obtain thread information from RTOS. Where: @var{startflag} (one hex
36537 digit) is one to indicate the first query and zero to indicate a
36538 subsequent query; @var{threadcount} (two hex digits) is the maximum
36539 number of threads the response packet can contain; and @var{nextthread}
36540 (eight hex digits), for subsequent queries (@var{startflag} is zero), is
36541 returned in the response as @var{argthread}.
36543 Don't use this packet; use the @samp{qfThreadInfo} query instead (see above).
36547 @item qM @var{count} @var{done} @var{argthread} @var{thread}@dots{}
36548 Where: @var{count} (two hex digits) is the number of threads being
36549 returned; @var{done} (one hex digit) is zero to indicate more threads
36550 and one indicates no further threads; @var{argthreadid} (eight hex
36551 digits) is @var{nextthread} from the request packet; @var{thread}@dots{}
36552 is a sequence of thread IDs, @var{threadid} (eight hex
36553 digits), from the target. See @code{remote.c:parse_threadlist_response()}.
36557 @cindex section offsets, remote request
36558 @cindex @samp{qOffsets} packet
36559 Get section offsets that the target used when relocating the downloaded
36564 @item Text=@var{xxx};Data=@var{yyy}@r{[};Bss=@var{zzz}@r{]}
36565 Relocate the @code{Text} section by @var{xxx} from its original address.
36566 Relocate the @code{Data} section by @var{yyy} from its original address.
36567 If the object file format provides segment information (e.g.@: @sc{elf}
36568 @samp{PT_LOAD} program headers), @value{GDBN} will relocate entire
36569 segments by the supplied offsets.
36571 @emph{Note: while a @code{Bss} offset may be included in the response,
36572 @value{GDBN} ignores this and instead applies the @code{Data} offset
36573 to the @code{Bss} section.}
36575 @item TextSeg=@var{xxx}@r{[};DataSeg=@var{yyy}@r{]}
36576 Relocate the first segment of the object file, which conventionally
36577 contains program code, to a starting address of @var{xxx}. If
36578 @samp{DataSeg} is specified, relocate the second segment, which
36579 conventionally contains modifiable data, to a starting address of
36580 @var{yyy}. @value{GDBN} will report an error if the object file
36581 does not contain segment information, or does not contain at least
36582 as many segments as mentioned in the reply. Extra segments are
36583 kept at fixed offsets relative to the last relocated segment.
36586 @item qP @var{mode} @var{thread-id}
36587 @cindex thread information, remote request
36588 @cindex @samp{qP} packet
36589 Returns information on @var{thread-id}. Where: @var{mode} is a hex
36590 encoded 32 bit mode; @var{thread-id} is a thread ID
36591 (@pxref{thread-id syntax}).
36593 Don't use this packet; use the @samp{qThreadExtraInfo} query instead
36596 Reply: see @code{remote.c:remote_unpack_thread_info_response()}.
36600 @cindex non-stop mode, remote request
36601 @cindex @samp{QNonStop} packet
36603 Enter non-stop (@samp{QNonStop:1}) or all-stop (@samp{QNonStop:0}) mode.
36604 @xref{Remote Non-Stop}, for more information.
36609 The request succeeded.
36612 An error occurred. The error number @var{nn} is given as hex digits.
36615 An empty reply indicates that @samp{QNonStop} is not supported by
36619 This packet is not probed by default; the remote stub must request it,
36620 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
36621 Use of this packet is controlled by the @code{set non-stop} command;
36622 @pxref{Non-Stop Mode}.
36624 @item QCatchSyscalls:1 @r{[};@var{sysno}@r{]}@dots{}
36625 @itemx QCatchSyscalls:0
36626 @cindex catch syscalls from inferior, remote request
36627 @cindex @samp{QCatchSyscalls} packet
36628 @anchor{QCatchSyscalls}
36629 Enable (@samp{QCatchSyscalls:1}) or disable (@samp{QCatchSyscalls:0})
36630 catching syscalls from the inferior process.
36632 For @samp{QCatchSyscalls:1}, each listed syscall @var{sysno} (encoded
36633 in hex) should be reported to @value{GDBN}. If no syscall @var{sysno}
36634 is listed, every system call should be reported.
36636 Note that if a syscall not in the list is reported, @value{GDBN} will
36637 still filter the event according to its own list from all corresponding
36638 @code{catch syscall} commands. However, it is more efficient to only
36639 report the requested syscalls.
36641 Multiple @samp{QCatchSyscalls:1} packets do not combine; any earlier
36642 @samp{QCatchSyscalls:1} list is completely replaced by the new list.
36644 If the inferior process execs, the state of @samp{QCatchSyscalls} is
36645 kept for the new process too. On targets where exec may affect syscall
36646 numbers, for example with exec between 32 and 64-bit processes, the
36647 client should send a new packet with the new syscall list.
36652 The request succeeded.
36655 An error occurred. @var{nn} are hex digits.
36658 An empty reply indicates that @samp{QCatchSyscalls} is not supported by
36662 Use of this packet is controlled by the @code{set remote catch-syscalls}
36663 command (@pxref{Remote Configuration, set remote catch-syscalls}).
36664 This packet is not probed by default; the remote stub must request it,
36665 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
36667 @item QPassSignals: @var{signal} @r{[};@var{signal}@r{]}@dots{}
36668 @cindex pass signals to inferior, remote request
36669 @cindex @samp{QPassSignals} packet
36670 @anchor{QPassSignals}
36671 Each listed @var{signal} should be passed directly to the inferior process.
36672 Signals are numbered identically to continue packets and stop replies
36673 (@pxref{Stop Reply Packets}). Each @var{signal} list item should be
36674 strictly greater than the previous item. These signals do not need to stop
36675 the inferior, or be reported to @value{GDBN}. All other signals should be
36676 reported to @value{GDBN}. Multiple @samp{QPassSignals} packets do not
36677 combine; any earlier @samp{QPassSignals} list is completely replaced by the
36678 new list. This packet improves performance when using @samp{handle
36679 @var{signal} nostop noprint pass}.
36684 The request succeeded.
36687 An error occurred. The error number @var{nn} is given as hex digits.
36690 An empty reply indicates that @samp{QPassSignals} is not supported by
36694 Use of this packet is controlled by the @code{set remote pass-signals}
36695 command (@pxref{Remote Configuration, set remote pass-signals}).
36696 This packet is not probed by default; the remote stub must request it,
36697 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
36699 @item QProgramSignals: @var{signal} @r{[};@var{signal}@r{]}@dots{}
36700 @cindex signals the inferior may see, remote request
36701 @cindex @samp{QProgramSignals} packet
36702 @anchor{QProgramSignals}
36703 Each listed @var{signal} may be delivered to the inferior process.
36704 Others should be silently discarded.
36706 In some cases, the remote stub may need to decide whether to deliver a
36707 signal to the program or not without @value{GDBN} involvement. One
36708 example of that is while detaching --- the program's threads may have
36709 stopped for signals that haven't yet had a chance of being reported to
36710 @value{GDBN}, and so the remote stub can use the signal list specified
36711 by this packet to know whether to deliver or ignore those pending
36714 This does not influence whether to deliver a signal as requested by a
36715 resumption packet (@pxref{vCont packet}).
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. Multiple
36720 @samp{QProgramSignals} packets do not combine; any earlier
36721 @samp{QProgramSignals} list is completely replaced by the new list.
36726 The request succeeded.
36729 An error occurred. The error number @var{nn} is given as hex digits.
36732 An empty reply indicates that @samp{QProgramSignals} is not supported
36736 Use of this packet is controlled by the @code{set remote program-signals}
36737 command (@pxref{Remote Configuration, set remote program-signals}).
36738 This packet is not probed by default; the remote stub must request it,
36739 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
36741 @anchor{QThreadEvents}
36742 @item QThreadEvents:1
36743 @itemx QThreadEvents:0
36744 @cindex thread create/exit events, remote request
36745 @cindex @samp{QThreadEvents} packet
36747 Enable (@samp{QThreadEvents:1}) or disable (@samp{QThreadEvents:0})
36748 reporting of thread create and exit events. @xref{thread create
36749 event}, for the reply specifications. For example, this is used in
36750 non-stop mode when @value{GDBN} stops a set of threads and
36751 synchronously waits for the their corresponding stop replies. Without
36752 exit events, if one of the threads exits, @value{GDBN} would hang
36753 forever not knowing that it should no longer expect a stop for that
36754 same thread. @value{GDBN} does not enable this feature unless the
36755 stub reports that it supports it by including @samp{QThreadEvents+} in
36756 its @samp{qSupported} reply.
36761 The request succeeded.
36764 An error occurred. The error number @var{nn} is given as hex digits.
36767 An empty reply indicates that @samp{QThreadEvents} is not supported by
36771 Use of this packet is controlled by the @code{set remote thread-events}
36772 command (@pxref{Remote Configuration, set remote thread-events}).
36774 @item qRcmd,@var{command}
36775 @cindex execute remote command, remote request
36776 @cindex @samp{qRcmd} packet
36777 @var{command} (hex encoded) is passed to the local interpreter for
36778 execution. Invalid commands should be reported using the output
36779 string. Before the final result packet, the target may also respond
36780 with a number of intermediate @samp{O@var{output}} console output
36781 packets. @emph{Implementors should note that providing access to a
36782 stubs's interpreter may have security implications}.
36787 A command response with no output.
36789 A command response with the hex encoded output string @var{OUTPUT}.
36791 Indicate a badly formed request.
36793 An empty reply indicates that @samp{qRcmd} is not recognized.
36796 (Note that the @code{qRcmd} packet's name is separated from the
36797 command by a @samp{,}, not a @samp{:}, contrary to the naming
36798 conventions above. Please don't use this packet as a model for new
36801 @item qSearch:memory:@var{address};@var{length};@var{search-pattern}
36802 @cindex searching memory, in remote debugging
36804 @cindex @samp{qSearch:memory} packet
36806 @cindex @samp{qSearch memory} packet
36807 @anchor{qSearch memory}
36808 Search @var{length} bytes at @var{address} for @var{search-pattern}.
36809 Both @var{address} and @var{length} are encoded in hex;
36810 @var{search-pattern} is a sequence of bytes, also hex encoded.
36815 The pattern was not found.
36817 The pattern was found at @var{address}.
36819 A badly formed request or an error was encountered while searching memory.
36821 An empty reply indicates that @samp{qSearch:memory} is not recognized.
36824 @item QStartNoAckMode
36825 @cindex @samp{QStartNoAckMode} packet
36826 @anchor{QStartNoAckMode}
36827 Request that the remote stub disable the normal @samp{+}/@samp{-}
36828 protocol acknowledgments (@pxref{Packet Acknowledgment}).
36833 The stub has switched to no-acknowledgment mode.
36834 @value{GDBN} acknowledges this reponse,
36835 but neither the stub nor @value{GDBN} shall send or expect further
36836 @samp{+}/@samp{-} acknowledgments in the current connection.
36838 An empty reply indicates that the stub does not support no-acknowledgment mode.
36841 @item qSupported @r{[}:@var{gdbfeature} @r{[};@var{gdbfeature}@r{]}@dots{} @r{]}
36842 @cindex supported packets, remote query
36843 @cindex features of the remote protocol
36844 @cindex @samp{qSupported} packet
36845 @anchor{qSupported}
36846 Tell the remote stub about features supported by @value{GDBN}, and
36847 query the stub for features it supports. This packet allows
36848 @value{GDBN} and the remote stub to take advantage of each others'
36849 features. @samp{qSupported} also consolidates multiple feature probes
36850 at startup, to improve @value{GDBN} performance---a single larger
36851 packet performs better than multiple smaller probe packets on
36852 high-latency links. Some features may enable behavior which must not
36853 be on by default, e.g.@: because it would confuse older clients or
36854 stubs. Other features may describe packets which could be
36855 automatically probed for, but are not. These features must be
36856 reported before @value{GDBN} will use them. This ``default
36857 unsupported'' behavior is not appropriate for all packets, but it
36858 helps to keep the initial connection time under control with new
36859 versions of @value{GDBN} which support increasing numbers of packets.
36863 @item @var{stubfeature} @r{[};@var{stubfeature}@r{]}@dots{}
36864 The stub supports or does not support each returned @var{stubfeature},
36865 depending on the form of each @var{stubfeature} (see below for the
36868 An empty reply indicates that @samp{qSupported} is not recognized,
36869 or that no features needed to be reported to @value{GDBN}.
36872 The allowed forms for each feature (either a @var{gdbfeature} in the
36873 @samp{qSupported} packet, or a @var{stubfeature} in the response)
36877 @item @var{name}=@var{value}
36878 The remote protocol feature @var{name} is supported, and associated
36879 with the specified @var{value}. The format of @var{value} depends
36880 on the feature, but it must not include a semicolon.
36882 The remote protocol feature @var{name} is supported, and does not
36883 need an associated value.
36885 The remote protocol feature @var{name} is not supported.
36887 The remote protocol feature @var{name} may be supported, and
36888 @value{GDBN} should auto-detect support in some other way when it is
36889 needed. This form will not be used for @var{gdbfeature} notifications,
36890 but may be used for @var{stubfeature} responses.
36893 Whenever the stub receives a @samp{qSupported} request, the
36894 supplied set of @value{GDBN} features should override any previous
36895 request. This allows @value{GDBN} to put the stub in a known
36896 state, even if the stub had previously been communicating with
36897 a different version of @value{GDBN}.
36899 The following values of @var{gdbfeature} (for the packet sent by @value{GDBN})
36904 This feature indicates whether @value{GDBN} supports multiprocess
36905 extensions to the remote protocol. @value{GDBN} does not use such
36906 extensions unless the stub also reports that it supports them by
36907 including @samp{multiprocess+} in its @samp{qSupported} reply.
36908 @xref{multiprocess extensions}, for details.
36911 This feature indicates that @value{GDBN} supports the XML target
36912 description. If the stub sees @samp{xmlRegisters=} with target
36913 specific strings separated by a comma, it will report register
36917 This feature indicates whether @value{GDBN} supports the
36918 @samp{qRelocInsn} packet (@pxref{Tracepoint Packets,,Relocate
36919 instruction reply packet}).
36922 This feature indicates whether @value{GDBN} supports the swbreak stop
36923 reason in stop replies. @xref{swbreak stop reason}, for details.
36926 This feature indicates whether @value{GDBN} supports the hwbreak stop
36927 reason in stop replies. @xref{swbreak stop reason}, for details.
36930 This feature indicates whether @value{GDBN} supports fork event
36931 extensions to the remote protocol. @value{GDBN} does not use such
36932 extensions unless the stub also reports that it supports them by
36933 including @samp{fork-events+} in its @samp{qSupported} reply.
36936 This feature indicates whether @value{GDBN} supports vfork event
36937 extensions to the remote protocol. @value{GDBN} does not use such
36938 extensions unless the stub also reports that it supports them by
36939 including @samp{vfork-events+} in its @samp{qSupported} reply.
36942 This feature indicates whether @value{GDBN} supports exec event
36943 extensions to the remote protocol. @value{GDBN} does not use such
36944 extensions unless the stub also reports that it supports them by
36945 including @samp{exec-events+} in its @samp{qSupported} reply.
36947 @item vContSupported
36948 This feature indicates whether @value{GDBN} wants to know the
36949 supported actions in the reply to @samp{vCont?} packet.
36952 Stubs should ignore any unknown values for
36953 @var{gdbfeature}. Any @value{GDBN} which sends a @samp{qSupported}
36954 packet supports receiving packets of unlimited length (earlier
36955 versions of @value{GDBN} may reject overly long responses). Additional values
36956 for @var{gdbfeature} may be defined in the future to let the stub take
36957 advantage of new features in @value{GDBN}, e.g.@: incompatible
36958 improvements in the remote protocol---the @samp{multiprocess} feature is
36959 an example of such a feature. The stub's reply should be independent
36960 of the @var{gdbfeature} entries sent by @value{GDBN}; first @value{GDBN}
36961 describes all the features it supports, and then the stub replies with
36962 all the features it supports.
36964 Similarly, @value{GDBN} will silently ignore unrecognized stub feature
36965 responses, as long as each response uses one of the standard forms.
36967 Some features are flags. A stub which supports a flag feature
36968 should respond with a @samp{+} form response. Other features
36969 require values, and the stub should respond with an @samp{=}
36972 Each feature has a default value, which @value{GDBN} will use if
36973 @samp{qSupported} is not available or if the feature is not mentioned
36974 in the @samp{qSupported} response. The default values are fixed; a
36975 stub is free to omit any feature responses that match the defaults.
36977 Not all features can be probed, but for those which can, the probing
36978 mechanism is useful: in some cases, a stub's internal
36979 architecture may not allow the protocol layer to know some information
36980 about the underlying target in advance. This is especially common in
36981 stubs which may be configured for multiple targets.
36983 These are the currently defined stub features and their properties:
36985 @multitable @columnfractions 0.35 0.2 0.12 0.2
36986 @c NOTE: The first row should be @headitem, but we do not yet require
36987 @c a new enough version of Texinfo (4.7) to use @headitem.
36989 @tab Value Required
36993 @item @samp{PacketSize}
36998 @item @samp{qXfer:auxv:read}
37003 @item @samp{qXfer:btrace:read}
37008 @item @samp{qXfer:btrace-conf:read}
37013 @item @samp{qXfer:exec-file:read}
37018 @item @samp{qXfer:features:read}
37023 @item @samp{qXfer:libraries:read}
37028 @item @samp{qXfer:libraries-svr4:read}
37033 @item @samp{augmented-libraries-svr4-read}
37038 @item @samp{qXfer:memory-map:read}
37043 @item @samp{qXfer:sdata:read}
37048 @item @samp{qXfer:spu:read}
37053 @item @samp{qXfer:spu:write}
37058 @item @samp{qXfer:siginfo:read}
37063 @item @samp{qXfer:siginfo:write}
37068 @item @samp{qXfer:threads:read}
37073 @item @samp{qXfer:traceframe-info:read}
37078 @item @samp{qXfer:uib:read}
37083 @item @samp{qXfer:fdpic:read}
37088 @item @samp{Qbtrace:off}
37093 @item @samp{Qbtrace:bts}
37098 @item @samp{Qbtrace:pt}
37103 @item @samp{Qbtrace-conf:bts:size}
37108 @item @samp{Qbtrace-conf:pt:size}
37113 @item @samp{QNonStop}
37118 @item @samp{QCatchSyscalls}
37123 @item @samp{QPassSignals}
37128 @item @samp{QStartNoAckMode}
37133 @item @samp{multiprocess}
37138 @item @samp{ConditionalBreakpoints}
37143 @item @samp{ConditionalTracepoints}
37148 @item @samp{ReverseContinue}
37153 @item @samp{ReverseStep}
37158 @item @samp{TracepointSource}
37163 @item @samp{QAgent}
37168 @item @samp{QAllow}
37173 @item @samp{QDisableRandomization}
37178 @item @samp{EnableDisableTracepoints}
37183 @item @samp{QTBuffer:size}
37188 @item @samp{tracenz}
37193 @item @samp{BreakpointCommands}
37198 @item @samp{swbreak}
37203 @item @samp{hwbreak}
37208 @item @samp{fork-events}
37213 @item @samp{vfork-events}
37218 @item @samp{exec-events}
37223 @item @samp{QThreadEvents}
37228 @item @samp{no-resumed}
37235 These are the currently defined stub features, in more detail:
37238 @cindex packet size, remote protocol
37239 @item PacketSize=@var{bytes}
37240 The remote stub can accept packets up to at least @var{bytes} in
37241 length. @value{GDBN} will send packets up to this size for bulk
37242 transfers, and will never send larger packets. This is a limit on the
37243 data characters in the packet, including the frame and checksum.
37244 There is no trailing NUL byte in a remote protocol packet; if the stub
37245 stores packets in a NUL-terminated format, it should allow an extra
37246 byte in its buffer for the NUL. If this stub feature is not supported,
37247 @value{GDBN} guesses based on the size of the @samp{g} packet response.
37249 @item qXfer:auxv:read
37250 The remote stub understands the @samp{qXfer:auxv:read} packet
37251 (@pxref{qXfer auxiliary vector read}).
37253 @item qXfer:btrace:read
37254 The remote stub understands the @samp{qXfer:btrace:read}
37255 packet (@pxref{qXfer btrace read}).
37257 @item qXfer:btrace-conf:read
37258 The remote stub understands the @samp{qXfer:btrace-conf:read}
37259 packet (@pxref{qXfer btrace-conf read}).
37261 @item qXfer:exec-file:read
37262 The remote stub understands the @samp{qXfer:exec-file:read} packet
37263 (@pxref{qXfer executable filename read}).
37265 @item qXfer:features:read
37266 The remote stub understands the @samp{qXfer:features:read} packet
37267 (@pxref{qXfer target description read}).
37269 @item qXfer:libraries:read
37270 The remote stub understands the @samp{qXfer:libraries:read} packet
37271 (@pxref{qXfer library list read}).
37273 @item qXfer:libraries-svr4:read
37274 The remote stub understands the @samp{qXfer:libraries-svr4:read} packet
37275 (@pxref{qXfer svr4 library list read}).
37277 @item augmented-libraries-svr4-read
37278 The remote stub understands the augmented form of the
37279 @samp{qXfer:libraries-svr4:read} packet
37280 (@pxref{qXfer svr4 library list read}).
37282 @item qXfer:memory-map:read
37283 The remote stub understands the @samp{qXfer:memory-map:read} packet
37284 (@pxref{qXfer memory map read}).
37286 @item qXfer:sdata:read
37287 The remote stub understands the @samp{qXfer:sdata:read} packet
37288 (@pxref{qXfer sdata read}).
37290 @item qXfer:spu:read
37291 The remote stub understands the @samp{qXfer:spu:read} packet
37292 (@pxref{qXfer spu read}).
37294 @item qXfer:spu:write
37295 The remote stub understands the @samp{qXfer:spu:write} packet
37296 (@pxref{qXfer spu write}).
37298 @item qXfer:siginfo:read
37299 The remote stub understands the @samp{qXfer:siginfo:read} packet
37300 (@pxref{qXfer siginfo read}).
37302 @item qXfer:siginfo:write
37303 The remote stub understands the @samp{qXfer:siginfo:write} packet
37304 (@pxref{qXfer siginfo write}).
37306 @item qXfer:threads:read
37307 The remote stub understands the @samp{qXfer:threads:read} packet
37308 (@pxref{qXfer threads read}).
37310 @item qXfer:traceframe-info:read
37311 The remote stub understands the @samp{qXfer:traceframe-info:read}
37312 packet (@pxref{qXfer traceframe info read}).
37314 @item qXfer:uib:read
37315 The remote stub understands the @samp{qXfer:uib:read}
37316 packet (@pxref{qXfer unwind info block}).
37318 @item qXfer:fdpic:read
37319 The remote stub understands the @samp{qXfer:fdpic:read}
37320 packet (@pxref{qXfer fdpic loadmap read}).
37323 The remote stub understands the @samp{QNonStop} packet
37324 (@pxref{QNonStop}).
37326 @item QCatchSyscalls
37327 The remote stub understands the @samp{QCatchSyscalls} packet
37328 (@pxref{QCatchSyscalls}).
37331 The remote stub understands the @samp{QPassSignals} packet
37332 (@pxref{QPassSignals}).
37334 @item QStartNoAckMode
37335 The remote stub understands the @samp{QStartNoAckMode} packet and
37336 prefers to operate in no-acknowledgment mode. @xref{Packet Acknowledgment}.
37339 @anchor{multiprocess extensions}
37340 @cindex multiprocess extensions, in remote protocol
37341 The remote stub understands the multiprocess extensions to the remote
37342 protocol syntax. The multiprocess extensions affect the syntax of
37343 thread IDs in both packets and replies (@pxref{thread-id syntax}), and
37344 add process IDs to the @samp{D} packet and @samp{W} and @samp{X}
37345 replies. Note that reporting this feature indicates support for the
37346 syntactic extensions only, not that the stub necessarily supports
37347 debugging of more than one process at a time. The stub must not use
37348 multiprocess extensions in packet replies unless @value{GDBN} has also
37349 indicated it supports them in its @samp{qSupported} request.
37351 @item qXfer:osdata:read
37352 The remote stub understands the @samp{qXfer:osdata:read} packet
37353 ((@pxref{qXfer osdata read}).
37355 @item ConditionalBreakpoints
37356 The target accepts and implements evaluation of conditional expressions
37357 defined for breakpoints. The target will only report breakpoint triggers
37358 when such conditions are true (@pxref{Conditions, ,Break Conditions}).
37360 @item ConditionalTracepoints
37361 The remote stub accepts and implements conditional expressions defined
37362 for tracepoints (@pxref{Tracepoint Conditions}).
37364 @item ReverseContinue
37365 The remote stub accepts and implements the reverse continue packet
37369 The remote stub accepts and implements the reverse step packet
37372 @item TracepointSource
37373 The remote stub understands the @samp{QTDPsrc} packet that supplies
37374 the source form of tracepoint definitions.
37377 The remote stub understands the @samp{QAgent} packet.
37380 The remote stub understands the @samp{QAllow} packet.
37382 @item QDisableRandomization
37383 The remote stub understands the @samp{QDisableRandomization} packet.
37385 @item StaticTracepoint
37386 @cindex static tracepoints, in remote protocol
37387 The remote stub supports static tracepoints.
37389 @item InstallInTrace
37390 @anchor{install tracepoint in tracing}
37391 The remote stub supports installing tracepoint in tracing.
37393 @item EnableDisableTracepoints
37394 The remote stub supports the @samp{QTEnable} (@pxref{QTEnable}) and
37395 @samp{QTDisable} (@pxref{QTDisable}) packets that allow tracepoints
37396 to be enabled and disabled while a trace experiment is running.
37398 @item QTBuffer:size
37399 The remote stub supports the @samp{QTBuffer:size} (@pxref{QTBuffer-size})
37400 packet that allows to change the size of the trace buffer.
37403 @cindex string tracing, in remote protocol
37404 The remote stub supports the @samp{tracenz} bytecode for collecting strings.
37405 See @ref{Bytecode Descriptions} for details about the bytecode.
37407 @item BreakpointCommands
37408 @cindex breakpoint commands, in remote protocol
37409 The remote stub supports running a breakpoint's command list itself,
37410 rather than reporting the hit to @value{GDBN}.
37413 The remote stub understands the @samp{Qbtrace:off} packet.
37416 The remote stub understands the @samp{Qbtrace:bts} packet.
37419 The remote stub understands the @samp{Qbtrace:pt} packet.
37421 @item Qbtrace-conf:bts:size
37422 The remote stub understands the @samp{Qbtrace-conf:bts:size} packet.
37424 @item Qbtrace-conf:pt:size
37425 The remote stub understands the @samp{Qbtrace-conf:pt:size} packet.
37428 The remote stub reports the @samp{swbreak} stop reason for memory
37432 The remote stub reports the @samp{hwbreak} stop reason for hardware
37436 The remote stub reports the @samp{fork} stop reason for fork events.
37439 The remote stub reports the @samp{vfork} stop reason for vfork events
37440 and vforkdone events.
37443 The remote stub reports the @samp{exec} stop reason for exec events.
37445 @item vContSupported
37446 The remote stub reports the supported actions in the reply to
37447 @samp{vCont?} packet.
37449 @item QThreadEvents
37450 The remote stub understands the @samp{QThreadEvents} packet.
37453 The remote stub reports the @samp{N} stop reply.
37458 @cindex symbol lookup, remote request
37459 @cindex @samp{qSymbol} packet
37460 Notify the target that @value{GDBN} is prepared to serve symbol lookup
37461 requests. Accept requests from the target for the values of symbols.
37466 The target does not need to look up any (more) symbols.
37467 @item qSymbol:@var{sym_name}
37468 The target requests the value of symbol @var{sym_name} (hex encoded).
37469 @value{GDBN} may provide the value by using the
37470 @samp{qSymbol:@var{sym_value}:@var{sym_name}} message, described
37474 @item qSymbol:@var{sym_value}:@var{sym_name}
37475 Set the value of @var{sym_name} to @var{sym_value}.
37477 @var{sym_name} (hex encoded) is the name of a symbol whose value the
37478 target has previously requested.
37480 @var{sym_value} (hex) is the value for symbol @var{sym_name}. If
37481 @value{GDBN} cannot supply a value for @var{sym_name}, then this field
37487 The target does not need to look up any (more) symbols.
37488 @item qSymbol:@var{sym_name}
37489 The target requests the value of a new symbol @var{sym_name} (hex
37490 encoded). @value{GDBN} will continue to supply the values of symbols
37491 (if available), until the target ceases to request them.
37496 @itemx QTDisconnected
37503 @itemx qTMinFTPILen
37505 @xref{Tracepoint Packets}.
37507 @item qThreadExtraInfo,@var{thread-id}
37508 @cindex thread attributes info, remote request
37509 @cindex @samp{qThreadExtraInfo} packet
37510 Obtain from the target OS a printable string description of thread
37511 attributes for the thread @var{thread-id}; see @ref{thread-id syntax},
37512 for the forms of @var{thread-id}. This
37513 string may contain anything that the target OS thinks is interesting
37514 for @value{GDBN} to tell the user about the thread. The string is
37515 displayed in @value{GDBN}'s @code{info threads} display. Some
37516 examples of possible thread extra info strings are @samp{Runnable}, or
37517 @samp{Blocked on Mutex}.
37521 @item @var{XX}@dots{}
37522 Where @samp{@var{XX}@dots{}} is a hex encoding of @sc{ascii} data,
37523 comprising the printable string containing the extra information about
37524 the thread's attributes.
37527 (Note that the @code{qThreadExtraInfo} packet's name is separated from
37528 the command by a @samp{,}, not a @samp{:}, contrary to the naming
37529 conventions above. Please don't use this packet as a model for new
37548 @xref{Tracepoint Packets}.
37550 @item qXfer:@var{object}:read:@var{annex}:@var{offset},@var{length}
37551 @cindex read special object, remote request
37552 @cindex @samp{qXfer} packet
37553 @anchor{qXfer read}
37554 Read uninterpreted bytes from the target's special data area
37555 identified by the keyword @var{object}. Request @var{length} bytes
37556 starting at @var{offset} bytes into the data. The content and
37557 encoding of @var{annex} is specific to @var{object}; it can supply
37558 additional details about what data to access.
37563 Data @var{data} (@pxref{Binary Data}) has been read from the
37564 target. There may be more data at a higher address (although
37565 it is permitted to return @samp{m} even for the last valid
37566 block of data, as long as at least one byte of data was read).
37567 It is possible for @var{data} to have fewer bytes than the @var{length} in the
37571 Data @var{data} (@pxref{Binary Data}) has been read from the target.
37572 There is no more data to be read. It is possible for @var{data} to
37573 have fewer bytes than the @var{length} in the request.
37576 The @var{offset} in the request is at the end of the data.
37577 There is no more data to be read.
37580 The request was malformed, or @var{annex} was invalid.
37583 The offset was invalid, or there was an error encountered reading the data.
37584 The @var{nn} part is a hex-encoded @code{errno} value.
37587 An empty reply indicates the @var{object} string was not recognized by
37588 the stub, or that the object does not support reading.
37591 Here are the specific requests of this form defined so far. All the
37592 @samp{qXfer:@var{object}:read:@dots{}} requests use the same reply
37593 formats, listed above.
37596 @item qXfer:auxv:read::@var{offset},@var{length}
37597 @anchor{qXfer auxiliary vector read}
37598 Access the target's @dfn{auxiliary vector}. @xref{OS Information,
37599 auxiliary vector}. Note @var{annex} must be empty.
37601 This packet is not probed by default; the remote stub must request it,
37602 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
37604 @item qXfer:btrace:read:@var{annex}:@var{offset},@var{length}
37605 @anchor{qXfer btrace read}
37607 Return a description of the current branch trace.
37608 @xref{Branch Trace Format}. The annex part of the generic @samp{qXfer}
37609 packet may have one of the following values:
37613 Returns all available branch trace.
37616 Returns all available branch trace if the branch trace changed since
37617 the last read request.
37620 Returns the new branch trace since the last read request. Adds a new
37621 block to the end of the trace that begins at zero and ends at the source
37622 location of the first branch in the trace buffer. This extra block is
37623 used to stitch traces together.
37625 If the trace buffer overflowed, returns an error indicating the overflow.
37628 This packet is not probed by default; the remote stub must request it
37629 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
37631 @item qXfer:btrace-conf:read::@var{offset},@var{length}
37632 @anchor{qXfer btrace-conf read}
37634 Return a description of the current branch trace configuration.
37635 @xref{Branch Trace Configuration Format}.
37637 This packet is not probed by default; the remote stub must request it
37638 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
37640 @item qXfer:exec-file:read:@var{annex}:@var{offset},@var{length}
37641 @anchor{qXfer executable filename read}
37642 Return the full absolute name of the file that was executed to create
37643 a process running on the remote system. The annex specifies the
37644 numeric process ID of the process to query, encoded as a hexadecimal
37645 number. If the annex part is empty the remote stub should return the
37646 filename corresponding to the currently executing process.
37648 This packet is not probed by default; the remote stub must request it,
37649 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
37651 @item qXfer:features:read:@var{annex}:@var{offset},@var{length}
37652 @anchor{qXfer target description read}
37653 Access the @dfn{target description}. @xref{Target Descriptions}. The
37654 annex specifies which XML document to access. The main description is
37655 always loaded from the @samp{target.xml} annex.
37657 This packet is not probed by default; the remote stub must request it,
37658 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
37660 @item qXfer:libraries:read:@var{annex}:@var{offset},@var{length}
37661 @anchor{qXfer library list read}
37662 Access the target's list of loaded libraries. @xref{Library List Format}.
37663 The annex part of the generic @samp{qXfer} packet must be empty
37664 (@pxref{qXfer read}).
37666 Targets which maintain a list of libraries in the program's memory do
37667 not need to implement this packet; it is designed for platforms where
37668 the operating system manages the list of loaded libraries.
37670 This packet is not probed by default; the remote stub must request it,
37671 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
37673 @item qXfer:libraries-svr4:read:@var{annex}:@var{offset},@var{length}
37674 @anchor{qXfer svr4 library list read}
37675 Access the target's list of loaded libraries when the target is an SVR4
37676 platform. @xref{Library List Format for SVR4 Targets}. The annex part
37677 of the generic @samp{qXfer} packet must be empty unless the remote
37678 stub indicated it supports the augmented form of this packet
37679 by supplying an appropriate @samp{qSupported} response
37680 (@pxref{qXfer read}, @ref{qSupported}).
37682 This packet is optional for better performance on SVR4 targets.
37683 @value{GDBN} uses memory read packets to read the SVR4 library list otherwise.
37685 This packet is not probed by default; the remote stub must request it,
37686 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
37688 If the remote stub indicates it supports the augmented form of this
37689 packet then the annex part of the generic @samp{qXfer} packet may
37690 contain a semicolon-separated list of @samp{@var{name}=@var{value}}
37691 arguments. The currently supported arguments are:
37694 @item start=@var{address}
37695 A hexadecimal number specifying the address of the @samp{struct
37696 link_map} to start reading the library list from. If unset or zero
37697 then the first @samp{struct link_map} in the library list will be
37698 chosen as the starting point.
37700 @item prev=@var{address}
37701 A hexadecimal number specifying the address of the @samp{struct
37702 link_map} immediately preceding the @samp{struct link_map}
37703 specified by the @samp{start} argument. If unset or zero then
37704 the remote stub will expect that no @samp{struct link_map}
37705 exists prior to the starting point.
37709 Arguments that are not understood by the remote stub will be silently
37712 @item qXfer:memory-map:read::@var{offset},@var{length}
37713 @anchor{qXfer memory map read}
37714 Access the target's @dfn{memory-map}. @xref{Memory Map Format}. The
37715 annex part of the generic @samp{qXfer} packet must be empty
37716 (@pxref{qXfer read}).
37718 This packet is not probed by default; the remote stub must request it,
37719 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
37721 @item qXfer:sdata:read::@var{offset},@var{length}
37722 @anchor{qXfer sdata read}
37724 Read contents of the extra collected static tracepoint marker
37725 information. The annex part of the generic @samp{qXfer} packet must
37726 be empty (@pxref{qXfer read}). @xref{Tracepoint Actions,,Tracepoint
37729 This packet is not probed by default; the remote stub must request it,
37730 by supplying an appropriate @samp{qSupported} response
37731 (@pxref{qSupported}).
37733 @item qXfer:siginfo:read::@var{offset},@var{length}
37734 @anchor{qXfer siginfo read}
37735 Read contents of the extra signal information on the target
37736 system. The annex part of the generic @samp{qXfer} packet must be
37737 empty (@pxref{qXfer read}).
37739 This packet is not probed by default; the remote stub must request it,
37740 by supplying an appropriate @samp{qSupported} response
37741 (@pxref{qSupported}).
37743 @item qXfer:spu:read:@var{annex}:@var{offset},@var{length}
37744 @anchor{qXfer spu read}
37745 Read contents of an @code{spufs} file on the target system. The
37746 annex specifies which file to read; it must be of the form
37747 @file{@var{id}/@var{name}}, where @var{id} specifies an SPU context ID
37748 in the target process, and @var{name} identifes the @code{spufs} file
37749 in that context to be accessed.
37751 This packet is not probed by default; the remote stub must request it,
37752 by supplying an appropriate @samp{qSupported} response
37753 (@pxref{qSupported}).
37755 @item qXfer:threads:read::@var{offset},@var{length}
37756 @anchor{qXfer threads read}
37757 Access the list of threads on target. @xref{Thread List Format}. The
37758 annex part of the generic @samp{qXfer} packet must be empty
37759 (@pxref{qXfer read}).
37761 This packet is not probed by default; the remote stub must request it,
37762 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
37764 @item qXfer:traceframe-info:read::@var{offset},@var{length}
37765 @anchor{qXfer traceframe info read}
37767 Return a description of the current traceframe's contents.
37768 @xref{Traceframe Info Format}. The annex part of the generic
37769 @samp{qXfer} packet must be empty (@pxref{qXfer read}).
37771 This packet is not probed by default; the remote stub must request it,
37772 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
37774 @item qXfer:uib:read:@var{pc}:@var{offset},@var{length}
37775 @anchor{qXfer unwind info block}
37777 Return the unwind information block for @var{pc}. This packet is used
37778 on OpenVMS/ia64 to ask the kernel unwind information.
37780 This packet is not probed by default.
37782 @item qXfer:fdpic:read:@var{annex}:@var{offset},@var{length}
37783 @anchor{qXfer fdpic loadmap read}
37784 Read contents of @code{loadmap}s on the target system. The
37785 annex, either @samp{exec} or @samp{interp}, specifies which @code{loadmap},
37786 executable @code{loadmap} or interpreter @code{loadmap} to read.
37788 This packet is not probed by default; the remote stub must request it,
37789 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
37791 @item qXfer:osdata:read::@var{offset},@var{length}
37792 @anchor{qXfer osdata read}
37793 Access the target's @dfn{operating system information}.
37794 @xref{Operating System Information}.
37798 @item qXfer:@var{object}:write:@var{annex}:@var{offset}:@var{data}@dots{}
37799 @cindex write data into object, remote request
37800 @anchor{qXfer write}
37801 Write uninterpreted bytes into the target's special data area
37802 identified by the keyword @var{object}, starting at @var{offset} bytes
37803 into the data. The binary-encoded data (@pxref{Binary Data}) to be
37804 written is given by @var{data}@dots{}. The content and encoding of @var{annex}
37805 is specific to @var{object}; it can supply additional details about what data
37811 @var{nn} (hex encoded) is the number of bytes written.
37812 This may be fewer bytes than supplied in the request.
37815 The request was malformed, or @var{annex} was invalid.
37818 The offset was invalid, or there was an error encountered writing the data.
37819 The @var{nn} part is a hex-encoded @code{errno} value.
37822 An empty reply indicates the @var{object} string was not
37823 recognized by the stub, or that the object does not support writing.
37826 Here are the specific requests of this form defined so far. All the
37827 @samp{qXfer:@var{object}:write:@dots{}} requests use the same reply
37828 formats, listed above.
37831 @item qXfer:siginfo:write::@var{offset}:@var{data}@dots{}
37832 @anchor{qXfer siginfo write}
37833 Write @var{data} to the extra signal information on the target system.
37834 The annex part of the generic @samp{qXfer} packet must be
37835 empty (@pxref{qXfer write}).
37837 This packet is not probed by default; the remote stub must request it,
37838 by supplying an appropriate @samp{qSupported} response
37839 (@pxref{qSupported}).
37841 @item qXfer:spu:write:@var{annex}:@var{offset}:@var{data}@dots{}
37842 @anchor{qXfer spu write}
37843 Write @var{data} to an @code{spufs} file on the target system. The
37844 annex specifies which file to write; it must be of the form
37845 @file{@var{id}/@var{name}}, where @var{id} specifies an SPU context ID
37846 in the target process, and @var{name} identifes the @code{spufs} file
37847 in that context to be accessed.
37849 This packet is not probed by default; the remote stub must request it,
37850 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
37853 @item qXfer:@var{object}:@var{operation}:@dots{}
37854 Requests of this form may be added in the future. When a stub does
37855 not recognize the @var{object} keyword, or its support for
37856 @var{object} does not recognize the @var{operation} keyword, the stub
37857 must respond with an empty packet.
37859 @item qAttached:@var{pid}
37860 @cindex query attached, remote request
37861 @cindex @samp{qAttached} packet
37862 Return an indication of whether the remote server attached to an
37863 existing process or created a new process. When the multiprocess
37864 protocol extensions are supported (@pxref{multiprocess extensions}),
37865 @var{pid} is an integer in hexadecimal format identifying the target
37866 process. Otherwise, @value{GDBN} will omit the @var{pid} field and
37867 the query packet will be simplified as @samp{qAttached}.
37869 This query is used, for example, to know whether the remote process
37870 should be detached or killed when a @value{GDBN} session is ended with
37871 the @code{quit} command.
37876 The remote server attached to an existing process.
37878 The remote server created a new process.
37880 A badly formed request or an error was encountered.
37884 Enable branch tracing for the current thread using Branch Trace Store.
37889 Branch tracing has been enabled.
37891 A badly formed request or an error was encountered.
37895 Enable branch tracing for the current thread using Intel Processor Trace.
37900 Branch tracing has been enabled.
37902 A badly formed request or an error was encountered.
37906 Disable branch tracing for the current thread.
37911 Branch tracing has been disabled.
37913 A badly formed request or an error was encountered.
37916 @item Qbtrace-conf:bts:size=@var{value}
37917 Set the requested ring buffer size for new threads that use the
37918 btrace recording method in bts format.
37923 The ring buffer size has been set.
37925 A badly formed request or an error was encountered.
37928 @item Qbtrace-conf:pt:size=@var{value}
37929 Set the requested ring buffer size for new threads that use the
37930 btrace recording method in pt format.
37935 The ring buffer size has been set.
37937 A badly formed request or an error was encountered.
37942 @node Architecture-Specific Protocol Details
37943 @section Architecture-Specific Protocol Details
37945 This section describes how the remote protocol is applied to specific
37946 target architectures. Also see @ref{Standard Target Features}, for
37947 details of XML target descriptions for each architecture.
37950 * ARM-Specific Protocol Details::
37951 * MIPS-Specific Protocol Details::
37954 @node ARM-Specific Protocol Details
37955 @subsection @acronym{ARM}-specific Protocol Details
37958 * ARM Breakpoint Kinds::
37961 @node ARM Breakpoint Kinds
37962 @subsubsection @acronym{ARM} Breakpoint Kinds
37963 @cindex breakpoint kinds, @acronym{ARM}
37965 These breakpoint kinds are defined for the @samp{Z0} and @samp{Z1} packets.
37970 16-bit Thumb mode breakpoint.
37973 32-bit Thumb mode (Thumb-2) breakpoint.
37976 32-bit @acronym{ARM} mode breakpoint.
37980 @node MIPS-Specific Protocol Details
37981 @subsection @acronym{MIPS}-specific Protocol Details
37984 * MIPS Register packet Format::
37985 * MIPS Breakpoint Kinds::
37988 @node MIPS Register packet Format
37989 @subsubsection @acronym{MIPS} Register Packet Format
37990 @cindex register packet format, @acronym{MIPS}
37992 The following @code{g}/@code{G} packets have previously been defined.
37993 In the below, some thirty-two bit registers are transferred as
37994 sixty-four bits. Those registers should be zero/sign extended (which?)
37995 to fill the space allocated. Register bytes are transferred in target
37996 byte order. The two nibbles within a register byte are transferred
37997 most-significant -- least-significant.
38002 All registers are transferred as thirty-two bit quantities in the order:
38003 32 general-purpose; sr; lo; hi; bad; cause; pc; 32 floating-point
38004 registers; fsr; fir; fp.
38007 All registers are transferred as sixty-four bit quantities (including
38008 thirty-two bit registers such as @code{sr}). The ordering is the same
38013 @node MIPS Breakpoint Kinds
38014 @subsubsection @acronym{MIPS} Breakpoint Kinds
38015 @cindex breakpoint kinds, @acronym{MIPS}
38017 These breakpoint kinds are defined for the @samp{Z0} and @samp{Z1} packets.
38022 16-bit @acronym{MIPS16} mode breakpoint.
38025 16-bit @acronym{microMIPS} mode breakpoint.
38028 32-bit standard @acronym{MIPS} mode breakpoint.
38031 32-bit @acronym{microMIPS} mode breakpoint.
38035 @node Tracepoint Packets
38036 @section Tracepoint Packets
38037 @cindex tracepoint packets
38038 @cindex packets, tracepoint
38040 Here we describe the packets @value{GDBN} uses to implement
38041 tracepoints (@pxref{Tracepoints}).
38045 @item QTDP:@var{n}:@var{addr}:@var{ena}:@var{step}:@var{pass}[:F@var{flen}][:X@var{len},@var{bytes}]@r{[}-@r{]}
38046 @cindex @samp{QTDP} packet
38047 Create a new tracepoint, number @var{n}, at @var{addr}. If @var{ena}
38048 is @samp{E}, then the tracepoint is enabled; if it is @samp{D}, then
38049 the tracepoint is disabled. The @var{step} gives the tracepoint's step
38050 count, and @var{pass} gives its pass count. If an @samp{F} is present,
38051 then the tracepoint is to be a fast tracepoint, and the @var{flen} is
38052 the number of bytes that the target should copy elsewhere to make room
38053 for the tracepoint. If an @samp{X} is present, it introduces a
38054 tracepoint condition, which consists of a hexadecimal length, followed
38055 by a comma and hex-encoded bytes, in a manner similar to action
38056 encodings as described below. If the trailing @samp{-} is present,
38057 further @samp{QTDP} packets will follow to specify this tracepoint's
38063 The packet was understood and carried out.
38065 @xref{Tracepoint Packets,,Relocate instruction reply packet}.
38067 The packet was not recognized.
38070 @item QTDP:-@var{n}:@var{addr}:@r{[}S@r{]}@var{action}@dots{}@r{[}-@r{]}
38071 Define actions to be taken when a tracepoint is hit. The @var{n} and
38072 @var{addr} must be the same as in the initial @samp{QTDP} packet for
38073 this tracepoint. This packet may only be sent immediately after
38074 another @samp{QTDP} packet that ended with a @samp{-}. If the
38075 trailing @samp{-} is present, further @samp{QTDP} packets will follow,
38076 specifying more actions for this tracepoint.
38078 In the series of action packets for a given tracepoint, at most one
38079 can have an @samp{S} before its first @var{action}. If such a packet
38080 is sent, it and the following packets define ``while-stepping''
38081 actions. Any prior packets define ordinary actions --- that is, those
38082 taken when the tracepoint is first hit. If no action packet has an
38083 @samp{S}, then all the packets in the series specify ordinary
38084 tracepoint actions.
38086 The @samp{@var{action}@dots{}} portion of the packet is a series of
38087 actions, concatenated without separators. Each action has one of the
38093 Collect the registers whose bits are set in @var{mask},
38094 a hexadecimal number whose @var{i}'th bit is set if register number
38095 @var{i} should be collected. (The least significant bit is numbered
38096 zero.) Note that @var{mask} may be any number of digits long; it may
38097 not fit in a 32-bit word.
38099 @item M @var{basereg},@var{offset},@var{len}
38100 Collect @var{len} bytes of memory starting at the address in register
38101 number @var{basereg}, plus @var{offset}. If @var{basereg} is
38102 @samp{-1}, then the range has a fixed address: @var{offset} is the
38103 address of the lowest byte to collect. The @var{basereg},
38104 @var{offset}, and @var{len} parameters are all unsigned hexadecimal
38105 values (the @samp{-1} value for @var{basereg} is a special case).
38107 @item X @var{len},@var{expr}
38108 Evaluate @var{expr}, whose length is @var{len}, and collect memory as
38109 it directs. The agent expression @var{expr} is as described in
38110 @ref{Agent Expressions}. Each byte of the expression is encoded as a
38111 two-digit hex number in the packet; @var{len} is the number of bytes
38112 in the expression (and thus one-half the number of hex digits in the
38117 Any number of actions may be packed together in a single @samp{QTDP}
38118 packet, as long as the packet does not exceed the maximum packet
38119 length (400 bytes, for many stubs). There may be only one @samp{R}
38120 action per tracepoint, and it must precede any @samp{M} or @samp{X}
38121 actions. Any registers referred to by @samp{M} and @samp{X} actions
38122 must be collected by a preceding @samp{R} action. (The
38123 ``while-stepping'' actions are treated as if they were attached to a
38124 separate tracepoint, as far as these restrictions are concerned.)
38129 The packet was understood and carried out.
38131 @xref{Tracepoint Packets,,Relocate instruction reply packet}.
38133 The packet was not recognized.
38136 @item QTDPsrc:@var{n}:@var{addr}:@var{type}:@var{start}:@var{slen}:@var{bytes}
38137 @cindex @samp{QTDPsrc} packet
38138 Specify a source string of tracepoint @var{n} at address @var{addr}.
38139 This is useful to get accurate reproduction of the tracepoints
38140 originally downloaded at the beginning of the trace run. The @var{type}
38141 is the name of the tracepoint part, such as @samp{cond} for the
38142 tracepoint's conditional expression (see below for a list of types), while
38143 @var{bytes} is the string, encoded in hexadecimal.
38145 @var{start} is the offset of the @var{bytes} within the overall source
38146 string, while @var{slen} is the total length of the source string.
38147 This is intended for handling source strings that are longer than will
38148 fit in a single packet.
38149 @c Add detailed example when this info is moved into a dedicated
38150 @c tracepoint descriptions section.
38152 The available string types are @samp{at} for the location,
38153 @samp{cond} for the conditional, and @samp{cmd} for an action command.
38154 @value{GDBN} sends a separate packet for each command in the action
38155 list, in the same order in which the commands are stored in the list.
38157 The target does not need to do anything with source strings except
38158 report them back as part of the replies to the @samp{qTfP}/@samp{qTsP}
38161 Although this packet is optional, and @value{GDBN} will only send it
38162 if the target replies with @samp{TracepointSource} @xref{General
38163 Query Packets}, it makes both disconnected tracing and trace files
38164 much easier to use. Otherwise the user must be careful that the
38165 tracepoints in effect while looking at trace frames are identical to
38166 the ones in effect during the trace run; even a small discrepancy
38167 could cause @samp{tdump} not to work, or a particular trace frame not
38170 @item QTDV:@var{n}:@var{value}:@var{builtin}:@var{name}
38171 @cindex define trace state variable, remote request
38172 @cindex @samp{QTDV} packet
38173 Create a new trace state variable, number @var{n}, with an initial
38174 value of @var{value}, which is a 64-bit signed integer. Both @var{n}
38175 and @var{value} are encoded as hexadecimal values. @value{GDBN} has
38176 the option of not using this packet for initial values of zero; the
38177 target should simply create the trace state variables as they are
38178 mentioned in expressions. The value @var{builtin} should be 1 (one)
38179 if the trace state variable is builtin and 0 (zero) if it is not builtin.
38180 @value{GDBN} only sets @var{builtin} to 1 if a previous @samp{qTfV} or
38181 @samp{qTsV} packet had it set. The contents of @var{name} is the
38182 hex-encoded name (without the leading @samp{$}) of the trace state
38185 @item QTFrame:@var{n}
38186 @cindex @samp{QTFrame} packet
38187 Select the @var{n}'th tracepoint frame from the buffer, and use the
38188 register and memory contents recorded there to answer subsequent
38189 request packets from @value{GDBN}.
38191 A successful reply from the stub indicates that the stub has found the
38192 requested frame. The response is a series of parts, concatenated
38193 without separators, describing the frame we selected. Each part has
38194 one of the following forms:
38198 The selected frame is number @var{n} in the trace frame buffer;
38199 @var{f} is a hexadecimal number. If @var{f} is @samp{-1}, then there
38200 was no frame matching the criteria in the request packet.
38203 The selected trace frame records a hit of tracepoint number @var{t};
38204 @var{t} is a hexadecimal number.
38208 @item QTFrame:pc:@var{addr}
38209 Like @samp{QTFrame:@var{n}}, but select the first tracepoint frame after the
38210 currently selected frame whose PC is @var{addr};
38211 @var{addr} is a hexadecimal number.
38213 @item QTFrame:tdp:@var{t}
38214 Like @samp{QTFrame:@var{n}}, but select the first tracepoint frame after the
38215 currently selected frame that is a hit of tracepoint @var{t}; @var{t}
38216 is a hexadecimal number.
38218 @item QTFrame:range:@var{start}:@var{end}
38219 Like @samp{QTFrame:@var{n}}, but select the first tracepoint frame after the
38220 currently selected frame whose PC is between @var{start} (inclusive)
38221 and @var{end} (inclusive); @var{start} and @var{end} are hexadecimal
38224 @item QTFrame:outside:@var{start}:@var{end}
38225 Like @samp{QTFrame:range:@var{start}:@var{end}}, but select the first
38226 frame @emph{outside} the given range of addresses (exclusive).
38229 @cindex @samp{qTMinFTPILen} packet
38230 This packet requests the minimum length of instruction at which a fast
38231 tracepoint (@pxref{Set Tracepoints}) may be placed. For instance, on
38232 the 32-bit x86 architecture, it is possible to use a 4-byte jump, but
38233 it depends on the target system being able to create trampolines in
38234 the first 64K of memory, which might or might not be possible for that
38235 system. So the reply to this packet will be 4 if it is able to
38242 The minimum instruction length is currently unknown.
38244 The minimum instruction length is @var{length}, where @var{length}
38245 is a hexadecimal number greater or equal to 1. A reply
38246 of 1 means that a fast tracepoint may be placed on any instruction
38247 regardless of size.
38249 An error has occurred.
38251 An empty reply indicates that the request is not supported by the stub.
38255 @cindex @samp{QTStart} packet
38256 Begin the tracepoint experiment. Begin collecting data from
38257 tracepoint hits in the trace frame buffer. This packet supports the
38258 @samp{qRelocInsn} reply (@pxref{Tracepoint Packets,,Relocate
38259 instruction reply packet}).
38262 @cindex @samp{QTStop} packet
38263 End the tracepoint experiment. Stop collecting trace frames.
38265 @item QTEnable:@var{n}:@var{addr}
38267 @cindex @samp{QTEnable} packet
38268 Enable tracepoint @var{n} at address @var{addr} in a started tracepoint
38269 experiment. If the tracepoint was previously disabled, then collection
38270 of data from it will resume.
38272 @item QTDisable:@var{n}:@var{addr}
38274 @cindex @samp{QTDisable} packet
38275 Disable tracepoint @var{n} at address @var{addr} in a started tracepoint
38276 experiment. No more data will be collected from the tracepoint unless
38277 @samp{QTEnable:@var{n}:@var{addr}} is subsequently issued.
38280 @cindex @samp{QTinit} packet
38281 Clear the table of tracepoints, and empty the trace frame buffer.
38283 @item QTro:@var{start1},@var{end1}:@var{start2},@var{end2}:@dots{}
38284 @cindex @samp{QTro} packet
38285 Establish the given ranges of memory as ``transparent''. The stub
38286 will answer requests for these ranges from memory's current contents,
38287 if they were not collected as part of the tracepoint hit.
38289 @value{GDBN} uses this to mark read-only regions of memory, like those
38290 containing program code. Since these areas never change, they should
38291 still have the same contents they did when the tracepoint was hit, so
38292 there's no reason for the stub to refuse to provide their contents.
38294 @item QTDisconnected:@var{value}
38295 @cindex @samp{QTDisconnected} packet
38296 Set the choice to what to do with the tracing run when @value{GDBN}
38297 disconnects from the target. A @var{value} of 1 directs the target to
38298 continue the tracing run, while 0 tells the target to stop tracing if
38299 @value{GDBN} is no longer in the picture.
38302 @cindex @samp{qTStatus} packet
38303 Ask the stub if there is a trace experiment running right now.
38305 The reply has the form:
38309 @item T@var{running}@r{[};@var{field}@r{]}@dots{}
38310 @var{running} is a single digit @code{1} if the trace is presently
38311 running, or @code{0} if not. It is followed by semicolon-separated
38312 optional fields that an agent may use to report additional status.
38316 If the trace is not running, the agent may report any of several
38317 explanations as one of the optional fields:
38322 No trace has been run yet.
38324 @item tstop[:@var{text}]:0
38325 The trace was stopped by a user-originated stop command. The optional
38326 @var{text} field is a user-supplied string supplied as part of the
38327 stop command (for instance, an explanation of why the trace was
38328 stopped manually). It is hex-encoded.
38331 The trace stopped because the trace buffer filled up.
38333 @item tdisconnected:0
38334 The trace stopped because @value{GDBN} disconnected from the target.
38336 @item tpasscount:@var{tpnum}
38337 The trace stopped because tracepoint @var{tpnum} exceeded its pass count.
38339 @item terror:@var{text}:@var{tpnum}
38340 The trace stopped because tracepoint @var{tpnum} had an error. The
38341 string @var{text} is available to describe the nature of the error
38342 (for instance, a divide by zero in the condition expression); it
38346 The trace stopped for some other reason.
38350 Additional optional fields supply statistical and other information.
38351 Although not required, they are extremely useful for users monitoring
38352 the progress of a trace run. If a trace has stopped, and these
38353 numbers are reported, they must reflect the state of the just-stopped
38358 @item tframes:@var{n}
38359 The number of trace frames in the buffer.
38361 @item tcreated:@var{n}
38362 The total number of trace frames created during the run. This may
38363 be larger than the trace frame count, if the buffer is circular.
38365 @item tsize:@var{n}
38366 The total size of the trace buffer, in bytes.
38368 @item tfree:@var{n}
38369 The number of bytes still unused in the buffer.
38371 @item circular:@var{n}
38372 The value of the circular trace buffer flag. @code{1} means that the
38373 trace buffer is circular and old trace frames will be discarded if
38374 necessary to make room, @code{0} means that the trace buffer is linear
38377 @item disconn:@var{n}
38378 The value of the disconnected tracing flag. @code{1} means that
38379 tracing will continue after @value{GDBN} disconnects, @code{0} means
38380 that the trace run will stop.
38384 @item qTP:@var{tp}:@var{addr}
38385 @cindex tracepoint status, remote request
38386 @cindex @samp{qTP} packet
38387 Ask the stub for the current state of tracepoint number @var{tp} at
38388 address @var{addr}.
38392 @item V@var{hits}:@var{usage}
38393 The tracepoint has been hit @var{hits} times so far during the trace
38394 run, and accounts for @var{usage} in the trace buffer. Note that
38395 @code{while-stepping} steps are not counted as separate hits, but the
38396 steps' space consumption is added into the usage number.
38400 @item qTV:@var{var}
38401 @cindex trace state variable value, remote request
38402 @cindex @samp{qTV} packet
38403 Ask the stub for the value of the trace state variable number @var{var}.
38408 The value of the variable is @var{value}. This will be the current
38409 value of the variable if the user is examining a running target, or a
38410 saved value if the variable was collected in the trace frame that the
38411 user is looking at. Note that multiple requests may result in
38412 different reply values, such as when requesting values while the
38413 program is running.
38416 The value of the variable is unknown. This would occur, for example,
38417 if the user is examining a trace frame in which the requested variable
38422 @cindex @samp{qTfP} packet
38424 @cindex @samp{qTsP} packet
38425 These packets request data about tracepoints that are being used by
38426 the target. @value{GDBN} sends @code{qTfP} to get the first piece
38427 of data, and multiple @code{qTsP} to get additional pieces. Replies
38428 to these packets generally take the form of the @code{QTDP} packets
38429 that define tracepoints. (FIXME add detailed syntax)
38432 @cindex @samp{qTfV} packet
38434 @cindex @samp{qTsV} packet
38435 These packets request data about trace state variables that are on the
38436 target. @value{GDBN} sends @code{qTfV} to get the first vari of data,
38437 and multiple @code{qTsV} to get additional variables. Replies to
38438 these packets follow the syntax of the @code{QTDV} packets that define
38439 trace state variables.
38445 @cindex @samp{qTfSTM} packet
38446 @cindex @samp{qTsSTM} packet
38447 These packets request data about static tracepoint markers that exist
38448 in the target program. @value{GDBN} sends @code{qTfSTM} to get the
38449 first piece of data, and multiple @code{qTsSTM} to get additional
38450 pieces. Replies to these packets take the following form:
38454 @item m @var{address}:@var{id}:@var{extra}
38456 @item m @var{address}:@var{id}:@var{extra},@var{address}:@var{id}:@var{extra}@dots{}
38457 a comma-separated list of markers
38459 (lower case letter @samp{L}) denotes end of list.
38461 An error occurred. The error number @var{nn} is given as hex digits.
38463 An empty reply indicates that the request is not supported by the
38467 The @var{address} is encoded in hex;
38468 @var{id} and @var{extra} are strings encoded in hex.
38470 In response to each query, the target will reply with a list of one or
38471 more markers, separated by commas. @value{GDBN} will respond to each
38472 reply with a request for more markers (using the @samp{qs} form of the
38473 query), until the target responds with @samp{l} (lower-case ell, for
38476 @item qTSTMat:@var{address}
38478 @cindex @samp{qTSTMat} packet
38479 This packets requests data about static tracepoint markers in the
38480 target program at @var{address}. Replies to this packet follow the
38481 syntax of the @samp{qTfSTM} and @code{qTsSTM} packets that list static
38482 tracepoint markers.
38484 @item QTSave:@var{filename}
38485 @cindex @samp{QTSave} packet
38486 This packet directs the target to save trace data to the file name
38487 @var{filename} in the target's filesystem. The @var{filename} is encoded
38488 as a hex string; the interpretation of the file name (relative vs
38489 absolute, wild cards, etc) is up to the target.
38491 @item qTBuffer:@var{offset},@var{len}
38492 @cindex @samp{qTBuffer} packet
38493 Return up to @var{len} bytes of the current contents of trace buffer,
38494 starting at @var{offset}. The trace buffer is treated as if it were
38495 a contiguous collection of traceframes, as per the trace file format.
38496 The reply consists as many hex-encoded bytes as the target can deliver
38497 in a packet; it is not an error to return fewer than were asked for.
38498 A reply consisting of just @code{l} indicates that no bytes are
38501 @item QTBuffer:circular:@var{value}
38502 This packet directs the target to use a circular trace buffer if
38503 @var{value} is 1, or a linear buffer if the value is 0.
38505 @item QTBuffer:size:@var{size}
38506 @anchor{QTBuffer-size}
38507 @cindex @samp{QTBuffer size} packet
38508 This packet directs the target to make the trace buffer be of size
38509 @var{size} if possible. A value of @code{-1} tells the target to
38510 use whatever size it prefers.
38512 @item QTNotes:@r{[}@var{type}:@var{text}@r{]}@r{[};@var{type}:@var{text}@r{]}@dots{}
38513 @cindex @samp{QTNotes} packet
38514 This packet adds optional textual notes to the trace run. Allowable
38515 types include @code{user}, @code{notes}, and @code{tstop}, the
38516 @var{text} fields are arbitrary strings, hex-encoded.
38520 @subsection Relocate instruction reply packet
38521 When installing fast tracepoints in memory, the target may need to
38522 relocate the instruction currently at the tracepoint address to a
38523 different address in memory. For most instructions, a simple copy is
38524 enough, but, for example, call instructions that implicitly push the
38525 return address on the stack, and relative branches or other
38526 PC-relative instructions require offset adjustment, so that the effect
38527 of executing the instruction at a different address is the same as if
38528 it had executed in the original location.
38530 In response to several of the tracepoint packets, the target may also
38531 respond with a number of intermediate @samp{qRelocInsn} request
38532 packets before the final result packet, to have @value{GDBN} handle
38533 this relocation operation. If a packet supports this mechanism, its
38534 documentation will explicitly say so. See for example the above
38535 descriptions for the @samp{QTStart} and @samp{QTDP} packets. The
38536 format of the request is:
38539 @item qRelocInsn:@var{from};@var{to}
38541 This requests @value{GDBN} to copy instruction at address @var{from}
38542 to address @var{to}, possibly adjusted so that executing the
38543 instruction at @var{to} has the same effect as executing it at
38544 @var{from}. @value{GDBN} writes the adjusted instruction to target
38545 memory starting at @var{to}.
38550 @item qRelocInsn:@var{adjusted_size}
38551 Informs the stub the relocation is complete. The @var{adjusted_size} is
38552 the length in bytes of resulting relocated instruction sequence.
38554 A badly formed request was detected, or an error was encountered while
38555 relocating the instruction.
38558 @node Host I/O Packets
38559 @section Host I/O Packets
38560 @cindex Host I/O, remote protocol
38561 @cindex file transfer, remote protocol
38563 The @dfn{Host I/O} packets allow @value{GDBN} to perform I/O
38564 operations on the far side of a remote link. For example, Host I/O is
38565 used to upload and download files to a remote target with its own
38566 filesystem. Host I/O uses the same constant values and data structure
38567 layout as the target-initiated File-I/O protocol. However, the
38568 Host I/O packets are structured differently. The target-initiated
38569 protocol relies on target memory to store parameters and buffers.
38570 Host I/O requests are initiated by @value{GDBN}, and the
38571 target's memory is not involved. @xref{File-I/O Remote Protocol
38572 Extension}, for more details on the target-initiated protocol.
38574 The Host I/O request packets all encode a single operation along with
38575 its arguments. They have this format:
38579 @item vFile:@var{operation}: @var{parameter}@dots{}
38580 @var{operation} is the name of the particular request; the target
38581 should compare the entire packet name up to the second colon when checking
38582 for a supported operation. The format of @var{parameter} depends on
38583 the operation. Numbers are always passed in hexadecimal. Negative
38584 numbers have an explicit minus sign (i.e.@: two's complement is not
38585 used). Strings (e.g.@: filenames) are encoded as a series of
38586 hexadecimal bytes. The last argument to a system call may be a
38587 buffer of escaped binary data (@pxref{Binary Data}).
38591 The valid responses to Host I/O packets are:
38595 @item F @var{result} [, @var{errno}] [; @var{attachment}]
38596 @var{result} is the integer value returned by this operation, usually
38597 non-negative for success and -1 for errors. If an error has occured,
38598 @var{errno} will be included in the result specifying a
38599 value defined by the File-I/O protocol (@pxref{Errno Values}). For
38600 operations which return data, @var{attachment} supplies the data as a
38601 binary buffer. Binary buffers in response packets are escaped in the
38602 normal way (@pxref{Binary Data}). See the individual packet
38603 documentation for the interpretation of @var{result} and
38607 An empty response indicates that this operation is not recognized.
38611 These are the supported Host I/O operations:
38614 @item vFile:open: @var{filename}, @var{flags}, @var{mode}
38615 Open a file at @var{filename} and return a file descriptor for it, or
38616 return -1 if an error occurs. The @var{filename} is a string,
38617 @var{flags} is an integer indicating a mask of open flags
38618 (@pxref{Open Flags}), and @var{mode} is an integer indicating a mask
38619 of mode bits to use if the file is created (@pxref{mode_t Values}).
38620 @xref{open}, for details of the open flags and mode values.
38622 @item vFile:close: @var{fd}
38623 Close the open file corresponding to @var{fd} and return 0, or
38624 -1 if an error occurs.
38626 @item vFile:pread: @var{fd}, @var{count}, @var{offset}
38627 Read data from the open file corresponding to @var{fd}. Up to
38628 @var{count} bytes will be read from the file, starting at @var{offset}
38629 relative to the start of the file. The target may read fewer bytes;
38630 common reasons include packet size limits and an end-of-file
38631 condition. The number of bytes read is returned. Zero should only be
38632 returned for a successful read at the end of the file, or if
38633 @var{count} was zero.
38635 The data read should be returned as a binary attachment on success.
38636 If zero bytes were read, the response should include an empty binary
38637 attachment (i.e.@: a trailing semicolon). The return value is the
38638 number of target bytes read; the binary attachment may be longer if
38639 some characters were escaped.
38641 @item vFile:pwrite: @var{fd}, @var{offset}, @var{data}
38642 Write @var{data} (a binary buffer) to the open file corresponding
38643 to @var{fd}. Start the write at @var{offset} from the start of the
38644 file. Unlike many @code{write} system calls, there is no
38645 separate @var{count} argument; the length of @var{data} in the
38646 packet is used. @samp{vFile:write} returns the number of bytes written,
38647 which may be shorter than the length of @var{data}, or -1 if an
38650 @item vFile:fstat: @var{fd}
38651 Get information about the open file corresponding to @var{fd}.
38652 On success the information is returned as a binary attachment
38653 and the return value is the size of this attachment in bytes.
38654 If an error occurs the return value is -1. The format of the
38655 returned binary attachment is as described in @ref{struct stat}.
38657 @item vFile:unlink: @var{filename}
38658 Delete the file at @var{filename} on the target. Return 0,
38659 or -1 if an error occurs. The @var{filename} is a string.
38661 @item vFile:readlink: @var{filename}
38662 Read value of symbolic link @var{filename} on the target. Return
38663 the number of bytes read, or -1 if an error occurs.
38665 The data read should be returned as a binary attachment on success.
38666 If zero bytes were read, the response should include an empty binary
38667 attachment (i.e.@: a trailing semicolon). The return value is the
38668 number of target bytes read; the binary attachment may be longer if
38669 some characters were escaped.
38671 @item vFile:setfs: @var{pid}
38672 Select the filesystem on which @code{vFile} operations with
38673 @var{filename} arguments will operate. This is required for
38674 @value{GDBN} to be able to access files on remote targets where
38675 the remote stub does not share a common filesystem with the
38678 If @var{pid} is nonzero, select the filesystem as seen by process
38679 @var{pid}. If @var{pid} is zero, select the filesystem as seen by
38680 the remote stub. Return 0 on success, or -1 if an error occurs.
38681 If @code{vFile:setfs:} indicates success, the selected filesystem
38682 remains selected until the next successful @code{vFile:setfs:}
38688 @section Interrupts
38689 @cindex interrupts (remote protocol)
38690 @anchor{interrupting remote targets}
38692 In all-stop mode, when a program on the remote target is running,
38693 @value{GDBN} may attempt to interrupt it by sending a @samp{Ctrl-C},
38694 @code{BREAK} or a @code{BREAK} followed by @code{g}, control of which
38695 is specified via @value{GDBN}'s @samp{interrupt-sequence}.
38697 The precise meaning of @code{BREAK} is defined by the transport
38698 mechanism and may, in fact, be undefined. @value{GDBN} does not
38699 currently define a @code{BREAK} mechanism for any of the network
38700 interfaces except for TCP, in which case @value{GDBN} sends the
38701 @code{telnet} BREAK sequence.
38703 @samp{Ctrl-C}, on the other hand, is defined and implemented for all
38704 transport mechanisms. It is represented by sending the single byte
38705 @code{0x03} without any of the usual packet overhead described in
38706 the Overview section (@pxref{Overview}). When a @code{0x03} byte is
38707 transmitted as part of a packet, it is considered to be packet data
38708 and does @emph{not} represent an interrupt. E.g., an @samp{X} packet
38709 (@pxref{X packet}), used for binary downloads, may include an unescaped
38710 @code{0x03} as part of its packet.
38712 @code{BREAK} followed by @code{g} is also known as Magic SysRq g.
38713 When Linux kernel receives this sequence from serial port,
38714 it stops execution and connects to gdb.
38716 In non-stop mode, because packet resumptions are asynchronous
38717 (@pxref{vCont packet}), @value{GDBN} is always free to send a remote
38718 command to the remote stub, even when the target is running. For that
38719 reason, @value{GDBN} instead sends a regular packet (@pxref{vCtrlC
38720 packet}) with the usual packet framing instead of the single byte
38723 Stubs are not required to recognize these interrupt mechanisms and the
38724 precise meaning associated with receipt of the interrupt is
38725 implementation defined. If the target supports debugging of multiple
38726 threads and/or processes, it should attempt to interrupt all
38727 currently-executing threads and processes.
38728 If the stub is successful at interrupting the
38729 running program, it should send one of the stop
38730 reply packets (@pxref{Stop Reply Packets}) to @value{GDBN} as a result
38731 of successfully stopping the program in all-stop mode, and a stop reply
38732 for each stopped thread in non-stop mode.
38733 Interrupts received while the
38734 program is stopped are queued and the program will be interrupted when
38735 it is resumed next time.
38737 @node Notification Packets
38738 @section Notification Packets
38739 @cindex notification packets
38740 @cindex packets, notification
38742 The @value{GDBN} remote serial protocol includes @dfn{notifications},
38743 packets that require no acknowledgment. Both the GDB and the stub
38744 may send notifications (although the only notifications defined at
38745 present are sent by the stub). Notifications carry information
38746 without incurring the round-trip latency of an acknowledgment, and so
38747 are useful for low-impact communications where occasional packet loss
38750 A notification packet has the form @samp{% @var{data} #
38751 @var{checksum}}, where @var{data} is the content of the notification,
38752 and @var{checksum} is a checksum of @var{data}, computed and formatted
38753 as for ordinary @value{GDBN} packets. A notification's @var{data}
38754 never contains @samp{$}, @samp{%} or @samp{#} characters. Upon
38755 receiving a notification, the recipient sends no @samp{+} or @samp{-}
38756 to acknowledge the notification's receipt or to report its corruption.
38758 Every notification's @var{data} begins with a name, which contains no
38759 colon characters, followed by a colon character.
38761 Recipients should silently ignore corrupted notifications and
38762 notifications they do not understand. Recipients should restart
38763 timeout periods on receipt of a well-formed notification, whether or
38764 not they understand it.
38766 Senders should only send the notifications described here when this
38767 protocol description specifies that they are permitted. In the
38768 future, we may extend the protocol to permit existing notifications in
38769 new contexts; this rule helps older senders avoid confusing newer
38772 (Older versions of @value{GDBN} ignore bytes received until they see
38773 the @samp{$} byte that begins an ordinary packet, so new stubs may
38774 transmit notifications without fear of confusing older clients. There
38775 are no notifications defined for @value{GDBN} to send at the moment, but we
38776 assume that most older stubs would ignore them, as well.)
38778 Each notification is comprised of three parts:
38780 @item @var{name}:@var{event}
38781 The notification packet is sent by the side that initiates the
38782 exchange (currently, only the stub does that), with @var{event}
38783 carrying the specific information about the notification, and
38784 @var{name} specifying the name of the notification.
38786 The acknowledge sent by the other side, usually @value{GDBN}, to
38787 acknowledge the exchange and request the event.
38790 The purpose of an asynchronous notification mechanism is to report to
38791 @value{GDBN} that something interesting happened in the remote stub.
38793 The remote stub may send notification @var{name}:@var{event}
38794 at any time, but @value{GDBN} acknowledges the notification when
38795 appropriate. The notification event is pending before @value{GDBN}
38796 acknowledges. Only one notification at a time may be pending; if
38797 additional events occur before @value{GDBN} has acknowledged the
38798 previous notification, they must be queued by the stub for later
38799 synchronous transmission in response to @var{ack} packets from
38800 @value{GDBN}. Because the notification mechanism is unreliable,
38801 the stub is permitted to resend a notification if it believes
38802 @value{GDBN} may not have received it.
38804 Specifically, notifications may appear when @value{GDBN} is not
38805 otherwise reading input from the stub, or when @value{GDBN} is
38806 expecting to read a normal synchronous response or a
38807 @samp{+}/@samp{-} acknowledgment to a packet it has sent.
38808 Notification packets are distinct from any other communication from
38809 the stub so there is no ambiguity.
38811 After receiving a notification, @value{GDBN} shall acknowledge it by
38812 sending a @var{ack} packet as a regular, synchronous request to the
38813 stub. Such acknowledgment is not required to happen immediately, as
38814 @value{GDBN} is permitted to send other, unrelated packets to the
38815 stub first, which the stub should process normally.
38817 Upon receiving a @var{ack} packet, if the stub has other queued
38818 events to report to @value{GDBN}, it shall respond by sending a
38819 normal @var{event}. @value{GDBN} shall then send another @var{ack}
38820 packet to solicit further responses; again, it is permitted to send
38821 other, unrelated packets as well which the stub should process
38824 If the stub receives a @var{ack} packet and there are no additional
38825 @var{event} to report, the stub shall return an @samp{OK} response.
38826 At this point, @value{GDBN} has finished processing a notification
38827 and the stub has completed sending any queued events. @value{GDBN}
38828 won't accept any new notifications until the final @samp{OK} is
38829 received . If further notification events occur, the stub shall send
38830 a new notification, @value{GDBN} shall accept the notification, and
38831 the process shall be repeated.
38833 The process of asynchronous notification can be illustrated by the
38836 <- @code{%Stop:T0505:98e7ffbf;04:4ce6ffbf;08:b1b6e54c;thread:p7526.7526;core:0;}
38839 <- @code{T0505:68f37db7;04:40f37db7;08:63850408;thread:p7526.7528;core:0;}
38841 <- @code{T0505:68e3fdb6;04:40e3fdb6;08:63850408;thread:p7526.7529;core:0;}
38846 The following notifications are defined:
38847 @multitable @columnfractions 0.12 0.12 0.38 0.38
38856 @tab @var{reply}. The @var{reply} has the form of a stop reply, as
38857 described in @ref{Stop Reply Packets}. Refer to @ref{Remote Non-Stop},
38858 for information on how these notifications are acknowledged by
38860 @tab Report an asynchronous stop event in non-stop mode.
38864 @node Remote Non-Stop
38865 @section Remote Protocol Support for Non-Stop Mode
38867 @value{GDBN}'s remote protocol supports non-stop debugging of
38868 multi-threaded programs, as described in @ref{Non-Stop Mode}. If the stub
38869 supports non-stop mode, it should report that to @value{GDBN} by including
38870 @samp{QNonStop+} in its @samp{qSupported} response (@pxref{qSupported}).
38872 @value{GDBN} typically sends a @samp{QNonStop} packet only when
38873 establishing a new connection with the stub. Entering non-stop mode
38874 does not alter the state of any currently-running threads, but targets
38875 must stop all threads in any already-attached processes when entering
38876 all-stop mode. @value{GDBN} uses the @samp{?} packet as necessary to
38877 probe the target state after a mode change.
38879 In non-stop mode, when an attached process encounters an event that
38880 would otherwise be reported with a stop reply, it uses the
38881 asynchronous notification mechanism (@pxref{Notification Packets}) to
38882 inform @value{GDBN}. In contrast to all-stop mode, where all threads
38883 in all processes are stopped when a stop reply is sent, in non-stop
38884 mode only the thread reporting the stop event is stopped. That is,
38885 when reporting a @samp{S} or @samp{T} response to indicate completion
38886 of a step operation, hitting a breakpoint, or a fault, only the
38887 affected thread is stopped; any other still-running threads continue
38888 to run. When reporting a @samp{W} or @samp{X} response, all running
38889 threads belonging to other attached processes continue to run.
38891 In non-stop mode, the target shall respond to the @samp{?} packet as
38892 follows. First, any incomplete stop reply notification/@samp{vStopped}
38893 sequence in progress is abandoned. The target must begin a new
38894 sequence reporting stop events for all stopped threads, whether or not
38895 it has previously reported those events to @value{GDBN}. The first
38896 stop reply is sent as a synchronous reply to the @samp{?} packet, and
38897 subsequent stop replies are sent as responses to @samp{vStopped} packets
38898 using the mechanism described above. The target must not send
38899 asynchronous stop reply notifications until the sequence is complete.
38900 If all threads are running when the target receives the @samp{?} packet,
38901 or if the target is not attached to any process, it shall respond
38904 If the stub supports non-stop mode, it should also support the
38905 @samp{swbreak} stop reason if software breakpoints are supported, and
38906 the @samp{hwbreak} stop reason if hardware breakpoints are supported
38907 (@pxref{swbreak stop reason}). This is because given the asynchronous
38908 nature of non-stop mode, between the time a thread hits a breakpoint
38909 and the time the event is finally processed by @value{GDBN}, the
38910 breakpoint may have already been removed from the target. Due to
38911 this, @value{GDBN} needs to be able to tell whether a trap stop was
38912 caused by a delayed breakpoint event, which should be ignored, as
38913 opposed to a random trap signal, which should be reported to the user.
38914 Note the @samp{swbreak} feature implies that the target is responsible
38915 for adjusting the PC when a software breakpoint triggers, if
38916 necessary, such as on the x86 architecture.
38918 @node Packet Acknowledgment
38919 @section Packet Acknowledgment
38921 @cindex acknowledgment, for @value{GDBN} remote
38922 @cindex packet acknowledgment, for @value{GDBN} remote
38923 By default, when either the host or the target machine receives a packet,
38924 the first response expected is an acknowledgment: either @samp{+} (to indicate
38925 the package was received correctly) or @samp{-} (to request retransmission).
38926 This mechanism allows the @value{GDBN} remote protocol to operate over
38927 unreliable transport mechanisms, such as a serial line.
38929 In cases where the transport mechanism is itself reliable (such as a pipe or
38930 TCP connection), the @samp{+}/@samp{-} acknowledgments are redundant.
38931 It may be desirable to disable them in that case to reduce communication
38932 overhead, or for other reasons. This can be accomplished by means of the
38933 @samp{QStartNoAckMode} packet; @pxref{QStartNoAckMode}.
38935 When in no-acknowledgment mode, neither the stub nor @value{GDBN} shall send or
38936 expect @samp{+}/@samp{-} protocol acknowledgments. The packet
38937 and response format still includes the normal checksum, as described in
38938 @ref{Overview}, but the checksum may be ignored by the receiver.
38940 If the stub supports @samp{QStartNoAckMode} and prefers to operate in
38941 no-acknowledgment mode, it should report that to @value{GDBN}
38942 by including @samp{QStartNoAckMode+} in its response to @samp{qSupported};
38943 @pxref{qSupported}.
38944 If @value{GDBN} also supports @samp{QStartNoAckMode} and it has not been
38945 disabled via the @code{set remote noack-packet off} command
38946 (@pxref{Remote Configuration}),
38947 @value{GDBN} may then send a @samp{QStartNoAckMode} packet to the stub.
38948 Only then may the stub actually turn off packet acknowledgments.
38949 @value{GDBN} sends a final @samp{+} acknowledgment of the stub's @samp{OK}
38950 response, which can be safely ignored by the stub.
38952 Note that @code{set remote noack-packet} command only affects negotiation
38953 between @value{GDBN} and the stub when subsequent connections are made;
38954 it does not affect the protocol acknowledgment state for any current
38956 Since @samp{+}/@samp{-} acknowledgments are enabled by default when a
38957 new connection is established,
38958 there is also no protocol request to re-enable the acknowledgments
38959 for the current connection, once disabled.
38964 Example sequence of a target being re-started. Notice how the restart
38965 does not get any direct output:
38970 @emph{target restarts}
38973 <- @code{T001:1234123412341234}
38977 Example sequence of a target being stepped by a single instruction:
38980 -> @code{G1445@dots{}}
38985 <- @code{T001:1234123412341234}
38989 <- @code{1455@dots{}}
38993 @node File-I/O Remote Protocol Extension
38994 @section File-I/O Remote Protocol Extension
38995 @cindex File-I/O remote protocol extension
38998 * File-I/O Overview::
38999 * Protocol Basics::
39000 * The F Request Packet::
39001 * The F Reply Packet::
39002 * The Ctrl-C Message::
39004 * List of Supported Calls::
39005 * Protocol-specific Representation of Datatypes::
39007 * File-I/O Examples::
39010 @node File-I/O Overview
39011 @subsection File-I/O Overview
39012 @cindex file-i/o overview
39014 The @dfn{File I/O remote protocol extension} (short: File-I/O) allows the
39015 target to use the host's file system and console I/O to perform various
39016 system calls. System calls on the target system are translated into a
39017 remote protocol packet to the host system, which then performs the needed
39018 actions and returns a response packet to the target system.
39019 This simulates file system operations even on targets that lack file systems.
39021 The protocol is defined to be independent of both the host and target systems.
39022 It uses its own internal representation of datatypes and values. Both
39023 @value{GDBN} and the target's @value{GDBN} stub are responsible for
39024 translating the system-dependent value representations into the internal
39025 protocol representations when data is transmitted.
39027 The communication is synchronous. A system call is possible only when
39028 @value{GDBN} is waiting for a response from the @samp{C}, @samp{c}, @samp{S}
39029 or @samp{s} packets. While @value{GDBN} handles the request for a system call,
39030 the target is stopped to allow deterministic access to the target's
39031 memory. Therefore File-I/O is not interruptible by target signals. On
39032 the other hand, it is possible to interrupt File-I/O by a user interrupt
39033 (@samp{Ctrl-C}) within @value{GDBN}.
39035 The target's request to perform a host system call does not finish
39036 the latest @samp{C}, @samp{c}, @samp{S} or @samp{s} action. That means,
39037 after finishing the system call, the target returns to continuing the
39038 previous activity (continue, step). No additional continue or step
39039 request from @value{GDBN} is required.
39042 (@value{GDBP}) continue
39043 <- target requests 'system call X'
39044 target is stopped, @value{GDBN} executes system call
39045 -> @value{GDBN} returns result
39046 ... target continues, @value{GDBN} returns to wait for the target
39047 <- target hits breakpoint and sends a Txx packet
39050 The protocol only supports I/O on the console and to regular files on
39051 the host file system. Character or block special devices, pipes,
39052 named pipes, sockets or any other communication method on the host
39053 system are not supported by this protocol.
39055 File I/O is not supported in non-stop mode.
39057 @node Protocol Basics
39058 @subsection Protocol Basics
39059 @cindex protocol basics, file-i/o
39061 The File-I/O protocol uses the @code{F} packet as the request as well
39062 as reply packet. Since a File-I/O system call can only occur when
39063 @value{GDBN} is waiting for a response from the continuing or stepping target,
39064 the File-I/O request is a reply that @value{GDBN} has to expect as a result
39065 of a previous @samp{C}, @samp{c}, @samp{S} or @samp{s} packet.
39066 This @code{F} packet contains all information needed to allow @value{GDBN}
39067 to call the appropriate host system call:
39071 A unique identifier for the requested system call.
39074 All parameters to the system call. Pointers are given as addresses
39075 in the target memory address space. Pointers to strings are given as
39076 pointer/length pair. Numerical values are given as they are.
39077 Numerical control flags are given in a protocol-specific representation.
39081 At this point, @value{GDBN} has to perform the following actions.
39085 If the parameters include pointer values to data needed as input to a
39086 system call, @value{GDBN} requests this data from the target with a
39087 standard @code{m} packet request. This additional communication has to be
39088 expected by the target implementation and is handled as any other @code{m}
39092 @value{GDBN} translates all value from protocol representation to host
39093 representation as needed. Datatypes are coerced into the host types.
39096 @value{GDBN} calls the system call.
39099 It then coerces datatypes back to protocol representation.
39102 If the system call is expected to return data in buffer space specified
39103 by pointer parameters to the call, the data is transmitted to the
39104 target using a @code{M} or @code{X} packet. This packet has to be expected
39105 by the target implementation and is handled as any other @code{M} or @code{X}
39110 Eventually @value{GDBN} replies with another @code{F} packet which contains all
39111 necessary information for the target to continue. This at least contains
39118 @code{errno}, if has been changed by the system call.
39125 After having done the needed type and value coercion, the target continues
39126 the latest continue or step action.
39128 @node The F Request Packet
39129 @subsection The @code{F} Request Packet
39130 @cindex file-i/o request packet
39131 @cindex @code{F} request packet
39133 The @code{F} request packet has the following format:
39136 @item F@var{call-id},@var{parameter@dots{}}
39138 @var{call-id} is the identifier to indicate the host system call to be called.
39139 This is just the name of the function.
39141 @var{parameter@dots{}} are the parameters to the system call.
39142 Parameters are hexadecimal integer values, either the actual values in case
39143 of scalar datatypes, pointers to target buffer space in case of compound
39144 datatypes and unspecified memory areas, or pointer/length pairs in case
39145 of string parameters. These are appended to the @var{call-id} as a
39146 comma-delimited list. All values are transmitted in ASCII
39147 string representation, pointer/length pairs separated by a slash.
39153 @node The F Reply Packet
39154 @subsection The @code{F} Reply Packet
39155 @cindex file-i/o reply packet
39156 @cindex @code{F} reply packet
39158 The @code{F} reply packet has the following format:
39162 @item F@var{retcode},@var{errno},@var{Ctrl-C flag};@var{call-specific attachment}
39164 @var{retcode} is the return code of the system call as hexadecimal value.
39166 @var{errno} is the @code{errno} set by the call, in protocol-specific
39168 This parameter can be omitted if the call was successful.
39170 @var{Ctrl-C flag} is only sent if the user requested a break. In this
39171 case, @var{errno} must be sent as well, even if the call was successful.
39172 The @var{Ctrl-C flag} itself consists of the character @samp{C}:
39179 or, if the call was interrupted before the host call has been performed:
39186 assuming 4 is the protocol-specific representation of @code{EINTR}.
39191 @node The Ctrl-C Message
39192 @subsection The @samp{Ctrl-C} Message
39193 @cindex ctrl-c message, in file-i/o protocol
39195 If the @samp{Ctrl-C} flag is set in the @value{GDBN}
39196 reply packet (@pxref{The F Reply Packet}),
39197 the target should behave as if it had
39198 gotten a break message. The meaning for the target is ``system call
39199 interrupted by @code{SIGINT}''. Consequentially, the target should actually stop
39200 (as with a break message) and return to @value{GDBN} with a @code{T02}
39203 It's important for the target to know in which
39204 state the system call was interrupted. There are two possible cases:
39208 The system call hasn't been performed on the host yet.
39211 The system call on the host has been finished.
39215 These two states can be distinguished by the target by the value of the
39216 returned @code{errno}. If it's the protocol representation of @code{EINTR}, the system
39217 call hasn't been performed. This is equivalent to the @code{EINTR} handling
39218 on POSIX systems. In any other case, the target may presume that the
39219 system call has been finished --- successfully or not --- and should behave
39220 as if the break message arrived right after the system call.
39222 @value{GDBN} must behave reliably. If the system call has not been called
39223 yet, @value{GDBN} may send the @code{F} reply immediately, setting @code{EINTR} as
39224 @code{errno} in the packet. If the system call on the host has been finished
39225 before the user requests a break, the full action must be finished by
39226 @value{GDBN}. This requires sending @code{M} or @code{X} packets as necessary.
39227 The @code{F} packet may only be sent when either nothing has happened
39228 or the full action has been completed.
39231 @subsection Console I/O
39232 @cindex console i/o as part of file-i/o
39234 By default and if not explicitly closed by the target system, the file
39235 descriptors 0, 1 and 2 are connected to the @value{GDBN} console. Output
39236 on the @value{GDBN} console is handled as any other file output operation
39237 (@code{write(1, @dots{})} or @code{write(2, @dots{})}). Console input is handled
39238 by @value{GDBN} so that after the target read request from file descriptor
39239 0 all following typing is buffered until either one of the following
39244 The user types @kbd{Ctrl-c}. The behaviour is as explained above, and the
39246 system call is treated as finished.
39249 The user presses @key{RET}. This is treated as end of input with a trailing
39253 The user types @kbd{Ctrl-d}. This is treated as end of input. No trailing
39254 character (neither newline nor @samp{Ctrl-D}) is appended to the input.
39258 If the user has typed more characters than fit in the buffer given to
39259 the @code{read} call, the trailing characters are buffered in @value{GDBN} until
39260 either another @code{read(0, @dots{})} is requested by the target, or debugging
39261 is stopped at the user's request.
39264 @node List of Supported Calls
39265 @subsection List of Supported Calls
39266 @cindex list of supported file-i/o calls
39283 @unnumberedsubsubsec open
39284 @cindex open, file-i/o system call
39289 int open(const char *pathname, int flags);
39290 int open(const char *pathname, int flags, mode_t mode);
39294 @samp{Fopen,@var{pathptr}/@var{len},@var{flags},@var{mode}}
39297 @var{flags} is the bitwise @code{OR} of the following values:
39301 If the file does not exist it will be created. The host
39302 rules apply as far as file ownership and time stamps
39306 When used with @code{O_CREAT}, if the file already exists it is
39307 an error and open() fails.
39310 If the file already exists and the open mode allows
39311 writing (@code{O_RDWR} or @code{O_WRONLY} is given) it will be
39312 truncated to zero length.
39315 The file is opened in append mode.
39318 The file is opened for reading only.
39321 The file is opened for writing only.
39324 The file is opened for reading and writing.
39328 Other bits are silently ignored.
39332 @var{mode} is the bitwise @code{OR} of the following values:
39336 User has read permission.
39339 User has write permission.
39342 Group has read permission.
39345 Group has write permission.
39348 Others have read permission.
39351 Others have write permission.
39355 Other bits are silently ignored.
39358 @item Return value:
39359 @code{open} returns the new file descriptor or -1 if an error
39366 @var{pathname} already exists and @code{O_CREAT} and @code{O_EXCL} were used.
39369 @var{pathname} refers to a directory.
39372 The requested access is not allowed.
39375 @var{pathname} was too long.
39378 A directory component in @var{pathname} does not exist.
39381 @var{pathname} refers to a device, pipe, named pipe or socket.
39384 @var{pathname} refers to a file on a read-only filesystem and
39385 write access was requested.
39388 @var{pathname} is an invalid pointer value.
39391 No space on device to create the file.
39394 The process already has the maximum number of files open.
39397 The limit on the total number of files open on the system
39401 The call was interrupted by the user.
39407 @unnumberedsubsubsec close
39408 @cindex close, file-i/o system call
39417 @samp{Fclose,@var{fd}}
39419 @item Return value:
39420 @code{close} returns zero on success, or -1 if an error occurred.
39426 @var{fd} isn't a valid open file descriptor.
39429 The call was interrupted by the user.
39435 @unnumberedsubsubsec read
39436 @cindex read, file-i/o system call
39441 int read(int fd, void *buf, unsigned int count);
39445 @samp{Fread,@var{fd},@var{bufptr},@var{count}}
39447 @item Return value:
39448 On success, the number of bytes read is returned.
39449 Zero indicates end of file. If count is zero, read
39450 returns zero as well. On error, -1 is returned.
39456 @var{fd} is not a valid file descriptor or is not open for
39460 @var{bufptr} is an invalid pointer value.
39463 The call was interrupted by the user.
39469 @unnumberedsubsubsec write
39470 @cindex write, file-i/o system call
39475 int write(int fd, const void *buf, unsigned int count);
39479 @samp{Fwrite,@var{fd},@var{bufptr},@var{count}}
39481 @item Return value:
39482 On success, the number of bytes written are returned.
39483 Zero indicates nothing was written. On error, -1
39490 @var{fd} is not a valid file descriptor or is not open for
39494 @var{bufptr} is an invalid pointer value.
39497 An attempt was made to write a file that exceeds the
39498 host-specific maximum file size allowed.
39501 No space on device to write the data.
39504 The call was interrupted by the user.
39510 @unnumberedsubsubsec lseek
39511 @cindex lseek, file-i/o system call
39516 long lseek (int fd, long offset, int flag);
39520 @samp{Flseek,@var{fd},@var{offset},@var{flag}}
39522 @var{flag} is one of:
39526 The offset is set to @var{offset} bytes.
39529 The offset is set to its current location plus @var{offset}
39533 The offset is set to the size of the file plus @var{offset}
39537 @item Return value:
39538 On success, the resulting unsigned offset in bytes from
39539 the beginning of the file is returned. Otherwise, a
39540 value of -1 is returned.
39546 @var{fd} is not a valid open file descriptor.
39549 @var{fd} is associated with the @value{GDBN} console.
39552 @var{flag} is not a proper value.
39555 The call was interrupted by the user.
39561 @unnumberedsubsubsec rename
39562 @cindex rename, file-i/o system call
39567 int rename(const char *oldpath, const char *newpath);
39571 @samp{Frename,@var{oldpathptr}/@var{len},@var{newpathptr}/@var{len}}
39573 @item Return value:
39574 On success, zero is returned. On error, -1 is returned.
39580 @var{newpath} is an existing directory, but @var{oldpath} is not a
39584 @var{newpath} is a non-empty directory.
39587 @var{oldpath} or @var{newpath} is a directory that is in use by some
39591 An attempt was made to make a directory a subdirectory
39595 A component used as a directory in @var{oldpath} or new
39596 path is not a directory. Or @var{oldpath} is a directory
39597 and @var{newpath} exists but is not a directory.
39600 @var{oldpathptr} or @var{newpathptr} are invalid pointer values.
39603 No access to the file or the path of the file.
39607 @var{oldpath} or @var{newpath} was too long.
39610 A directory component in @var{oldpath} or @var{newpath} does not exist.
39613 The file is on a read-only filesystem.
39616 The device containing the file has no room for the new
39620 The call was interrupted by the user.
39626 @unnumberedsubsubsec unlink
39627 @cindex unlink, file-i/o system call
39632 int unlink(const char *pathname);
39636 @samp{Funlink,@var{pathnameptr}/@var{len}}
39638 @item Return value:
39639 On success, zero is returned. On error, -1 is returned.
39645 No access to the file or the path of the file.
39648 The system does not allow unlinking of directories.
39651 The file @var{pathname} cannot be unlinked because it's
39652 being used by another process.
39655 @var{pathnameptr} is an invalid pointer value.
39658 @var{pathname} was too long.
39661 A directory component in @var{pathname} does not exist.
39664 A component of the path is not a directory.
39667 The file is on a read-only filesystem.
39670 The call was interrupted by the user.
39676 @unnumberedsubsubsec stat/fstat
39677 @cindex fstat, file-i/o system call
39678 @cindex stat, file-i/o system call
39683 int stat(const char *pathname, struct stat *buf);
39684 int fstat(int fd, struct stat *buf);
39688 @samp{Fstat,@var{pathnameptr}/@var{len},@var{bufptr}}@*
39689 @samp{Ffstat,@var{fd},@var{bufptr}}
39691 @item Return value:
39692 On success, zero is returned. On error, -1 is returned.
39698 @var{fd} is not a valid open file.
39701 A directory component in @var{pathname} does not exist or the
39702 path is an empty string.
39705 A component of the path is not a directory.
39708 @var{pathnameptr} is an invalid pointer value.
39711 No access to the file or the path of the file.
39714 @var{pathname} was too long.
39717 The call was interrupted by the user.
39723 @unnumberedsubsubsec gettimeofday
39724 @cindex gettimeofday, file-i/o system call
39729 int gettimeofday(struct timeval *tv, void *tz);
39733 @samp{Fgettimeofday,@var{tvptr},@var{tzptr}}
39735 @item Return value:
39736 On success, 0 is returned, -1 otherwise.
39742 @var{tz} is a non-NULL pointer.
39745 @var{tvptr} and/or @var{tzptr} is an invalid pointer value.
39751 @unnumberedsubsubsec isatty
39752 @cindex isatty, file-i/o system call
39757 int isatty(int fd);
39761 @samp{Fisatty,@var{fd}}
39763 @item Return value:
39764 Returns 1 if @var{fd} refers to the @value{GDBN} console, 0 otherwise.
39770 The call was interrupted by the user.
39775 Note that the @code{isatty} call is treated as a special case: it returns
39776 1 to the target if the file descriptor is attached
39777 to the @value{GDBN} console, 0 otherwise. Implementing through system calls
39778 would require implementing @code{ioctl} and would be more complex than
39783 @unnumberedsubsubsec system
39784 @cindex system, file-i/o system call
39789 int system(const char *command);
39793 @samp{Fsystem,@var{commandptr}/@var{len}}
39795 @item Return value:
39796 If @var{len} is zero, the return value indicates whether a shell is
39797 available. A zero return value indicates a shell is not available.
39798 For non-zero @var{len}, the value returned is -1 on error and the
39799 return status of the command otherwise. Only the exit status of the
39800 command is returned, which is extracted from the host's @code{system}
39801 return value by calling @code{WEXITSTATUS(retval)}. In case
39802 @file{/bin/sh} could not be executed, 127 is returned.
39808 The call was interrupted by the user.
39813 @value{GDBN} takes over the full task of calling the necessary host calls
39814 to perform the @code{system} call. The return value of @code{system} on
39815 the host is simplified before it's returned
39816 to the target. Any termination signal information from the child process
39817 is discarded, and the return value consists
39818 entirely of the exit status of the called command.
39820 Due to security concerns, the @code{system} call is by default refused
39821 by @value{GDBN}. The user has to allow this call explicitly with the
39822 @code{set remote system-call-allowed 1} command.
39825 @item set remote system-call-allowed
39826 @kindex set remote system-call-allowed
39827 Control whether to allow the @code{system} calls in the File I/O
39828 protocol for the remote target. The default is zero (disabled).
39830 @item show remote system-call-allowed
39831 @kindex show remote system-call-allowed
39832 Show whether the @code{system} calls are allowed in the File I/O
39836 @node Protocol-specific Representation of Datatypes
39837 @subsection Protocol-specific Representation of Datatypes
39838 @cindex protocol-specific representation of datatypes, in file-i/o protocol
39841 * Integral Datatypes::
39843 * Memory Transfer::
39848 @node Integral Datatypes
39849 @unnumberedsubsubsec Integral Datatypes
39850 @cindex integral datatypes, in file-i/o protocol
39852 The integral datatypes used in the system calls are @code{int},
39853 @code{unsigned int}, @code{long}, @code{unsigned long},
39854 @code{mode_t}, and @code{time_t}.
39856 @code{int}, @code{unsigned int}, @code{mode_t} and @code{time_t} are
39857 implemented as 32 bit values in this protocol.
39859 @code{long} and @code{unsigned long} are implemented as 64 bit types.
39861 @xref{Limits}, for corresponding MIN and MAX values (similar to those
39862 in @file{limits.h}) to allow range checking on host and target.
39864 @code{time_t} datatypes are defined as seconds since the Epoch.
39866 All integral datatypes transferred as part of a memory read or write of a
39867 structured datatype e.g.@: a @code{struct stat} have to be given in big endian
39870 @node Pointer Values
39871 @unnumberedsubsubsec Pointer Values
39872 @cindex pointer values, in file-i/o protocol
39874 Pointers to target data are transmitted as they are. An exception
39875 is made for pointers to buffers for which the length isn't
39876 transmitted as part of the function call, namely strings. Strings
39877 are transmitted as a pointer/length pair, both as hex values, e.g.@:
39884 which is a pointer to data of length 18 bytes at position 0x1aaf.
39885 The length is defined as the full string length in bytes, including
39886 the trailing null byte. For example, the string @code{"hello world"}
39887 at address 0x123456 is transmitted as
39893 @node Memory Transfer
39894 @unnumberedsubsubsec Memory Transfer
39895 @cindex memory transfer, in file-i/o protocol
39897 Structured data which is transferred using a memory read or write (for
39898 example, a @code{struct stat}) is expected to be in a protocol-specific format
39899 with all scalar multibyte datatypes being big endian. Translation to
39900 this representation needs to be done both by the target before the @code{F}
39901 packet is sent, and by @value{GDBN} before
39902 it transfers memory to the target. Transferred pointers to structured
39903 data should point to the already-coerced data at any time.
39907 @unnumberedsubsubsec struct stat
39908 @cindex struct stat, in file-i/o protocol
39910 The buffer of type @code{struct stat} used by the target and @value{GDBN}
39911 is defined as follows:
39915 unsigned int st_dev; /* device */
39916 unsigned int st_ino; /* inode */
39917 mode_t st_mode; /* protection */
39918 unsigned int st_nlink; /* number of hard links */
39919 unsigned int st_uid; /* user ID of owner */
39920 unsigned int st_gid; /* group ID of owner */
39921 unsigned int st_rdev; /* device type (if inode device) */
39922 unsigned long st_size; /* total size, in bytes */
39923 unsigned long st_blksize; /* blocksize for filesystem I/O */
39924 unsigned long st_blocks; /* number of blocks allocated */
39925 time_t st_atime; /* time of last access */
39926 time_t st_mtime; /* time of last modification */
39927 time_t st_ctime; /* time of last change */
39931 The integral datatypes conform to the definitions given in the
39932 appropriate section (see @ref{Integral Datatypes}, for details) so this
39933 structure is of size 64 bytes.
39935 The values of several fields have a restricted meaning and/or
39941 A value of 0 represents a file, 1 the console.
39944 No valid meaning for the target. Transmitted unchanged.
39947 Valid mode bits are described in @ref{Constants}. Any other
39948 bits have currently no meaning for the target.
39953 No valid meaning for the target. Transmitted unchanged.
39958 These values have a host and file system dependent
39959 accuracy. Especially on Windows hosts, the file system may not
39960 support exact timing values.
39963 The target gets a @code{struct stat} of the above representation and is
39964 responsible for coercing it to the target representation before
39967 Note that due to size differences between the host, target, and protocol
39968 representations of @code{struct stat} members, these members could eventually
39969 get truncated on the target.
39971 @node struct timeval
39972 @unnumberedsubsubsec struct timeval
39973 @cindex struct timeval, in file-i/o protocol
39975 The buffer of type @code{struct timeval} used by the File-I/O protocol
39976 is defined as follows:
39980 time_t tv_sec; /* second */
39981 long tv_usec; /* microsecond */
39985 The integral datatypes conform to the definitions given in the
39986 appropriate section (see @ref{Integral Datatypes}, for details) so this
39987 structure is of size 8 bytes.
39990 @subsection Constants
39991 @cindex constants, in file-i/o protocol
39993 The following values are used for the constants inside of the
39994 protocol. @value{GDBN} and target are responsible for translating these
39995 values before and after the call as needed.
40006 @unnumberedsubsubsec Open Flags
40007 @cindex open flags, in file-i/o protocol
40009 All values are given in hexadecimal representation.
40021 @node mode_t Values
40022 @unnumberedsubsubsec mode_t Values
40023 @cindex mode_t values, in file-i/o protocol
40025 All values are given in octal representation.
40042 @unnumberedsubsubsec Errno Values
40043 @cindex errno values, in file-i/o protocol
40045 All values are given in decimal representation.
40070 @code{EUNKNOWN} is used as a fallback error value if a host system returns
40071 any error value not in the list of supported error numbers.
40074 @unnumberedsubsubsec Lseek Flags
40075 @cindex lseek flags, in file-i/o protocol
40084 @unnumberedsubsubsec Limits
40085 @cindex limits, in file-i/o protocol
40087 All values are given in decimal representation.
40090 INT_MIN -2147483648
40092 UINT_MAX 4294967295
40093 LONG_MIN -9223372036854775808
40094 LONG_MAX 9223372036854775807
40095 ULONG_MAX 18446744073709551615
40098 @node File-I/O Examples
40099 @subsection File-I/O Examples
40100 @cindex file-i/o examples
40102 Example sequence of a write call, file descriptor 3, buffer is at target
40103 address 0x1234, 6 bytes should be written:
40106 <- @code{Fwrite,3,1234,6}
40107 @emph{request memory read from target}
40110 @emph{return "6 bytes written"}
40114 Example sequence of a read call, file descriptor 3, buffer is at target
40115 address 0x1234, 6 bytes should be read:
40118 <- @code{Fread,3,1234,6}
40119 @emph{request memory write to target}
40120 -> @code{X1234,6:XXXXXX}
40121 @emph{return "6 bytes read"}
40125 Example sequence of a read call, call fails on the host due to invalid
40126 file descriptor (@code{EBADF}):
40129 <- @code{Fread,3,1234,6}
40133 Example sequence of a read call, user presses @kbd{Ctrl-c} before syscall on
40137 <- @code{Fread,3,1234,6}
40142 Example sequence of a read call, user presses @kbd{Ctrl-c} after syscall on
40146 <- @code{Fread,3,1234,6}
40147 -> @code{X1234,6:XXXXXX}
40151 @node Library List Format
40152 @section Library List Format
40153 @cindex library list format, remote protocol
40155 On some platforms, a dynamic loader (e.g.@: @file{ld.so}) runs in the
40156 same process as your application to manage libraries. In this case,
40157 @value{GDBN} can use the loader's symbol table and normal memory
40158 operations to maintain a list of shared libraries. On other
40159 platforms, the operating system manages loaded libraries.
40160 @value{GDBN} can not retrieve the list of currently loaded libraries
40161 through memory operations, so it uses the @samp{qXfer:libraries:read}
40162 packet (@pxref{qXfer library list read}) instead. The remote stub
40163 queries the target's operating system and reports which libraries
40166 The @samp{qXfer:libraries:read} packet returns an XML document which
40167 lists loaded libraries and their offsets. Each library has an
40168 associated name and one or more segment or section base addresses,
40169 which report where the library was loaded in memory.
40171 For the common case of libraries that are fully linked binaries, the
40172 library should have a list of segments. If the target supports
40173 dynamic linking of a relocatable object file, its library XML element
40174 should instead include a list of allocated sections. The segment or
40175 section bases are start addresses, not relocation offsets; they do not
40176 depend on the library's link-time base addresses.
40178 @value{GDBN} must be linked with the Expat library to support XML
40179 library lists. @xref{Expat}.
40181 A simple memory map, with one loaded library relocated by a single
40182 offset, looks like this:
40186 <library name="/lib/libc.so.6">
40187 <segment address="0x10000000"/>
40192 Another simple memory map, with one loaded library with three
40193 allocated sections (.text, .data, .bss), looks like this:
40197 <library name="sharedlib.o">
40198 <section address="0x10000000"/>
40199 <section address="0x20000000"/>
40200 <section address="0x30000000"/>
40205 The format of a library list is described by this DTD:
40208 <!-- library-list: Root element with versioning -->
40209 <!ELEMENT library-list (library)*>
40210 <!ATTLIST library-list version CDATA #FIXED "1.0">
40211 <!ELEMENT library (segment*, section*)>
40212 <!ATTLIST library name CDATA #REQUIRED>
40213 <!ELEMENT segment EMPTY>
40214 <!ATTLIST segment address CDATA #REQUIRED>
40215 <!ELEMENT section EMPTY>
40216 <!ATTLIST section address CDATA #REQUIRED>
40219 In addition, segments and section descriptors cannot be mixed within a
40220 single library element, and you must supply at least one segment or
40221 section for each library.
40223 @node Library List Format for SVR4 Targets
40224 @section Library List Format for SVR4 Targets
40225 @cindex library list format, remote protocol
40227 On SVR4 platforms @value{GDBN} can use the symbol table of a dynamic loader
40228 (e.g.@: @file{ld.so}) and normal memory operations to maintain a list of
40229 shared libraries. Still a special library list provided by this packet is
40230 more efficient for the @value{GDBN} remote protocol.
40232 The @samp{qXfer:libraries-svr4:read} packet returns an XML document which lists
40233 loaded libraries and their SVR4 linker parameters. For each library on SVR4
40234 target, the following parameters are reported:
40238 @code{name}, the absolute file name from the @code{l_name} field of
40239 @code{struct link_map}.
40241 @code{lm} with address of @code{struct link_map} used for TLS
40242 (Thread Local Storage) access.
40244 @code{l_addr}, the displacement as read from the field @code{l_addr} of
40245 @code{struct link_map}. For prelinked libraries this is not an absolute
40246 memory address. It is a displacement of absolute memory address against
40247 address the file was prelinked to during the library load.
40249 @code{l_ld}, which is memory address of the @code{PT_DYNAMIC} segment
40252 Additionally the single @code{main-lm} attribute specifies address of
40253 @code{struct link_map} used for the main executable. This parameter is used
40254 for TLS access and its presence is optional.
40256 @value{GDBN} must be linked with the Expat library to support XML
40257 SVR4 library lists. @xref{Expat}.
40259 A simple memory map, with two loaded libraries (which do not use prelink),
40263 <library-list-svr4 version="1.0" main-lm="0xe4f8f8">
40264 <library name="/lib/ld-linux.so.2" lm="0xe4f51c" l_addr="0xe2d000"
40266 <library name="/lib/libc.so.6" lm="0xe4fbe8" l_addr="0x154000"
40268 </library-list-svr>
40271 The format of an SVR4 library list is described by this DTD:
40274 <!-- library-list-svr4: Root element with versioning -->
40275 <!ELEMENT library-list-svr4 (library)*>
40276 <!ATTLIST library-list-svr4 version CDATA #FIXED "1.0">
40277 <!ATTLIST library-list-svr4 main-lm CDATA #IMPLIED>
40278 <!ELEMENT library EMPTY>
40279 <!ATTLIST library name CDATA #REQUIRED>
40280 <!ATTLIST library lm CDATA #REQUIRED>
40281 <!ATTLIST library l_addr CDATA #REQUIRED>
40282 <!ATTLIST library l_ld CDATA #REQUIRED>
40285 @node Memory Map Format
40286 @section Memory Map Format
40287 @cindex memory map format
40289 To be able to write into flash memory, @value{GDBN} needs to obtain a
40290 memory map from the target. This section describes the format of the
40293 The memory map is obtained using the @samp{qXfer:memory-map:read}
40294 (@pxref{qXfer memory map read}) packet and is an XML document that
40295 lists memory regions.
40297 @value{GDBN} must be linked with the Expat library to support XML
40298 memory maps. @xref{Expat}.
40300 The top-level structure of the document is shown below:
40303 <?xml version="1.0"?>
40304 <!DOCTYPE memory-map
40305 PUBLIC "+//IDN gnu.org//DTD GDB Memory Map V1.0//EN"
40306 "http://sourceware.org/gdb/gdb-memory-map.dtd">
40312 Each region can be either:
40317 A region of RAM starting at @var{addr} and extending for @var{length}
40321 <memory type="ram" start="@var{addr}" length="@var{length}"/>
40326 A region of read-only memory:
40329 <memory type="rom" start="@var{addr}" length="@var{length}"/>
40334 A region of flash memory, with erasure blocks @var{blocksize}
40338 <memory type="flash" start="@var{addr}" length="@var{length}">
40339 <property name="blocksize">@var{blocksize}</property>
40345 Regions must not overlap. @value{GDBN} assumes that areas of memory not covered
40346 by the memory map are RAM, and uses the ordinary @samp{M} and @samp{X}
40347 packets to write to addresses in such ranges.
40349 The formal DTD for memory map format is given below:
40352 <!-- ................................................... -->
40353 <!-- Memory Map XML DTD ................................ -->
40354 <!-- File: memory-map.dtd .............................. -->
40355 <!-- .................................... .............. -->
40356 <!-- memory-map.dtd -->
40357 <!-- memory-map: Root element with versioning -->
40358 <!ELEMENT memory-map (memory | property)>
40359 <!ATTLIST memory-map version CDATA #FIXED "1.0.0">
40360 <!ELEMENT memory (property)>
40361 <!-- memory: Specifies a memory region,
40362 and its type, or device. -->
40363 <!ATTLIST memory type CDATA #REQUIRED
40364 start CDATA #REQUIRED
40365 length CDATA #REQUIRED
40366 device CDATA #IMPLIED>
40367 <!-- property: Generic attribute tag -->
40368 <!ELEMENT property (#PCDATA | property)*>
40369 <!ATTLIST property name CDATA #REQUIRED>
40372 @node Thread List Format
40373 @section Thread List Format
40374 @cindex thread list format
40376 To efficiently update the list of threads and their attributes,
40377 @value{GDBN} issues the @samp{qXfer:threads:read} packet
40378 (@pxref{qXfer threads read}) and obtains the XML document with
40379 the following structure:
40382 <?xml version="1.0"?>
40384 <thread id="id" core="0" name="name">
40385 ... description ...
40390 Each @samp{thread} element must have the @samp{id} attribute that
40391 identifies the thread (@pxref{thread-id syntax}). The
40392 @samp{core} attribute, if present, specifies which processor core
40393 the thread was last executing on. The @samp{name} attribute, if
40394 present, specifies the human-readable name of the thread. The content
40395 of the of @samp{thread} element is interpreted as human-readable
40396 auxiliary information.
40398 @node Traceframe Info Format
40399 @section Traceframe Info Format
40400 @cindex traceframe info format
40402 To be able to know which objects in the inferior can be examined when
40403 inspecting a tracepoint hit, @value{GDBN} needs to obtain the list of
40404 memory ranges, registers and trace state variables that have been
40405 collected in a traceframe.
40407 This list is obtained using the @samp{qXfer:traceframe-info:read}
40408 (@pxref{qXfer traceframe info read}) packet and is an XML document.
40410 @value{GDBN} must be linked with the Expat library to support XML
40411 traceframe info discovery. @xref{Expat}.
40413 The top-level structure of the document is shown below:
40416 <?xml version="1.0"?>
40417 <!DOCTYPE traceframe-info
40418 PUBLIC "+//IDN gnu.org//DTD GDB Memory Map V1.0//EN"
40419 "http://sourceware.org/gdb/gdb-traceframe-info.dtd">
40425 Each traceframe block can be either:
40430 A region of collected memory starting at @var{addr} and extending for
40431 @var{length} bytes from there:
40434 <memory start="@var{addr}" length="@var{length}"/>
40438 A block indicating trace state variable numbered @var{number} has been
40442 <tvar id="@var{number}"/>
40447 The formal DTD for the traceframe info format is given below:
40450 <!ELEMENT traceframe-info (memory | tvar)* >
40451 <!ATTLIST traceframe-info version CDATA #FIXED "1.0">
40453 <!ELEMENT memory EMPTY>
40454 <!ATTLIST memory start CDATA #REQUIRED
40455 length CDATA #REQUIRED>
40457 <!ATTLIST tvar id CDATA #REQUIRED>
40460 @node Branch Trace Format
40461 @section Branch Trace Format
40462 @cindex branch trace format
40464 In order to display the branch trace of an inferior thread,
40465 @value{GDBN} needs to obtain the list of branches. This list is
40466 represented as list of sequential code blocks that are connected via
40467 branches. The code in each block has been executed sequentially.
40469 This list is obtained using the @samp{qXfer:btrace:read}
40470 (@pxref{qXfer btrace read}) packet and is an XML document.
40472 @value{GDBN} must be linked with the Expat library to support XML
40473 traceframe info discovery. @xref{Expat}.
40475 The top-level structure of the document is shown below:
40478 <?xml version="1.0"?>
40480 PUBLIC "+//IDN gnu.org//DTD GDB Branch Trace V1.0//EN"
40481 "http://sourceware.org/gdb/gdb-btrace.dtd">
40490 A block of sequentially executed instructions starting at @var{begin}
40491 and ending at @var{end}:
40494 <block begin="@var{begin}" end="@var{end}"/>
40499 The formal DTD for the branch trace format is given below:
40502 <!ELEMENT btrace (block* | pt) >
40503 <!ATTLIST btrace version CDATA #FIXED "1.0">
40505 <!ELEMENT block EMPTY>
40506 <!ATTLIST block begin CDATA #REQUIRED
40507 end CDATA #REQUIRED>
40509 <!ELEMENT pt (pt-config?, raw?)>
40511 <!ELEMENT pt-config (cpu?)>
40513 <!ELEMENT cpu EMPTY>
40514 <!ATTLIST cpu vendor CDATA #REQUIRED
40515 family CDATA #REQUIRED
40516 model CDATA #REQUIRED
40517 stepping CDATA #REQUIRED>
40519 <!ELEMENT raw (#PCDATA)>
40522 @node Branch Trace Configuration Format
40523 @section Branch Trace Configuration Format
40524 @cindex branch trace configuration format
40526 For each inferior thread, @value{GDBN} can obtain the branch trace
40527 configuration using the @samp{qXfer:btrace-conf:read}
40528 (@pxref{qXfer btrace-conf read}) packet.
40530 The configuration describes the branch trace format and configuration
40531 settings for that format. The following information is described:
40535 This thread uses the @dfn{Branch Trace Store} (@acronym{BTS}) format.
40538 The size of the @acronym{BTS} ring buffer in bytes.
40541 This thread uses the @dfn{Intel Processor Trace} (@acronym{Intel
40545 The size of the @acronym{Intel PT} ring buffer in bytes.
40549 @value{GDBN} must be linked with the Expat library to support XML
40550 branch trace configuration discovery. @xref{Expat}.
40552 The formal DTD for the branch trace configuration format is given below:
40555 <!ELEMENT btrace-conf (bts?, pt?)>
40556 <!ATTLIST btrace-conf version CDATA #FIXED "1.0">
40558 <!ELEMENT bts EMPTY>
40559 <!ATTLIST bts size CDATA #IMPLIED>
40561 <!ELEMENT pt EMPTY>
40562 <!ATTLIST pt size CDATA #IMPLIED>
40565 @include agentexpr.texi
40567 @node Target Descriptions
40568 @appendix Target Descriptions
40569 @cindex target descriptions
40571 One of the challenges of using @value{GDBN} to debug embedded systems
40572 is that there are so many minor variants of each processor
40573 architecture in use. It is common practice for vendors to start with
40574 a standard processor core --- ARM, PowerPC, or @acronym{MIPS}, for example ---
40575 and then make changes to adapt it to a particular market niche. Some
40576 architectures have hundreds of variants, available from dozens of
40577 vendors. This leads to a number of problems:
40581 With so many different customized processors, it is difficult for
40582 the @value{GDBN} maintainers to keep up with the changes.
40584 Since individual variants may have short lifetimes or limited
40585 audiences, it may not be worthwhile to carry information about every
40586 variant in the @value{GDBN} source tree.
40588 When @value{GDBN} does support the architecture of the embedded system
40589 at hand, the task of finding the correct architecture name to give the
40590 @command{set architecture} command can be error-prone.
40593 To address these problems, the @value{GDBN} remote protocol allows a
40594 target system to not only identify itself to @value{GDBN}, but to
40595 actually describe its own features. This lets @value{GDBN} support
40596 processor variants it has never seen before --- to the extent that the
40597 descriptions are accurate, and that @value{GDBN} understands them.
40599 @value{GDBN} must be linked with the Expat library to support XML
40600 target descriptions. @xref{Expat}.
40603 * Retrieving Descriptions:: How descriptions are fetched from a target.
40604 * Target Description Format:: The contents of a target description.
40605 * Predefined Target Types:: Standard types available for target
40607 * Enum Target Types:: How to define enum target types.
40608 * Standard Target Features:: Features @value{GDBN} knows about.
40611 @node Retrieving Descriptions
40612 @section Retrieving Descriptions
40614 Target descriptions can be read from the target automatically, or
40615 specified by the user manually. The default behavior is to read the
40616 description from the target. @value{GDBN} retrieves it via the remote
40617 protocol using @samp{qXfer} requests (@pxref{General Query Packets,
40618 qXfer}). The @var{annex} in the @samp{qXfer} packet will be
40619 @samp{target.xml}. The contents of the @samp{target.xml} annex are an
40620 XML document, of the form described in @ref{Target Description
40623 Alternatively, you can specify a file to read for the target description.
40624 If a file is set, the target will not be queried. The commands to
40625 specify a file are:
40628 @cindex set tdesc filename
40629 @item set tdesc filename @var{path}
40630 Read the target description from @var{path}.
40632 @cindex unset tdesc filename
40633 @item unset tdesc filename
40634 Do not read the XML target description from a file. @value{GDBN}
40635 will use the description supplied by the current target.
40637 @cindex show tdesc filename
40638 @item show tdesc filename
40639 Show the filename to read for a target description, if any.
40643 @node Target Description Format
40644 @section Target Description Format
40645 @cindex target descriptions, XML format
40647 A target description annex is an @uref{http://www.w3.org/XML/, XML}
40648 document which complies with the Document Type Definition provided in
40649 the @value{GDBN} sources in @file{gdb/features/gdb-target.dtd}. This
40650 means you can use generally available tools like @command{xmllint} to
40651 check that your feature descriptions are well-formed and valid.
40652 However, to help people unfamiliar with XML write descriptions for
40653 their targets, we also describe the grammar here.
40655 Target descriptions can identify the architecture of the remote target
40656 and (for some architectures) provide information about custom register
40657 sets. They can also identify the OS ABI of the remote target.
40658 @value{GDBN} can use this information to autoconfigure for your
40659 target, or to warn you if you connect to an unsupported target.
40661 Here is a simple target description:
40664 <target version="1.0">
40665 <architecture>i386:x86-64</architecture>
40670 This minimal description only says that the target uses
40671 the x86-64 architecture.
40673 A target description has the following overall form, with [ ] marking
40674 optional elements and @dots{} marking repeatable elements. The elements
40675 are explained further below.
40678 <?xml version="1.0"?>
40679 <!DOCTYPE target SYSTEM "gdb-target.dtd">
40680 <target version="1.0">
40681 @r{[}@var{architecture}@r{]}
40682 @r{[}@var{osabi}@r{]}
40683 @r{[}@var{compatible}@r{]}
40684 @r{[}@var{feature}@dots{}@r{]}
40689 The description is generally insensitive to whitespace and line
40690 breaks, under the usual common-sense rules. The XML version
40691 declaration and document type declaration can generally be omitted
40692 (@value{GDBN} does not require them), but specifying them may be
40693 useful for XML validation tools. The @samp{version} attribute for
40694 @samp{<target>} may also be omitted, but we recommend
40695 including it; if future versions of @value{GDBN} use an incompatible
40696 revision of @file{gdb-target.dtd}, they will detect and report
40697 the version mismatch.
40699 @subsection Inclusion
40700 @cindex target descriptions, inclusion
40703 @cindex <xi:include>
40706 It can sometimes be valuable to split a target description up into
40707 several different annexes, either for organizational purposes, or to
40708 share files between different possible target descriptions. You can
40709 divide a description into multiple files by replacing any element of
40710 the target description with an inclusion directive of the form:
40713 <xi:include href="@var{document}"/>
40717 When @value{GDBN} encounters an element of this form, it will retrieve
40718 the named XML @var{document}, and replace the inclusion directive with
40719 the contents of that document. If the current description was read
40720 using @samp{qXfer}, then so will be the included document;
40721 @var{document} will be interpreted as the name of an annex. If the
40722 current description was read from a file, @value{GDBN} will look for
40723 @var{document} as a file in the same directory where it found the
40724 original description.
40726 @subsection Architecture
40727 @cindex <architecture>
40729 An @samp{<architecture>} element has this form:
40732 <architecture>@var{arch}</architecture>
40735 @var{arch} is one of the architectures from the set accepted by
40736 @code{set architecture} (@pxref{Targets, ,Specifying a Debugging Target}).
40739 @cindex @code{<osabi>}
40741 This optional field was introduced in @value{GDBN} version 7.0.
40742 Previous versions of @value{GDBN} ignore it.
40744 An @samp{<osabi>} element has this form:
40747 <osabi>@var{abi-name}</osabi>
40750 @var{abi-name} is an OS ABI name from the same selection accepted by
40751 @w{@code{set osabi}} (@pxref{ABI, ,Configuring the Current ABI}).
40753 @subsection Compatible Architecture
40754 @cindex @code{<compatible>}
40756 This optional field was introduced in @value{GDBN} version 7.0.
40757 Previous versions of @value{GDBN} ignore it.
40759 A @samp{<compatible>} element has this form:
40762 <compatible>@var{arch}</compatible>
40765 @var{arch} is one of the architectures from the set accepted by
40766 @code{set architecture} (@pxref{Targets, ,Specifying a Debugging Target}).
40768 A @samp{<compatible>} element is used to specify that the target
40769 is able to run binaries in some other than the main target architecture
40770 given by the @samp{<architecture>} element. For example, on the
40771 Cell Broadband Engine, the main architecture is @code{powerpc:common}
40772 or @code{powerpc:common64}, but the system is able to run binaries
40773 in the @code{spu} architecture as well. The way to describe this
40774 capability with @samp{<compatible>} is as follows:
40777 <architecture>powerpc:common</architecture>
40778 <compatible>spu</compatible>
40781 @subsection Features
40784 Each @samp{<feature>} describes some logical portion of the target
40785 system. Features are currently used to describe available CPU
40786 registers and the types of their contents. A @samp{<feature>} element
40790 <feature name="@var{name}">
40791 @r{[}@var{type}@dots{}@r{]}
40797 Each feature's name should be unique within the description. The name
40798 of a feature does not matter unless @value{GDBN} has some special
40799 knowledge of the contents of that feature; if it does, the feature
40800 should have its standard name. @xref{Standard Target Features}.
40804 Any register's value is a collection of bits which @value{GDBN} must
40805 interpret. The default interpretation is a two's complement integer,
40806 but other types can be requested by name in the register description.
40807 Some predefined types are provided by @value{GDBN} (@pxref{Predefined
40808 Target Types}), and the description can define additional composite
40811 Each type element must have an @samp{id} attribute, which gives
40812 a unique (within the containing @samp{<feature>}) name to the type.
40813 Types must be defined before they are used.
40816 Some targets offer vector registers, which can be treated as arrays
40817 of scalar elements. These types are written as @samp{<vector>} elements,
40818 specifying the array element type, @var{type}, and the number of elements,
40822 <vector id="@var{id}" type="@var{type}" count="@var{count}"/>
40826 If a register's value is usefully viewed in multiple ways, define it
40827 with a union type containing the useful representations. The
40828 @samp{<union>} element contains one or more @samp{<field>} elements,
40829 each of which has a @var{name} and a @var{type}:
40832 <union id="@var{id}">
40833 <field name="@var{name}" type="@var{type}"/>
40840 If a register's value is composed from several separate values, define
40841 it with either a structure type or a flags type.
40842 A flags type may only contain bitfields.
40843 A structure type may either contain only bitfields or contain no bitfields.
40844 If the value contains only bitfields, its total size in bytes must be
40847 Non-bitfield values have a @var{name} and @var{type}.
40850 <struct id="@var{id}">
40851 <field name="@var{name}" type="@var{type}"/>
40856 Both @var{name} and @var{type} values are required.
40857 No implicit padding is added.
40859 Bitfield values have a @var{name}, @var{start}, @var{end} and @var{type}.
40862 <struct id="@var{id}" size="@var{size}">
40863 <field name="@var{name}" start="@var{start}" end="@var{end}" type="@var{type}"/>
40869 <flags id="@var{id}" size="@var{size}">
40870 <field name="@var{name}" start="@var{start}" end="@var{end}" type="@var{type}"/>
40875 The @var{name} value is required.
40876 Bitfield values may be named with the empty string, @samp{""},
40877 in which case the field is ``filler'' and its value is not printed.
40878 Not all bits need to be specified, so ``filler'' fields are optional.
40880 The @var{start} and @var{end} values are required, and @var{type}
40882 The field's @var{start} must be less than or equal to its @var{end},
40883 and zero represents the least significant bit.
40885 The default value of @var{type} is @code{bool} for single bit fields,
40886 and an unsigned integer otherwise.
40888 Which to choose? Structures or flags?
40890 Registers defined with @samp{flags} have these advantages over
40891 defining them with @samp{struct}:
40895 Arithmetic may be performed on them as if they were integers.
40897 They are printed in a more readable fashion.
40900 Registers defined with @samp{struct} have one advantage over
40901 defining them with @samp{flags}:
40905 One can fetch individual fields like in @samp{C}.
40908 (gdb) print $my_struct_reg.field3
40914 @subsection Registers
40917 Each register is represented as an element with this form:
40920 <reg name="@var{name}"
40921 bitsize="@var{size}"
40922 @r{[}regnum="@var{num}"@r{]}
40923 @r{[}save-restore="@var{save-restore}"@r{]}
40924 @r{[}type="@var{type}"@r{]}
40925 @r{[}group="@var{group}"@r{]}/>
40929 The components are as follows:
40934 The register's name; it must be unique within the target description.
40937 The register's size, in bits.
40940 The register's number. If omitted, a register's number is one greater
40941 than that of the previous register (either in the current feature or in
40942 a preceding feature); the first register in the target description
40943 defaults to zero. This register number is used to read or write
40944 the register; e.g.@: it is used in the remote @code{p} and @code{P}
40945 packets, and registers appear in the @code{g} and @code{G} packets
40946 in order of increasing register number.
40949 Whether the register should be preserved across inferior function
40950 calls; this must be either @code{yes} or @code{no}. The default is
40951 @code{yes}, which is appropriate for most registers except for
40952 some system control registers; this is not related to the target's
40956 The type of the register. It may be a predefined type, a type
40957 defined in the current feature, or one of the special types @code{int}
40958 and @code{float}. @code{int} is an integer type of the correct size
40959 for @var{bitsize}, and @code{float} is a floating point type (in the
40960 architecture's normal floating point format) of the correct size for
40961 @var{bitsize}. The default is @code{int}.
40964 The register group to which this register belongs. It must
40965 be either @code{general}, @code{float}, or @code{vector}. If no
40966 @var{group} is specified, @value{GDBN} will not display the register
40967 in @code{info registers}.
40971 @node Predefined Target Types
40972 @section Predefined Target Types
40973 @cindex target descriptions, predefined types
40975 Type definitions in the self-description can build up composite types
40976 from basic building blocks, but can not define fundamental types. Instead,
40977 standard identifiers are provided by @value{GDBN} for the fundamental
40978 types. The currently supported types are:
40983 Boolean type, occupying a single bit.
40990 Signed integer types holding the specified number of bits.
40997 Unsigned integer types holding the specified number of bits.
41001 Pointers to unspecified code and data. The program counter and
41002 any dedicated return address register may be marked as code
41003 pointers; printing a code pointer converts it into a symbolic
41004 address. The stack pointer and any dedicated address registers
41005 may be marked as data pointers.
41008 Single precision IEEE floating point.
41011 Double precision IEEE floating point.
41014 The 12-byte extended precision format used by ARM FPA registers.
41017 The 10-byte extended precision format used by x87 registers.
41020 32bit @sc{eflags} register used by x86.
41023 32bit @sc{mxcsr} register used by x86.
41027 @node Enum Target Types
41028 @section Enum Target Types
41029 @cindex target descriptions, enum types
41031 Enum target types are useful in @samp{struct} and @samp{flags}
41032 register descriptions. @xref{Target Description Format}.
41034 Enum types have a name, size and a list of name/value pairs.
41037 <enum id="@var{id}" size="@var{size}">
41038 <evalue name="@var{name}" value="@var{value}"/>
41043 Enums must be defined before they are used.
41046 <enum id="levels_type" size="4">
41047 <evalue name="low" value="0"/>
41048 <evalue name="high" value="1"/>
41050 <flags id="flags_type" size="4">
41051 <field name="X" start="0"/>
41052 <field name="LEVEL" start="1" end="1" type="levels_type"/>
41054 <reg name="flags" bitsize="32" type="flags_type"/>
41057 Given that description, a value of 3 for the @samp{flags} register
41058 would be printed as:
41061 (gdb) info register flags
41062 flags 0x3 [ X LEVEL=high ]
41065 @node Standard Target Features
41066 @section Standard Target Features
41067 @cindex target descriptions, standard features
41069 A target description must contain either no registers or all the
41070 target's registers. If the description contains no registers, then
41071 @value{GDBN} will assume a default register layout, selected based on
41072 the architecture. If the description contains any registers, the
41073 default layout will not be used; the standard registers must be
41074 described in the target description, in such a way that @value{GDBN}
41075 can recognize them.
41077 This is accomplished by giving specific names to feature elements
41078 which contain standard registers. @value{GDBN} will look for features
41079 with those names and verify that they contain the expected registers;
41080 if any known feature is missing required registers, or if any required
41081 feature is missing, @value{GDBN} will reject the target
41082 description. You can add additional registers to any of the
41083 standard features --- @value{GDBN} will display them just as if
41084 they were added to an unrecognized feature.
41086 This section lists the known features and their expected contents.
41087 Sample XML documents for these features are included in the
41088 @value{GDBN} source tree, in the directory @file{gdb/features}.
41090 Names recognized by @value{GDBN} should include the name of the
41091 company or organization which selected the name, and the overall
41092 architecture to which the feature applies; so e.g.@: the feature
41093 containing ARM core registers is named @samp{org.gnu.gdb.arm.core}.
41095 The names of registers are not case sensitive for the purpose
41096 of recognizing standard features, but @value{GDBN} will only display
41097 registers using the capitalization used in the description.
41100 * AArch64 Features::
41104 * MicroBlaze Features::
41108 * Nios II Features::
41109 * PowerPC Features::
41110 * S/390 and System z Features::
41116 @node AArch64 Features
41117 @subsection AArch64 Features
41118 @cindex target descriptions, AArch64 features
41120 The @samp{org.gnu.gdb.aarch64.core} feature is required for AArch64
41121 targets. It should contain registers @samp{x0} through @samp{x30},
41122 @samp{sp}, @samp{pc}, and @samp{cpsr}.
41124 The @samp{org.gnu.gdb.aarch64.fpu} feature is optional. If present,
41125 it should contain registers @samp{v0} through @samp{v31}, @samp{fpsr},
41129 @subsection ARC Features
41130 @cindex target descriptions, ARC Features
41132 ARC processors are highly configurable, so even core registers and their number
41133 are not completely predetermined. In addition flags and PC registers which are
41134 important to @value{GDBN} are not ``core'' registers in ARC. It is required
41135 that one of the core registers features is present.
41136 @samp{org.gnu.gdb.arc.aux-minimal} feature is mandatory.
41138 The @samp{org.gnu.gdb.arc.core.v2} feature is required for ARC EM and ARC HS
41139 targets with a normal register file. It should contain registers @samp{r0}
41140 through @samp{r25}, @samp{gp}, @samp{fp}, @samp{sp}, @samp{r30}, @samp{blink},
41141 @samp{lp_count} and @samp{pcl}. This feature may contain register @samp{ilink}
41142 and any of extension core registers @samp{r32} through @samp{r59/acch}.
41143 @samp{ilink} and extension core registers are not available to read/write, when
41144 debugging GNU/Linux applications, thus @samp{ilink} is made optional.
41146 The @samp{org.gnu.gdb.arc.core-reduced.v2} feature is required for ARC EM and
41147 ARC HS targets with a reduced register file. It should contain registers
41148 @samp{r0} through @samp{r3}, @samp{r10} through @samp{r15}, @samp{gp},
41149 @samp{fp}, @samp{sp}, @samp{r30}, @samp{blink}, @samp{lp_count} and @samp{pcl}.
41150 This feature may contain register @samp{ilink} and any of extension core
41151 registers @samp{r32} through @samp{r59/acch}.
41153 The @samp{org.gnu.gdb.arc.core.arcompact} feature is required for ARCompact
41154 targets with a normal register file. It should contain registers @samp{r0}
41155 through @samp{r25}, @samp{gp}, @samp{fp}, @samp{sp}, @samp{r30}, @samp{blink},
41156 @samp{lp_count} and @samp{pcl}. This feature may contain registers
41157 @samp{ilink1}, @samp{ilink2} and any of extension core registers @samp{r32}
41158 through @samp{r59/acch}. @samp{ilink1} and @samp{ilink2} and extension core
41159 registers are not available when debugging GNU/Linux applications. The only
41160 difference with @samp{org.gnu.gdb.arc.core.v2} feature is in the names of
41161 @samp{ilink1} and @samp{ilink2} registers and that @samp{r30} is mandatory in
41162 ARC v2, but @samp{ilink2} is optional on ARCompact.
41164 The @samp{org.gnu.gdb.arc.aux-minimal} feature is required for all ARC
41165 targets. It should contain registers @samp{pc} and @samp{status32}.
41168 @subsection ARM Features
41169 @cindex target descriptions, ARM features
41171 The @samp{org.gnu.gdb.arm.core} feature is required for non-M-profile
41173 It should contain registers @samp{r0} through @samp{r13}, @samp{sp},
41174 @samp{lr}, @samp{pc}, and @samp{cpsr}.
41176 For M-profile targets (e.g. Cortex-M3), the @samp{org.gnu.gdb.arm.core}
41177 feature is replaced by @samp{org.gnu.gdb.arm.m-profile}. It should contain
41178 registers @samp{r0} through @samp{r13}, @samp{sp}, @samp{lr}, @samp{pc},
41181 The @samp{org.gnu.gdb.arm.fpa} feature is optional. If present, it
41182 should contain registers @samp{f0} through @samp{f7} and @samp{fps}.
41184 The @samp{org.gnu.gdb.xscale.iwmmxt} feature is optional. If present,
41185 it should contain at least registers @samp{wR0} through @samp{wR15} and
41186 @samp{wCGR0} through @samp{wCGR3}. The @samp{wCID}, @samp{wCon},
41187 @samp{wCSSF}, and @samp{wCASF} registers are optional.
41189 The @samp{org.gnu.gdb.arm.vfp} feature is optional. If present, it
41190 should contain at least registers @samp{d0} through @samp{d15}. If
41191 they are present, @samp{d16} through @samp{d31} should also be included.
41192 @value{GDBN} will synthesize the single-precision registers from
41193 halves of the double-precision registers.
41195 The @samp{org.gnu.gdb.arm.neon} feature is optional. It does not
41196 need to contain registers; it instructs @value{GDBN} to display the
41197 VFP double-precision registers as vectors and to synthesize the
41198 quad-precision registers from pairs of double-precision registers.
41199 If this feature is present, @samp{org.gnu.gdb.arm.vfp} must also
41200 be present and include 32 double-precision registers.
41202 @node i386 Features
41203 @subsection i386 Features
41204 @cindex target descriptions, i386 features
41206 The @samp{org.gnu.gdb.i386.core} feature is required for i386/amd64
41207 targets. It should describe the following registers:
41211 @samp{eax} through @samp{edi} plus @samp{eip} for i386
41213 @samp{rax} through @samp{r15} plus @samp{rip} for amd64
41215 @samp{eflags}, @samp{cs}, @samp{ss}, @samp{ds}, @samp{es},
41216 @samp{fs}, @samp{gs}
41218 @samp{st0} through @samp{st7}
41220 @samp{fctrl}, @samp{fstat}, @samp{ftag}, @samp{fiseg}, @samp{fioff},
41221 @samp{foseg}, @samp{fooff} and @samp{fop}
41224 The register sets may be different, depending on the target.
41226 The @samp{org.gnu.gdb.i386.sse} feature is optional. It should
41227 describe registers:
41231 @samp{xmm0} through @samp{xmm7} for i386
41233 @samp{xmm0} through @samp{xmm15} for amd64
41238 The @samp{org.gnu.gdb.i386.avx} feature is optional and requires the
41239 @samp{org.gnu.gdb.i386.sse} feature. It should
41240 describe the upper 128 bits of @sc{ymm} registers:
41244 @samp{ymm0h} through @samp{ymm7h} for i386
41246 @samp{ymm0h} through @samp{ymm15h} for amd64
41249 The @samp{org.gnu.gdb.i386.mpx} is an optional feature representing Intel
41250 Memory Protection Extension (MPX). It should describe the following registers:
41254 @samp{bnd0raw} through @samp{bnd3raw} for i386 and amd64.
41256 @samp{bndcfgu} and @samp{bndstatus} for i386 and amd64.
41259 The @samp{org.gnu.gdb.i386.linux} feature is optional. It should
41260 describe a single register, @samp{orig_eax}.
41262 The @samp{org.gnu.gdb.i386.segments} feature is optional. It should
41263 describe two system registers: @samp{fs_base} and @samp{gs_base}.
41265 The @samp{org.gnu.gdb.i386.avx512} feature is optional and requires the
41266 @samp{org.gnu.gdb.i386.avx} feature. It should
41267 describe additional @sc{xmm} registers:
41271 @samp{xmm16h} through @samp{xmm31h}, only valid for amd64.
41274 It should describe the upper 128 bits of additional @sc{ymm} registers:
41278 @samp{ymm16h} through @samp{ymm31h}, only valid for amd64.
41282 describe the upper 256 bits of @sc{zmm} registers:
41286 @samp{zmm0h} through @samp{zmm7h} for i386.
41288 @samp{zmm0h} through @samp{zmm15h} for amd64.
41292 describe the additional @sc{zmm} registers:
41296 @samp{zmm16h} through @samp{zmm31h}, only valid for amd64.
41299 The @samp{org.gnu.gdb.i386.pkeys} feature is optional. It should
41300 describe a single register, @samp{pkru}. It is a 32-bit register
41301 valid for i386 and amd64.
41303 @node MicroBlaze Features
41304 @subsection MicroBlaze Features
41305 @cindex target descriptions, MicroBlaze features
41307 The @samp{org.gnu.gdb.microblaze.core} feature is required for MicroBlaze
41308 targets. It should contain registers @samp{r0} through @samp{r31},
41309 @samp{rpc}, @samp{rmsr}, @samp{rear}, @samp{resr}, @samp{rfsr}, @samp{rbtr},
41310 @samp{rpvr}, @samp{rpvr1} through @samp{rpvr11}, @samp{redr}, @samp{rpid},
41311 @samp{rzpr}, @samp{rtlbx}, @samp{rtlbsx}, @samp{rtlblo}, and @samp{rtlbhi}.
41313 The @samp{org.gnu.gdb.microblaze.stack-protect} feature is optional.
41314 If present, it should contain registers @samp{rshr} and @samp{rslr}
41316 @node MIPS Features
41317 @subsection @acronym{MIPS} Features
41318 @cindex target descriptions, @acronym{MIPS} features
41320 The @samp{org.gnu.gdb.mips.cpu} feature is required for @acronym{MIPS} targets.
41321 It should contain registers @samp{r0} through @samp{r31}, @samp{lo},
41322 @samp{hi}, and @samp{pc}. They may be 32-bit or 64-bit depending
41325 The @samp{org.gnu.gdb.mips.cp0} feature is also required. It should
41326 contain at least the @samp{status}, @samp{badvaddr}, and @samp{cause}
41327 registers. They may be 32-bit or 64-bit depending on the target.
41329 The @samp{org.gnu.gdb.mips.fpu} feature is currently required, though
41330 it may be optional in a future version of @value{GDBN}. It should
41331 contain registers @samp{f0} through @samp{f31}, @samp{fcsr}, and
41332 @samp{fir}. They may be 32-bit or 64-bit depending on the target.
41334 The @samp{org.gnu.gdb.mips.dsp} feature is optional. It should
41335 contain registers @samp{hi1} through @samp{hi3}, @samp{lo1} through
41336 @samp{lo3}, and @samp{dspctl}. The @samp{dspctl} register should
41337 be 32-bit and the rest may be 32-bit or 64-bit depending on the target.
41339 The @samp{org.gnu.gdb.mips.linux} feature is optional. It should
41340 contain a single register, @samp{restart}, which is used by the
41341 Linux kernel to control restartable syscalls.
41343 @node M68K Features
41344 @subsection M68K Features
41345 @cindex target descriptions, M68K features
41348 @item @samp{org.gnu.gdb.m68k.core}
41349 @itemx @samp{org.gnu.gdb.coldfire.core}
41350 @itemx @samp{org.gnu.gdb.fido.core}
41351 One of those features must be always present.
41352 The feature that is present determines which flavor of m68k is
41353 used. The feature that is present should contain registers
41354 @samp{d0} through @samp{d7}, @samp{a0} through @samp{a5}, @samp{fp},
41355 @samp{sp}, @samp{ps} and @samp{pc}.
41357 @item @samp{org.gnu.gdb.coldfire.fp}
41358 This feature is optional. If present, it should contain registers
41359 @samp{fp0} through @samp{fp7}, @samp{fpcontrol}, @samp{fpstatus} and
41363 @node NDS32 Features
41364 @subsection NDS32 Features
41365 @cindex target descriptions, NDS32 features
41367 The @samp{org.gnu.gdb.nds32.core} feature is required for NDS32
41368 targets. It should contain at least registers @samp{r0} through
41369 @samp{r10}, @samp{r15}, @samp{fp}, @samp{gp}, @samp{lp}, @samp{sp},
41372 The @samp{org.gnu.gdb.nds32.fpu} feature is optional. If present,
41373 it should contain 64-bit double-precision floating-point registers
41374 @samp{fd0} through @emph{fdN}, which should be @samp{fd3}, @samp{fd7},
41375 @samp{fd15}, or @samp{fd31} based on the FPU configuration implemented.
41377 @emph{Note:} The first sixteen 64-bit double-precision floating-point
41378 registers are overlapped with the thirty-two 32-bit single-precision
41379 floating-point registers. The 32-bit single-precision registers, if
41380 not being listed explicitly, will be synthesized from halves of the
41381 overlapping 64-bit double-precision registers. Listing 32-bit
41382 single-precision registers explicitly is deprecated, and the
41383 support to it could be totally removed some day.
41385 @node Nios II Features
41386 @subsection Nios II Features
41387 @cindex target descriptions, Nios II features
41389 The @samp{org.gnu.gdb.nios2.cpu} feature is required for Nios II
41390 targets. It should contain the 32 core registers (@samp{zero},
41391 @samp{at}, @samp{r2} through @samp{r23}, @samp{et} through @samp{ra}),
41392 @samp{pc}, and the 16 control registers (@samp{status} through
41395 @node PowerPC Features
41396 @subsection PowerPC Features
41397 @cindex target descriptions, PowerPC features
41399 The @samp{org.gnu.gdb.power.core} feature is required for PowerPC
41400 targets. It should contain registers @samp{r0} through @samp{r31},
41401 @samp{pc}, @samp{msr}, @samp{cr}, @samp{lr}, @samp{ctr}, and
41402 @samp{xer}. They may be 32-bit or 64-bit depending on the target.
41404 The @samp{org.gnu.gdb.power.fpu} feature is optional. It should
41405 contain registers @samp{f0} through @samp{f31} and @samp{fpscr}.
41407 The @samp{org.gnu.gdb.power.altivec} feature is optional. It should
41408 contain registers @samp{vr0} through @samp{vr31}, @samp{vscr},
41411 The @samp{org.gnu.gdb.power.vsx} feature is optional. It should
41412 contain registers @samp{vs0h} through @samp{vs31h}. @value{GDBN}
41413 will combine these registers with the floating point registers
41414 (@samp{f0} through @samp{f31}) and the altivec registers (@samp{vr0}
41415 through @samp{vr31}) to present the 128-bit wide registers @samp{vs0}
41416 through @samp{vs63}, the set of vector registers for POWER7.
41418 The @samp{org.gnu.gdb.power.spe} feature is optional. It should
41419 contain registers @samp{ev0h} through @samp{ev31h}, @samp{acc}, and
41420 @samp{spefscr}. SPE targets should provide 32-bit registers in
41421 @samp{org.gnu.gdb.power.core} and provide the upper halves in
41422 @samp{ev0h} through @samp{ev31h}. @value{GDBN} will combine
41423 these to present registers @samp{ev0} through @samp{ev31} to the
41426 @node S/390 and System z Features
41427 @subsection S/390 and System z Features
41428 @cindex target descriptions, S/390 features
41429 @cindex target descriptions, System z features
41431 The @samp{org.gnu.gdb.s390.core} feature is required for S/390 and
41432 System z targets. It should contain the PSW and the 16 general
41433 registers. In particular, System z targets should provide the 64-bit
41434 registers @samp{pswm}, @samp{pswa}, and @samp{r0} through @samp{r15}.
41435 S/390 targets should provide the 32-bit versions of these registers.
41436 A System z target that runs in 31-bit addressing mode should provide
41437 32-bit versions of @samp{pswm} and @samp{pswa}, as well as the general
41438 register's upper halves @samp{r0h} through @samp{r15h}, and their
41439 lower halves @samp{r0l} through @samp{r15l}.
41441 The @samp{org.gnu.gdb.s390.fpr} feature is required. It should
41442 contain the 64-bit registers @samp{f0} through @samp{f15}, and
41445 The @samp{org.gnu.gdb.s390.acr} feature is required. It should
41446 contain the 32-bit registers @samp{acr0} through @samp{acr15}.
41448 The @samp{org.gnu.gdb.s390.linux} feature is optional. It should
41449 contain the register @samp{orig_r2}, which is 64-bit wide on System z
41450 targets and 32-bit otherwise. In addition, the feature may contain
41451 the @samp{last_break} register, whose width depends on the addressing
41452 mode, as well as the @samp{system_call} register, which is always
41455 The @samp{org.gnu.gdb.s390.tdb} feature is optional. It should
41456 contain the 64-bit registers @samp{tdb0}, @samp{tac}, @samp{tct},
41457 @samp{atia}, and @samp{tr0} through @samp{tr15}.
41459 The @samp{org.gnu.gdb.s390.vx} feature is optional. It should contain
41460 64-bit wide registers @samp{v0l} through @samp{v15l}, which will be
41461 combined by @value{GDBN} with the floating point registers @samp{f0}
41462 through @samp{f15} to present the 128-bit wide vector registers
41463 @samp{v0} through @samp{v15}. In addition, this feature should
41464 contain the 128-bit wide vector registers @samp{v16} through
41467 @node Sparc Features
41468 @subsection Sparc Features
41469 @cindex target descriptions, sparc32 features
41470 @cindex target descriptions, sparc64 features
41471 The @samp{org.gnu.gdb.sparc.cpu} feature is required for sparc32/sparc64
41472 targets. It should describe the following registers:
41476 @samp{g0} through @samp{g7}
41478 @samp{o0} through @samp{o7}
41480 @samp{l0} through @samp{l7}
41482 @samp{i0} through @samp{i7}
41485 They may be 32-bit or 64-bit depending on the target.
41487 Also the @samp{org.gnu.gdb.sparc.fpu} feature is required for sparc32/sparc64
41488 targets. It should describe the following registers:
41492 @samp{f0} through @samp{f31}
41494 @samp{f32} through @samp{f62} for sparc64
41497 The @samp{org.gnu.gdb.sparc.cp0} feature is required for sparc32/sparc64
41498 targets. It should describe the following registers:
41502 @samp{y}, @samp{psr}, @samp{wim}, @samp{tbr}, @samp{pc}, @samp{npc},
41503 @samp{fsr}, and @samp{csr} for sparc32
41505 @samp{pc}, @samp{npc}, @samp{state}, @samp{fsr}, @samp{fprs}, and @samp{y}
41509 @node TIC6x Features
41510 @subsection TMS320C6x Features
41511 @cindex target descriptions, TIC6x features
41512 @cindex target descriptions, TMS320C6x features
41513 The @samp{org.gnu.gdb.tic6x.core} feature is required for TMS320C6x
41514 targets. It should contain registers @samp{A0} through @samp{A15},
41515 registers @samp{B0} through @samp{B15}, @samp{CSR} and @samp{PC}.
41517 The @samp{org.gnu.gdb.tic6x.gp} feature is optional. It should
41518 contain registers @samp{A16} through @samp{A31} and @samp{B16}
41519 through @samp{B31}.
41521 The @samp{org.gnu.gdb.tic6x.c6xp} feature is optional. It should
41522 contain registers @samp{TSR}, @samp{ILC} and @samp{RILC}.
41524 @node Operating System Information
41525 @appendix Operating System Information
41526 @cindex operating system information
41532 Users of @value{GDBN} often wish to obtain information about the state of
41533 the operating system running on the target---for example the list of
41534 processes, or the list of open files. This section describes the
41535 mechanism that makes it possible. This mechanism is similar to the
41536 target features mechanism (@pxref{Target Descriptions}), but focuses
41537 on a different aspect of target.
41539 Operating system information is retrived from the target via the
41540 remote protocol, using @samp{qXfer} requests (@pxref{qXfer osdata
41541 read}). The object name in the request should be @samp{osdata}, and
41542 the @var{annex} identifies the data to be fetched.
41545 @appendixsection Process list
41546 @cindex operating system information, process list
41548 When requesting the process list, the @var{annex} field in the
41549 @samp{qXfer} request should be @samp{processes}. The returned data is
41550 an XML document. The formal syntax of this document is defined in
41551 @file{gdb/features/osdata.dtd}.
41553 An example document is:
41556 <?xml version="1.0"?>
41557 <!DOCTYPE target SYSTEM "osdata.dtd">
41558 <osdata type="processes">
41560 <column name="pid">1</column>
41561 <column name="user">root</column>
41562 <column name="command">/sbin/init</column>
41563 <column name="cores">1,2,3</column>
41568 Each item should include a column whose name is @samp{pid}. The value
41569 of that column should identify the process on the target. The
41570 @samp{user} and @samp{command} columns are optional, and will be
41571 displayed by @value{GDBN}. The @samp{cores} column, if present,
41572 should contain a comma-separated list of cores that this process
41573 is running on. Target may provide additional columns,
41574 which @value{GDBN} currently ignores.
41576 @node Trace File Format
41577 @appendix Trace File Format
41578 @cindex trace file format
41580 The trace file comes in three parts: a header, a textual description
41581 section, and a trace frame section with binary data.
41583 The header has the form @code{\x7fTRACE0\n}. The first byte is
41584 @code{0x7f} so as to indicate that the file contains binary data,
41585 while the @code{0} is a version number that may have different values
41588 The description section consists of multiple lines of @sc{ascii} text
41589 separated by newline characters (@code{0xa}). The lines may include a
41590 variety of optional descriptive or context-setting information, such
41591 as tracepoint definitions or register set size. @value{GDBN} will
41592 ignore any line that it does not recognize. An empty line marks the end
41597 Specifies the size of a register block in bytes. This is equal to the
41598 size of a @code{g} packet payload in the remote protocol. @var{size}
41599 is an ascii decimal number. There should be only one such line in
41600 a single trace file.
41602 @item status @var{status}
41603 Trace status. @var{status} has the same format as a @code{qTStatus}
41604 remote packet reply. There should be only one such line in a single trace
41607 @item tp @var{payload}
41608 Tracepoint definition. The @var{payload} has the same format as
41609 @code{qTfP}/@code{qTsP} remote packet reply payload. A single tracepoint
41610 may take multiple lines of definition, corresponding to the multiple
41613 @item tsv @var{payload}
41614 Trace state variable definition. The @var{payload} has the same format as
41615 @code{qTfV}/@code{qTsV} remote packet reply payload. A single variable
41616 may take multiple lines of definition, corresponding to the multiple
41619 @item tdesc @var{payload}
41620 Target description in XML format. The @var{payload} is a single line of
41621 the XML file. All such lines should be concatenated together to get
41622 the original XML file. This file is in the same format as @code{qXfer}
41623 @code{features} payload, and corresponds to the main @code{target.xml}
41624 file. Includes are not allowed.
41628 The trace frame section consists of a number of consecutive frames.
41629 Each frame begins with a two-byte tracepoint number, followed by a
41630 four-byte size giving the amount of data in the frame. The data in
41631 the frame consists of a number of blocks, each introduced by a
41632 character indicating its type (at least register, memory, and trace
41633 state variable). The data in this section is raw binary, not a
41634 hexadecimal or other encoding; its endianness matches the target's
41637 @c FIXME bi-arch may require endianness/arch info in description section
41640 @item R @var{bytes}
41641 Register block. The number and ordering of bytes matches that of a
41642 @code{g} packet in the remote protocol. Note that these are the
41643 actual bytes, in target order, not a hexadecimal encoding.
41645 @item M @var{address} @var{length} @var{bytes}...
41646 Memory block. This is a contiguous block of memory, at the 8-byte
41647 address @var{address}, with a 2-byte length @var{length}, followed by
41648 @var{length} bytes.
41650 @item V @var{number} @var{value}
41651 Trace state variable block. This records the 8-byte signed value
41652 @var{value} of trace state variable numbered @var{number}.
41656 Future enhancements of the trace file format may include additional types
41659 @node Index Section Format
41660 @appendix @code{.gdb_index} section format
41661 @cindex .gdb_index section format
41662 @cindex index section format
41664 This section documents the index section that is created by @code{save
41665 gdb-index} (@pxref{Index Files}). The index section is
41666 DWARF-specific; some knowledge of DWARF is assumed in this
41669 The mapped index file format is designed to be directly
41670 @code{mmap}able on any architecture. In most cases, a datum is
41671 represented using a little-endian 32-bit integer value, called an
41672 @code{offset_type}. Big endian machines must byte-swap the values
41673 before using them. Exceptions to this rule are noted. The data is
41674 laid out such that alignment is always respected.
41676 A mapped index consists of several areas, laid out in order.
41680 The file header. This is a sequence of values, of @code{offset_type}
41681 unless otherwise noted:
41685 The version number, currently 8. Versions 1, 2 and 3 are obsolete.
41686 Version 4 uses a different hashing function from versions 5 and 6.
41687 Version 6 includes symbols for inlined functions, whereas versions 4
41688 and 5 do not. Version 7 adds attributes to the CU indices in the
41689 symbol table. Version 8 specifies that symbols from DWARF type units
41690 (@samp{DW_TAG_type_unit}) refer to the type unit's symbol table and not the
41691 compilation unit (@samp{DW_TAG_comp_unit}) using the type.
41693 @value{GDBN} will only read version 4, 5, or 6 indices
41694 by specifying @code{set use-deprecated-index-sections on}.
41695 GDB has a workaround for potentially broken version 7 indices so it is
41696 currently not flagged as deprecated.
41699 The offset, from the start of the file, of the CU list.
41702 The offset, from the start of the file, of the types CU list. Note
41703 that this area can be empty, in which case this offset will be equal
41704 to the next offset.
41707 The offset, from the start of the file, of the address area.
41710 The offset, from the start of the file, of the symbol table.
41713 The offset, from the start of the file, of the constant pool.
41717 The CU list. This is a sequence of pairs of 64-bit little-endian
41718 values, sorted by the CU offset. The first element in each pair is
41719 the offset of a CU in the @code{.debug_info} section. The second
41720 element in each pair is the length of that CU. References to a CU
41721 elsewhere in the map are done using a CU index, which is just the
41722 0-based index into this table. Note that if there are type CUs, then
41723 conceptually CUs and type CUs form a single list for the purposes of
41727 The types CU list. This is a sequence of triplets of 64-bit
41728 little-endian values. In a triplet, the first value is the CU offset,
41729 the second value is the type offset in the CU, and the third value is
41730 the type signature. The types CU list is not sorted.
41733 The address area. The address area consists of a sequence of address
41734 entries. Each address entry has three elements:
41738 The low address. This is a 64-bit little-endian value.
41741 The high address. This is a 64-bit little-endian value. Like
41742 @code{DW_AT_high_pc}, the value is one byte beyond the end.
41745 The CU index. This is an @code{offset_type} value.
41749 The symbol table. This is an open-addressed hash table. The size of
41750 the hash table is always a power of 2.
41752 Each slot in the hash table consists of a pair of @code{offset_type}
41753 values. The first value is the offset of the symbol's name in the
41754 constant pool. The second value is the offset of the CU vector in the
41757 If both values are 0, then this slot in the hash table is empty. This
41758 is ok because while 0 is a valid constant pool index, it cannot be a
41759 valid index for both a string and a CU vector.
41761 The hash value for a table entry is computed by applying an
41762 iterative hash function to the symbol's name. Starting with an
41763 initial value of @code{r = 0}, each (unsigned) character @samp{c} in
41764 the string is incorporated into the hash using the formula depending on the
41769 The formula is @code{r = r * 67 + c - 113}.
41771 @item Versions 5 to 7
41772 The formula is @code{r = r * 67 + tolower (c) - 113}.
41775 The terminating @samp{\0} is not incorporated into the hash.
41777 The step size used in the hash table is computed via
41778 @code{((hash * 17) & (size - 1)) | 1}, where @samp{hash} is the hash
41779 value, and @samp{size} is the size of the hash table. The step size
41780 is used to find the next candidate slot when handling a hash
41783 The names of C@t{++} symbols in the hash table are canonicalized. We
41784 don't currently have a simple description of the canonicalization
41785 algorithm; if you intend to create new index sections, you must read
41789 The constant pool. This is simply a bunch of bytes. It is organized
41790 so that alignment is correct: CU vectors are stored first, followed by
41793 A CU vector in the constant pool is a sequence of @code{offset_type}
41794 values. The first value is the number of CU indices in the vector.
41795 Each subsequent value is the index and symbol attributes of a CU in
41796 the CU list. This element in the hash table is used to indicate which
41797 CUs define the symbol and how the symbol is used.
41798 See below for the format of each CU index+attributes entry.
41800 A string in the constant pool is zero-terminated.
41803 Attributes were added to CU index values in @code{.gdb_index} version 7.
41804 If a symbol has multiple uses within a CU then there is one
41805 CU index+attributes value for each use.
41807 The format of each CU index+attributes entry is as follows
41813 This is the index of the CU in the CU list.
41815 These bits are reserved for future purposes and must be zero.
41817 The kind of the symbol in the CU.
41821 This value is reserved and should not be used.
41822 By reserving zero the full @code{offset_type} value is backwards compatible
41823 with previous versions of the index.
41825 The symbol is a type.
41827 The symbol is a variable or an enum value.
41829 The symbol is a function.
41831 Any other kind of symbol.
41833 These values are reserved.
41837 This bit is zero if the value is global and one if it is static.
41839 The determination of whether a symbol is global or static is complicated.
41840 The authorative reference is the file @file{dwarf2read.c} in
41841 @value{GDBN} sources.
41845 This pseudo-code describes the computation of a symbol's kind and
41846 global/static attributes in the index.
41849 is_external = get_attribute (die, DW_AT_external);
41850 language = get_attribute (cu_die, DW_AT_language);
41853 case DW_TAG_typedef:
41854 case DW_TAG_base_type:
41855 case DW_TAG_subrange_type:
41859 case DW_TAG_enumerator:
41861 is_static = language != CPLUS;
41863 case DW_TAG_subprogram:
41865 is_static = ! (is_external || language == ADA);
41867 case DW_TAG_constant:
41869 is_static = ! is_external;
41871 case DW_TAG_variable:
41873 is_static = ! is_external;
41875 case DW_TAG_namespace:
41879 case DW_TAG_class_type:
41880 case DW_TAG_interface_type:
41881 case DW_TAG_structure_type:
41882 case DW_TAG_union_type:
41883 case DW_TAG_enumeration_type:
41885 is_static = language != CPLUS;
41893 @appendix Manual pages
41897 * gdb man:: The GNU Debugger man page
41898 * gdbserver man:: Remote Server for the GNU Debugger man page
41899 * gcore man:: Generate a core file of a running program
41900 * gdbinit man:: gdbinit scripts
41906 @c man title gdb The GNU Debugger
41908 @c man begin SYNOPSIS gdb
41909 gdb [@option{-help}] [@option{-nh}] [@option{-nx}] [@option{-q}]
41910 [@option{-batch}] [@option{-cd=}@var{dir}] [@option{-f}]
41911 [@option{-b}@w{ }@var{bps}]
41912 [@option{-tty=}@var{dev}] [@option{-s} @var{symfile}]
41913 [@option{-e}@w{ }@var{prog}] [@option{-se}@w{ }@var{prog}]
41914 [@option{-c}@w{ }@var{core}] [@option{-p}@w{ }@var{procID}]
41915 [@option{-x}@w{ }@var{cmds}] [@option{-d}@w{ }@var{dir}]
41916 [@var{prog}|@var{prog} @var{procID}|@var{prog} @var{core}]
41919 @c man begin DESCRIPTION gdb
41920 The purpose of a debugger such as @value{GDBN} is to allow you to see what is
41921 going on ``inside'' another program while it executes -- or what another
41922 program was doing at the moment it crashed.
41924 @value{GDBN} can do four main kinds of things (plus other things in support of
41925 these) to help you catch bugs in the act:
41929 Start your program, specifying anything that might affect its behavior.
41932 Make your program stop on specified conditions.
41935 Examine what has happened, when your program has stopped.
41938 Change things in your program, so you can experiment with correcting the
41939 effects of one bug and go on to learn about another.
41942 You can use @value{GDBN} to debug programs written in C, C@t{++}, Fortran and
41945 @value{GDBN} is invoked with the shell command @code{gdb}. Once started, it reads
41946 commands from the terminal until you tell it to exit with the @value{GDBN}
41947 command @code{quit}. You can get online help from @value{GDBN} itself
41948 by using the command @code{help}.
41950 You can run @code{gdb} with no arguments or options; but the most
41951 usual way to start @value{GDBN} is with one argument or two, specifying an
41952 executable program as the argument:
41958 You can also start with both an executable program and a core file specified:
41964 You can, instead, specify a process ID as a second argument, if you want
41965 to debug a running process:
41973 would attach @value{GDBN} to process @code{1234} (unless you also have a file
41974 named @file{1234}; @value{GDBN} does check for a core file first).
41975 With option @option{-p} you can omit the @var{program} filename.
41977 Here are some of the most frequently needed @value{GDBN} commands:
41979 @c pod2man highlights the right hand side of the @item lines.
41981 @item break [@var{file}:]@var{function}
41982 Set a breakpoint at @var{function} (in @var{file}).
41984 @item run [@var{arglist}]
41985 Start your program (with @var{arglist}, if specified).
41988 Backtrace: display the program stack.
41990 @item print @var{expr}
41991 Display the value of an expression.
41994 Continue running your program (after stopping, e.g. at a breakpoint).
41997 Execute next program line (after stopping); step @emph{over} any
41998 function calls in the line.
42000 @item edit [@var{file}:]@var{function}
42001 look at the program line where it is presently stopped.
42003 @item list [@var{file}:]@var{function}
42004 type the text of the program in the vicinity of where it is presently stopped.
42007 Execute next program line (after stopping); step @emph{into} any
42008 function calls in the line.
42010 @item help [@var{name}]
42011 Show information about @value{GDBN} command @var{name}, or general information
42012 about using @value{GDBN}.
42015 Exit from @value{GDBN}.
42019 For full details on @value{GDBN},
42020 see @cite{Using GDB: A Guide to the GNU Source-Level Debugger},
42021 by Richard M. Stallman and Roland H. Pesch. The same text is available online
42022 as the @code{gdb} entry in the @code{info} program.
42026 @c man begin OPTIONS gdb
42027 Any arguments other than options specify an executable
42028 file and core file (or process ID); that is, the first argument
42029 encountered with no
42030 associated option flag is equivalent to a @option{-se} option, and the second,
42031 if any, is equivalent to a @option{-c} option if it's the name of a file.
42033 both long and short forms; both are shown here. The long forms are also
42034 recognized if you truncate them, so long as enough of the option is
42035 present to be unambiguous. (If you prefer, you can flag option
42036 arguments with @option{+} rather than @option{-}, though we illustrate the
42037 more usual convention.)
42039 All the options and command line arguments you give are processed
42040 in sequential order. The order makes a difference when the @option{-x}
42046 List all options, with brief explanations.
42048 @item -symbols=@var{file}
42049 @itemx -s @var{file}
42050 Read symbol table from file @var{file}.
42053 Enable writing into executable and core files.
42055 @item -exec=@var{file}
42056 @itemx -e @var{file}
42057 Use file @var{file} as the executable file to execute when
42058 appropriate, and for examining pure data in conjunction with a core
42061 @item -se=@var{file}
42062 Read symbol table from file @var{file} and use it as the executable
42065 @item -core=@var{file}
42066 @itemx -c @var{file}
42067 Use file @var{file} as a core dump to examine.
42069 @item -command=@var{file}
42070 @itemx -x @var{file}
42071 Execute @value{GDBN} commands from file @var{file}.
42073 @item -ex @var{command}
42074 Execute given @value{GDBN} @var{command}.
42076 @item -directory=@var{directory}
42077 @itemx -d @var{directory}
42078 Add @var{directory} to the path to search for source files.
42081 Do not execute commands from @file{~/.gdbinit}.
42085 Do not execute commands from any @file{.gdbinit} initialization files.
42089 ``Quiet''. Do not print the introductory and copyright messages. These
42090 messages are also suppressed in batch mode.
42093 Run in batch mode. Exit with status @code{0} after processing all the command
42094 files specified with @option{-x} (and @file{.gdbinit}, if not inhibited).
42095 Exit with nonzero status if an error occurs in executing the @value{GDBN}
42096 commands in the command files.
42098 Batch mode may be useful for running @value{GDBN} as a filter, for example to
42099 download and run a program on another computer; in order to make this
42100 more useful, the message
42103 Program exited normally.
42107 (which is ordinarily issued whenever a program running under @value{GDBN} control
42108 terminates) is not issued when running in batch mode.
42110 @item -cd=@var{directory}
42111 Run @value{GDBN} using @var{directory} as its working directory,
42112 instead of the current directory.
42116 Emacs sets this option when it runs @value{GDBN} as a subprocess. It tells
42117 @value{GDBN} to output the full file name and line number in a standard,
42118 recognizable fashion each time a stack frame is displayed (which
42119 includes each time the program stops). This recognizable format looks
42120 like two @samp{\032} characters, followed by the file name, line number
42121 and character position separated by colons, and a newline. The
42122 Emacs-to-@value{GDBN} interface program uses the two @samp{\032}
42123 characters as a signal to display the source code for the frame.
42126 Set the line speed (baud rate or bits per second) of any serial
42127 interface used by @value{GDBN} for remote debugging.
42129 @item -tty=@var{device}
42130 Run using @var{device} for your program's standard input and output.
42134 @c man begin SEEALSO gdb
42136 The full documentation for @value{GDBN} is maintained as a Texinfo manual.
42137 If the @code{info} and @code{gdb} programs and @value{GDBN}'s Texinfo
42138 documentation are properly installed at your site, the command
42145 should give you access to the complete manual.
42147 @cite{Using GDB: A Guide to the GNU Source-Level Debugger},
42148 Richard M. Stallman and Roland H. Pesch, July 1991.
42152 @node gdbserver man
42153 @heading gdbserver man
42155 @c man title gdbserver Remote Server for the GNU Debugger
42157 @c man begin SYNOPSIS gdbserver
42158 gdbserver @var{comm} @var{prog} [@var{args}@dots{}]
42160 gdbserver --attach @var{comm} @var{pid}
42162 gdbserver --multi @var{comm}
42166 @c man begin DESCRIPTION gdbserver
42167 @command{gdbserver} is a program that allows you to run @value{GDBN} on a different machine
42168 than the one which is running the program being debugged.
42171 @subheading Usage (server (target) side)
42174 Usage (server (target) side):
42177 First, you need to have a copy of the program you want to debug put onto
42178 the target system. The program can be stripped to save space if needed, as
42179 @command{gdbserver} doesn't care about symbols. All symbol handling is taken care of by
42180 the @value{GDBN} running on the host system.
42182 To use the server, you log on to the target system, and run the @command{gdbserver}
42183 program. You must tell it (a) how to communicate with @value{GDBN}, (b) the name of
42184 your program, and (c) its arguments. The general syntax is:
42187 target> gdbserver @var{comm} @var{program} [@var{args} ...]
42190 For example, using a serial port, you might say:
42194 @c @file would wrap it as F</dev/com1>.
42195 target> gdbserver /dev/com1 emacs foo.txt
42198 target> gdbserver @file{/dev/com1} emacs foo.txt
42202 This tells @command{gdbserver} to debug emacs with an argument of foo.txt, and
42203 to communicate with @value{GDBN} via @file{/dev/com1}. @command{gdbserver} now
42204 waits patiently for the host @value{GDBN} to communicate with it.
42206 To use a TCP connection, you could say:
42209 target> gdbserver host:2345 emacs foo.txt
42212 This says pretty much the same thing as the last example, except that we are
42213 going to communicate with the @code{host} @value{GDBN} via TCP. The @code{host:2345} argument means
42214 that we are expecting to see a TCP connection from @code{host} to local TCP port
42215 2345. (Currently, the @code{host} part is ignored.) You can choose any number you
42216 want for the port number as long as it does not conflict with any existing TCP
42217 ports on the target system. This same port number must be used in the host
42218 @value{GDBN}s @code{target remote} command, which will be described shortly. Note that if
42219 you chose a port number that conflicts with another service, @command{gdbserver} will
42220 print an error message and exit.
42222 @command{gdbserver} can also attach to running programs.
42223 This is accomplished via the @option{--attach} argument. The syntax is:
42226 target> gdbserver --attach @var{comm} @var{pid}
42229 @var{pid} is the process ID of a currently running process. It isn't
42230 necessary to point @command{gdbserver} at a binary for the running process.
42232 To start @code{gdbserver} without supplying an initial command to run
42233 or process ID to attach, use the @option{--multi} command line option.
42234 In such case you should connect using @kbd{target extended-remote} to start
42235 the program you want to debug.
42238 target> gdbserver --multi @var{comm}
42242 @subheading Usage (host side)
42248 You need an unstripped copy of the target program on your host system, since
42249 @value{GDBN} needs to examine it's symbol tables and such. Start up @value{GDBN} as you normally
42250 would, with the target program as the first argument. (You may need to use the
42251 @option{--baud} option if the serial line is running at anything except 9600 baud.)
42252 That is @code{gdb TARGET-PROG}, or @code{gdb --baud BAUD TARGET-PROG}. After that, the only
42253 new command you need to know about is @code{target remote}
42254 (or @code{target extended-remote}). Its argument is either
42255 a device name (usually a serial device, like @file{/dev/ttyb}), or a @code{HOST:PORT}
42256 descriptor. For example:
42260 @c @file would wrap it as F</dev/ttyb>.
42261 (gdb) target remote /dev/ttyb
42264 (gdb) target remote @file{/dev/ttyb}
42269 communicates with the server via serial line @file{/dev/ttyb}, and:
42272 (gdb) target remote the-target:2345
42276 communicates via a TCP connection to port 2345 on host `the-target', where
42277 you previously started up @command{gdbserver} with the same port number. Note that for
42278 TCP connections, you must start up @command{gdbserver} prior to using the `target remote'
42279 command, otherwise you may get an error that looks something like
42280 `Connection refused'.
42282 @command{gdbserver} can also debug multiple inferiors at once,
42285 the @value{GDBN} manual in node @code{Inferiors and Programs}
42286 -- shell command @code{info -f gdb -n 'Inferiors and Programs'}.
42289 @ref{Inferiors and Programs}.
42291 In such case use the @code{extended-remote} @value{GDBN} command variant:
42294 (gdb) target extended-remote the-target:2345
42297 The @command{gdbserver} option @option{--multi} may or may not be used in such
42301 @c man begin OPTIONS gdbserver
42302 There are three different modes for invoking @command{gdbserver}:
42307 Debug a specific program specified by its program name:
42310 gdbserver @var{comm} @var{prog} [@var{args}@dots{}]
42313 The @var{comm} parameter specifies how should the server communicate
42314 with @value{GDBN}; it is either a device name (to use a serial line),
42315 a TCP port number (@code{:1234}), or @code{-} or @code{stdio} to use
42316 stdin/stdout of @code{gdbserver}. Specify the name of the program to
42317 debug in @var{prog}. Any remaining arguments will be passed to the
42318 program verbatim. When the program exits, @value{GDBN} will close the
42319 connection, and @code{gdbserver} will exit.
42322 Debug a specific program by specifying the process ID of a running
42326 gdbserver --attach @var{comm} @var{pid}
42329 The @var{comm} parameter is as described above. Supply the process ID
42330 of a running program in @var{pid}; @value{GDBN} will do everything
42331 else. Like with the previous mode, when the process @var{pid} exits,
42332 @value{GDBN} will close the connection, and @code{gdbserver} will exit.
42335 Multi-process mode -- debug more than one program/process:
42338 gdbserver --multi @var{comm}
42341 In this mode, @value{GDBN} can instruct @command{gdbserver} which
42342 command(s) to run. Unlike the other 2 modes, @value{GDBN} will not
42343 close the connection when a process being debugged exits, so you can
42344 debug several processes in the same session.
42347 In each of the modes you may specify these options:
42352 List all options, with brief explanations.
42355 This option causes @command{gdbserver} to print its version number and exit.
42358 @command{gdbserver} will attach to a running program. The syntax is:
42361 target> gdbserver --attach @var{comm} @var{pid}
42364 @var{pid} is the process ID of a currently running process. It isn't
42365 necessary to point @command{gdbserver} at a binary for the running process.
42368 To start @code{gdbserver} without supplying an initial command to run
42369 or process ID to attach, use this command line option.
42370 Then you can connect using @kbd{target extended-remote} and start
42371 the program you want to debug. The syntax is:
42374 target> gdbserver --multi @var{comm}
42378 Instruct @code{gdbserver} to display extra status information about the debugging
42380 This option is intended for @code{gdbserver} development and for bug reports to
42383 @item --remote-debug
42384 Instruct @code{gdbserver} to display remote protocol debug output.
42385 This option is intended for @code{gdbserver} development and for bug reports to
42388 @item --debug-format=option1@r{[},option2,...@r{]}
42389 Instruct @code{gdbserver} to include extra information in each line
42390 of debugging output.
42391 @xref{Other Command-Line Arguments for gdbserver}.
42394 Specify a wrapper to launch programs
42395 for debugging. The option should be followed by the name of the
42396 wrapper, then any command-line arguments to pass to the wrapper, then
42397 @kbd{--} indicating the end of the wrapper arguments.
42400 By default, @command{gdbserver} keeps the listening TCP port open, so that
42401 additional connections are possible. However, if you start @code{gdbserver}
42402 with the @option{--once} option, it will stop listening for any further
42403 connection attempts after connecting to the first @value{GDBN} session.
42405 @c --disable-packet is not documented for users.
42407 @c --disable-randomization and --no-disable-randomization are superseded by
42408 @c QDisableRandomization.
42413 @c man begin SEEALSO gdbserver
42415 The full documentation for @value{GDBN} is maintained as a Texinfo manual.
42416 If the @code{info} and @code{gdb} programs and @value{GDBN}'s Texinfo
42417 documentation are properly installed at your site, the command
42423 should give you access to the complete manual.
42425 @cite{Using GDB: A Guide to the GNU Source-Level Debugger},
42426 Richard M. Stallman and Roland H. Pesch, July 1991.
42433 @c man title gcore Generate a core file of a running program
42436 @c man begin SYNOPSIS gcore
42437 gcore [-o @var{filename}] @var{pid}
42441 @c man begin DESCRIPTION gcore
42442 Generate a core dump of a running program with process ID @var{pid}.
42443 Produced file is equivalent to a kernel produced core file as if the process
42444 crashed (and if @kbd{ulimit -c} were used to set up an appropriate core dump
42445 limit). Unlike after a crash, after @command{gcore} the program remains
42446 running without any change.
42449 @c man begin OPTIONS gcore
42451 @item -o @var{filename}
42452 The optional argument
42453 @var{filename} specifies the file name where to put the core dump.
42454 If not specified, the file name defaults to @file{core.@var{pid}},
42455 where @var{pid} is the running program process ID.
42459 @c man begin SEEALSO gcore
42461 The full documentation for @value{GDBN} is maintained as a Texinfo manual.
42462 If the @code{info} and @code{gdb} programs and @value{GDBN}'s Texinfo
42463 documentation are properly installed at your site, the command
42470 should give you access to the complete manual.
42472 @cite{Using GDB: A Guide to the GNU Source-Level Debugger},
42473 Richard M. Stallman and Roland H. Pesch, July 1991.
42480 @c man title gdbinit GDB initialization scripts
42483 @c man begin SYNOPSIS gdbinit
42484 @ifset SYSTEM_GDBINIT
42485 @value{SYSTEM_GDBINIT}
42494 @c man begin DESCRIPTION gdbinit
42495 These files contain @value{GDBN} commands to automatically execute during
42496 @value{GDBN} startup. The lines of contents are canned sequences of commands,
42499 the @value{GDBN} manual in node @code{Sequences}
42500 -- shell command @code{info -f gdb -n Sequences}.
42506 Please read more in
42508 the @value{GDBN} manual in node @code{Startup}
42509 -- shell command @code{info -f gdb -n Startup}.
42516 @ifset SYSTEM_GDBINIT
42517 @item @value{SYSTEM_GDBINIT}
42519 @ifclear SYSTEM_GDBINIT
42520 @item (not enabled with @code{--with-system-gdbinit} during compilation)
42522 System-wide initialization file. It is executed unless user specified
42523 @value{GDBN} option @code{-nx} or @code{-n}.
42526 the @value{GDBN} manual in node @code{System-wide configuration}
42527 -- shell command @code{info -f gdb -n 'System-wide configuration'}.
42530 @ref{System-wide configuration}.
42534 User initialization file. It is executed unless user specified
42535 @value{GDBN} options @code{-nx}, @code{-n} or @code{-nh}.
42538 Initialization file for current directory. It may need to be enabled with
42539 @value{GDBN} security command @code{set auto-load local-gdbinit}.
42542 the @value{GDBN} manual in node @code{Init File in the Current Directory}
42543 -- shell command @code{info -f gdb -n 'Init File in the Current Directory'}.
42546 @ref{Init File in the Current Directory}.
42551 @c man begin SEEALSO gdbinit
42553 gdb(1), @code{info -f gdb -n Startup}
42555 The full documentation for @value{GDBN} is maintained as a Texinfo manual.
42556 If the @code{info} and @code{gdb} programs and @value{GDBN}'s Texinfo
42557 documentation are properly installed at your site, the command
42563 should give you access to the complete manual.
42565 @cite{Using GDB: A Guide to the GNU Source-Level Debugger},
42566 Richard M. Stallman and Roland H. Pesch, July 1991.
42572 @node GNU Free Documentation License
42573 @appendix GNU Free Documentation License
42576 @node Concept Index
42577 @unnumbered Concept Index
42581 @node Command and Variable Index
42582 @unnumbered Command, Variable, and Function Index
42587 % I think something like @@colophon should be in texinfo. In the
42589 \long\def\colophon{\hbox to0pt{}\vfill
42590 \centerline{The body of this manual is set in}
42591 \centerline{\fontname\tenrm,}
42592 \centerline{with headings in {\bf\fontname\tenbf}}
42593 \centerline{and examples in {\tt\fontname\tentt}.}
42594 \centerline{{\it\fontname\tenit\/},}
42595 \centerline{{\bf\fontname\tenbf}, and}
42596 \centerline{{\sl\fontname\tensl\/}}
42597 \centerline{are used for emphasis.}\vfill}
42599 % Blame: doc@@cygnus.com, 1991.