1 \input texinfo @c -*-texinfo-*-
2 @c Copyright (C) 1988-2019 Free Software Foundation, Inc.
5 @c makeinfo ignores cmds prev to setfilename, so its arg cannot make use
6 @c of @set vars. However, you can override filename with makeinfo -o.
13 @settitle Debugging with @value{GDBN}
14 @setchapternewpage odd
23 @c To avoid file-name clashes between index.html and Index.html, when
24 @c the manual is produced on a Posix host and then moved to a
25 @c case-insensitive filesystem (e.g., MS-Windows), we separate the
26 @c indices into two: Concept Index and all the rest.
30 @c readline appendices use @vindex, @findex and @ftable,
31 @c annotate.texi and gdbmi use @findex.
34 @c !!set GDB manual's edition---not the same as GDB version!
35 @c This is updated by GNU Press.
38 @c !!set GDB edit command default editor
41 @c THIS MANUAL REQUIRES TEXINFO 4.0 OR LATER.
43 @c This is a dir.info fragment to support semi-automated addition of
44 @c manuals to an info tree.
45 @dircategory Software development
47 * Gdb: (gdb). The GNU debugger.
48 * gdbserver: (gdb) Server. The GNU debugging server.
52 @c man begin COPYRIGHT
53 Copyright @copyright{} 1988-2019 Free Software Foundation, Inc.
55 Permission is granted to copy, distribute and/or modify this document
56 under the terms of the GNU Free Documentation License, Version 1.3 or
57 any later version published by the Free Software Foundation; with the
58 Invariant Sections being ``Free Software'' and ``Free Software Needs
59 Free Documentation'', with the Front-Cover Texts being ``A GNU Manual,''
60 and with the Back-Cover Texts as in (a) below.
62 (a) The FSF's Back-Cover Text is: ``You are free to copy and modify
63 this GNU Manual. Buying copies from GNU Press supports the FSF in
64 developing GNU and promoting software freedom.''
69 This file documents the @sc{gnu} debugger @value{GDBN}.
71 This is the @value{EDITION} Edition, of @cite{Debugging with
72 @value{GDBN}: the @sc{gnu} Source-Level Debugger} for @value{GDBN}
73 @ifset VERSION_PACKAGE
74 @value{VERSION_PACKAGE}
76 Version @value{GDBVN}.
82 @title Debugging with @value{GDBN}
83 @subtitle The @sc{gnu} Source-Level Debugger
85 @subtitle @value{EDITION} Edition, for @value{GDBN} version @value{GDBVN}
86 @ifset VERSION_PACKAGE
88 @subtitle @value{VERSION_PACKAGE}
90 @author Richard Stallman, Roland Pesch, Stan Shebs, et al.
94 \hfill (Send bugs and comments on @value{GDBN} to @value{BUGURL}.)\par
95 \hfill {\it Debugging with @value{GDBN}}\par
96 \hfill \TeX{}info \texinfoversion\par
100 @vskip 0pt plus 1filll
101 Published by the Free Software Foundation @*
102 51 Franklin Street, Fifth Floor,
103 Boston, MA 02110-1301, USA@*
104 ISBN 978-0-9831592-3-0 @*
111 @node Top, Summary, (dir), (dir)
113 @top Debugging with @value{GDBN}
115 This file describes @value{GDBN}, the @sc{gnu} symbolic debugger.
117 This is the @value{EDITION} Edition, for @value{GDBN}
118 @ifset VERSION_PACKAGE
119 @value{VERSION_PACKAGE}
121 Version @value{GDBVN}.
123 Copyright (C) 1988-2019 Free Software Foundation, Inc.
125 This edition of the GDB manual is dedicated to the memory of Fred
126 Fish. Fred was a long-standing contributor to GDB and to Free
127 software in general. We will miss him.
130 * Summary:: Summary of @value{GDBN}
131 * Sample Session:: A sample @value{GDBN} session
133 * Invocation:: Getting in and out of @value{GDBN}
134 * Commands:: @value{GDBN} commands
135 * Running:: Running programs under @value{GDBN}
136 * Stopping:: Stopping and continuing
137 * Reverse Execution:: Running programs backward
138 * Process Record and Replay:: Recording inferior's execution and replaying it
139 * Stack:: Examining the stack
140 * Source:: Examining source files
141 * Data:: Examining data
142 * Optimized Code:: Debugging optimized code
143 * Macros:: Preprocessor Macros
144 * Tracepoints:: Debugging remote targets non-intrusively
145 * Overlays:: Debugging programs that use overlays
147 * Languages:: Using @value{GDBN} with different languages
149 * Symbols:: Examining the symbol table
150 * Altering:: Altering execution
151 * GDB Files:: @value{GDBN} files
152 * Targets:: Specifying a debugging target
153 * Remote Debugging:: Debugging remote programs
154 * Configurations:: Configuration-specific information
155 * Controlling GDB:: Controlling @value{GDBN}
156 * Extending GDB:: Extending @value{GDBN}
157 * Interpreters:: Command Interpreters
158 * TUI:: @value{GDBN} Text User Interface
159 * Emacs:: Using @value{GDBN} under @sc{gnu} Emacs
160 * GDB/MI:: @value{GDBN}'s Machine Interface.
161 * Annotations:: @value{GDBN}'s annotation interface.
162 * JIT Interface:: Using the JIT debugging interface.
163 * In-Process Agent:: In-Process Agent
165 * GDB Bugs:: Reporting bugs in @value{GDBN}
167 @ifset SYSTEM_READLINE
168 * Command Line Editing: (rluserman). Command Line Editing
169 * Using History Interactively: (history). Using History Interactively
171 @ifclear SYSTEM_READLINE
172 * Command Line Editing:: Command Line Editing
173 * Using History Interactively:: Using History Interactively
175 * In Memoriam:: In Memoriam
176 * Formatting Documentation:: How to format and print @value{GDBN} documentation
177 * Installing GDB:: Installing GDB
178 * Maintenance Commands:: Maintenance Commands
179 * Remote Protocol:: GDB Remote Serial Protocol
180 * Agent Expressions:: The GDB Agent Expression Mechanism
181 * Target Descriptions:: How targets can describe themselves to
183 * Operating System Information:: Getting additional information from
185 * Linux kernel ptrace restrictions:: Restrictions sometimes
187 kernel on @code{ptrace}
188 * Trace File Format:: GDB trace file format
189 * Index Section Format:: .gdb_index section format
190 * Man Pages:: Manual pages
191 * Copying:: GNU General Public License says
192 how you can copy and share GDB
193 * GNU Free Documentation License:: The license for this documentation
194 * Concept Index:: Index of @value{GDBN} concepts
195 * Command and Variable Index:: Index of @value{GDBN} commands, variables,
196 functions, and Python data types
204 @unnumbered Summary of @value{GDBN}
206 The purpose of a debugger such as @value{GDBN} is to allow you to see what is
207 going on ``inside'' another program while it executes---or what another
208 program was doing at the moment it crashed.
210 @value{GDBN} can do four main kinds of things (plus other things in support of
211 these) to help you catch bugs in the act:
215 Start your program, specifying anything that might affect its behavior.
218 Make your program stop on specified conditions.
221 Examine what has happened, when your program has stopped.
224 Change things in your program, so you can experiment with correcting the
225 effects of one bug and go on to learn about another.
228 You can use @value{GDBN} to debug programs written in C and C@t{++}.
229 For more information, see @ref{Supported Languages,,Supported Languages}.
230 For more information, see @ref{C,,C and C++}.
232 Support for D is partial. For information on D, see
236 Support for Modula-2 is partial. For information on Modula-2, see
237 @ref{Modula-2,,Modula-2}.
239 Support for OpenCL C is partial. For information on OpenCL C, see
240 @ref{OpenCL C,,OpenCL C}.
243 Debugging Pascal programs which use sets, subranges, file variables, or
244 nested functions does not currently work. @value{GDBN} does not support
245 entering expressions, printing values, or similar features using Pascal
249 @value{GDBN} can be used to debug programs written in Fortran, although
250 it may be necessary to refer to some variables with a trailing
253 @value{GDBN} can be used to debug programs written in Objective-C,
254 using either the Apple/NeXT or the GNU Objective-C runtime.
257 * Free Software:: Freely redistributable software
258 * Free Documentation:: Free Software Needs Free Documentation
259 * Contributors:: Contributors to GDB
263 @unnumberedsec Free Software
265 @value{GDBN} is @dfn{free software}, protected by the @sc{gnu}
266 General Public License
267 (GPL). The GPL gives you the freedom to copy or adapt a licensed
268 program---but every person getting a copy also gets with it the
269 freedom to modify that copy (which means that they must get access to
270 the source code), and the freedom to distribute further copies.
271 Typical software companies use copyrights to limit your freedoms; the
272 Free Software Foundation uses the GPL to preserve these freedoms.
274 Fundamentally, the General Public License is a license which says that
275 you have these freedoms and that you cannot take these freedoms away
278 @node Free Documentation
279 @unnumberedsec Free Software Needs Free Documentation
281 The biggest deficiency in the free software community today is not in
282 the software---it is the lack of good free documentation that we can
283 include with the free software. Many of our most important
284 programs do not come with free reference manuals and free introductory
285 texts. Documentation is an essential part of any software package;
286 when an important free software package does not come with a free
287 manual and a free tutorial, that is a major gap. We have many such
290 Consider Perl, for instance. The tutorial manuals that people
291 normally use are non-free. How did this come about? Because the
292 authors of those manuals published them with restrictive terms---no
293 copying, no modification, source files not available---which exclude
294 them from the free software world.
296 That wasn't the first time this sort of thing happened, and it was far
297 from the last. Many times we have heard a GNU user eagerly describe a
298 manual that he is writing, his intended contribution to the community,
299 only to learn that he had ruined everything by signing a publication
300 contract to make it non-free.
302 Free documentation, like free software, is a matter of freedom, not
303 price. The problem with the non-free manual is not that publishers
304 charge a price for printed copies---that in itself is fine. (The Free
305 Software Foundation sells printed copies of manuals, too.) The
306 problem is the restrictions on the use of the manual. Free manuals
307 are available in source code form, and give you permission to copy and
308 modify. Non-free manuals do not allow this.
310 The criteria of freedom for a free manual are roughly the same as for
311 free software. Redistribution (including the normal kinds of
312 commercial redistribution) must be permitted, so that the manual can
313 accompany every copy of the program, both on-line and on paper.
315 Permission for modification of the technical content is crucial too.
316 When people modify the software, adding or changing features, if they
317 are conscientious they will change the manual too---so they can
318 provide accurate and clear documentation for the modified program. A
319 manual that leaves you no choice but to write a new manual to document
320 a changed version of the program is not really available to our
323 Some kinds of limits on the way modification is handled are
324 acceptable. For example, requirements to preserve the original
325 author's copyright notice, the distribution terms, or the list of
326 authors, are ok. It is also no problem to require modified versions
327 to include notice that they were modified. Even entire sections that
328 may not be deleted or changed are acceptable, as long as they deal
329 with nontechnical topics (like this one). These kinds of restrictions
330 are acceptable because they don't obstruct the community's normal use
333 However, it must be possible to modify all the @emph{technical}
334 content of the manual, and then distribute the result in all the usual
335 media, through all the usual channels. Otherwise, the restrictions
336 obstruct the use of the manual, it is not free, and we need another
337 manual to replace it.
339 Please spread the word about this issue. Our community continues to
340 lose manuals to proprietary publishing. If we spread the word that
341 free software needs free reference manuals and free tutorials, perhaps
342 the next person who wants to contribute by writing documentation will
343 realize, before it is too late, that only free manuals contribute to
344 the free software community.
346 If you are writing documentation, please insist on publishing it under
347 the GNU Free Documentation License or another free documentation
348 license. Remember that this decision requires your approval---you
349 don't have to let the publisher decide. Some commercial publishers
350 will use a free license if you insist, but they will not propose the
351 option; it is up to you to raise the issue and say firmly that this is
352 what you want. If the publisher you are dealing with refuses, please
353 try other publishers. If you're not sure whether a proposed license
354 is free, write to @email{licensing@@gnu.org}.
356 You can encourage commercial publishers to sell more free, copylefted
357 manuals and tutorials by buying them, and particularly by buying
358 copies from the publishers that paid for their writing or for major
359 improvements. Meanwhile, try to avoid buying non-free documentation
360 at all. Check the distribution terms of a manual before you buy it,
361 and insist that whoever seeks your business must respect your freedom.
362 Check the history of the book, and try to reward the publishers that
363 have paid or pay the authors to work on it.
365 The Free Software Foundation maintains a list of free documentation
366 published by other publishers, at
367 @url{http://www.fsf.org/doc/other-free-books.html}.
370 @unnumberedsec Contributors to @value{GDBN}
372 Richard Stallman was the original author of @value{GDBN}, and of many
373 other @sc{gnu} programs. Many others have contributed to its
374 development. This section attempts to credit major contributors. One
375 of the virtues of free software is that everyone is free to contribute
376 to it; with regret, we cannot actually acknowledge everyone here. The
377 file @file{ChangeLog} in the @value{GDBN} distribution approximates a
378 blow-by-blow account.
380 Changes much prior to version 2.0 are lost in the mists of time.
383 @emph{Plea:} Additions to this section are particularly welcome. If you
384 or your friends (or enemies, to be evenhanded) have been unfairly
385 omitted from this list, we would like to add your names!
388 So that they may not regard their many labors as thankless, we
389 particularly thank those who shepherded @value{GDBN} through major
391 Andrew Cagney (releases 6.3, 6.2, 6.1, 6.0, 5.3, 5.2, 5.1 and 5.0);
392 Jim Blandy (release 4.18);
393 Jason Molenda (release 4.17);
394 Stan Shebs (release 4.14);
395 Fred Fish (releases 4.16, 4.15, 4.13, 4.12, 4.11, 4.10, and 4.9);
396 Stu Grossman and John Gilmore (releases 4.8, 4.7, 4.6, 4.5, and 4.4);
397 John Gilmore (releases 4.3, 4.2, 4.1, 4.0, and 3.9);
398 Jim Kingdon (releases 3.5, 3.4, and 3.3);
399 and Randy Smith (releases 3.2, 3.1, and 3.0).
401 Richard Stallman, assisted at various times by Peter TerMaat, Chris
402 Hanson, and Richard Mlynarik, handled releases through 2.8.
404 Michael Tiemann is the author of most of the @sc{gnu} C@t{++} support
405 in @value{GDBN}, with significant additional contributions from Per
406 Bothner and Daniel Berlin. James Clark wrote the @sc{gnu} C@t{++}
407 demangler. Early work on C@t{++} was by Peter TerMaat (who also did
408 much general update work leading to release 3.0).
410 @value{GDBN} uses the BFD subroutine library to examine multiple
411 object-file formats; BFD was a joint project of David V.
412 Henkel-Wallace, Rich Pixley, Steve Chamberlain, and John Gilmore.
414 David Johnson wrote the original COFF support; Pace Willison did
415 the original support for encapsulated COFF.
417 Brent Benson of Harris Computer Systems contributed DWARF 2 support.
419 Adam de Boor and Bradley Davis contributed the ISI Optimum V support.
420 Per Bothner, Noboyuki Hikichi, and Alessandro Forin contributed MIPS
422 Jean-Daniel Fekete contributed Sun 386i support.
423 Chris Hanson improved the HP9000 support.
424 Noboyuki Hikichi and Tomoyuki Hasei contributed Sony/News OS 3 support.
425 David Johnson contributed Encore Umax support.
426 Jyrki Kuoppala contributed Altos 3068 support.
427 Jeff Law contributed HP PA and SOM support.
428 Keith Packard contributed NS32K support.
429 Doug Rabson contributed Acorn Risc Machine support.
430 Bob Rusk contributed Harris Nighthawk CX-UX support.
431 Chris Smith contributed Convex support (and Fortran debugging).
432 Jonathan Stone contributed Pyramid support.
433 Michael Tiemann contributed SPARC support.
434 Tim Tucker contributed support for the Gould NP1 and Gould Powernode.
435 Pace Willison contributed Intel 386 support.
436 Jay Vosburgh contributed Symmetry support.
437 Marko Mlinar contributed OpenRISC 1000 support.
439 Andreas Schwab contributed M68K @sc{gnu}/Linux support.
441 Rich Schaefer and Peter Schauer helped with support of SunOS shared
444 Jay Fenlason and Roland McGrath ensured that @value{GDBN} and GAS agree
445 about several machine instruction sets.
447 Patrick Duval, Ted Goldstein, Vikram Koka and Glenn Engel helped develop
448 remote debugging. Intel Corporation, Wind River Systems, AMD, and ARM
449 contributed remote debugging modules for the i960, VxWorks, A29K UDI,
450 and RDI targets, respectively.
452 Brian Fox is the author of the readline libraries providing
453 command-line editing and command history.
455 Andrew Beers of SUNY Buffalo wrote the language-switching code, the
456 Modula-2 support, and contributed the Languages chapter of this manual.
458 Fred Fish wrote most of the support for Unix System Vr4.
459 He also enhanced the command-completion support to cover C@t{++} overloaded
462 Hitachi America (now Renesas America), Ltd. sponsored the support for
463 H8/300, H8/500, and Super-H processors.
465 NEC sponsored the support for the v850, Vr4xxx, and Vr5xxx processors.
467 Mitsubishi (now Renesas) sponsored the support for D10V, D30V, and M32R/D
470 Toshiba sponsored the support for the TX39 Mips processor.
472 Matsushita sponsored the support for the MN10200 and MN10300 processors.
474 Fujitsu sponsored the support for SPARClite and FR30 processors.
476 Kung Hsu, Jeff Law, and Rick Sladkey added support for hardware
479 Michael Snyder added support for tracepoints.
481 Stu Grossman wrote gdbserver.
483 Jim Kingdon, Peter Schauer, Ian Taylor, and Stu Grossman made
484 nearly innumerable bug fixes and cleanups throughout @value{GDBN}.
486 The following people at the Hewlett-Packard Company contributed
487 support for the PA-RISC 2.0 architecture, HP-UX 10.20, 10.30, and 11.0
488 (narrow mode), HP's implementation of kernel threads, HP's aC@t{++}
489 compiler, and the Text User Interface (nee Terminal User Interface):
490 Ben Krepp, Richard Title, John Bishop, Susan Macchia, Kathy Mann,
491 Satish Pai, India Paul, Steve Rehrauer, and Elena Zannoni. Kim Haase
492 provided HP-specific information in this manual.
494 DJ Delorie ported @value{GDBN} to MS-DOS, for the DJGPP project.
495 Robert Hoehne made significant contributions to the DJGPP port.
497 Cygnus Solutions has sponsored @value{GDBN} maintenance and much of its
498 development since 1991. Cygnus engineers who have worked on @value{GDBN}
499 fulltime include Mark Alexander, Jim Blandy, Per Bothner, Kevin
500 Buettner, Edith Epstein, Chris Faylor, Fred Fish, Martin Hunt, Jim
501 Ingham, John Gilmore, Stu Grossman, Kung Hsu, Jim Kingdon, John Metzler,
502 Fernando Nasser, Geoffrey Noer, Dawn Perchik, Rich Pixley, Zdenek
503 Radouch, Keith Seitz, Stan Shebs, David Taylor, and Elena Zannoni. In
504 addition, Dave Brolley, Ian Carmichael, Steve Chamberlain, Nick Clifton,
505 JT Conklin, Stan Cox, DJ Delorie, Ulrich Drepper, Frank Eigler, Doug
506 Evans, Sean Fagan, David Henkel-Wallace, Richard Henderson, Jeff
507 Holcomb, Jeff Law, Jim Lemke, Tom Lord, Bob Manson, Michael Meissner,
508 Jason Merrill, Catherine Moore, Drew Moseley, Ken Raeburn, Gavin
509 Romig-Koch, Rob Savoye, Jamie Smith, Mike Stump, Ian Taylor, Angela
510 Thomas, Michael Tiemann, Tom Tromey, Ron Unrau, Jim Wilson, and David
511 Zuhn have made contributions both large and small.
513 Andrew Cagney, Fernando Nasser, and Elena Zannoni, while working for
514 Cygnus Solutions, implemented the original @sc{gdb/mi} interface.
516 Jim Blandy added support for preprocessor macros, while working for Red
519 Andrew Cagney designed @value{GDBN}'s architecture vector. Many
520 people including Andrew Cagney, Stephane Carrez, Randolph Chung, Nick
521 Duffek, Richard Henderson, Mark Kettenis, Grace Sainsbury, Kei
522 Sakamoto, Yoshinori Sato, Michael Snyder, Andreas Schwab, Jason
523 Thorpe, Corinna Vinschen, Ulrich Weigand, and Elena Zannoni, helped
524 with the migration of old architectures to this new framework.
526 Andrew Cagney completely re-designed and re-implemented @value{GDBN}'s
527 unwinder framework, this consisting of a fresh new design featuring
528 frame IDs, independent frame sniffers, and the sentinel frame. Mark
529 Kettenis implemented the @sc{dwarf 2} unwinder, Jeff Johnston the
530 libunwind unwinder, and Andrew Cagney the dummy, sentinel, tramp, and
531 trad unwinders. The architecture-specific changes, each involving a
532 complete rewrite of the architecture's frame code, were carried out by
533 Jim Blandy, Joel Brobecker, Kevin Buettner, Andrew Cagney, Stephane
534 Carrez, Randolph Chung, Orjan Friberg, Richard Henderson, Daniel
535 Jacobowitz, Jeff Johnston, Mark Kettenis, Theodore A. Roth, Kei
536 Sakamoto, Yoshinori Sato, Michael Snyder, Corinna Vinschen, and Ulrich
539 Christian Zankel, Ross Morley, Bob Wilson, and Maxim Grigoriev from
540 Tensilica, Inc.@: contributed support for Xtensa processors. Others
541 who have worked on the Xtensa port of @value{GDBN} in the past include
542 Steve Tjiang, John Newlin, and Scott Foehner.
544 Michael Eager and staff of Xilinx, Inc., contributed support for the
545 Xilinx MicroBlaze architecture.
547 Initial support for the FreeBSD/mips target and native configuration
548 was developed by SRI International and the University of Cambridge
549 Computer Laboratory under DARPA/AFRL contract FA8750-10-C-0237
550 ("CTSRD"), as part of the DARPA CRASH research programme.
552 Initial support for the FreeBSD/riscv target and native configuration
553 was developed by SRI International and the University of Cambridge
554 Computer Laboratory (Department of Computer Science and Technology)
555 under DARPA contract HR0011-18-C-0016 ("ECATS"), as part of the DARPA
556 SSITH research programme.
558 The original port to the OpenRISC 1000 is believed to be due to
559 Alessandro Forin and Per Bothner. More recent ports have been the work
560 of Jeremy Bennett, Franck Jullien, Stefan Wallentowitz and
564 @chapter A Sample @value{GDBN} Session
566 You can use this manual at your leisure to read all about @value{GDBN}.
567 However, a handful of commands are enough to get started using the
568 debugger. This chapter illustrates those commands.
571 In this sample session, we emphasize user input like this: @b{input},
572 to make it easier to pick out from the surrounding output.
575 @c FIXME: this example may not be appropriate for some configs, where
576 @c FIXME...primary interest is in remote use.
578 One of the preliminary versions of @sc{gnu} @code{m4} (a generic macro
579 processor) exhibits the following bug: sometimes, when we change its
580 quote strings from the default, the commands used to capture one macro
581 definition within another stop working. In the following short @code{m4}
582 session, we define a macro @code{foo} which expands to @code{0000}; we
583 then use the @code{m4} built-in @code{defn} to define @code{bar} as the
584 same thing. However, when we change the open quote string to
585 @code{<QUOTE>} and the close quote string to @code{<UNQUOTE>}, the same
586 procedure fails to define a new synonym @code{baz}:
595 @b{define(bar,defn(`foo'))}
599 @b{changequote(<QUOTE>,<UNQUOTE>)}
601 @b{define(baz,defn(<QUOTE>foo<UNQUOTE>))}
604 m4: End of input: 0: fatal error: EOF in string
608 Let us use @value{GDBN} to try to see what is going on.
611 $ @b{@value{GDBP} m4}
612 @c FIXME: this falsifies the exact text played out, to permit smallbook
613 @c FIXME... format to come out better.
614 @value{GDBN} is free software and you are welcome to distribute copies
615 of it under certain conditions; type "show copying" to see
617 There is absolutely no warranty for @value{GDBN}; type "show warranty"
620 @value{GDBN} @value{GDBVN}, Copyright 1999 Free Software Foundation, Inc...
625 @value{GDBN} reads only enough symbol data to know where to find the
626 rest when needed; as a result, the first prompt comes up very quickly.
627 We now tell @value{GDBN} to use a narrower display width than usual, so
628 that examples fit in this manual.
631 (@value{GDBP}) @b{set width 70}
635 We need to see how the @code{m4} built-in @code{changequote} works.
636 Having looked at the source, we know the relevant subroutine is
637 @code{m4_changequote}, so we set a breakpoint there with the @value{GDBN}
638 @code{break} command.
641 (@value{GDBP}) @b{break m4_changequote}
642 Breakpoint 1 at 0x62f4: file builtin.c, line 879.
646 Using the @code{run} command, we start @code{m4} running under @value{GDBN}
647 control; as long as control does not reach the @code{m4_changequote}
648 subroutine, the program runs as usual:
651 (@value{GDBP}) @b{run}
652 Starting program: /work/Editorial/gdb/gnu/m4/m4
660 To trigger the breakpoint, we call @code{changequote}. @value{GDBN}
661 suspends execution of @code{m4}, displaying information about the
662 context where it stops.
665 @b{changequote(<QUOTE>,<UNQUOTE>)}
667 Breakpoint 1, m4_changequote (argc=3, argv=0x33c70)
669 879 if (bad_argc(TOKEN_DATA_TEXT(argv[0]),argc,1,3))
673 Now we use the command @code{n} (@code{next}) to advance execution to
674 the next line of the current function.
678 882 set_quotes((argc >= 2) ? TOKEN_DATA_TEXT(argv[1])\
683 @code{set_quotes} looks like a promising subroutine. We can go into it
684 by using the command @code{s} (@code{step}) instead of @code{next}.
685 @code{step} goes to the next line to be executed in @emph{any}
686 subroutine, so it steps into @code{set_quotes}.
690 set_quotes (lq=0x34c78 "<QUOTE>", rq=0x34c88 "<UNQUOTE>")
692 530 if (lquote != def_lquote)
696 The display that shows the subroutine where @code{m4} is now
697 suspended (and its arguments) is called a stack frame display. It
698 shows a summary of the stack. We can use the @code{backtrace}
699 command (which can also be spelled @code{bt}), to see where we are
700 in the stack as a whole: the @code{backtrace} command displays a
701 stack frame for each active subroutine.
704 (@value{GDBP}) @b{bt}
705 #0 set_quotes (lq=0x34c78 "<QUOTE>", rq=0x34c88 "<UNQUOTE>")
707 #1 0x6344 in m4_changequote (argc=3, argv=0x33c70)
709 #2 0x8174 in expand_macro (sym=0x33320) at macro.c:242
710 #3 0x7a88 in expand_token (obs=0x0, t=209696, td=0xf7fffa30)
712 #4 0x79dc in expand_input () at macro.c:40
713 #5 0x2930 in main (argc=0, argv=0xf7fffb20) at m4.c:195
717 We step through a few more lines to see what happens. The first two
718 times, we can use @samp{s}; the next two times we use @code{n} to avoid
719 falling into the @code{xstrdup} subroutine.
723 0x3b5c 532 if (rquote != def_rquote)
725 0x3b80 535 lquote = (lq == nil || *lq == '\0') ? \
726 def_lquote : xstrdup(lq);
728 536 rquote = (rq == nil || *rq == '\0') ? def_rquote\
731 538 len_lquote = strlen(rquote);
735 The last line displayed looks a little odd; we can examine the variables
736 @code{lquote} and @code{rquote} to see if they are in fact the new left
737 and right quotes we specified. We use the command @code{p}
738 (@code{print}) to see their values.
741 (@value{GDBP}) @b{p lquote}
742 $1 = 0x35d40 "<QUOTE>"
743 (@value{GDBP}) @b{p rquote}
744 $2 = 0x35d50 "<UNQUOTE>"
748 @code{lquote} and @code{rquote} are indeed the new left and right quotes.
749 To look at some context, we can display ten lines of source
750 surrounding the current line with the @code{l} (@code{list}) command.
756 535 lquote = (lq == nil || *lq == '\0') ? def_lquote\
758 536 rquote = (rq == nil || *rq == '\0') ? def_rquote\
761 538 len_lquote = strlen(rquote);
762 539 len_rquote = strlen(lquote);
769 Let us step past the two lines that set @code{len_lquote} and
770 @code{len_rquote}, and then examine the values of those variables.
774 539 len_rquote = strlen(lquote);
777 (@value{GDBP}) @b{p len_lquote}
779 (@value{GDBP}) @b{p len_rquote}
784 That certainly looks wrong, assuming @code{len_lquote} and
785 @code{len_rquote} are meant to be the lengths of @code{lquote} and
786 @code{rquote} respectively. We can set them to better values using
787 the @code{p} command, since it can print the value of
788 any expression---and that expression can include subroutine calls and
792 (@value{GDBP}) @b{p len_lquote=strlen(lquote)}
794 (@value{GDBP}) @b{p len_rquote=strlen(rquote)}
799 Is that enough to fix the problem of using the new quotes with the
800 @code{m4} built-in @code{defn}? We can allow @code{m4} to continue
801 executing with the @code{c} (@code{continue}) command, and then try the
802 example that caused trouble initially:
808 @b{define(baz,defn(<QUOTE>foo<UNQUOTE>))}
815 Success! The new quotes now work just as well as the default ones. The
816 problem seems to have been just the two typos defining the wrong
817 lengths. We allow @code{m4} exit by giving it an EOF as input:
821 Program exited normally.
825 The message @samp{Program exited normally.} is from @value{GDBN}; it
826 indicates @code{m4} has finished executing. We can end our @value{GDBN}
827 session with the @value{GDBN} @code{quit} command.
830 (@value{GDBP}) @b{quit}
834 @chapter Getting In and Out of @value{GDBN}
836 This chapter discusses how to start @value{GDBN}, and how to get out of it.
840 type @samp{@value{GDBP}} to start @value{GDBN}.
842 type @kbd{quit} or @kbd{Ctrl-d} to exit.
846 * Invoking GDB:: How to start @value{GDBN}
847 * Quitting GDB:: How to quit @value{GDBN}
848 * Shell Commands:: How to use shell commands inside @value{GDBN}
849 * Logging Output:: How to log @value{GDBN}'s output to a file
853 @section Invoking @value{GDBN}
855 Invoke @value{GDBN} by running the program @code{@value{GDBP}}. Once started,
856 @value{GDBN} reads commands from the terminal until you tell it to exit.
858 You can also run @code{@value{GDBP}} with a variety of arguments and options,
859 to specify more of your debugging environment at the outset.
861 The command-line options described here are designed
862 to cover a variety of situations; in some environments, some of these
863 options may effectively be unavailable.
865 The most usual way to start @value{GDBN} is with one argument,
866 specifying an executable program:
869 @value{GDBP} @var{program}
873 You can also start with both an executable program and a core file
877 @value{GDBP} @var{program} @var{core}
880 You can, instead, specify a process ID as a second argument or use option
881 @code{-p}, if you want to debug a running process:
884 @value{GDBP} @var{program} 1234
889 would attach @value{GDBN} to process @code{1234}. With option @option{-p} you
890 can omit the @var{program} filename.
892 Taking advantage of the second command-line argument requires a fairly
893 complete operating system; when you use @value{GDBN} as a remote
894 debugger attached to a bare board, there may not be any notion of
895 ``process'', and there is often no way to get a core dump. @value{GDBN}
896 will warn you if it is unable to attach or to read core dumps.
898 You can optionally have @code{@value{GDBP}} pass any arguments after the
899 executable file to the inferior using @code{--args}. This option stops
902 @value{GDBP} --args gcc -O2 -c foo.c
904 This will cause @code{@value{GDBP}} to debug @code{gcc}, and to set
905 @code{gcc}'s command-line arguments (@pxref{Arguments}) to @samp{-O2 -c foo.c}.
907 You can run @code{@value{GDBP}} without printing the front material, which describes
908 @value{GDBN}'s non-warranty, by specifying @code{--silent}
909 (or @code{-q}/@code{--quiet}):
912 @value{GDBP} --silent
916 You can further control how @value{GDBN} starts up by using command-line
917 options. @value{GDBN} itself can remind you of the options available.
927 to display all available options and briefly describe their use
928 (@samp{@value{GDBP} -h} is a shorter equivalent).
930 All options and command line arguments you give are processed
931 in sequential order. The order makes a difference when the
932 @samp{-x} option is used.
936 * File Options:: Choosing files
937 * Mode Options:: Choosing modes
938 * Startup:: What @value{GDBN} does during startup
942 @subsection Choosing Files
944 When @value{GDBN} starts, it reads any arguments other than options as
945 specifying an executable file and core file (or process ID). This is
946 the same as if the arguments were specified by the @samp{-se} and
947 @samp{-c} (or @samp{-p}) options respectively. (@value{GDBN} reads the
948 first argument that does not have an associated option flag as
949 equivalent to the @samp{-se} option followed by that argument; and the
950 second argument that does not have an associated option flag, if any, as
951 equivalent to the @samp{-c}/@samp{-p} option followed by that argument.)
952 If the second argument begins with a decimal digit, @value{GDBN} will
953 first attempt to attach to it as a process, and if that fails, attempt
954 to open it as a corefile. If you have a corefile whose name begins with
955 a digit, you can prevent @value{GDBN} from treating it as a pid by
956 prefixing it with @file{./}, e.g.@: @file{./12345}.
958 If @value{GDBN} has not been configured to included core file support,
959 such as for most embedded targets, then it will complain about a second
960 argument and ignore it.
962 Many options have both long and short forms; both are shown in the
963 following list. @value{GDBN} also recognizes the long forms if you truncate
964 them, so long as enough of the option is present to be unambiguous.
965 (If you prefer, you can flag option arguments with @samp{--} rather
966 than @samp{-}, though we illustrate the more usual convention.)
968 @c NOTE: the @cindex entries here use double dashes ON PURPOSE. This
969 @c way, both those who look for -foo and --foo in the index, will find
973 @item -symbols @var{file}
975 @cindex @code{--symbols}
977 Read symbol table from file @var{file}.
979 @item -exec @var{file}
981 @cindex @code{--exec}
983 Use file @var{file} as the executable file to execute when appropriate,
984 and for examining pure data in conjunction with a core dump.
988 Read symbol table from file @var{file} and use it as the executable
991 @item -core @var{file}
993 @cindex @code{--core}
995 Use file @var{file} as a core dump to examine.
997 @item -pid @var{number}
998 @itemx -p @var{number}
1001 Connect to process ID @var{number}, as with the @code{attach} command.
1003 @item -command @var{file}
1004 @itemx -x @var{file}
1005 @cindex @code{--command}
1007 Execute commands from file @var{file}. The contents of this file is
1008 evaluated exactly as the @code{source} command would.
1009 @xref{Command Files,, Command files}.
1011 @item -eval-command @var{command}
1012 @itemx -ex @var{command}
1013 @cindex @code{--eval-command}
1015 Execute a single @value{GDBN} command.
1017 This option may be used multiple times to call multiple commands. It may
1018 also be interleaved with @samp{-command} as required.
1021 @value{GDBP} -ex 'target sim' -ex 'load' \
1022 -x setbreakpoints -ex 'run' a.out
1025 @item -init-command @var{file}
1026 @itemx -ix @var{file}
1027 @cindex @code{--init-command}
1029 Execute commands from file @var{file} before loading the inferior (but
1030 after loading gdbinit files).
1033 @item -init-eval-command @var{command}
1034 @itemx -iex @var{command}
1035 @cindex @code{--init-eval-command}
1037 Execute a single @value{GDBN} command before loading the inferior (but
1038 after loading gdbinit files).
1041 @item -directory @var{directory}
1042 @itemx -d @var{directory}
1043 @cindex @code{--directory}
1045 Add @var{directory} to the path to search for source and script files.
1049 @cindex @code{--readnow}
1051 Read each symbol file's entire symbol table immediately, rather than
1052 the default, which is to read it incrementally as it is needed.
1053 This makes startup slower, but makes future operations faster.
1056 @anchor{--readnever}
1057 @cindex @code{--readnever}, command-line option
1058 Do not read each symbol file's symbolic debug information. This makes
1059 startup faster but at the expense of not being able to perform
1060 symbolic debugging. DWARF unwind information is also not read,
1061 meaning backtraces may become incomplete or inaccurate. One use of
1062 this is when a user simply wants to do the following sequence: attach,
1063 dump core, detach. Loading the debugging information in this case is
1064 an unnecessary cause of delay.
1068 @subsection Choosing Modes
1070 You can run @value{GDBN} in various alternative modes---for example, in
1071 batch mode or quiet mode.
1079 Do not execute commands found in any initialization file.
1080 There are three init files, loaded in the following order:
1083 @item @file{system.gdbinit}
1084 This is the system-wide init file.
1085 Its location is specified with the @code{--with-system-gdbinit}
1086 configure option (@pxref{System-wide configuration}).
1087 It is loaded first when @value{GDBN} starts, before command line options
1088 have been processed.
1089 @item @file{~/.gdbinit}
1090 This is the init file in your home directory.
1091 It is loaded next, after @file{system.gdbinit}, and before
1092 command options have been processed.
1093 @item @file{./.gdbinit}
1094 This is the init file in the current directory.
1095 It is loaded last, after command line options other than @code{-x} and
1096 @code{-ex} have been processed. Command line options @code{-x} and
1097 @code{-ex} are processed last, after @file{./.gdbinit} has been loaded.
1100 For further documentation on startup processing, @xref{Startup}.
1101 For documentation on how to write command files,
1102 @xref{Command Files,,Command Files}.
1107 Do not execute commands found in @file{~/.gdbinit}, the init file
1108 in your home directory.
1114 @cindex @code{--quiet}
1115 @cindex @code{--silent}
1117 ``Quiet''. Do not print the introductory and copyright messages. These
1118 messages are also suppressed in batch mode.
1121 @cindex @code{--batch}
1122 Run in batch mode. Exit with status @code{0} after processing all the
1123 command files specified with @samp{-x} (and all commands from
1124 initialization files, if not inhibited with @samp{-n}). Exit with
1125 nonzero status if an error occurs in executing the @value{GDBN} commands
1126 in the command files. Batch mode also disables pagination, sets unlimited
1127 terminal width and height @pxref{Screen Size}, and acts as if @kbd{set confirm
1128 off} were in effect (@pxref{Messages/Warnings}).
1130 Batch mode may be useful for running @value{GDBN} as a filter, for
1131 example to download and run a program on another computer; in order to
1132 make this more useful, the message
1135 Program exited normally.
1139 (which is ordinarily issued whenever a program running under
1140 @value{GDBN} control terminates) is not issued when running in batch
1144 @cindex @code{--batch-silent}
1145 Run in batch mode exactly like @samp{-batch}, but totally silently. All
1146 @value{GDBN} output to @code{stdout} is prevented (@code{stderr} is
1147 unaffected). This is much quieter than @samp{-silent} and would be useless
1148 for an interactive session.
1150 This is particularly useful when using targets that give @samp{Loading section}
1151 messages, for example.
1153 Note that targets that give their output via @value{GDBN}, as opposed to
1154 writing directly to @code{stdout}, will also be made silent.
1156 @item -return-child-result
1157 @cindex @code{--return-child-result}
1158 The return code from @value{GDBN} will be the return code from the child
1159 process (the process being debugged), with the following exceptions:
1163 @value{GDBN} exits abnormally. E.g., due to an incorrect argument or an
1164 internal error. In this case the exit code is the same as it would have been
1165 without @samp{-return-child-result}.
1167 The user quits with an explicit value. E.g., @samp{quit 1}.
1169 The child process never runs, or is not allowed to terminate, in which case
1170 the exit code will be -1.
1173 This option is useful in conjunction with @samp{-batch} or @samp{-batch-silent},
1174 when @value{GDBN} is being used as a remote program loader or simulator
1179 @cindex @code{--nowindows}
1181 ``No windows''. If @value{GDBN} comes with a graphical user interface
1182 (GUI) built in, then this option tells @value{GDBN} to only use the command-line
1183 interface. If no GUI is available, this option has no effect.
1187 @cindex @code{--windows}
1189 If @value{GDBN} includes a GUI, then this option requires it to be
1192 @item -cd @var{directory}
1194 Run @value{GDBN} using @var{directory} as its working directory,
1195 instead of the current directory.
1197 @item -data-directory @var{directory}
1198 @itemx -D @var{directory}
1199 @cindex @code{--data-directory}
1201 Run @value{GDBN} using @var{directory} as its data directory.
1202 The data directory is where @value{GDBN} searches for its
1203 auxiliary files. @xref{Data Files}.
1207 @cindex @code{--fullname}
1209 @sc{gnu} Emacs sets this option when it runs @value{GDBN} as a
1210 subprocess. It tells @value{GDBN} to output the full file name and line
1211 number in a standard, recognizable fashion each time a stack frame is
1212 displayed (which includes each time your program stops). This
1213 recognizable format looks like two @samp{\032} characters, followed by
1214 the file name, line number and character position separated by colons,
1215 and a newline. The Emacs-to-@value{GDBN} interface program uses the two
1216 @samp{\032} characters as a signal to display the source code for the
1219 @item -annotate @var{level}
1220 @cindex @code{--annotate}
1221 This option sets the @dfn{annotation level} inside @value{GDBN}. Its
1222 effect is identical to using @samp{set annotate @var{level}}
1223 (@pxref{Annotations}). The annotation @var{level} controls how much
1224 information @value{GDBN} prints together with its prompt, values of
1225 expressions, source lines, and other types of output. Level 0 is the
1226 normal, level 1 is for use when @value{GDBN} is run as a subprocess of
1227 @sc{gnu} Emacs, level 3 is the maximum annotation suitable for programs
1228 that control @value{GDBN}, and level 2 has been deprecated.
1230 The annotation mechanism has largely been superseded by @sc{gdb/mi}
1234 @cindex @code{--args}
1235 Change interpretation of command line so that arguments following the
1236 executable file are passed as command line arguments to the inferior.
1237 This option stops option processing.
1239 @item -baud @var{bps}
1241 @cindex @code{--baud}
1243 Set the line speed (baud rate or bits per second) of any serial
1244 interface used by @value{GDBN} for remote debugging.
1246 @item -l @var{timeout}
1248 Set the timeout (in seconds) of any communication used by @value{GDBN}
1249 for remote debugging.
1251 @item -tty @var{device}
1252 @itemx -t @var{device}
1253 @cindex @code{--tty}
1255 Run using @var{device} for your program's standard input and output.
1256 @c FIXME: kingdon thinks there is more to -tty. Investigate.
1258 @c resolve the situation of these eventually
1260 @cindex @code{--tui}
1261 Activate the @dfn{Text User Interface} when starting. The Text User
1262 Interface manages several text windows on the terminal, showing
1263 source, assembly, registers and @value{GDBN} command outputs
1264 (@pxref{TUI, ,@value{GDBN} Text User Interface}). Do not use this
1265 option if you run @value{GDBN} from Emacs (@pxref{Emacs, ,
1266 Using @value{GDBN} under @sc{gnu} Emacs}).
1268 @item -interpreter @var{interp}
1269 @cindex @code{--interpreter}
1270 Use the interpreter @var{interp} for interface with the controlling
1271 program or device. This option is meant to be set by programs which
1272 communicate with @value{GDBN} using it as a back end.
1273 @xref{Interpreters, , Command Interpreters}.
1275 @samp{--interpreter=mi} (or @samp{--interpreter=mi3}) causes
1276 @value{GDBN} to use the @dfn{@sc{gdb/mi} interface} version 3 (@pxref{GDB/MI, ,
1277 The @sc{gdb/mi} Interface}) included since @value{GDBN} version 9.1. @sc{gdb/mi}
1278 version 2 (@code{mi2}), included in @value{GDBN} 6.0 and version 1 (@code{mi1}),
1279 included in @value{GDBN} 5.3, are also available. Earlier @sc{gdb/mi}
1280 interfaces are no longer supported.
1283 @cindex @code{--write}
1284 Open the executable and core files for both reading and writing. This
1285 is equivalent to the @samp{set write on} command inside @value{GDBN}
1289 @cindex @code{--statistics}
1290 This option causes @value{GDBN} to print statistics about time and
1291 memory usage after it completes each command and returns to the prompt.
1294 @cindex @code{--version}
1295 This option causes @value{GDBN} to print its version number and
1296 no-warranty blurb, and exit.
1298 @item -configuration
1299 @cindex @code{--configuration}
1300 This option causes @value{GDBN} to print details about its build-time
1301 configuration parameters, and then exit. These details can be
1302 important when reporting @value{GDBN} bugs (@pxref{GDB Bugs}).
1307 @subsection What @value{GDBN} Does During Startup
1308 @cindex @value{GDBN} startup
1310 Here's the description of what @value{GDBN} does during session startup:
1314 Sets up the command interpreter as specified by the command line
1315 (@pxref{Mode Options, interpreter}).
1319 Reads the system-wide @dfn{init file} (if @option{--with-system-gdbinit} was
1320 used when building @value{GDBN}; @pxref{System-wide configuration,
1321 ,System-wide configuration and settings}) and executes all the commands in
1324 @anchor{Home Directory Init File}
1326 Reads the init file (if any) in your home directory@footnote{On
1327 DOS/Windows systems, the home directory is the one pointed to by the
1328 @code{HOME} environment variable.} and executes all the commands in
1331 @anchor{Option -init-eval-command}
1333 Executes commands and command files specified by the @samp{-iex} and
1334 @samp{-ix} options in their specified order. Usually you should use the
1335 @samp{-ex} and @samp{-x} options instead, but this way you can apply
1336 settings before @value{GDBN} init files get executed and before inferior
1340 Processes command line options and operands.
1342 @anchor{Init File in the Current Directory during Startup}
1344 Reads and executes the commands from init file (if any) in the current
1345 working directory as long as @samp{set auto-load local-gdbinit} is set to
1346 @samp{on} (@pxref{Init File in the Current Directory}).
1347 This is only done if the current directory is
1348 different from your home directory. Thus, you can have more than one
1349 init file, one generic in your home directory, and another, specific
1350 to the program you are debugging, in the directory where you invoke
1354 If the command line specified a program to debug, or a process to
1355 attach to, or a core file, @value{GDBN} loads any auto-loaded
1356 scripts provided for the program or for its loaded shared libraries.
1357 @xref{Auto-loading}.
1359 If you wish to disable the auto-loading during startup,
1360 you must do something like the following:
1363 $ gdb -iex "set auto-load python-scripts off" myprogram
1366 Option @samp{-ex} does not work because the auto-loading is then turned
1370 Executes commands and command files specified by the @samp{-ex} and
1371 @samp{-x} options in their specified order. @xref{Command Files}, for
1372 more details about @value{GDBN} command files.
1375 Reads the command history recorded in the @dfn{history file}.
1376 @xref{Command History}, for more details about the command history and the
1377 files where @value{GDBN} records it.
1380 Init files use the same syntax as @dfn{command files} (@pxref{Command
1381 Files}) and are processed by @value{GDBN} in the same way. The init
1382 file in your home directory can set options (such as @samp{set
1383 complaints}) that affect subsequent processing of command line options
1384 and operands. Init files are not executed if you use the @samp{-nx}
1385 option (@pxref{Mode Options, ,Choosing Modes}).
1387 To display the list of init files loaded by gdb at startup, you
1388 can use @kbd{gdb --help}.
1390 @cindex init file name
1391 @cindex @file{.gdbinit}
1392 @cindex @file{gdb.ini}
1393 The @value{GDBN} init files are normally called @file{.gdbinit}.
1394 The DJGPP port of @value{GDBN} uses the name @file{gdb.ini}, due to
1395 the limitations of file names imposed by DOS filesystems. The Windows
1396 port of @value{GDBN} uses the standard name, but if it finds a
1397 @file{gdb.ini} file in your home directory, it warns you about that
1398 and suggests to rename the file to the standard name.
1402 @section Quitting @value{GDBN}
1403 @cindex exiting @value{GDBN}
1404 @cindex leaving @value{GDBN}
1407 @kindex quit @r{[}@var{expression}@r{]}
1408 @kindex q @r{(@code{quit})}
1409 @item quit @r{[}@var{expression}@r{]}
1411 To exit @value{GDBN}, use the @code{quit} command (abbreviated
1412 @code{q}), or type an end-of-file character (usually @kbd{Ctrl-d}). If you
1413 do not supply @var{expression}, @value{GDBN} will terminate normally;
1414 otherwise it will terminate using the result of @var{expression} as the
1419 An interrupt (often @kbd{Ctrl-c}) does not exit from @value{GDBN}, but rather
1420 terminates the action of any @value{GDBN} command that is in progress and
1421 returns to @value{GDBN} command level. It is safe to type the interrupt
1422 character at any time because @value{GDBN} does not allow it to take effect
1423 until a time when it is safe.
1425 If you have been using @value{GDBN} to control an attached process or
1426 device, you can release it with the @code{detach} command
1427 (@pxref{Attach, ,Debugging an Already-running Process}).
1429 @node Shell Commands
1430 @section Shell Commands
1432 If you need to execute occasional shell commands during your
1433 debugging session, there is no need to leave or suspend @value{GDBN}; you can
1434 just use the @code{shell} command.
1439 @cindex shell escape
1440 @item shell @var{command-string}
1441 @itemx !@var{command-string}
1442 Invoke a standard shell to execute @var{command-string}.
1443 Note that no space is needed between @code{!} and @var{command-string}.
1444 If it exists, the environment variable @code{SHELL} determines which
1445 shell to run. Otherwise @value{GDBN} uses the default shell
1446 (@file{/bin/sh} on Unix systems, @file{COMMAND.COM} on MS-DOS, etc.).
1449 The utility @code{make} is often needed in development environments.
1450 You do not have to use the @code{shell} command for this purpose in
1455 @cindex calling make
1456 @item make @var{make-args}
1457 Execute the @code{make} program with the specified
1458 arguments. This is equivalent to @samp{shell make @var{make-args}}.
1464 @cindex send the output of a gdb command to a shell command
1466 @item pipe [@var{command}] | @var{shell_command}
1467 @itemx | [@var{command}] | @var{shell_command}
1468 @itemx pipe -d @var{delim} @var{command} @var{delim} @var{shell_command}
1469 @itemx | -d @var{delim} @var{command} @var{delim} @var{shell_command}
1470 Executes @var{command} and sends its output to @var{shell_command}.
1471 Note that no space is needed around @code{|}.
1472 If no @var{command} is provided, the last command executed is repeated.
1474 In case the @var{command} contains a @code{|}, the option @code{-d @var{delim}}
1475 can be used to specify an alternate delimiter string @var{delim} that separates
1476 the @var{command} from the @var{shell_command}.
1509 (gdb) | -d ! echo this contains a | char\n ! sed -e 's/|/PIPE/'
1510 this contains a PIPE char
1511 (gdb) | -d xxx echo this contains a | char!\n xxx sed -e 's/|/PIPE/'
1512 this contains a PIPE char!
1518 The convenience variables @code{$_shell_exitcode} and @code{$_shell_exitsignal}
1519 can be used to examine the exit status of the last shell command launched
1520 by @code{shell}, @code{make}, @code{pipe} and @code{|}.
1521 @xref{Convenience Vars,, Convenience Variables}.
1523 @node Logging Output
1524 @section Logging Output
1525 @cindex logging @value{GDBN} output
1526 @cindex save @value{GDBN} output to a file
1528 You may want to save the output of @value{GDBN} commands to a file.
1529 There are several commands to control @value{GDBN}'s logging.
1533 @item set logging on
1535 @item set logging off
1537 @cindex logging file name
1538 @item set logging file @var{file}
1539 Change the name of the current logfile. The default logfile is @file{gdb.txt}.
1540 @item set logging overwrite [on|off]
1541 By default, @value{GDBN} will append to the logfile. Set @code{overwrite} if
1542 you want @code{set logging on} to overwrite the logfile instead.
1543 @item set logging redirect [on|off]
1544 By default, @value{GDBN} output will go to both the terminal and the logfile.
1545 Set @code{redirect} if you want output to go only to the log file.
1546 @item set logging debugredirect [on|off]
1547 By default, @value{GDBN} debug output will go to both the terminal and the logfile.
1548 Set @code{debugredirect} if you want debug output to go only to the log file.
1549 @kindex show logging
1551 Show the current values of the logging settings.
1554 You can also redirect the output of a @value{GDBN} command to a
1555 shell command. @xref{pipe}.
1557 @chapter @value{GDBN} Commands
1559 You can abbreviate a @value{GDBN} command to the first few letters of the command
1560 name, if that abbreviation is unambiguous; and you can repeat certain
1561 @value{GDBN} commands by typing just @key{RET}. You can also use the @key{TAB}
1562 key to get @value{GDBN} to fill out the rest of a word in a command (or to
1563 show you the alternatives available, if there is more than one possibility).
1566 * Command Syntax:: How to give commands to @value{GDBN}
1567 * Command Settings:: How to change default behavior of commands
1568 * Completion:: Command completion
1569 * Command Options:: Command options
1570 * Help:: How to ask @value{GDBN} for help
1573 @node Command Syntax
1574 @section Command Syntax
1576 A @value{GDBN} command is a single line of input. There is no limit on
1577 how long it can be. It starts with a command name, which is followed by
1578 arguments whose meaning depends on the command name. For example, the
1579 command @code{step} accepts an argument which is the number of times to
1580 step, as in @samp{step 5}. You can also use the @code{step} command
1581 with no arguments. Some commands do not allow any arguments.
1583 @cindex abbreviation
1584 @value{GDBN} command names may always be truncated if that abbreviation is
1585 unambiguous. Other possible command abbreviations are listed in the
1586 documentation for individual commands. In some cases, even ambiguous
1587 abbreviations are allowed; for example, @code{s} is specially defined as
1588 equivalent to @code{step} even though there are other commands whose
1589 names start with @code{s}. You can test abbreviations by using them as
1590 arguments to the @code{help} command.
1592 @cindex repeating commands
1593 @kindex RET @r{(repeat last command)}
1594 A blank line as input to @value{GDBN} (typing just @key{RET}) means to
1595 repeat the previous command. Certain commands (for example, @code{run})
1596 will not repeat this way; these are commands whose unintentional
1597 repetition might cause trouble and which you are unlikely to want to
1598 repeat. User-defined commands can disable this feature; see
1599 @ref{Define, dont-repeat}.
1601 The @code{list} and @code{x} commands, when you repeat them with
1602 @key{RET}, construct new arguments rather than repeating
1603 exactly as typed. This permits easy scanning of source or memory.
1605 @value{GDBN} can also use @key{RET} in another way: to partition lengthy
1606 output, in a way similar to the common utility @code{more}
1607 (@pxref{Screen Size,,Screen Size}). Since it is easy to press one
1608 @key{RET} too many in this situation, @value{GDBN} disables command
1609 repetition after any command that generates this sort of display.
1611 @kindex # @r{(a comment)}
1613 Any text from a @kbd{#} to the end of the line is a comment; it does
1614 nothing. This is useful mainly in command files (@pxref{Command
1615 Files,,Command Files}).
1617 @cindex repeating command sequences
1618 @kindex Ctrl-o @r{(operate-and-get-next)}
1619 The @kbd{Ctrl-o} binding is useful for repeating a complex sequence of
1620 commands. This command accepts the current line, like @key{RET}, and
1621 then fetches the next line relative to the current line from the history
1625 @node Command Settings
1626 @section Command Settings
1627 @cindex default behavior of commands, changing
1628 @cindex default settings, changing
1630 Many commands change their behavior according to command-specific
1631 variables or settings. These settings can be changed with the
1632 @code{set} subcommands. For example, the @code{print} command
1633 (@pxref{Data, ,Examining Data}) prints arrays differently depending on
1634 settings changeable with the commands @code{set print elements
1635 NUMBER-OF-ELEMENTS} and @code{set print array-indexes}, among others.
1637 You can change these settings to your preference in the gdbinit files
1638 loaded at @value{GDBN} startup. @xref{Startup}.
1640 The settings can also be changed interactively during the debugging
1641 session. For example, to change the limit of array elements to print,
1642 you can do the following:
1644 (@value{GDBN}) set print elements 10
1645 (@value{GDBN}) print some_array
1646 $1 = @{0, 10, 20, 30, 40, 50, 60, 70, 80, 90...@}
1649 The above @code{set print elements 10} command changes the number of
1650 elements to print from the default of 200 to 10. If you only intend
1651 this limit of 10 to be used for printing @code{some_array}, then you
1652 must restore the limit back to 200, with @code{set print elements
1655 Some commands allow overriding settings with command options. For
1656 example, the @code{print} command supports a number of options that
1657 allow overriding relevant global print settings as set by @code{set
1658 print} subcommands. @xref{print options}. The example above could be
1661 (@value{GDBN}) print -elements 10 -- some_array
1662 $1 = @{0, 10, 20, 30, 40, 50, 60, 70, 80, 90...@}
1665 Alternatively, you can use the @code{with} command to change a setting
1666 temporarily, for the duration of a command invocation.
1669 @kindex with command
1670 @kindex w @r{(@code{with})}
1672 @cindex temporarily change settings
1673 @item with @var{setting} [@var{value}] [-- @var{command}]
1674 @itemx w @var{setting} [@var{value}] [-- @var{command}]
1675 Temporarily set @var{setting} to @var{value} for the duration of
1678 @var{setting} is any setting you can change with the @code{set}
1679 subcommands. @var{value} is the value to assign to @code{setting}
1680 while running @code{command}.
1682 If no @var{command} is provided, the last command executed is
1685 If a @var{command} is provided, it must be preceded by a double dash
1686 (@code{--}) separator. This is required because some settings accept
1687 free-form arguments, such as expressions or filenames.
1689 For example, the command
1691 (@value{GDBN}) with print array on -- print some_array
1694 is equivalent to the following 3 commands:
1696 (@value{GDBN}) set print array on
1697 (@value{GDBN}) print some_array
1698 (@value{GDBN}) set print array off
1701 The @code{with} command is particularly useful when you want to
1702 override a setting while running user-defined commands, or commands
1703 defined in Python or Guile. @xref{Extending GDB,, Extending GDB}.
1706 (@value{GDBN}) with print pretty on -- my_complex_command
1709 To change several settings for the same command, you can nest
1710 @code{with} commands. For example, @code{with language ada -- with
1711 print elements 10} temporarily changes the language to Ada and sets a
1712 limit of 10 elements to print for arrays and strings.
1717 @section Command Completion
1720 @cindex word completion
1721 @value{GDBN} can fill in the rest of a word in a command for you, if there is
1722 only one possibility; it can also show you what the valid possibilities
1723 are for the next word in a command, at any time. This works for @value{GDBN}
1724 commands, @value{GDBN} subcommands, command options, and the names of symbols
1727 Press the @key{TAB} key whenever you want @value{GDBN} to fill out the rest
1728 of a word. If there is only one possibility, @value{GDBN} fills in the
1729 word, and waits for you to finish the command (or press @key{RET} to
1730 enter it). For example, if you type
1732 @c FIXME "@key" does not distinguish its argument sufficiently to permit
1733 @c complete accuracy in these examples; space introduced for clarity.
1734 @c If texinfo enhancements make it unnecessary, it would be nice to
1735 @c replace " @key" by "@key" in the following...
1737 (@value{GDBP}) info bre @key{TAB}
1741 @value{GDBN} fills in the rest of the word @samp{breakpoints}, since that is
1742 the only @code{info} subcommand beginning with @samp{bre}:
1745 (@value{GDBP}) info breakpoints
1749 You can either press @key{RET} at this point, to run the @code{info
1750 breakpoints} command, or backspace and enter something else, if
1751 @samp{breakpoints} does not look like the command you expected. (If you
1752 were sure you wanted @code{info breakpoints} in the first place, you
1753 might as well just type @key{RET} immediately after @samp{info bre},
1754 to exploit command abbreviations rather than command completion).
1756 If there is more than one possibility for the next word when you press
1757 @key{TAB}, @value{GDBN} sounds a bell. You can either supply more
1758 characters and try again, or just press @key{TAB} a second time;
1759 @value{GDBN} displays all the possible completions for that word. For
1760 example, you might want to set a breakpoint on a subroutine whose name
1761 begins with @samp{make_}, but when you type @kbd{b make_@key{TAB}} @value{GDBN}
1762 just sounds the bell. Typing @key{TAB} again displays all the
1763 function names in your program that begin with those characters, for
1767 (@value{GDBP}) b make_ @key{TAB}
1768 @exdent @value{GDBN} sounds bell; press @key{TAB} again, to see:
1769 make_a_section_from_file make_environ
1770 make_abs_section make_function_type
1771 make_blockvector make_pointer_type
1772 make_cleanup make_reference_type
1773 make_command make_symbol_completion_list
1774 (@value{GDBP}) b make_
1778 After displaying the available possibilities, @value{GDBN} copies your
1779 partial input (@samp{b make_} in the example) so you can finish the
1782 If you just want to see the list of alternatives in the first place, you
1783 can press @kbd{M-?} rather than pressing @key{TAB} twice. @kbd{M-?}
1784 means @kbd{@key{META} ?}. You can type this either by holding down a
1785 key designated as the @key{META} shift on your keyboard (if there is
1786 one) while typing @kbd{?}, or as @key{ESC} followed by @kbd{?}.
1788 If the number of possible completions is large, @value{GDBN} will
1789 print as much of the list as it has collected, as well as a message
1790 indicating that the list may be truncated.
1793 (@value{GDBP}) b m@key{TAB}@key{TAB}
1795 <... the rest of the possible completions ...>
1796 *** List may be truncated, max-completions reached. ***
1801 This behavior can be controlled with the following commands:
1804 @kindex set max-completions
1805 @item set max-completions @var{limit}
1806 @itemx set max-completions unlimited
1807 Set the maximum number of completion candidates. @value{GDBN} will
1808 stop looking for more completions once it collects this many candidates.
1809 This is useful when completing on things like function names as collecting
1810 all the possible candidates can be time consuming.
1811 The default value is 200. A value of zero disables tab-completion.
1812 Note that setting either no limit or a very large limit can make
1814 @kindex show max-completions
1815 @item show max-completions
1816 Show the maximum number of candidates that @value{GDBN} will collect and show
1820 @cindex quotes in commands
1821 @cindex completion of quoted strings
1822 Sometimes the string you need, while logically a ``word'', may contain
1823 parentheses or other characters that @value{GDBN} normally excludes from
1824 its notion of a word. To permit word completion to work in this
1825 situation, you may enclose words in @code{'} (single quote marks) in
1826 @value{GDBN} commands.
1828 A likely situation where you might need this is in typing an
1829 expression that involves a C@t{++} symbol name with template
1830 parameters. This is because when completing expressions, GDB treats
1831 the @samp{<} character as word delimiter, assuming that it's the
1832 less-than comparison operator (@pxref{C Operators, , C and C@t{++}
1835 For example, when you want to call a C@t{++} template function
1836 interactively using the @code{print} or @code{call} commands, you may
1837 need to distinguish whether you mean the version of @code{name} that
1838 was specialized for @code{int}, @code{name<int>()}, or the version
1839 that was specialized for @code{float}, @code{name<float>()}. To use
1840 the word-completion facilities in this situation, type a single quote
1841 @code{'} at the beginning of the function name. This alerts
1842 @value{GDBN} that it may need to consider more information than usual
1843 when you press @key{TAB} or @kbd{M-?} to request word completion:
1846 (@value{GDBP}) p 'func< @kbd{M-?}
1847 func<int>() func<float>()
1848 (@value{GDBP}) p 'func<
1851 When setting breakpoints however (@pxref{Specify Location}), you don't
1852 usually need to type a quote before the function name, because
1853 @value{GDBN} understands that you want to set a breakpoint on a
1857 (@value{GDBP}) b func< @kbd{M-?}
1858 func<int>() func<float>()
1859 (@value{GDBP}) b func<
1862 This is true even in the case of typing the name of C@t{++} overloaded
1863 functions (multiple definitions of the same function, distinguished by
1864 argument type). For example, when you want to set a breakpoint you
1865 don't need to distinguish whether you mean the version of @code{name}
1866 that takes an @code{int} parameter, @code{name(int)}, or the version
1867 that takes a @code{float} parameter, @code{name(float)}.
1870 (@value{GDBP}) b bubble( @kbd{M-?}
1871 bubble(int) bubble(double)
1872 (@value{GDBP}) b bubble(dou @kbd{M-?}
1876 See @ref{quoting names} for a description of other scenarios that
1879 For more information about overloaded functions, see @ref{C Plus Plus
1880 Expressions, ,C@t{++} Expressions}. You can use the command @code{set
1881 overload-resolution off} to disable overload resolution;
1882 see @ref{Debugging C Plus Plus, ,@value{GDBN} Features for C@t{++}}.
1884 @cindex completion of structure field names
1885 @cindex structure field name completion
1886 @cindex completion of union field names
1887 @cindex union field name completion
1888 When completing in an expression which looks up a field in a
1889 structure, @value{GDBN} also tries@footnote{The completer can be
1890 confused by certain kinds of invalid expressions. Also, it only
1891 examines the static type of the expression, not the dynamic type.} to
1892 limit completions to the field names available in the type of the
1896 (@value{GDBP}) p gdb_stdout.@kbd{M-?}
1897 magic to_fputs to_rewind
1898 to_data to_isatty to_write
1899 to_delete to_put to_write_async_safe
1904 This is because the @code{gdb_stdout} is a variable of the type
1905 @code{struct ui_file} that is defined in @value{GDBN} sources as
1912 ui_file_flush_ftype *to_flush;
1913 ui_file_write_ftype *to_write;
1914 ui_file_write_async_safe_ftype *to_write_async_safe;
1915 ui_file_fputs_ftype *to_fputs;
1916 ui_file_read_ftype *to_read;
1917 ui_file_delete_ftype *to_delete;
1918 ui_file_isatty_ftype *to_isatty;
1919 ui_file_rewind_ftype *to_rewind;
1920 ui_file_put_ftype *to_put;
1925 @node Command Options
1926 @section Command options
1928 @cindex command options
1929 Some commands accept options starting with a leading dash. For
1930 example, @code{print -pretty}. Similarly to command names, you can
1931 abbreviate a @value{GDBN} option to the first few letters of the
1932 option name, if that abbreviation is unambiguous, and you can also use
1933 the @key{TAB} key to get @value{GDBN} to fill out the rest of a word
1934 in an option (or to show you the alternatives available, if there is
1935 more than one possibility).
1937 @cindex command options, raw input
1938 Some commands take raw input as argument. For example, the print
1939 command processes arbitrary expressions in any of the languages
1940 supported by @value{GDBN}. With such commands, because raw input may
1941 start with a leading dash that would be confused with an option or any
1942 of its abbreviations, e.g.@: @code{print -r} (short for @code{print
1943 -raw} or printing negative @code{r}?), if you specify any command
1944 option, then you must use a double-dash (@code{--}) delimiter to
1945 indicate the end of options.
1947 @cindex command options, boolean
1949 Some options are described as accepting an argument which can be
1950 either @code{on} or @code{off}. These are known as @dfn{boolean
1951 options}. Similarly to boolean settings commands---@code{on} and
1952 @code{off} are the typical values, but any of @code{1}, @code{yes} and
1953 @code{enable} can also be used as ``true'' value, and any of @code{0},
1954 @code{no} and @code{disable} can also be used as ``false'' value. You
1955 can also omit a ``true'' value, as it is implied by default.
1957 For example, these are equivalent:
1960 (@value{GDBP}) print -object on -pretty off -element unlimited -- *myptr
1961 (@value{GDBP}) p -o -p 0 -e u -- *myptr
1964 You can discover the set of options some command accepts by completing
1965 on @code{-} after the command name. For example:
1968 (@value{GDBP}) print -@key{TAB}@key{TAB}
1969 -address -max-depth -repeats -vtbl
1970 -array -null-stop -static-members
1971 -array-indexes -object -symbol
1972 -elements -pretty -union
1975 Completion will in some cases guide you with a suggestion of what kind
1976 of argument an option expects. For example:
1979 (@value{GDBP}) print -elements @key{TAB}@key{TAB}
1983 Here, the option expects a number (e.g., @code{100}), not literal
1984 @code{NUMBER}. Such metasyntactical arguments are always presented in
1987 (For more on using the @code{print} command, see @ref{Data, ,Examining
1991 @section Getting Help
1992 @cindex online documentation
1995 You can always ask @value{GDBN} itself for information on its commands,
1996 using the command @code{help}.
1999 @kindex h @r{(@code{help})}
2002 You can use @code{help} (abbreviated @code{h}) with no arguments to
2003 display a short list of named classes of commands:
2007 List of classes of commands:
2009 aliases -- Aliases of other commands
2010 breakpoints -- Making program stop at certain points
2011 data -- Examining data
2012 files -- Specifying and examining files
2013 internals -- Maintenance commands
2014 obscure -- Obscure features
2015 running -- Running the program
2016 stack -- Examining the stack
2017 status -- Status inquiries
2018 support -- Support facilities
2019 tracepoints -- Tracing of program execution without
2020 stopping the program
2021 user-defined -- User-defined commands
2023 Type "help" followed by a class name for a list of
2024 commands in that class.
2025 Type "help" followed by command name for full
2027 Command name abbreviations are allowed if unambiguous.
2030 @c the above line break eliminates huge line overfull...
2032 @item help @var{class}
2033 Using one of the general help classes as an argument, you can get a
2034 list of the individual commands in that class. For example, here is the
2035 help display for the class @code{status}:
2038 (@value{GDBP}) help status
2043 @c Line break in "show" line falsifies real output, but needed
2044 @c to fit in smallbook page size.
2045 info -- Generic command for showing things
2046 about the program being debugged
2047 show -- Generic command for showing things
2050 Type "help" followed by command name for full
2052 Command name abbreviations are allowed if unambiguous.
2056 @item help @var{command}
2057 With a command name as @code{help} argument, @value{GDBN} displays a
2058 short paragraph on how to use that command.
2061 @item apropos [-v] @var{regexp}
2062 The @code{apropos} command searches through all of the @value{GDBN}
2063 commands, and their documentation, for the regular expression specified in
2064 @var{args}. It prints out all matches found. The optional flag @samp{-v},
2065 which stands for @samp{verbose}, indicates to output the full documentation
2066 of the matching commands and highlight the parts of the documentation
2067 matching @var{regexp}. For example:
2078 alias -- Define a new command that is an alias of an existing command
2079 aliases -- Aliases of other commands
2080 d -- Delete some breakpoints or auto-display expressions
2081 del -- Delete some breakpoints or auto-display expressions
2082 delete -- Delete some breakpoints or auto-display expressions
2090 apropos -v cut.*thread apply
2094 results in the below output, where @samp{cut for 'thread apply}
2095 is highlighted if styling is enabled.
2099 taas -- Apply a command to all threads (ignoring errors
2102 shortcut for 'thread apply all -s COMMAND'
2104 tfaas -- Apply a command to all frames of all threads
2105 (ignoring errors and empty output).
2106 Usage: tfaas COMMAND
2107 shortcut for 'thread apply all -s frame apply all -s COMMAND'
2112 @item complete @var{args}
2113 The @code{complete @var{args}} command lists all the possible completions
2114 for the beginning of a command. Use @var{args} to specify the beginning of the
2115 command you want completed. For example:
2121 @noindent results in:
2132 @noindent This is intended for use by @sc{gnu} Emacs.
2135 In addition to @code{help}, you can use the @value{GDBN} commands @code{info}
2136 and @code{show} to inquire about the state of your program, or the state
2137 of @value{GDBN} itself. Each command supports many topics of inquiry; this
2138 manual introduces each of them in the appropriate context. The listings
2139 under @code{info} and under @code{show} in the Command, Variable, and
2140 Function Index point to all the sub-commands. @xref{Command and Variable
2146 @kindex i @r{(@code{info})}
2148 This command (abbreviated @code{i}) is for describing the state of your
2149 program. For example, you can show the arguments passed to a function
2150 with @code{info args}, list the registers currently in use with @code{info
2151 registers}, or list the breakpoints you have set with @code{info breakpoints}.
2152 You can get a complete list of the @code{info} sub-commands with
2153 @w{@code{help info}}.
2157 You can assign the result of an expression to an environment variable with
2158 @code{set}. For example, you can set the @value{GDBN} prompt to a $-sign with
2159 @code{set prompt $}.
2163 In contrast to @code{info}, @code{show} is for describing the state of
2164 @value{GDBN} itself.
2165 You can change most of the things you can @code{show}, by using the
2166 related command @code{set}; for example, you can control what number
2167 system is used for displays with @code{set radix}, or simply inquire
2168 which is currently in use with @code{show radix}.
2171 To display all the settable parameters and their current
2172 values, you can use @code{show} with no arguments; you may also use
2173 @code{info set}. Both commands produce the same display.
2174 @c FIXME: "info set" violates the rule that "info" is for state of
2175 @c FIXME...program. Ck w/ GNU: "info set" to be called something else,
2176 @c FIXME...or change desc of rule---eg "state of prog and debugging session"?
2180 Here are several miscellaneous @code{show} subcommands, all of which are
2181 exceptional in lacking corresponding @code{set} commands:
2184 @kindex show version
2185 @cindex @value{GDBN} version number
2187 Show what version of @value{GDBN} is running. You should include this
2188 information in @value{GDBN} bug-reports. If multiple versions of
2189 @value{GDBN} are in use at your site, you may need to determine which
2190 version of @value{GDBN} you are running; as @value{GDBN} evolves, new
2191 commands are introduced, and old ones may wither away. Also, many
2192 system vendors ship variant versions of @value{GDBN}, and there are
2193 variant versions of @value{GDBN} in @sc{gnu}/Linux distributions as well.
2194 The version number is the same as the one announced when you start
2197 @kindex show copying
2198 @kindex info copying
2199 @cindex display @value{GDBN} copyright
2202 Display information about permission for copying @value{GDBN}.
2204 @kindex show warranty
2205 @kindex info warranty
2207 @itemx info warranty
2208 Display the @sc{gnu} ``NO WARRANTY'' statement, or a warranty,
2209 if your version of @value{GDBN} comes with one.
2211 @kindex show configuration
2212 @item show configuration
2213 Display detailed information about the way @value{GDBN} was configured
2214 when it was built. This displays the optional arguments passed to the
2215 @file{configure} script and also configuration parameters detected
2216 automatically by @command{configure}. When reporting a @value{GDBN}
2217 bug (@pxref{GDB Bugs}), it is important to include this information in
2223 @chapter Running Programs Under @value{GDBN}
2225 When you run a program under @value{GDBN}, you must first generate
2226 debugging information when you compile it.
2228 You may start @value{GDBN} with its arguments, if any, in an environment
2229 of your choice. If you are doing native debugging, you may redirect
2230 your program's input and output, debug an already running process, or
2231 kill a child process.
2234 * Compilation:: Compiling for debugging
2235 * Starting:: Starting your program
2236 * Arguments:: Your program's arguments
2237 * Environment:: Your program's environment
2239 * Working Directory:: Your program's working directory
2240 * Input/Output:: Your program's input and output
2241 * Attach:: Debugging an already-running process
2242 * Kill Process:: Killing the child process
2244 * Inferiors and Programs:: Debugging multiple inferiors and programs
2245 * Threads:: Debugging programs with multiple threads
2246 * Forks:: Debugging forks
2247 * Checkpoint/Restart:: Setting a @emph{bookmark} to return to later
2251 @section Compiling for Debugging
2253 In order to debug a program effectively, you need to generate
2254 debugging information when you compile it. This debugging information
2255 is stored in the object file; it describes the data type of each
2256 variable or function and the correspondence between source line numbers
2257 and addresses in the executable code.
2259 To request debugging information, specify the @samp{-g} option when you run
2262 Programs that are to be shipped to your customers are compiled with
2263 optimizations, using the @samp{-O} compiler option. However, some
2264 compilers are unable to handle the @samp{-g} and @samp{-O} options
2265 together. Using those compilers, you cannot generate optimized
2266 executables containing debugging information.
2268 @value{NGCC}, the @sc{gnu} C/C@t{++} compiler, supports @samp{-g} with or
2269 without @samp{-O}, making it possible to debug optimized code. We
2270 recommend that you @emph{always} use @samp{-g} whenever you compile a
2271 program. You may think your program is correct, but there is no sense
2272 in pushing your luck. For more information, see @ref{Optimized Code}.
2274 Older versions of the @sc{gnu} C compiler permitted a variant option
2275 @w{@samp{-gg}} for debugging information. @value{GDBN} no longer supports this
2276 format; if your @sc{gnu} C compiler has this option, do not use it.
2278 @value{GDBN} knows about preprocessor macros and can show you their
2279 expansion (@pxref{Macros}). Most compilers do not include information
2280 about preprocessor macros in the debugging information if you specify
2281 the @option{-g} flag alone. Version 3.1 and later of @value{NGCC},
2282 the @sc{gnu} C compiler, provides macro information if you are using
2283 the DWARF debugging format, and specify the option @option{-g3}.
2285 @xref{Debugging Options,,Options for Debugging Your Program or GCC,
2286 gcc, Using the @sc{gnu} Compiler Collection (GCC)}, for more
2287 information on @value{NGCC} options affecting debug information.
2289 You will have the best debugging experience if you use the latest
2290 version of the DWARF debugging format that your compiler supports.
2291 DWARF is currently the most expressive and best supported debugging
2292 format in @value{GDBN}.
2296 @section Starting your Program
2302 @kindex r @r{(@code{run})}
2305 Use the @code{run} command to start your program under @value{GDBN}.
2306 You must first specify the program name with an argument to
2307 @value{GDBN} (@pxref{Invocation, ,Getting In and Out of
2308 @value{GDBN}}), or by using the @code{file} or @code{exec-file}
2309 command (@pxref{Files, ,Commands to Specify Files}).
2313 If you are running your program in an execution environment that
2314 supports processes, @code{run} creates an inferior process and makes
2315 that process run your program. In some environments without processes,
2316 @code{run} jumps to the start of your program. Other targets,
2317 like @samp{remote}, are always running. If you get an error
2318 message like this one:
2321 The "remote" target does not support "run".
2322 Try "help target" or "continue".
2326 then use @code{continue} to run your program. You may need @code{load}
2327 first (@pxref{load}).
2329 The execution of a program is affected by certain information it
2330 receives from its superior. @value{GDBN} provides ways to specify this
2331 information, which you must do @emph{before} starting your program. (You
2332 can change it after starting your program, but such changes only affect
2333 your program the next time you start it.) This information may be
2334 divided into four categories:
2337 @item The @emph{arguments.}
2338 Specify the arguments to give your program as the arguments of the
2339 @code{run} command. If a shell is available on your target, the shell
2340 is used to pass the arguments, so that you may use normal conventions
2341 (such as wildcard expansion or variable substitution) in describing
2343 In Unix systems, you can control which shell is used with the
2344 @code{SHELL} environment variable. If you do not define @code{SHELL},
2345 @value{GDBN} uses the default shell (@file{/bin/sh}). You can disable
2346 use of any shell with the @code{set startup-with-shell} command (see
2349 @item The @emph{environment.}
2350 Your program normally inherits its environment from @value{GDBN}, but you can
2351 use the @value{GDBN} commands @code{set environment} and @code{unset
2352 environment} to change parts of the environment that affect
2353 your program. @xref{Environment, ,Your Program's Environment}.
2355 @item The @emph{working directory.}
2356 You can set your program's working directory with the command
2357 @kbd{set cwd}. If you do not set any working directory with this
2358 command, your program will inherit @value{GDBN}'s working directory if
2359 native debugging, or the remote server's working directory if remote
2360 debugging. @xref{Working Directory, ,Your Program's Working
2363 @item The @emph{standard input and output.}
2364 Your program normally uses the same device for standard input and
2365 standard output as @value{GDBN} is using. You can redirect input and output
2366 in the @code{run} command line, or you can use the @code{tty} command to
2367 set a different device for your program.
2368 @xref{Input/Output, ,Your Program's Input and Output}.
2371 @emph{Warning:} While input and output redirection work, you cannot use
2372 pipes to pass the output of the program you are debugging to another
2373 program; if you attempt this, @value{GDBN} is likely to wind up debugging the
2377 When you issue the @code{run} command, your program begins to execute
2378 immediately. @xref{Stopping, ,Stopping and Continuing}, for discussion
2379 of how to arrange for your program to stop. Once your program has
2380 stopped, you may call functions in your program, using the @code{print}
2381 or @code{call} commands. @xref{Data, ,Examining Data}.
2383 If the modification time of your symbol file has changed since the last
2384 time @value{GDBN} read its symbols, @value{GDBN} discards its symbol
2385 table, and reads it again. When it does this, @value{GDBN} tries to retain
2386 your current breakpoints.
2391 @cindex run to main procedure
2392 The name of the main procedure can vary from language to language.
2393 With C or C@t{++}, the main procedure name is always @code{main}, but
2394 other languages such as Ada do not require a specific name for their
2395 main procedure. The debugger provides a convenient way to start the
2396 execution of the program and to stop at the beginning of the main
2397 procedure, depending on the language used.
2399 The @samp{start} command does the equivalent of setting a temporary
2400 breakpoint at the beginning of the main procedure and then invoking
2401 the @samp{run} command.
2403 @cindex elaboration phase
2404 Some programs contain an @dfn{elaboration} phase where some startup code is
2405 executed before the main procedure is called. This depends on the
2406 languages used to write your program. In C@t{++}, for instance,
2407 constructors for static and global objects are executed before
2408 @code{main} is called. It is therefore possible that the debugger stops
2409 before reaching the main procedure. However, the temporary breakpoint
2410 will remain to halt execution.
2412 Specify the arguments to give to your program as arguments to the
2413 @samp{start} command. These arguments will be given verbatim to the
2414 underlying @samp{run} command. Note that the same arguments will be
2415 reused if no argument is provided during subsequent calls to
2416 @samp{start} or @samp{run}.
2418 It is sometimes necessary to debug the program during elaboration. In
2419 these cases, using the @code{start} command would stop the execution
2420 of your program too late, as the program would have already completed
2421 the elaboration phase. Under these circumstances, either insert
2422 breakpoints in your elaboration code before running your program or
2423 use the @code{starti} command.
2427 @cindex run to first instruction
2428 The @samp{starti} command does the equivalent of setting a temporary
2429 breakpoint at the first instruction of a program's execution and then
2430 invoking the @samp{run} command. For programs containing an
2431 elaboration phase, the @code{starti} command will stop execution at
2432 the start of the elaboration phase.
2434 @anchor{set exec-wrapper}
2435 @kindex set exec-wrapper
2436 @item set exec-wrapper @var{wrapper}
2437 @itemx show exec-wrapper
2438 @itemx unset exec-wrapper
2439 When @samp{exec-wrapper} is set, the specified wrapper is used to
2440 launch programs for debugging. @value{GDBN} starts your program
2441 with a shell command of the form @kbd{exec @var{wrapper}
2442 @var{program}}. Quoting is added to @var{program} and its
2443 arguments, but not to @var{wrapper}, so you should add quotes if
2444 appropriate for your shell. The wrapper runs until it executes
2445 your program, and then @value{GDBN} takes control.
2447 You can use any program that eventually calls @code{execve} with
2448 its arguments as a wrapper. Several standard Unix utilities do
2449 this, e.g.@: @code{env} and @code{nohup}. Any Unix shell script ending
2450 with @code{exec "$@@"} will also work.
2452 For example, you can use @code{env} to pass an environment variable to
2453 the debugged program, without setting the variable in your shell's
2457 (@value{GDBP}) set exec-wrapper env 'LD_PRELOAD=libtest.so'
2461 This command is available when debugging locally on most targets, excluding
2462 @sc{djgpp}, Cygwin, MS Windows, and QNX Neutrino.
2464 @kindex set startup-with-shell
2465 @anchor{set startup-with-shell}
2466 @item set startup-with-shell
2467 @itemx set startup-with-shell on
2468 @itemx set startup-with-shell off
2469 @itemx show startup-with-shell
2470 On Unix systems, by default, if a shell is available on your target,
2471 @value{GDBN}) uses it to start your program. Arguments of the
2472 @code{run} command are passed to the shell, which does variable
2473 substitution, expands wildcard characters and performs redirection of
2474 I/O. In some circumstances, it may be useful to disable such use of a
2475 shell, for example, when debugging the shell itself or diagnosing
2476 startup failures such as:
2480 Starting program: ./a.out
2481 During startup program terminated with signal SIGSEGV, Segmentation fault.
2485 which indicates the shell or the wrapper specified with
2486 @samp{exec-wrapper} crashed, not your program. Most often, this is
2487 caused by something odd in your shell's non-interactive mode
2488 initialization file---such as @file{.cshrc} for C-shell,
2489 $@file{.zshenv} for the Z shell, or the file specified in the
2490 @samp{BASH_ENV} environment variable for BASH.
2492 @anchor{set auto-connect-native-target}
2493 @kindex set auto-connect-native-target
2494 @item set auto-connect-native-target
2495 @itemx set auto-connect-native-target on
2496 @itemx set auto-connect-native-target off
2497 @itemx show auto-connect-native-target
2499 By default, if not connected to any target yet (e.g., with
2500 @code{target remote}), the @code{run} command starts your program as a
2501 native process under @value{GDBN}, on your local machine. If you're
2502 sure you don't want to debug programs on your local machine, you can
2503 tell @value{GDBN} to not connect to the native target automatically
2504 with the @code{set auto-connect-native-target off} command.
2506 If @code{on}, which is the default, and if @value{GDBN} is not
2507 connected to a target already, the @code{run} command automaticaly
2508 connects to the native target, if one is available.
2510 If @code{off}, and if @value{GDBN} is not connected to a target
2511 already, the @code{run} command fails with an error:
2515 Don't know how to run. Try "help target".
2518 If @value{GDBN} is already connected to a target, @value{GDBN} always
2519 uses it with the @code{run} command.
2521 In any case, you can explicitly connect to the native target with the
2522 @code{target native} command. For example,
2525 (@value{GDBP}) set auto-connect-native-target off
2527 Don't know how to run. Try "help target".
2528 (@value{GDBP}) target native
2530 Starting program: ./a.out
2531 [Inferior 1 (process 10421) exited normally]
2534 In case you connected explicitly to the @code{native} target,
2535 @value{GDBN} remains connected even if all inferiors exit, ready for
2536 the next @code{run} command. Use the @code{disconnect} command to
2539 Examples of other commands that likewise respect the
2540 @code{auto-connect-native-target} setting: @code{attach}, @code{info
2541 proc}, @code{info os}.
2543 @kindex set disable-randomization
2544 @item set disable-randomization
2545 @itemx set disable-randomization on
2546 This option (enabled by default in @value{GDBN}) will turn off the native
2547 randomization of the virtual address space of the started program. This option
2548 is useful for multiple debugging sessions to make the execution better
2549 reproducible and memory addresses reusable across debugging sessions.
2551 This feature is implemented only on certain targets, including @sc{gnu}/Linux.
2552 On @sc{gnu}/Linux you can get the same behavior using
2555 (@value{GDBP}) set exec-wrapper setarch `uname -m` -R
2558 @item set disable-randomization off
2559 Leave the behavior of the started executable unchanged. Some bugs rear their
2560 ugly heads only when the program is loaded at certain addresses. If your bug
2561 disappears when you run the program under @value{GDBN}, that might be because
2562 @value{GDBN} by default disables the address randomization on platforms, such
2563 as @sc{gnu}/Linux, which do that for stand-alone programs. Use @kbd{set
2564 disable-randomization off} to try to reproduce such elusive bugs.
2566 On targets where it is available, virtual address space randomization
2567 protects the programs against certain kinds of security attacks. In these
2568 cases the attacker needs to know the exact location of a concrete executable
2569 code. Randomizing its location makes it impossible to inject jumps misusing
2570 a code at its expected addresses.
2572 Prelinking shared libraries provides a startup performance advantage but it
2573 makes addresses in these libraries predictable for privileged processes by
2574 having just unprivileged access at the target system. Reading the shared
2575 library binary gives enough information for assembling the malicious code
2576 misusing it. Still even a prelinked shared library can get loaded at a new
2577 random address just requiring the regular relocation process during the
2578 startup. Shared libraries not already prelinked are always loaded at
2579 a randomly chosen address.
2581 Position independent executables (PIE) contain position independent code
2582 similar to the shared libraries and therefore such executables get loaded at
2583 a randomly chosen address upon startup. PIE executables always load even
2584 already prelinked shared libraries at a random address. You can build such
2585 executable using @command{gcc -fPIE -pie}.
2587 Heap (malloc storage), stack and custom mmap areas are always placed randomly
2588 (as long as the randomization is enabled).
2590 @item show disable-randomization
2591 Show the current setting of the explicit disable of the native randomization of
2592 the virtual address space of the started program.
2597 @section Your Program's Arguments
2599 @cindex arguments (to your program)
2600 The arguments to your program can be specified by the arguments of the
2602 They are passed to a shell, which expands wildcard characters and
2603 performs redirection of I/O, and thence to your program. Your
2604 @code{SHELL} environment variable (if it exists) specifies what shell
2605 @value{GDBN} uses. If you do not define @code{SHELL}, @value{GDBN} uses
2606 the default shell (@file{/bin/sh} on Unix).
2608 On non-Unix systems, the program is usually invoked directly by
2609 @value{GDBN}, which emulates I/O redirection via the appropriate system
2610 calls, and the wildcard characters are expanded by the startup code of
2611 the program, not by the shell.
2613 @code{run} with no arguments uses the same arguments used by the previous
2614 @code{run}, or those set by the @code{set args} command.
2619 Specify the arguments to be used the next time your program is run. If
2620 @code{set args} has no arguments, @code{run} executes your program
2621 with no arguments. Once you have run your program with arguments,
2622 using @code{set args} before the next @code{run} is the only way to run
2623 it again without arguments.
2627 Show the arguments to give your program when it is started.
2631 @section Your Program's Environment
2633 @cindex environment (of your program)
2634 The @dfn{environment} consists of a set of environment variables and
2635 their values. Environment variables conventionally record such things as
2636 your user name, your home directory, your terminal type, and your search
2637 path for programs to run. Usually you set up environment variables with
2638 the shell and they are inherited by all the other programs you run. When
2639 debugging, it can be useful to try running your program with a modified
2640 environment without having to start @value{GDBN} over again.
2644 @item path @var{directory}
2645 Add @var{directory} to the front of the @code{PATH} environment variable
2646 (the search path for executables) that will be passed to your program.
2647 The value of @code{PATH} used by @value{GDBN} does not change.
2648 You may specify several directory names, separated by whitespace or by a
2649 system-dependent separator character (@samp{:} on Unix, @samp{;} on
2650 MS-DOS and MS-Windows). If @var{directory} is already in the path, it
2651 is moved to the front, so it is searched sooner.
2653 You can use the string @samp{$cwd} to refer to whatever is the current
2654 working directory at the time @value{GDBN} searches the path. If you
2655 use @samp{.} instead, it refers to the directory where you executed the
2656 @code{path} command. @value{GDBN} replaces @samp{.} in the
2657 @var{directory} argument (with the current path) before adding
2658 @var{directory} to the search path.
2659 @c 'path' is explicitly nonrepeatable, but RMS points out it is silly to
2660 @c document that, since repeating it would be a no-op.
2664 Display the list of search paths for executables (the @code{PATH}
2665 environment variable).
2667 @kindex show environment
2668 @item show environment @r{[}@var{varname}@r{]}
2669 Print the value of environment variable @var{varname} to be given to
2670 your program when it starts. If you do not supply @var{varname},
2671 print the names and values of all environment variables to be given to
2672 your program. You can abbreviate @code{environment} as @code{env}.
2674 @kindex set environment
2675 @anchor{set environment}
2676 @item set environment @var{varname} @r{[}=@var{value}@r{]}
2677 Set environment variable @var{varname} to @var{value}. The value
2678 changes for your program (and the shell @value{GDBN} uses to launch
2679 it), not for @value{GDBN} itself. The @var{value} may be any string; the
2680 values of environment variables are just strings, and any
2681 interpretation is supplied by your program itself. The @var{value}
2682 parameter is optional; if it is eliminated, the variable is set to a
2684 @c "any string" here does not include leading, trailing
2685 @c blanks. Gnu asks: does anyone care?
2687 For example, this command:
2694 tells the debugged program, when subsequently run, that its user is named
2695 @samp{foo}. (The spaces around @samp{=} are used for clarity here; they
2696 are not actually required.)
2698 Note that on Unix systems, @value{GDBN} runs your program via a shell,
2699 which also inherits the environment set with @code{set environment}.
2700 If necessary, you can avoid that by using the @samp{env} program as a
2701 wrapper instead of using @code{set environment}. @xref{set
2702 exec-wrapper}, for an example doing just that.
2704 Environment variables that are set by the user are also transmitted to
2705 @command{gdbserver} to be used when starting the remote inferior.
2706 @pxref{QEnvironmentHexEncoded}.
2708 @kindex unset environment
2709 @anchor{unset environment}
2710 @item unset environment @var{varname}
2711 Remove variable @var{varname} from the environment to be passed to your
2712 program. This is different from @samp{set env @var{varname} =};
2713 @code{unset environment} removes the variable from the environment,
2714 rather than assigning it an empty value.
2716 Environment variables that are unset by the user are also unset on
2717 @command{gdbserver} when starting the remote inferior.
2718 @pxref{QEnvironmentUnset}.
2721 @emph{Warning:} On Unix systems, @value{GDBN} runs your program using
2722 the shell indicated by your @code{SHELL} environment variable if it
2723 exists (or @code{/bin/sh} if not). If your @code{SHELL} variable
2724 names a shell that runs an initialization file when started
2725 non-interactively---such as @file{.cshrc} for C-shell, $@file{.zshenv}
2726 for the Z shell, or the file specified in the @samp{BASH_ENV}
2727 environment variable for BASH---any variables you set in that file
2728 affect your program. You may wish to move setting of environment
2729 variables to files that are only run when you sign on, such as
2730 @file{.login} or @file{.profile}.
2732 @node Working Directory
2733 @section Your Program's Working Directory
2735 @cindex working directory (of your program)
2736 Each time you start your program with @code{run}, the inferior will be
2737 initialized with the current working directory specified by the
2738 @kbd{set cwd} command. If no directory has been specified by this
2739 command, then the inferior will inherit @value{GDBN}'s current working
2740 directory as its working directory if native debugging, or it will
2741 inherit the remote server's current working directory if remote
2746 @cindex change inferior's working directory
2747 @anchor{set cwd command}
2748 @item set cwd @r{[}@var{directory}@r{]}
2749 Set the inferior's working directory to @var{directory}, which will be
2750 @code{glob}-expanded in order to resolve tildes (@file{~}). If no
2751 argument has been specified, the command clears the setting and resets
2752 it to an empty state. This setting has no effect on @value{GDBN}'s
2753 working directory, and it only takes effect the next time you start
2754 the inferior. The @file{~} in @var{directory} is a short for the
2755 @dfn{home directory}, usually pointed to by the @env{HOME} environment
2756 variable. On MS-Windows, if @env{HOME} is not defined, @value{GDBN}
2757 uses the concatenation of @env{HOMEDRIVE} and @env{HOMEPATH} as
2760 You can also change @value{GDBN}'s current working directory by using
2761 the @code{cd} command.
2765 @cindex show inferior's working directory
2767 Show the inferior's working directory. If no directory has been
2768 specified by @kbd{set cwd}, then the default inferior's working
2769 directory is the same as @value{GDBN}'s working directory.
2772 @cindex change @value{GDBN}'s working directory
2774 @item cd @r{[}@var{directory}@r{]}
2775 Set the @value{GDBN} working directory to @var{directory}. If not
2776 given, @var{directory} uses @file{'~'}.
2778 The @value{GDBN} working directory serves as a default for the
2779 commands that specify files for @value{GDBN} to operate on.
2780 @xref{Files, ,Commands to Specify Files}.
2781 @xref{set cwd command}.
2785 Print the @value{GDBN} working directory.
2788 It is generally impossible to find the current working directory of
2789 the process being debugged (since a program can change its directory
2790 during its run). If you work on a system where @value{GDBN} supports
2791 the @code{info proc} command (@pxref{Process Information}), you can
2792 use the @code{info proc} command to find out the
2793 current working directory of the debuggee.
2796 @section Your Program's Input and Output
2801 By default, the program you run under @value{GDBN} does input and output to
2802 the same terminal that @value{GDBN} uses. @value{GDBN} switches the terminal
2803 to its own terminal modes to interact with you, but it records the terminal
2804 modes your program was using and switches back to them when you continue
2805 running your program.
2808 @kindex info terminal
2810 Displays information recorded by @value{GDBN} about the terminal modes your
2814 You can redirect your program's input and/or output using shell
2815 redirection with the @code{run} command. For example,
2822 starts your program, diverting its output to the file @file{outfile}.
2825 @cindex controlling terminal
2826 Another way to specify where your program should do input and output is
2827 with the @code{tty} command. This command accepts a file name as
2828 argument, and causes this file to be the default for future @code{run}
2829 commands. It also resets the controlling terminal for the child
2830 process, for future @code{run} commands. For example,
2837 directs that processes started with subsequent @code{run} commands
2838 default to do input and output on the terminal @file{/dev/ttyb} and have
2839 that as their controlling terminal.
2841 An explicit redirection in @code{run} overrides the @code{tty} command's
2842 effect on the input/output device, but not its effect on the controlling
2845 When you use the @code{tty} command or redirect input in the @code{run}
2846 command, only the input @emph{for your program} is affected. The input
2847 for @value{GDBN} still comes from your terminal. @code{tty} is an alias
2848 for @code{set inferior-tty}.
2850 @cindex inferior tty
2851 @cindex set inferior controlling terminal
2852 You can use the @code{show inferior-tty} command to tell @value{GDBN} to
2853 display the name of the terminal that will be used for future runs of your
2857 @item set inferior-tty [ @var{tty} ]
2858 @kindex set inferior-tty
2859 Set the tty for the program being debugged to @var{tty}. Omitting @var{tty}
2860 restores the default behavior, which is to use the same terminal as
2863 @item show inferior-tty
2864 @kindex show inferior-tty
2865 Show the current tty for the program being debugged.
2869 @section Debugging an Already-running Process
2874 @item attach @var{process-id}
2875 This command attaches to a running process---one that was started
2876 outside @value{GDBN}. (@code{info files} shows your active
2877 targets.) The command takes as argument a process ID. The usual way to
2878 find out the @var{process-id} of a Unix process is with the @code{ps} utility,
2879 or with the @samp{jobs -l} shell command.
2881 @code{attach} does not repeat if you press @key{RET} a second time after
2882 executing the command.
2885 To use @code{attach}, your program must be running in an environment
2886 which supports processes; for example, @code{attach} does not work for
2887 programs on bare-board targets that lack an operating system. You must
2888 also have permission to send the process a signal.
2890 When you use @code{attach}, the debugger finds the program running in
2891 the process first by looking in the current working directory, then (if
2892 the program is not found) by using the source file search path
2893 (@pxref{Source Path, ,Specifying Source Directories}). You can also use
2894 the @code{file} command to load the program. @xref{Files, ,Commands to
2897 The first thing @value{GDBN} does after arranging to debug the specified
2898 process is to stop it. You can examine and modify an attached process
2899 with all the @value{GDBN} commands that are ordinarily available when
2900 you start processes with @code{run}. You can insert breakpoints; you
2901 can step and continue; you can modify storage. If you would rather the
2902 process continue running, you may use the @code{continue} command after
2903 attaching @value{GDBN} to the process.
2908 When you have finished debugging the attached process, you can use the
2909 @code{detach} command to release it from @value{GDBN} control. Detaching
2910 the process continues its execution. After the @code{detach} command,
2911 that process and @value{GDBN} become completely independent once more, and you
2912 are ready to @code{attach} another process or start one with @code{run}.
2913 @code{detach} does not repeat if you press @key{RET} again after
2914 executing the command.
2917 If you exit @value{GDBN} while you have an attached process, you detach
2918 that process. If you use the @code{run} command, you kill that process.
2919 By default, @value{GDBN} asks for confirmation if you try to do either of these
2920 things; you can control whether or not you need to confirm by using the
2921 @code{set confirm} command (@pxref{Messages/Warnings, ,Optional Warnings and
2925 @section Killing the Child Process
2930 Kill the child process in which your program is running under @value{GDBN}.
2933 This command is useful if you wish to debug a core dump instead of a
2934 running process. @value{GDBN} ignores any core dump file while your program
2937 On some operating systems, a program cannot be executed outside @value{GDBN}
2938 while you have breakpoints set on it inside @value{GDBN}. You can use the
2939 @code{kill} command in this situation to permit running your program
2940 outside the debugger.
2942 The @code{kill} command is also useful if you wish to recompile and
2943 relink your program, since on many systems it is impossible to modify an
2944 executable file while it is running in a process. In this case, when you
2945 next type @code{run}, @value{GDBN} notices that the file has changed, and
2946 reads the symbol table again (while trying to preserve your current
2947 breakpoint settings).
2949 @node Inferiors and Programs
2950 @section Debugging Multiple Inferiors and Programs
2952 @value{GDBN} lets you run and debug multiple programs in a single
2953 session. In addition, @value{GDBN} on some systems may let you run
2954 several programs simultaneously (otherwise you have to exit from one
2955 before starting another). In the most general case, you can have
2956 multiple threads of execution in each of multiple processes, launched
2957 from multiple executables.
2960 @value{GDBN} represents the state of each program execution with an
2961 object called an @dfn{inferior}. An inferior typically corresponds to
2962 a process, but is more general and applies also to targets that do not
2963 have processes. Inferiors may be created before a process runs, and
2964 may be retained after a process exits. Inferiors have unique
2965 identifiers that are different from process ids. Usually each
2966 inferior will also have its own distinct address space, although some
2967 embedded targets may have several inferiors running in different parts
2968 of a single address space. Each inferior may in turn have multiple
2969 threads running in it.
2971 To find out what inferiors exist at any moment, use @w{@code{info
2975 @kindex info inferiors [ @var{id}@dots{} ]
2976 @item info inferiors
2977 Print a list of all inferiors currently being managed by @value{GDBN}.
2978 By default all inferiors are printed, but the argument @var{id}@dots{}
2979 -- a space separated list of inferior numbers -- can be used to limit
2980 the display to just the requested inferiors.
2982 @value{GDBN} displays for each inferior (in this order):
2986 the inferior number assigned by @value{GDBN}
2989 the target system's inferior identifier
2992 the name of the executable the inferior is running.
2997 An asterisk @samp{*} preceding the @value{GDBN} inferior number
2998 indicates the current inferior.
3002 @c end table here to get a little more width for example
3005 (@value{GDBP}) info inferiors
3006 Num Description Executable
3007 2 process 2307 hello
3008 * 1 process 3401 goodbye
3011 To switch focus between inferiors, use the @code{inferior} command:
3014 @kindex inferior @var{infno}
3015 @item inferior @var{infno}
3016 Make inferior number @var{infno} the current inferior. The argument
3017 @var{infno} is the inferior number assigned by @value{GDBN}, as shown
3018 in the first field of the @samp{info inferiors} display.
3021 @vindex $_inferior@r{, convenience variable}
3022 The debugger convenience variable @samp{$_inferior} contains the
3023 number of the current inferior. You may find this useful in writing
3024 breakpoint conditional expressions, command scripts, and so forth.
3025 @xref{Convenience Vars,, Convenience Variables}, for general
3026 information on convenience variables.
3028 You can get multiple executables into a debugging session via the
3029 @code{add-inferior} and @w{@code{clone-inferior}} commands. On some
3030 systems @value{GDBN} can add inferiors to the debug session
3031 automatically by following calls to @code{fork} and @code{exec}. To
3032 remove inferiors from the debugging session use the
3033 @w{@code{remove-inferiors}} command.
3036 @kindex add-inferior
3037 @item add-inferior [ -copies @var{n} ] [ -exec @var{executable} ]
3038 Adds @var{n} inferiors to be run using @var{executable} as the
3039 executable; @var{n} defaults to 1. If no executable is specified,
3040 the inferiors begins empty, with no program. You can still assign or
3041 change the program assigned to the inferior at any time by using the
3042 @code{file} command with the executable name as its argument.
3044 @kindex clone-inferior
3045 @item clone-inferior [ -copies @var{n} ] [ @var{infno} ]
3046 Adds @var{n} inferiors ready to execute the same program as inferior
3047 @var{infno}; @var{n} defaults to 1, and @var{infno} defaults to the
3048 number of the current inferior. This is a convenient command when you
3049 want to run another instance of the inferior you are debugging.
3052 (@value{GDBP}) info inferiors
3053 Num Description Executable
3054 * 1 process 29964 helloworld
3055 (@value{GDBP}) clone-inferior
3058 (@value{GDBP}) info inferiors
3059 Num Description Executable
3061 * 1 process 29964 helloworld
3064 You can now simply switch focus to inferior 2 and run it.
3066 @kindex remove-inferiors
3067 @item remove-inferiors @var{infno}@dots{}
3068 Removes the inferior or inferiors @var{infno}@dots{}. It is not
3069 possible to remove an inferior that is running with this command. For
3070 those, use the @code{kill} or @code{detach} command first.
3074 To quit debugging one of the running inferiors that is not the current
3075 inferior, you can either detach from it by using the @w{@code{detach
3076 inferior}} command (allowing it to run independently), or kill it
3077 using the @w{@code{kill inferiors}} command:
3080 @kindex detach inferiors @var{infno}@dots{}
3081 @item detach inferior @var{infno}@dots{}
3082 Detach from the inferior or inferiors identified by @value{GDBN}
3083 inferior number(s) @var{infno}@dots{}. Note that the inferior's entry
3084 still stays on the list of inferiors shown by @code{info inferiors},
3085 but its Description will show @samp{<null>}.
3087 @kindex kill inferiors @var{infno}@dots{}
3088 @item kill inferiors @var{infno}@dots{}
3089 Kill the inferior or inferiors identified by @value{GDBN} inferior
3090 number(s) @var{infno}@dots{}. Note that the inferior's entry still
3091 stays on the list of inferiors shown by @code{info inferiors}, but its
3092 Description will show @samp{<null>}.
3095 After the successful completion of a command such as @code{detach},
3096 @code{detach inferiors}, @code{kill} or @code{kill inferiors}, or after
3097 a normal process exit, the inferior is still valid and listed with
3098 @code{info inferiors}, ready to be restarted.
3101 To be notified when inferiors are started or exit under @value{GDBN}'s
3102 control use @w{@code{set print inferior-events}}:
3105 @kindex set print inferior-events
3106 @cindex print messages on inferior start and exit
3107 @item set print inferior-events
3108 @itemx set print inferior-events on
3109 @itemx set print inferior-events off
3110 The @code{set print inferior-events} command allows you to enable or
3111 disable printing of messages when @value{GDBN} notices that new
3112 inferiors have started or that inferiors have exited or have been
3113 detached. By default, these messages will not be printed.
3115 @kindex show print inferior-events
3116 @item show print inferior-events
3117 Show whether messages will be printed when @value{GDBN} detects that
3118 inferiors have started, exited or have been detached.
3121 Many commands will work the same with multiple programs as with a
3122 single program: e.g., @code{print myglobal} will simply display the
3123 value of @code{myglobal} in the current inferior.
3126 Occasionaly, when debugging @value{GDBN} itself, it may be useful to
3127 get more info about the relationship of inferiors, programs, address
3128 spaces in a debug session. You can do that with the @w{@code{maint
3129 info program-spaces}} command.
3132 @kindex maint info program-spaces
3133 @item maint info program-spaces
3134 Print a list of all program spaces currently being managed by
3137 @value{GDBN} displays for each program space (in this order):
3141 the program space number assigned by @value{GDBN}
3144 the name of the executable loaded into the program space, with e.g.,
3145 the @code{file} command.
3150 An asterisk @samp{*} preceding the @value{GDBN} program space number
3151 indicates the current program space.
3153 In addition, below each program space line, @value{GDBN} prints extra
3154 information that isn't suitable to display in tabular form. For
3155 example, the list of inferiors bound to the program space.
3158 (@value{GDBP}) maint info program-spaces
3162 Bound inferiors: ID 1 (process 21561)
3165 Here we can see that no inferior is running the program @code{hello},
3166 while @code{process 21561} is running the program @code{goodbye}. On
3167 some targets, it is possible that multiple inferiors are bound to the
3168 same program space. The most common example is that of debugging both
3169 the parent and child processes of a @code{vfork} call. For example,
3172 (@value{GDBP}) maint info program-spaces
3175 Bound inferiors: ID 2 (process 18050), ID 1 (process 18045)
3178 Here, both inferior 2 and inferior 1 are running in the same program
3179 space as a result of inferior 1 having executed a @code{vfork} call.
3183 @section Debugging Programs with Multiple Threads
3185 @cindex threads of execution
3186 @cindex multiple threads
3187 @cindex switching threads
3188 In some operating systems, such as GNU/Linux and Solaris, a single program
3189 may have more than one @dfn{thread} of execution. The precise semantics
3190 of threads differ from one operating system to another, but in general
3191 the threads of a single program are akin to multiple processes---except
3192 that they share one address space (that is, they can all examine and
3193 modify the same variables). On the other hand, each thread has its own
3194 registers and execution stack, and perhaps private memory.
3196 @value{GDBN} provides these facilities for debugging multi-thread
3200 @item automatic notification of new threads
3201 @item @samp{thread @var{thread-id}}, a command to switch among threads
3202 @item @samp{info threads}, a command to inquire about existing threads
3203 @item @samp{thread apply [@var{thread-id-list} | all] @var{args}},
3204 a command to apply a command to a list of threads
3205 @item thread-specific breakpoints
3206 @item @samp{set print thread-events}, which controls printing of
3207 messages on thread start and exit.
3208 @item @samp{set libthread-db-search-path @var{path}}, which lets
3209 the user specify which @code{libthread_db} to use if the default choice
3210 isn't compatible with the program.
3213 @cindex focus of debugging
3214 @cindex current thread
3215 The @value{GDBN} thread debugging facility allows you to observe all
3216 threads while your program runs---but whenever @value{GDBN} takes
3217 control, one thread in particular is always the focus of debugging.
3218 This thread is called the @dfn{current thread}. Debugging commands show
3219 program information from the perspective of the current thread.
3221 @cindex @code{New} @var{systag} message
3222 @cindex thread identifier (system)
3223 @c FIXME-implementors!! It would be more helpful if the [New...] message
3224 @c included GDB's numeric thread handle, so you could just go to that
3225 @c thread without first checking `info threads'.
3226 Whenever @value{GDBN} detects a new thread in your program, it displays
3227 the target system's identification for the thread with a message in the
3228 form @samp{[New @var{systag}]}, where @var{systag} is a thread identifier
3229 whose form varies depending on the particular system. For example, on
3230 @sc{gnu}/Linux, you might see
3233 [New Thread 0x41e02940 (LWP 25582)]
3237 when @value{GDBN} notices a new thread. In contrast, on other systems,
3238 the @var{systag} is simply something like @samp{process 368}, with no
3241 @c FIXME!! (1) Does the [New...] message appear even for the very first
3242 @c thread of a program, or does it only appear for the
3243 @c second---i.e.@: when it becomes obvious we have a multithread
3245 @c (2) *Is* there necessarily a first thread always? Or do some
3246 @c multithread systems permit starting a program with multiple
3247 @c threads ab initio?
3249 @anchor{thread numbers}
3250 @cindex thread number, per inferior
3251 @cindex thread identifier (GDB)
3252 For debugging purposes, @value{GDBN} associates its own thread number
3253 ---always a single integer---with each thread of an inferior. This
3254 number is unique between all threads of an inferior, but not unique
3255 between threads of different inferiors.
3257 @cindex qualified thread ID
3258 You can refer to a given thread in an inferior using the qualified
3259 @var{inferior-num}.@var{thread-num} syntax, also known as
3260 @dfn{qualified thread ID}, with @var{inferior-num} being the inferior
3261 number and @var{thread-num} being the thread number of the given
3262 inferior. For example, thread @code{2.3} refers to thread number 3 of
3263 inferior 2. If you omit @var{inferior-num} (e.g., @code{thread 3}),
3264 then @value{GDBN} infers you're referring to a thread of the current
3267 Until you create a second inferior, @value{GDBN} does not show the
3268 @var{inferior-num} part of thread IDs, even though you can always use
3269 the full @var{inferior-num}.@var{thread-num} form to refer to threads
3270 of inferior 1, the initial inferior.
3272 @anchor{thread ID lists}
3273 @cindex thread ID lists
3274 Some commands accept a space-separated @dfn{thread ID list} as
3275 argument. A list element can be:
3279 A thread ID as shown in the first field of the @samp{info threads}
3280 display, with or without an inferior qualifier. E.g., @samp{2.1} or
3284 A range of thread numbers, again with or without an inferior
3285 qualifier, as in @var{inf}.@var{thr1}-@var{thr2} or
3286 @var{thr1}-@var{thr2}. E.g., @samp{1.2-4} or @samp{2-4}.
3289 All threads of an inferior, specified with a star wildcard, with or
3290 without an inferior qualifier, as in @var{inf}.@code{*} (e.g.,
3291 @samp{1.*}) or @code{*}. The former refers to all threads of the
3292 given inferior, and the latter form without an inferior qualifier
3293 refers to all threads of the current inferior.
3297 For example, if the current inferior is 1, and inferior 7 has one
3298 thread with ID 7.1, the thread list @samp{1 2-3 4.5 6.7-9 7.*}
3299 includes threads 1 to 3 of inferior 1, thread 5 of inferior 4, threads
3300 7 to 9 of inferior 6 and all threads of inferior 7. That is, in
3301 expanded qualified form, the same as @samp{1.1 1.2 1.3 4.5 6.7 6.8 6.9
3305 @anchor{global thread numbers}
3306 @cindex global thread number
3307 @cindex global thread identifier (GDB)
3308 In addition to a @emph{per-inferior} number, each thread is also
3309 assigned a unique @emph{global} number, also known as @dfn{global
3310 thread ID}, a single integer. Unlike the thread number component of
3311 the thread ID, no two threads have the same global ID, even when
3312 you're debugging multiple inferiors.
3314 From @value{GDBN}'s perspective, a process always has at least one
3315 thread. In other words, @value{GDBN} assigns a thread number to the
3316 program's ``main thread'' even if the program is not multi-threaded.
3318 @vindex $_thread@r{, convenience variable}
3319 @vindex $_gthread@r{, convenience variable}
3320 The debugger convenience variables @samp{$_thread} and
3321 @samp{$_gthread} contain, respectively, the per-inferior thread number
3322 and the global thread number of the current thread. You may find this
3323 useful in writing breakpoint conditional expressions, command scripts,
3324 and so forth. @xref{Convenience Vars,, Convenience Variables}, for
3325 general information on convenience variables.
3327 If @value{GDBN} detects the program is multi-threaded, it augments the
3328 usual message about stopping at a breakpoint with the ID and name of
3329 the thread that hit the breakpoint.
3332 Thread 2 "client" hit Breakpoint 1, send_message () at client.c:68
3335 Likewise when the program receives a signal:
3338 Thread 1 "main" received signal SIGINT, Interrupt.
3342 @kindex info threads
3343 @item info threads @r{[}@var{thread-id-list}@r{]}
3345 Display information about one or more threads. With no arguments
3346 displays information about all threads. You can specify the list of
3347 threads that you want to display using the thread ID list syntax
3348 (@pxref{thread ID lists}).
3350 @value{GDBN} displays for each thread (in this order):
3354 the per-inferior thread number assigned by @value{GDBN}
3357 the global thread number assigned by @value{GDBN}, if the @samp{-gid}
3358 option was specified
3361 the target system's thread identifier (@var{systag})
3364 the thread's name, if one is known. A thread can either be named by
3365 the user (see @code{thread name}, below), or, in some cases, by the
3369 the current stack frame summary for that thread
3373 An asterisk @samp{*} to the left of the @value{GDBN} thread number
3374 indicates the current thread.
3378 @c end table here to get a little more width for example
3381 (@value{GDBP}) info threads
3383 * 1 process 35 thread 13 main (argc=1, argv=0x7ffffff8)
3384 2 process 35 thread 23 0x34e5 in sigpause ()
3385 3 process 35 thread 27 0x34e5 in sigpause ()
3389 If you're debugging multiple inferiors, @value{GDBN} displays thread
3390 IDs using the qualified @var{inferior-num}.@var{thread-num} format.
3391 Otherwise, only @var{thread-num} is shown.
3393 If you specify the @samp{-gid} option, @value{GDBN} displays a column
3394 indicating each thread's global thread ID:
3397 (@value{GDBP}) info threads
3398 Id GId Target Id Frame
3399 1.1 1 process 35 thread 13 main (argc=1, argv=0x7ffffff8)
3400 1.2 3 process 35 thread 23 0x34e5 in sigpause ()
3401 1.3 4 process 35 thread 27 0x34e5 in sigpause ()
3402 * 2.1 2 process 65 thread 1 main (argc=1, argv=0x7ffffff8)
3405 On Solaris, you can display more information about user threads with a
3406 Solaris-specific command:
3409 @item maint info sol-threads
3410 @kindex maint info sol-threads
3411 @cindex thread info (Solaris)
3412 Display info on Solaris user threads.
3416 @kindex thread @var{thread-id}
3417 @item thread @var{thread-id}
3418 Make thread ID @var{thread-id} the current thread. The command
3419 argument @var{thread-id} is the @value{GDBN} thread ID, as shown in
3420 the first field of the @samp{info threads} display, with or without an
3421 inferior qualifier (e.g., @samp{2.1} or @samp{1}).
3423 @value{GDBN} responds by displaying the system identifier of the
3424 thread you selected, and its current stack frame summary:
3427 (@value{GDBP}) thread 2
3428 [Switching to thread 2 (Thread 0xb7fdab70 (LWP 12747))]
3429 #0 some_function (ignore=0x0) at example.c:8
3430 8 printf ("hello\n");
3434 As with the @samp{[New @dots{}]} message, the form of the text after
3435 @samp{Switching to} depends on your system's conventions for identifying
3438 @anchor{thread apply all}
3439 @kindex thread apply
3440 @cindex apply command to several threads
3441 @item thread apply [@var{thread-id-list} | all [-ascending]] [@var{flag}]@dots{} @var{command}
3442 The @code{thread apply} command allows you to apply the named
3443 @var{command} to one or more threads. Specify the threads that you
3444 want affected using the thread ID list syntax (@pxref{thread ID
3445 lists}), or specify @code{all} to apply to all threads. To apply a
3446 command to all threads in descending order, type @kbd{thread apply all
3447 @var{command}}. To apply a command to all threads in ascending order,
3448 type @kbd{thread apply all -ascending @var{command}}.
3450 The @var{flag} arguments control what output to produce and how to handle
3451 errors raised when applying @var{command} to a thread. @var{flag}
3452 must start with a @code{-} directly followed by one letter in
3453 @code{qcs}. If several flags are provided, they must be given
3454 individually, such as @code{-c -q}.
3456 By default, @value{GDBN} displays some thread information before the
3457 output produced by @var{command}, and an error raised during the
3458 execution of a @var{command} will abort @code{thread apply}. The
3459 following flags can be used to fine-tune this behavior:
3463 The flag @code{-c}, which stands for @samp{continue}, causes any
3464 errors in @var{command} to be displayed, and the execution of
3465 @code{thread apply} then continues.
3467 The flag @code{-s}, which stands for @samp{silent}, causes any errors
3468 or empty output produced by a @var{command} to be silently ignored.
3469 That is, the execution continues, but the thread information and errors
3472 The flag @code{-q} (@samp{quiet}) disables printing the thread
3476 Flags @code{-c} and @code{-s} cannot be used together.
3479 @cindex apply command to all threads (ignoring errors and empty output)
3480 @item taas [@var{option}]@dots{} @var{command}
3481 Shortcut for @code{thread apply all -s [@var{option}]@dots{} @var{command}}.
3482 Applies @var{command} on all threads, ignoring errors and empty output.
3484 The @code{taas} command accepts the same options as the @code{thread
3485 apply all} command. @xref{thread apply all}.
3488 @cindex apply a command to all frames of all threads (ignoring errors and empty output)
3489 @item tfaas [@var{option}]@dots{} @var{command}
3490 Shortcut for @code{thread apply all -s -- frame apply all -s [@var{option}]@dots{} @var{command}}.
3491 Applies @var{command} on all frames of all threads, ignoring errors
3492 and empty output. Note that the flag @code{-s} is specified twice:
3493 The first @code{-s} ensures that @code{thread apply} only shows the thread
3494 information of the threads for which @code{frame apply} produces
3495 some output. The second @code{-s} is needed to ensure that @code{frame
3496 apply} shows the frame information of a frame only if the
3497 @var{command} successfully produced some output.
3499 It can for example be used to print a local variable or a function
3500 argument without knowing the thread or frame where this variable or argument
3503 (@value{GDBP}) tfaas p some_local_var_i_do_not_remember_where_it_is
3506 The @code{tfaas} command accepts the same options as the @code{frame
3507 apply} command. @xref{frame apply}.
3510 @cindex name a thread
3511 @item thread name [@var{name}]
3512 This command assigns a name to the current thread. If no argument is
3513 given, any existing user-specified name is removed. The thread name
3514 appears in the @samp{info threads} display.
3516 On some systems, such as @sc{gnu}/Linux, @value{GDBN} is able to
3517 determine the name of the thread as given by the OS. On these
3518 systems, a name specified with @samp{thread name} will override the
3519 system-give name, and removing the user-specified name will cause
3520 @value{GDBN} to once again display the system-specified name.
3523 @cindex search for a thread
3524 @item thread find [@var{regexp}]
3525 Search for and display thread ids whose name or @var{systag}
3526 matches the supplied regular expression.
3528 As well as being the complement to the @samp{thread name} command,
3529 this command also allows you to identify a thread by its target
3530 @var{systag}. For instance, on @sc{gnu}/Linux, the target @var{systag}
3534 (@value{GDBN}) thread find 26688
3535 Thread 4 has target id 'Thread 0x41e02940 (LWP 26688)'
3536 (@value{GDBN}) info thread 4
3538 4 Thread 0x41e02940 (LWP 26688) 0x00000031ca6cd372 in select ()
3541 @kindex set print thread-events
3542 @cindex print messages on thread start and exit
3543 @item set print thread-events
3544 @itemx set print thread-events on
3545 @itemx set print thread-events off
3546 The @code{set print thread-events} command allows you to enable or
3547 disable printing of messages when @value{GDBN} notices that new threads have
3548 started or that threads have exited. By default, these messages will
3549 be printed if detection of these events is supported by the target.
3550 Note that these messages cannot be disabled on all targets.
3552 @kindex show print thread-events
3553 @item show print thread-events
3554 Show whether messages will be printed when @value{GDBN} detects that threads
3555 have started and exited.
3558 @xref{Thread Stops,,Stopping and Starting Multi-thread Programs}, for
3559 more information about how @value{GDBN} behaves when you stop and start
3560 programs with multiple threads.
3562 @xref{Set Watchpoints,,Setting Watchpoints}, for information about
3563 watchpoints in programs with multiple threads.
3565 @anchor{set libthread-db-search-path}
3567 @kindex set libthread-db-search-path
3568 @cindex search path for @code{libthread_db}
3569 @item set libthread-db-search-path @r{[}@var{path}@r{]}
3570 If this variable is set, @var{path} is a colon-separated list of
3571 directories @value{GDBN} will use to search for @code{libthread_db}.
3572 If you omit @var{path}, @samp{libthread-db-search-path} will be reset to
3573 its default value (@code{$sdir:$pdir} on @sc{gnu}/Linux and Solaris systems).
3574 Internally, the default value comes from the @code{LIBTHREAD_DB_SEARCH_PATH}
3577 On @sc{gnu}/Linux and Solaris systems, @value{GDBN} uses a ``helper''
3578 @code{libthread_db} library to obtain information about threads in the
3579 inferior process. @value{GDBN} will use @samp{libthread-db-search-path}
3580 to find @code{libthread_db}. @value{GDBN} also consults first if inferior
3581 specific thread debugging library loading is enabled
3582 by @samp{set auto-load libthread-db} (@pxref{libthread_db.so.1 file}).
3584 A special entry @samp{$sdir} for @samp{libthread-db-search-path}
3585 refers to the default system directories that are
3586 normally searched for loading shared libraries. The @samp{$sdir} entry
3587 is the only kind not needing to be enabled by @samp{set auto-load libthread-db}
3588 (@pxref{libthread_db.so.1 file}).
3590 A special entry @samp{$pdir} for @samp{libthread-db-search-path}
3591 refers to the directory from which @code{libpthread}
3592 was loaded in the inferior process.
3594 For any @code{libthread_db} library @value{GDBN} finds in above directories,
3595 @value{GDBN} attempts to initialize it with the current inferior process.
3596 If this initialization fails (which could happen because of a version
3597 mismatch between @code{libthread_db} and @code{libpthread}), @value{GDBN}
3598 will unload @code{libthread_db}, and continue with the next directory.
3599 If none of @code{libthread_db} libraries initialize successfully,
3600 @value{GDBN} will issue a warning and thread debugging will be disabled.
3602 Setting @code{libthread-db-search-path} is currently implemented
3603 only on some platforms.
3605 @kindex show libthread-db-search-path
3606 @item show libthread-db-search-path
3607 Display current libthread_db search path.
3609 @kindex set debug libthread-db
3610 @kindex show debug libthread-db
3611 @cindex debugging @code{libthread_db}
3612 @item set debug libthread-db
3613 @itemx show debug libthread-db
3614 Turns on or off display of @code{libthread_db}-related events.
3615 Use @code{1} to enable, @code{0} to disable.
3619 @section Debugging Forks
3621 @cindex fork, debugging programs which call
3622 @cindex multiple processes
3623 @cindex processes, multiple
3624 On most systems, @value{GDBN} has no special support for debugging
3625 programs which create additional processes using the @code{fork}
3626 function. When a program forks, @value{GDBN} will continue to debug the
3627 parent process and the child process will run unimpeded. If you have
3628 set a breakpoint in any code which the child then executes, the child
3629 will get a @code{SIGTRAP} signal which (unless it catches the signal)
3630 will cause it to terminate.
3632 However, if you want to debug the child process there is a workaround
3633 which isn't too painful. Put a call to @code{sleep} in the code which
3634 the child process executes after the fork. It may be useful to sleep
3635 only if a certain environment variable is set, or a certain file exists,
3636 so that the delay need not occur when you don't want to run @value{GDBN}
3637 on the child. While the child is sleeping, use the @code{ps} program to
3638 get its process ID. Then tell @value{GDBN} (a new invocation of
3639 @value{GDBN} if you are also debugging the parent process) to attach to
3640 the child process (@pxref{Attach}). From that point on you can debug
3641 the child process just like any other process which you attached to.
3643 On some systems, @value{GDBN} provides support for debugging programs
3644 that create additional processes using the @code{fork} or @code{vfork}
3645 functions. On @sc{gnu}/Linux platforms, this feature is supported
3646 with kernel version 2.5.46 and later.
3648 The fork debugging commands are supported in native mode and when
3649 connected to @code{gdbserver} in either @code{target remote} mode or
3650 @code{target extended-remote} mode.
3652 By default, when a program forks, @value{GDBN} will continue to debug
3653 the parent process and the child process will run unimpeded.
3655 If you want to follow the child process instead of the parent process,
3656 use the command @w{@code{set follow-fork-mode}}.
3659 @kindex set follow-fork-mode
3660 @item set follow-fork-mode @var{mode}
3661 Set the debugger response to a program call of @code{fork} or
3662 @code{vfork}. A call to @code{fork} or @code{vfork} creates a new
3663 process. The @var{mode} argument can be:
3667 The original process is debugged after a fork. The child process runs
3668 unimpeded. This is the default.
3671 The new process is debugged after a fork. The parent process runs
3676 @kindex show follow-fork-mode
3677 @item show follow-fork-mode
3678 Display the current debugger response to a @code{fork} or @code{vfork} call.
3681 @cindex debugging multiple processes
3682 On Linux, if you want to debug both the parent and child processes, use the
3683 command @w{@code{set detach-on-fork}}.
3686 @kindex set detach-on-fork
3687 @item set detach-on-fork @var{mode}
3688 Tells gdb whether to detach one of the processes after a fork, or
3689 retain debugger control over them both.
3693 The child process (or parent process, depending on the value of
3694 @code{follow-fork-mode}) will be detached and allowed to run
3695 independently. This is the default.
3698 Both processes will be held under the control of @value{GDBN}.
3699 One process (child or parent, depending on the value of
3700 @code{follow-fork-mode}) is debugged as usual, while the other
3705 @kindex show detach-on-fork
3706 @item show detach-on-fork
3707 Show whether detach-on-fork mode is on/off.
3710 If you choose to set @samp{detach-on-fork} mode off, then @value{GDBN}
3711 will retain control of all forked processes (including nested forks).
3712 You can list the forked processes under the control of @value{GDBN} by
3713 using the @w{@code{info inferiors}} command, and switch from one fork
3714 to another by using the @code{inferior} command (@pxref{Inferiors and
3715 Programs, ,Debugging Multiple Inferiors and Programs}).
3717 To quit debugging one of the forked processes, you can either detach
3718 from it by using the @w{@code{detach inferiors}} command (allowing it
3719 to run independently), or kill it using the @w{@code{kill inferiors}}
3720 command. @xref{Inferiors and Programs, ,Debugging Multiple Inferiors
3723 If you ask to debug a child process and a @code{vfork} is followed by an
3724 @code{exec}, @value{GDBN} executes the new target up to the first
3725 breakpoint in the new target. If you have a breakpoint set on
3726 @code{main} in your original program, the breakpoint will also be set on
3727 the child process's @code{main}.
3729 On some systems, when a child process is spawned by @code{vfork}, you
3730 cannot debug the child or parent until an @code{exec} call completes.
3732 If you issue a @code{run} command to @value{GDBN} after an @code{exec}
3733 call executes, the new target restarts. To restart the parent
3734 process, use the @code{file} command with the parent executable name
3735 as its argument. By default, after an @code{exec} call executes,
3736 @value{GDBN} discards the symbols of the previous executable image.
3737 You can change this behaviour with the @w{@code{set follow-exec-mode}}
3741 @kindex set follow-exec-mode
3742 @item set follow-exec-mode @var{mode}
3744 Set debugger response to a program call of @code{exec}. An
3745 @code{exec} call replaces the program image of a process.
3747 @code{follow-exec-mode} can be:
3751 @value{GDBN} creates a new inferior and rebinds the process to this
3752 new inferior. The program the process was running before the
3753 @code{exec} call can be restarted afterwards by restarting the
3759 (@value{GDBP}) info inferiors
3761 Id Description Executable
3764 process 12020 is executing new program: prog2
3765 Program exited normally.
3766 (@value{GDBP}) info inferiors
3767 Id Description Executable
3773 @value{GDBN} keeps the process bound to the same inferior. The new
3774 executable image replaces the previous executable loaded in the
3775 inferior. Restarting the inferior after the @code{exec} call, with
3776 e.g., the @code{run} command, restarts the executable the process was
3777 running after the @code{exec} call. This is the default mode.
3782 (@value{GDBP}) info inferiors
3783 Id Description Executable
3786 process 12020 is executing new program: prog2
3787 Program exited normally.
3788 (@value{GDBP}) info inferiors
3789 Id Description Executable
3796 @code{follow-exec-mode} is supported in native mode and
3797 @code{target extended-remote} mode.
3799 You can use the @code{catch} command to make @value{GDBN} stop whenever
3800 a @code{fork}, @code{vfork}, or @code{exec} call is made. @xref{Set
3801 Catchpoints, ,Setting Catchpoints}.
3803 @node Checkpoint/Restart
3804 @section Setting a @emph{Bookmark} to Return to Later
3809 @cindex snapshot of a process
3810 @cindex rewind program state
3812 On certain operating systems@footnote{Currently, only
3813 @sc{gnu}/Linux.}, @value{GDBN} is able to save a @dfn{snapshot} of a
3814 program's state, called a @dfn{checkpoint}, and come back to it
3817 Returning to a checkpoint effectively undoes everything that has
3818 happened in the program since the @code{checkpoint} was saved. This
3819 includes changes in memory, registers, and even (within some limits)
3820 system state. Effectively, it is like going back in time to the
3821 moment when the checkpoint was saved.
3823 Thus, if you're stepping thru a program and you think you're
3824 getting close to the point where things go wrong, you can save
3825 a checkpoint. Then, if you accidentally go too far and miss
3826 the critical statement, instead of having to restart your program
3827 from the beginning, you can just go back to the checkpoint and
3828 start again from there.
3830 This can be especially useful if it takes a lot of time or
3831 steps to reach the point where you think the bug occurs.
3833 To use the @code{checkpoint}/@code{restart} method of debugging:
3838 Save a snapshot of the debugged program's current execution state.
3839 The @code{checkpoint} command takes no arguments, but each checkpoint
3840 is assigned a small integer id, similar to a breakpoint id.
3842 @kindex info checkpoints
3843 @item info checkpoints
3844 List the checkpoints that have been saved in the current debugging
3845 session. For each checkpoint, the following information will be
3852 @item Source line, or label
3855 @kindex restart @var{checkpoint-id}
3856 @item restart @var{checkpoint-id}
3857 Restore the program state that was saved as checkpoint number
3858 @var{checkpoint-id}. All program variables, registers, stack frames
3859 etc.@: will be returned to the values that they had when the checkpoint
3860 was saved. In essence, gdb will ``wind back the clock'' to the point
3861 in time when the checkpoint was saved.
3863 Note that breakpoints, @value{GDBN} variables, command history etc.
3864 are not affected by restoring a checkpoint. In general, a checkpoint
3865 only restores things that reside in the program being debugged, not in
3868 @kindex delete checkpoint @var{checkpoint-id}
3869 @item delete checkpoint @var{checkpoint-id}
3870 Delete the previously-saved checkpoint identified by @var{checkpoint-id}.
3874 Returning to a previously saved checkpoint will restore the user state
3875 of the program being debugged, plus a significant subset of the system
3876 (OS) state, including file pointers. It won't ``un-write'' data from
3877 a file, but it will rewind the file pointer to the previous location,
3878 so that the previously written data can be overwritten. For files
3879 opened in read mode, the pointer will also be restored so that the
3880 previously read data can be read again.
3882 Of course, characters that have been sent to a printer (or other
3883 external device) cannot be ``snatched back'', and characters received
3884 from eg.@: a serial device can be removed from internal program buffers,
3885 but they cannot be ``pushed back'' into the serial pipeline, ready to
3886 be received again. Similarly, the actual contents of files that have
3887 been changed cannot be restored (at this time).
3889 However, within those constraints, you actually can ``rewind'' your
3890 program to a previously saved point in time, and begin debugging it
3891 again --- and you can change the course of events so as to debug a
3892 different execution path this time.
3894 @cindex checkpoints and process id
3895 Finally, there is one bit of internal program state that will be
3896 different when you return to a checkpoint --- the program's process
3897 id. Each checkpoint will have a unique process id (or @var{pid}),
3898 and each will be different from the program's original @var{pid}.
3899 If your program has saved a local copy of its process id, this could
3900 potentially pose a problem.
3902 @subsection A Non-obvious Benefit of Using Checkpoints
3904 On some systems such as @sc{gnu}/Linux, address space randomization
3905 is performed on new processes for security reasons. This makes it
3906 difficult or impossible to set a breakpoint, or watchpoint, on an
3907 absolute address if you have to restart the program, since the
3908 absolute location of a symbol will change from one execution to the
3911 A checkpoint, however, is an @emph{identical} copy of a process.
3912 Therefore if you create a checkpoint at (eg.@:) the start of main,
3913 and simply return to that checkpoint instead of restarting the
3914 process, you can avoid the effects of address randomization and
3915 your symbols will all stay in the same place.
3918 @chapter Stopping and Continuing
3920 The principal purposes of using a debugger are so that you can stop your
3921 program before it terminates; or so that, if your program runs into
3922 trouble, you can investigate and find out why.
3924 Inside @value{GDBN}, your program may stop for any of several reasons,
3925 such as a signal, a breakpoint, or reaching a new line after a
3926 @value{GDBN} command such as @code{step}. You may then examine and
3927 change variables, set new breakpoints or remove old ones, and then
3928 continue execution. Usually, the messages shown by @value{GDBN} provide
3929 ample explanation of the status of your program---but you can also
3930 explicitly request this information at any time.
3933 @kindex info program
3935 Display information about the status of your program: whether it is
3936 running or not, what process it is, and why it stopped.
3940 * Breakpoints:: Breakpoints, watchpoints, and catchpoints
3941 * Continuing and Stepping:: Resuming execution
3942 * Skipping Over Functions and Files::
3943 Skipping over functions and files
3945 * Thread Stops:: Stopping and starting multi-thread programs
3949 @section Breakpoints, Watchpoints, and Catchpoints
3952 A @dfn{breakpoint} makes your program stop whenever a certain point in
3953 the program is reached. For each breakpoint, you can add conditions to
3954 control in finer detail whether your program stops. You can set
3955 breakpoints with the @code{break} command and its variants (@pxref{Set
3956 Breaks, ,Setting Breakpoints}), to specify the place where your program
3957 should stop by line number, function name or exact address in the
3960 On some systems, you can set breakpoints in shared libraries before
3961 the executable is run.
3964 @cindex data breakpoints
3965 @cindex memory tracing
3966 @cindex breakpoint on memory address
3967 @cindex breakpoint on variable modification
3968 A @dfn{watchpoint} is a special breakpoint that stops your program
3969 when the value of an expression changes. The expression may be a value
3970 of a variable, or it could involve values of one or more variables
3971 combined by operators, such as @samp{a + b}. This is sometimes called
3972 @dfn{data breakpoints}. You must use a different command to set
3973 watchpoints (@pxref{Set Watchpoints, ,Setting Watchpoints}), but aside
3974 from that, you can manage a watchpoint like any other breakpoint: you
3975 enable, disable, and delete both breakpoints and watchpoints using the
3978 You can arrange to have values from your program displayed automatically
3979 whenever @value{GDBN} stops at a breakpoint. @xref{Auto Display,,
3983 @cindex breakpoint on events
3984 A @dfn{catchpoint} is another special breakpoint that stops your program
3985 when a certain kind of event occurs, such as the throwing of a C@t{++}
3986 exception or the loading of a library. As with watchpoints, you use a
3987 different command to set a catchpoint (@pxref{Set Catchpoints, ,Setting
3988 Catchpoints}), but aside from that, you can manage a catchpoint like any
3989 other breakpoint. (To stop when your program receives a signal, use the
3990 @code{handle} command; see @ref{Signals, ,Signals}.)
3992 @cindex breakpoint numbers
3993 @cindex numbers for breakpoints
3994 @value{GDBN} assigns a number to each breakpoint, watchpoint, or
3995 catchpoint when you create it; these numbers are successive integers
3996 starting with one. In many of the commands for controlling various
3997 features of breakpoints you use the breakpoint number to say which
3998 breakpoint you want to change. Each breakpoint may be @dfn{enabled} or
3999 @dfn{disabled}; if disabled, it has no effect on your program until you
4002 @cindex breakpoint ranges
4003 @cindex breakpoint lists
4004 @cindex ranges of breakpoints
4005 @cindex lists of breakpoints
4006 Some @value{GDBN} commands accept a space-separated list of breakpoints
4007 on which to operate. A list element can be either a single breakpoint number,
4008 like @samp{5}, or a range of such numbers, like @samp{5-7}.
4009 When a breakpoint list is given to a command, all breakpoints in that list
4013 * Set Breaks:: Setting breakpoints
4014 * Set Watchpoints:: Setting watchpoints
4015 * Set Catchpoints:: Setting catchpoints
4016 * Delete Breaks:: Deleting breakpoints
4017 * Disabling:: Disabling breakpoints
4018 * Conditions:: Break conditions
4019 * Break Commands:: Breakpoint command lists
4020 * Dynamic Printf:: Dynamic printf
4021 * Save Breakpoints:: How to save breakpoints in a file
4022 * Static Probe Points:: Listing static probe points
4023 * Error in Breakpoints:: ``Cannot insert breakpoints''
4024 * Breakpoint-related Warnings:: ``Breakpoint address adjusted...''
4028 @subsection Setting Breakpoints
4030 @c FIXME LMB what does GDB do if no code on line of breakpt?
4031 @c consider in particular declaration with/without initialization.
4033 @c FIXME 2 is there stuff on this already? break at fun start, already init?
4036 @kindex b @r{(@code{break})}
4037 @vindex $bpnum@r{, convenience variable}
4038 @cindex latest breakpoint
4039 Breakpoints are set with the @code{break} command (abbreviated
4040 @code{b}). The debugger convenience variable @samp{$bpnum} records the
4041 number of the breakpoint you've set most recently; see @ref{Convenience
4042 Vars,, Convenience Variables}, for a discussion of what you can do with
4043 convenience variables.
4046 @item break @var{location}
4047 Set a breakpoint at the given @var{location}, which can specify a
4048 function name, a line number, or an address of an instruction.
4049 (@xref{Specify Location}, for a list of all the possible ways to
4050 specify a @var{location}.) The breakpoint will stop your program just
4051 before it executes any of the code in the specified @var{location}.
4053 When using source languages that permit overloading of symbols, such as
4054 C@t{++}, a function name may refer to more than one possible place to break.
4055 @xref{Ambiguous Expressions,,Ambiguous Expressions}, for a discussion of
4058 It is also possible to insert a breakpoint that will stop the program
4059 only if a specific thread (@pxref{Thread-Specific Breakpoints})
4060 or a specific task (@pxref{Ada Tasks}) hits that breakpoint.
4063 When called without any arguments, @code{break} sets a breakpoint at
4064 the next instruction to be executed in the selected stack frame
4065 (@pxref{Stack, ,Examining the Stack}). In any selected frame but the
4066 innermost, this makes your program stop as soon as control
4067 returns to that frame. This is similar to the effect of a
4068 @code{finish} command in the frame inside the selected frame---except
4069 that @code{finish} does not leave an active breakpoint. If you use
4070 @code{break} without an argument in the innermost frame, @value{GDBN} stops
4071 the next time it reaches the current location; this may be useful
4074 @value{GDBN} normally ignores breakpoints when it resumes execution, until at
4075 least one instruction has been executed. If it did not do this, you
4076 would be unable to proceed past a breakpoint without first disabling the
4077 breakpoint. This rule applies whether or not the breakpoint already
4078 existed when your program stopped.
4080 @item break @dots{} if @var{cond}
4081 Set a breakpoint with condition @var{cond}; evaluate the expression
4082 @var{cond} each time the breakpoint is reached, and stop only if the
4083 value is nonzero---that is, if @var{cond} evaluates as true.
4084 @samp{@dots{}} stands for one of the possible arguments described
4085 above (or no argument) specifying where to break. @xref{Conditions,
4086 ,Break Conditions}, for more information on breakpoint conditions.
4089 @item tbreak @var{args}
4090 Set a breakpoint enabled only for one stop. The @var{args} are the
4091 same as for the @code{break} command, and the breakpoint is set in the same
4092 way, but the breakpoint is automatically deleted after the first time your
4093 program stops there. @xref{Disabling, ,Disabling Breakpoints}.
4096 @cindex hardware breakpoints
4097 @item hbreak @var{args}
4098 Set a hardware-assisted breakpoint. The @var{args} are the same as for the
4099 @code{break} command and the breakpoint is set in the same way, but the
4100 breakpoint requires hardware support and some target hardware may not
4101 have this support. The main purpose of this is EPROM/ROM code
4102 debugging, so you can set a breakpoint at an instruction without
4103 changing the instruction. This can be used with the new trap-generation
4104 provided by SPARClite DSU and most x86-based targets. These targets
4105 will generate traps when a program accesses some data or instruction
4106 address that is assigned to the debug registers. However the hardware
4107 breakpoint registers can take a limited number of breakpoints. For
4108 example, on the DSU, only two data breakpoints can be set at a time, and
4109 @value{GDBN} will reject this command if more than two are used. Delete
4110 or disable unused hardware breakpoints before setting new ones
4111 (@pxref{Disabling, ,Disabling Breakpoints}).
4112 @xref{Conditions, ,Break Conditions}.
4113 For remote targets, you can restrict the number of hardware
4114 breakpoints @value{GDBN} will use, see @ref{set remote
4115 hardware-breakpoint-limit}.
4118 @item thbreak @var{args}
4119 Set a hardware-assisted breakpoint enabled only for one stop. The @var{args}
4120 are the same as for the @code{hbreak} command and the breakpoint is set in
4121 the same way. However, like the @code{tbreak} command,
4122 the breakpoint is automatically deleted after the
4123 first time your program stops there. Also, like the @code{hbreak}
4124 command, the breakpoint requires hardware support and some target hardware
4125 may not have this support. @xref{Disabling, ,Disabling Breakpoints}.
4126 See also @ref{Conditions, ,Break Conditions}.
4129 @cindex regular expression
4130 @cindex breakpoints at functions matching a regexp
4131 @cindex set breakpoints in many functions
4132 @item rbreak @var{regex}
4133 Set breakpoints on all functions matching the regular expression
4134 @var{regex}. This command sets an unconditional breakpoint on all
4135 matches, printing a list of all breakpoints it set. Once these
4136 breakpoints are set, they are treated just like the breakpoints set with
4137 the @code{break} command. You can delete them, disable them, or make
4138 them conditional the same way as any other breakpoint.
4140 In programs using different languages, @value{GDBN} chooses the syntax
4141 to print the list of all breakpoints it sets according to the
4142 @samp{set language} value: using @samp{set language auto}
4143 (see @ref{Automatically, ,Set Language Automatically}) means to use the
4144 language of the breakpoint's function, other values mean to use
4145 the manually specified language (see @ref{Manually, ,Set Language Manually}).
4147 The syntax of the regular expression is the standard one used with tools
4148 like @file{grep}. Note that this is different from the syntax used by
4149 shells, so for instance @code{foo*} matches all functions that include
4150 an @code{fo} followed by zero or more @code{o}s. There is an implicit
4151 @code{.*} leading and trailing the regular expression you supply, so to
4152 match only functions that begin with @code{foo}, use @code{^foo}.
4154 @cindex non-member C@t{++} functions, set breakpoint in
4155 When debugging C@t{++} programs, @code{rbreak} is useful for setting
4156 breakpoints on overloaded functions that are not members of any special
4159 @cindex set breakpoints on all functions
4160 The @code{rbreak} command can be used to set breakpoints in
4161 @strong{all} the functions in a program, like this:
4164 (@value{GDBP}) rbreak .
4167 @item rbreak @var{file}:@var{regex}
4168 If @code{rbreak} is called with a filename qualification, it limits
4169 the search for functions matching the given regular expression to the
4170 specified @var{file}. This can be used, for example, to set breakpoints on
4171 every function in a given file:
4174 (@value{GDBP}) rbreak file.c:.
4177 The colon separating the filename qualifier from the regex may
4178 optionally be surrounded by spaces.
4180 @kindex info breakpoints
4181 @cindex @code{$_} and @code{info breakpoints}
4182 @item info breakpoints @r{[}@var{list}@dots{}@r{]}
4183 @itemx info break @r{[}@var{list}@dots{}@r{]}
4184 Print a table of all breakpoints, watchpoints, and catchpoints set and
4185 not deleted. Optional argument @var{n} means print information only
4186 about the specified breakpoint(s) (or watchpoint(s) or catchpoint(s)).
4187 For each breakpoint, following columns are printed:
4190 @item Breakpoint Numbers
4192 Breakpoint, watchpoint, or catchpoint.
4194 Whether the breakpoint is marked to be disabled or deleted when hit.
4195 @item Enabled or Disabled
4196 Enabled breakpoints are marked with @samp{y}. @samp{n} marks breakpoints
4197 that are not enabled.
4199 Where the breakpoint is in your program, as a memory address. For a
4200 pending breakpoint whose address is not yet known, this field will
4201 contain @samp{<PENDING>}. Such breakpoint won't fire until a shared
4202 library that has the symbol or line referred by breakpoint is loaded.
4203 See below for details. A breakpoint with several locations will
4204 have @samp{<MULTIPLE>} in this field---see below for details.
4206 Where the breakpoint is in the source for your program, as a file and
4207 line number. For a pending breakpoint, the original string passed to
4208 the breakpoint command will be listed as it cannot be resolved until
4209 the appropriate shared library is loaded in the future.
4213 If a breakpoint is conditional, there are two evaluation modes: ``host'' and
4214 ``target''. If mode is ``host'', breakpoint condition evaluation is done by
4215 @value{GDBN} on the host's side. If it is ``target'', then the condition
4216 is evaluated by the target. The @code{info break} command shows
4217 the condition on the line following the affected breakpoint, together with
4218 its condition evaluation mode in between parentheses.
4220 Breakpoint commands, if any, are listed after that. A pending breakpoint is
4221 allowed to have a condition specified for it. The condition is not parsed for
4222 validity until a shared library is loaded that allows the pending
4223 breakpoint to resolve to a valid location.
4226 @code{info break} with a breakpoint
4227 number @var{n} as argument lists only that breakpoint. The
4228 convenience variable @code{$_} and the default examining-address for
4229 the @code{x} command are set to the address of the last breakpoint
4230 listed (@pxref{Memory, ,Examining Memory}).
4233 @code{info break} displays a count of the number of times the breakpoint
4234 has been hit. This is especially useful in conjunction with the
4235 @code{ignore} command. You can ignore a large number of breakpoint
4236 hits, look at the breakpoint info to see how many times the breakpoint
4237 was hit, and then run again, ignoring one less than that number. This
4238 will get you quickly to the last hit of that breakpoint.
4241 For a breakpoints with an enable count (xref) greater than 1,
4242 @code{info break} also displays that count.
4246 @value{GDBN} allows you to set any number of breakpoints at the same place in
4247 your program. There is nothing silly or meaningless about this. When
4248 the breakpoints are conditional, this is even useful
4249 (@pxref{Conditions, ,Break Conditions}).
4251 @cindex multiple locations, breakpoints
4252 @cindex breakpoints, multiple locations
4253 It is possible that a breakpoint corresponds to several locations
4254 in your program. Examples of this situation are:
4258 Multiple functions in the program may have the same name.
4261 For a C@t{++} constructor, the @value{NGCC} compiler generates several
4262 instances of the function body, used in different cases.
4265 For a C@t{++} template function, a given line in the function can
4266 correspond to any number of instantiations.
4269 For an inlined function, a given source line can correspond to
4270 several places where that function is inlined.
4273 In all those cases, @value{GDBN} will insert a breakpoint at all
4274 the relevant locations.
4276 A breakpoint with multiple locations is displayed in the breakpoint
4277 table using several rows---one header row, followed by one row for
4278 each breakpoint location. The header row has @samp{<MULTIPLE>} in the
4279 address column. The rows for individual locations contain the actual
4280 addresses for locations, and show the functions to which those
4281 locations belong. The number column for a location is of the form
4282 @var{breakpoint-number}.@var{location-number}.
4287 Num Type Disp Enb Address What
4288 1 breakpoint keep y <MULTIPLE>
4290 breakpoint already hit 1 time
4291 1.1 y 0x080486a2 in void foo<int>() at t.cc:8
4292 1.2 y 0x080486ca in void foo<double>() at t.cc:8
4295 You cannot delete the individual locations from a breakpoint. However,
4296 each location can be individually enabled or disabled by passing
4297 @var{breakpoint-number}.@var{location-number} as argument to the
4298 @code{enable} and @code{disable} commands. It's also possible to
4299 @code{enable} and @code{disable} a range of @var{location-number}
4300 locations using a @var{breakpoint-number} and two @var{location-number}s,
4301 in increasing order, separated by a hyphen, like
4302 @kbd{@var{breakpoint-number}.@var{location-number1}-@var{location-number2}},
4303 in which case @value{GDBN} acts on all the locations in the range (inclusive).
4304 Disabling or enabling the parent breakpoint (@pxref{Disabling}) affects
4305 all of the locations that belong to that breakpoint.
4307 @cindex pending breakpoints
4308 It's quite common to have a breakpoint inside a shared library.
4309 Shared libraries can be loaded and unloaded explicitly,
4310 and possibly repeatedly, as the program is executed. To support
4311 this use case, @value{GDBN} updates breakpoint locations whenever
4312 any shared library is loaded or unloaded. Typically, you would
4313 set a breakpoint in a shared library at the beginning of your
4314 debugging session, when the library is not loaded, and when the
4315 symbols from the library are not available. When you try to set
4316 breakpoint, @value{GDBN} will ask you if you want to set
4317 a so called @dfn{pending breakpoint}---breakpoint whose address
4318 is not yet resolved.
4320 After the program is run, whenever a new shared library is loaded,
4321 @value{GDBN} reevaluates all the breakpoints. When a newly loaded
4322 shared library contains the symbol or line referred to by some
4323 pending breakpoint, that breakpoint is resolved and becomes an
4324 ordinary breakpoint. When a library is unloaded, all breakpoints
4325 that refer to its symbols or source lines become pending again.
4327 This logic works for breakpoints with multiple locations, too. For
4328 example, if you have a breakpoint in a C@t{++} template function, and
4329 a newly loaded shared library has an instantiation of that template,
4330 a new location is added to the list of locations for the breakpoint.
4332 Except for having unresolved address, pending breakpoints do not
4333 differ from regular breakpoints. You can set conditions or commands,
4334 enable and disable them and perform other breakpoint operations.
4336 @value{GDBN} provides some additional commands for controlling what
4337 happens when the @samp{break} command cannot resolve breakpoint
4338 address specification to an address:
4340 @kindex set breakpoint pending
4341 @kindex show breakpoint pending
4343 @item set breakpoint pending auto
4344 This is the default behavior. When @value{GDBN} cannot find the breakpoint
4345 location, it queries you whether a pending breakpoint should be created.
4347 @item set breakpoint pending on
4348 This indicates that an unrecognized breakpoint location should automatically
4349 result in a pending breakpoint being created.
4351 @item set breakpoint pending off
4352 This indicates that pending breakpoints are not to be created. Any
4353 unrecognized breakpoint location results in an error. This setting does
4354 not affect any pending breakpoints previously created.
4356 @item show breakpoint pending
4357 Show the current behavior setting for creating pending breakpoints.
4360 The settings above only affect the @code{break} command and its
4361 variants. Once breakpoint is set, it will be automatically updated
4362 as shared libraries are loaded and unloaded.
4364 @cindex automatic hardware breakpoints
4365 For some targets, @value{GDBN} can automatically decide if hardware or
4366 software breakpoints should be used, depending on whether the
4367 breakpoint address is read-only or read-write. This applies to
4368 breakpoints set with the @code{break} command as well as to internal
4369 breakpoints set by commands like @code{next} and @code{finish}. For
4370 breakpoints set with @code{hbreak}, @value{GDBN} will always use hardware
4373 You can control this automatic behaviour with the following commands:
4375 @kindex set breakpoint auto-hw
4376 @kindex show breakpoint auto-hw
4378 @item set breakpoint auto-hw on
4379 This is the default behavior. When @value{GDBN} sets a breakpoint, it
4380 will try to use the target memory map to decide if software or hardware
4381 breakpoint must be used.
4383 @item set breakpoint auto-hw off
4384 This indicates @value{GDBN} should not automatically select breakpoint
4385 type. If the target provides a memory map, @value{GDBN} will warn when
4386 trying to set software breakpoint at a read-only address.
4389 @value{GDBN} normally implements breakpoints by replacing the program code
4390 at the breakpoint address with a special instruction, which, when
4391 executed, given control to the debugger. By default, the program
4392 code is so modified only when the program is resumed. As soon as
4393 the program stops, @value{GDBN} restores the original instructions. This
4394 behaviour guards against leaving breakpoints inserted in the
4395 target should gdb abrubptly disconnect. However, with slow remote
4396 targets, inserting and removing breakpoint can reduce the performance.
4397 This behavior can be controlled with the following commands::
4399 @kindex set breakpoint always-inserted
4400 @kindex show breakpoint always-inserted
4402 @item set breakpoint always-inserted off
4403 All breakpoints, including newly added by the user, are inserted in
4404 the target only when the target is resumed. All breakpoints are
4405 removed from the target when it stops. This is the default mode.
4407 @item set breakpoint always-inserted on
4408 Causes all breakpoints to be inserted in the target at all times. If
4409 the user adds a new breakpoint, or changes an existing breakpoint, the
4410 breakpoints in the target are updated immediately. A breakpoint is
4411 removed from the target only when breakpoint itself is deleted.
4414 @value{GDBN} handles conditional breakpoints by evaluating these conditions
4415 when a breakpoint breaks. If the condition is true, then the process being
4416 debugged stops, otherwise the process is resumed.
4418 If the target supports evaluating conditions on its end, @value{GDBN} may
4419 download the breakpoint, together with its conditions, to it.
4421 This feature can be controlled via the following commands:
4423 @kindex set breakpoint condition-evaluation
4424 @kindex show breakpoint condition-evaluation
4426 @item set breakpoint condition-evaluation host
4427 This option commands @value{GDBN} to evaluate the breakpoint
4428 conditions on the host's side. Unconditional breakpoints are sent to
4429 the target which in turn receives the triggers and reports them back to GDB
4430 for condition evaluation. This is the standard evaluation mode.
4432 @item set breakpoint condition-evaluation target
4433 This option commands @value{GDBN} to download breakpoint conditions
4434 to the target at the moment of their insertion. The target
4435 is responsible for evaluating the conditional expression and reporting
4436 breakpoint stop events back to @value{GDBN} whenever the condition
4437 is true. Due to limitations of target-side evaluation, some conditions
4438 cannot be evaluated there, e.g., conditions that depend on local data
4439 that is only known to the host. Examples include
4440 conditional expressions involving convenience variables, complex types
4441 that cannot be handled by the agent expression parser and expressions
4442 that are too long to be sent over to the target, specially when the
4443 target is a remote system. In these cases, the conditions will be
4444 evaluated by @value{GDBN}.
4446 @item set breakpoint condition-evaluation auto
4447 This is the default mode. If the target supports evaluating breakpoint
4448 conditions on its end, @value{GDBN} will download breakpoint conditions to
4449 the target (limitations mentioned previously apply). If the target does
4450 not support breakpoint condition evaluation, then @value{GDBN} will fallback
4451 to evaluating all these conditions on the host's side.
4455 @cindex negative breakpoint numbers
4456 @cindex internal @value{GDBN} breakpoints
4457 @value{GDBN} itself sometimes sets breakpoints in your program for
4458 special purposes, such as proper handling of @code{longjmp} (in C
4459 programs). These internal breakpoints are assigned negative numbers,
4460 starting with @code{-1}; @samp{info breakpoints} does not display them.
4461 You can see these breakpoints with the @value{GDBN} maintenance command
4462 @samp{maint info breakpoints} (@pxref{maint info breakpoints}).
4465 @node Set Watchpoints
4466 @subsection Setting Watchpoints
4468 @cindex setting watchpoints
4469 You can use a watchpoint to stop execution whenever the value of an
4470 expression changes, without having to predict a particular place where
4471 this may happen. (This is sometimes called a @dfn{data breakpoint}.)
4472 The expression may be as simple as the value of a single variable, or
4473 as complex as many variables combined by operators. Examples include:
4477 A reference to the value of a single variable.
4480 An address cast to an appropriate data type. For example,
4481 @samp{*(int *)0x12345678} will watch a 4-byte region at the specified
4482 address (assuming an @code{int} occupies 4 bytes).
4485 An arbitrarily complex expression, such as @samp{a*b + c/d}. The
4486 expression can use any operators valid in the program's native
4487 language (@pxref{Languages}).
4490 You can set a watchpoint on an expression even if the expression can
4491 not be evaluated yet. For instance, you can set a watchpoint on
4492 @samp{*global_ptr} before @samp{global_ptr} is initialized.
4493 @value{GDBN} will stop when your program sets @samp{global_ptr} and
4494 the expression produces a valid value. If the expression becomes
4495 valid in some other way than changing a variable (e.g.@: if the memory
4496 pointed to by @samp{*global_ptr} becomes readable as the result of a
4497 @code{malloc} call), @value{GDBN} may not stop until the next time
4498 the expression changes.
4500 @cindex software watchpoints
4501 @cindex hardware watchpoints
4502 Depending on your system, watchpoints may be implemented in software or
4503 hardware. @value{GDBN} does software watchpointing by single-stepping your
4504 program and testing the variable's value each time, which is hundreds of
4505 times slower than normal execution. (But this may still be worth it, to
4506 catch errors where you have no clue what part of your program is the
4509 On some systems, such as most PowerPC or x86-based targets,
4510 @value{GDBN} includes support for hardware watchpoints, which do not
4511 slow down the running of your program.
4515 @item watch @r{[}-l@r{|}-location@r{]} @var{expr} @r{[}thread @var{thread-id}@r{]} @r{[}mask @var{maskvalue}@r{]}
4516 Set a watchpoint for an expression. @value{GDBN} will break when the
4517 expression @var{expr} is written into by the program and its value
4518 changes. The simplest (and the most popular) use of this command is
4519 to watch the value of a single variable:
4522 (@value{GDBP}) watch foo
4525 If the command includes a @code{@r{[}thread @var{thread-id}@r{]}}
4526 argument, @value{GDBN} breaks only when the thread identified by
4527 @var{thread-id} changes the value of @var{expr}. If any other threads
4528 change the value of @var{expr}, @value{GDBN} will not break. Note
4529 that watchpoints restricted to a single thread in this way only work
4530 with Hardware Watchpoints.
4532 Ordinarily a watchpoint respects the scope of variables in @var{expr}
4533 (see below). The @code{-location} argument tells @value{GDBN} to
4534 instead watch the memory referred to by @var{expr}. In this case,
4535 @value{GDBN} will evaluate @var{expr}, take the address of the result,
4536 and watch the memory at that address. The type of the result is used
4537 to determine the size of the watched memory. If the expression's
4538 result does not have an address, then @value{GDBN} will print an
4541 The @code{@r{[}mask @var{maskvalue}@r{]}} argument allows creation
4542 of masked watchpoints, if the current architecture supports this
4543 feature (e.g., PowerPC Embedded architecture, see @ref{PowerPC
4544 Embedded}.) A @dfn{masked watchpoint} specifies a mask in addition
4545 to an address to watch. The mask specifies that some bits of an address
4546 (the bits which are reset in the mask) should be ignored when matching
4547 the address accessed by the inferior against the watchpoint address.
4548 Thus, a masked watchpoint watches many addresses simultaneously---those
4549 addresses whose unmasked bits are identical to the unmasked bits in the
4550 watchpoint address. The @code{mask} argument implies @code{-location}.
4554 (@value{GDBP}) watch foo mask 0xffff00ff
4555 (@value{GDBP}) watch *0xdeadbeef mask 0xffffff00
4559 @item rwatch @r{[}-l@r{|}-location@r{]} @var{expr} @r{[}thread @var{thread-id}@r{]} @r{[}mask @var{maskvalue}@r{]}
4560 Set a watchpoint that will break when the value of @var{expr} is read
4564 @item awatch @r{[}-l@r{|}-location@r{]} @var{expr} @r{[}thread @var{thread-id}@r{]} @r{[}mask @var{maskvalue}@r{]}
4565 Set a watchpoint that will break when @var{expr} is either read from
4566 or written into by the program.
4568 @kindex info watchpoints @r{[}@var{list}@dots{}@r{]}
4569 @item info watchpoints @r{[}@var{list}@dots{}@r{]}
4570 This command prints a list of watchpoints, using the same format as
4571 @code{info break} (@pxref{Set Breaks}).
4574 If you watch for a change in a numerically entered address you need to
4575 dereference it, as the address itself is just a constant number which will
4576 never change. @value{GDBN} refuses to create a watchpoint that watches
4577 a never-changing value:
4580 (@value{GDBP}) watch 0x600850
4581 Cannot watch constant value 0x600850.
4582 (@value{GDBP}) watch *(int *) 0x600850
4583 Watchpoint 1: *(int *) 6293584
4586 @value{GDBN} sets a @dfn{hardware watchpoint} if possible. Hardware
4587 watchpoints execute very quickly, and the debugger reports a change in
4588 value at the exact instruction where the change occurs. If @value{GDBN}
4589 cannot set a hardware watchpoint, it sets a software watchpoint, which
4590 executes more slowly and reports the change in value at the next
4591 @emph{statement}, not the instruction, after the change occurs.
4593 @cindex use only software watchpoints
4594 You can force @value{GDBN} to use only software watchpoints with the
4595 @kbd{set can-use-hw-watchpoints 0} command. With this variable set to
4596 zero, @value{GDBN} will never try to use hardware watchpoints, even if
4597 the underlying system supports them. (Note that hardware-assisted
4598 watchpoints that were set @emph{before} setting
4599 @code{can-use-hw-watchpoints} to zero will still use the hardware
4600 mechanism of watching expression values.)
4603 @item set can-use-hw-watchpoints
4604 @kindex set can-use-hw-watchpoints
4605 Set whether or not to use hardware watchpoints.
4607 @item show can-use-hw-watchpoints
4608 @kindex show can-use-hw-watchpoints
4609 Show the current mode of using hardware watchpoints.
4612 For remote targets, you can restrict the number of hardware
4613 watchpoints @value{GDBN} will use, see @ref{set remote
4614 hardware-breakpoint-limit}.
4616 When you issue the @code{watch} command, @value{GDBN} reports
4619 Hardware watchpoint @var{num}: @var{expr}
4623 if it was able to set a hardware watchpoint.
4625 Currently, the @code{awatch} and @code{rwatch} commands can only set
4626 hardware watchpoints, because accesses to data that don't change the
4627 value of the watched expression cannot be detected without examining
4628 every instruction as it is being executed, and @value{GDBN} does not do
4629 that currently. If @value{GDBN} finds that it is unable to set a
4630 hardware breakpoint with the @code{awatch} or @code{rwatch} command, it
4631 will print a message like this:
4634 Expression cannot be implemented with read/access watchpoint.
4637 Sometimes, @value{GDBN} cannot set a hardware watchpoint because the
4638 data type of the watched expression is wider than what a hardware
4639 watchpoint on the target machine can handle. For example, some systems
4640 can only watch regions that are up to 4 bytes wide; on such systems you
4641 cannot set hardware watchpoints for an expression that yields a
4642 double-precision floating-point number (which is typically 8 bytes
4643 wide). As a work-around, it might be possible to break the large region
4644 into a series of smaller ones and watch them with separate watchpoints.
4646 If you set too many hardware watchpoints, @value{GDBN} might be unable
4647 to insert all of them when you resume the execution of your program.
4648 Since the precise number of active watchpoints is unknown until such
4649 time as the program is about to be resumed, @value{GDBN} might not be
4650 able to warn you about this when you set the watchpoints, and the
4651 warning will be printed only when the program is resumed:
4654 Hardware watchpoint @var{num}: Could not insert watchpoint
4658 If this happens, delete or disable some of the watchpoints.
4660 Watching complex expressions that reference many variables can also
4661 exhaust the resources available for hardware-assisted watchpoints.
4662 That's because @value{GDBN} needs to watch every variable in the
4663 expression with separately allocated resources.
4665 If you call a function interactively using @code{print} or @code{call},
4666 any watchpoints you have set will be inactive until @value{GDBN} reaches another
4667 kind of breakpoint or the call completes.
4669 @value{GDBN} automatically deletes watchpoints that watch local
4670 (automatic) variables, or expressions that involve such variables, when
4671 they go out of scope, that is, when the execution leaves the block in
4672 which these variables were defined. In particular, when the program
4673 being debugged terminates, @emph{all} local variables go out of scope,
4674 and so only watchpoints that watch global variables remain set. If you
4675 rerun the program, you will need to set all such watchpoints again. One
4676 way of doing that would be to set a code breakpoint at the entry to the
4677 @code{main} function and when it breaks, set all the watchpoints.
4679 @cindex watchpoints and threads
4680 @cindex threads and watchpoints
4681 In multi-threaded programs, watchpoints will detect changes to the
4682 watched expression from every thread.
4685 @emph{Warning:} In multi-threaded programs, software watchpoints
4686 have only limited usefulness. If @value{GDBN} creates a software
4687 watchpoint, it can only watch the value of an expression @emph{in a
4688 single thread}. If you are confident that the expression can only
4689 change due to the current thread's activity (and if you are also
4690 confident that no other thread can become current), then you can use
4691 software watchpoints as usual. However, @value{GDBN} may not notice
4692 when a non-current thread's activity changes the expression. (Hardware
4693 watchpoints, in contrast, watch an expression in all threads.)
4696 @xref{set remote hardware-watchpoint-limit}.
4698 @node Set Catchpoints
4699 @subsection Setting Catchpoints
4700 @cindex catchpoints, setting
4701 @cindex exception handlers
4702 @cindex event handling
4704 You can use @dfn{catchpoints} to cause the debugger to stop for certain
4705 kinds of program events, such as C@t{++} exceptions or the loading of a
4706 shared library. Use the @code{catch} command to set a catchpoint.
4710 @item catch @var{event}
4711 Stop when @var{event} occurs. The @var{event} can be any of the following:
4714 @item throw @r{[}@var{regexp}@r{]}
4715 @itemx rethrow @r{[}@var{regexp}@r{]}
4716 @itemx catch @r{[}@var{regexp}@r{]}
4718 @kindex catch rethrow
4720 @cindex stop on C@t{++} exceptions
4721 The throwing, re-throwing, or catching of a C@t{++} exception.
4723 If @var{regexp} is given, then only exceptions whose type matches the
4724 regular expression will be caught.
4726 @vindex $_exception@r{, convenience variable}
4727 The convenience variable @code{$_exception} is available at an
4728 exception-related catchpoint, on some systems. This holds the
4729 exception being thrown.
4731 There are currently some limitations to C@t{++} exception handling in
4736 The support for these commands is system-dependent. Currently, only
4737 systems using the @samp{gnu-v3} C@t{++} ABI (@pxref{ABI}) are
4741 The regular expression feature and the @code{$_exception} convenience
4742 variable rely on the presence of some SDT probes in @code{libstdc++}.
4743 If these probes are not present, then these features cannot be used.
4744 These probes were first available in the GCC 4.8 release, but whether
4745 or not they are available in your GCC also depends on how it was
4749 The @code{$_exception} convenience variable is only valid at the
4750 instruction at which an exception-related catchpoint is set.
4753 When an exception-related catchpoint is hit, @value{GDBN} stops at a
4754 location in the system library which implements runtime exception
4755 support for C@t{++}, usually @code{libstdc++}. You can use @code{up}
4756 (@pxref{Selection}) to get to your code.
4759 If you call a function interactively, @value{GDBN} normally returns
4760 control to you when the function has finished executing. If the call
4761 raises an exception, however, the call may bypass the mechanism that
4762 returns control to you and cause your program either to abort or to
4763 simply continue running until it hits a breakpoint, catches a signal
4764 that @value{GDBN} is listening for, or exits. This is the case even if
4765 you set a catchpoint for the exception; catchpoints on exceptions are
4766 disabled within interactive calls. @xref{Calling}, for information on
4767 controlling this with @code{set unwind-on-terminating-exception}.
4770 You cannot raise an exception interactively.
4773 You cannot install an exception handler interactively.
4776 @item exception @r{[}@var{name}@r{]}
4777 @kindex catch exception
4778 @cindex Ada exception catching
4779 @cindex catch Ada exceptions
4780 An Ada exception being raised. If an exception name is specified
4781 at the end of the command (eg @code{catch exception Program_Error}),
4782 the debugger will stop only when this specific exception is raised.
4783 Otherwise, the debugger stops execution when any Ada exception is raised.
4785 When inserting an exception catchpoint on a user-defined exception whose
4786 name is identical to one of the exceptions defined by the language, the
4787 fully qualified name must be used as the exception name. Otherwise,
4788 @value{GDBN} will assume that it should stop on the pre-defined exception
4789 rather than the user-defined one. For instance, assuming an exception
4790 called @code{Constraint_Error} is defined in package @code{Pck}, then
4791 the command to use to catch such exceptions is @kbd{catch exception
4792 Pck.Constraint_Error}.
4794 @item exception unhandled
4795 @kindex catch exception unhandled
4796 An exception that was raised but is not handled by the program.
4798 @item handlers @r{[}@var{name}@r{]}
4799 @kindex catch handlers
4800 @cindex Ada exception handlers catching
4801 @cindex catch Ada exceptions when handled
4802 An Ada exception being handled. If an exception name is
4803 specified at the end of the command
4804 (eg @kbd{catch handlers Program_Error}), the debugger will stop
4805 only when this specific exception is handled.
4806 Otherwise, the debugger stops execution when any Ada exception is handled.
4808 When inserting a handlers catchpoint on a user-defined
4809 exception whose name is identical to one of the exceptions
4810 defined by the language, the fully qualified name must be used
4811 as the exception name. Otherwise, @value{GDBN} will assume that it
4812 should stop on the pre-defined exception rather than the
4813 user-defined one. For instance, assuming an exception called
4814 @code{Constraint_Error} is defined in package @code{Pck}, then the
4815 command to use to catch such exceptions handling is
4816 @kbd{catch handlers Pck.Constraint_Error}.
4819 @kindex catch assert
4820 A failed Ada assertion.
4824 @cindex break on fork/exec
4825 A call to @code{exec}.
4827 @anchor{catch syscall}
4829 @itemx syscall @r{[}@var{name} @r{|} @var{number} @r{|} @r{group:}@var{groupname} @r{|} @r{g:}@var{groupname}@r{]} @dots{}
4830 @kindex catch syscall
4831 @cindex break on a system call.
4832 A call to or return from a system call, a.k.a.@: @dfn{syscall}. A
4833 syscall is a mechanism for application programs to request a service
4834 from the operating system (OS) or one of the OS system services.
4835 @value{GDBN} can catch some or all of the syscalls issued by the
4836 debuggee, and show the related information for each syscall. If no
4837 argument is specified, calls to and returns from all system calls
4840 @var{name} can be any system call name that is valid for the
4841 underlying OS. Just what syscalls are valid depends on the OS. On
4842 GNU and Unix systems, you can find the full list of valid syscall
4843 names on @file{/usr/include/asm/unistd.h}.
4845 @c For MS-Windows, the syscall names and the corresponding numbers
4846 @c can be found, e.g., on this URL:
4847 @c http://www.metasploit.com/users/opcode/syscalls.html
4848 @c but we don't support Windows syscalls yet.
4850 Normally, @value{GDBN} knows in advance which syscalls are valid for
4851 each OS, so you can use the @value{GDBN} command-line completion
4852 facilities (@pxref{Completion,, command completion}) to list the
4855 You may also specify the system call numerically. A syscall's
4856 number is the value passed to the OS's syscall dispatcher to
4857 identify the requested service. When you specify the syscall by its
4858 name, @value{GDBN} uses its database of syscalls to convert the name
4859 into the corresponding numeric code, but using the number directly
4860 may be useful if @value{GDBN}'s database does not have the complete
4861 list of syscalls on your system (e.g., because @value{GDBN} lags
4862 behind the OS upgrades).
4864 You may specify a group of related syscalls to be caught at once using
4865 the @code{group:} syntax (@code{g:} is a shorter equivalent). For
4866 instance, on some platforms @value{GDBN} allows you to catch all
4867 network related syscalls, by passing the argument @code{group:network}
4868 to @code{catch syscall}. Note that not all syscall groups are
4869 available in every system. You can use the command completion
4870 facilities (@pxref{Completion,, command completion}) to list the
4871 syscall groups available on your environment.
4873 The example below illustrates how this command works if you don't provide
4877 (@value{GDBP}) catch syscall
4878 Catchpoint 1 (syscall)
4880 Starting program: /tmp/catch-syscall
4882 Catchpoint 1 (call to syscall 'close'), \
4883 0xffffe424 in __kernel_vsyscall ()
4887 Catchpoint 1 (returned from syscall 'close'), \
4888 0xffffe424 in __kernel_vsyscall ()
4892 Here is an example of catching a system call by name:
4895 (@value{GDBP}) catch syscall chroot
4896 Catchpoint 1 (syscall 'chroot' [61])
4898 Starting program: /tmp/catch-syscall
4900 Catchpoint 1 (call to syscall 'chroot'), \
4901 0xffffe424 in __kernel_vsyscall ()
4905 Catchpoint 1 (returned from syscall 'chroot'), \
4906 0xffffe424 in __kernel_vsyscall ()
4910 An example of specifying a system call numerically. In the case
4911 below, the syscall number has a corresponding entry in the XML
4912 file, so @value{GDBN} finds its name and prints it:
4915 (@value{GDBP}) catch syscall 252
4916 Catchpoint 1 (syscall(s) 'exit_group')
4918 Starting program: /tmp/catch-syscall
4920 Catchpoint 1 (call to syscall 'exit_group'), \
4921 0xffffe424 in __kernel_vsyscall ()
4925 Program exited normally.
4929 Here is an example of catching a syscall group:
4932 (@value{GDBP}) catch syscall group:process
4933 Catchpoint 1 (syscalls 'exit' [1] 'fork' [2] 'waitpid' [7]
4934 'execve' [11] 'wait4' [114] 'clone' [120] 'vfork' [190]
4935 'exit_group' [252] 'waitid' [284] 'unshare' [310])
4937 Starting program: /tmp/catch-syscall
4939 Catchpoint 1 (call to syscall fork), 0x00007ffff7df4e27 in open64 ()
4940 from /lib64/ld-linux-x86-64.so.2
4946 However, there can be situations when there is no corresponding name
4947 in XML file for that syscall number. In this case, @value{GDBN} prints
4948 a warning message saying that it was not able to find the syscall name,
4949 but the catchpoint will be set anyway. See the example below:
4952 (@value{GDBP}) catch syscall 764
4953 warning: The number '764' does not represent a known syscall.
4954 Catchpoint 2 (syscall 764)
4958 If you configure @value{GDBN} using the @samp{--without-expat} option,
4959 it will not be able to display syscall names. Also, if your
4960 architecture does not have an XML file describing its system calls,
4961 you will not be able to see the syscall names. It is important to
4962 notice that these two features are used for accessing the syscall
4963 name database. In either case, you will see a warning like this:
4966 (@value{GDBP}) catch syscall
4967 warning: Could not open "syscalls/i386-linux.xml"
4968 warning: Could not load the syscall XML file 'syscalls/i386-linux.xml'.
4969 GDB will not be able to display syscall names.
4970 Catchpoint 1 (syscall)
4974 Of course, the file name will change depending on your architecture and system.
4976 Still using the example above, you can also try to catch a syscall by its
4977 number. In this case, you would see something like:
4980 (@value{GDBP}) catch syscall 252
4981 Catchpoint 1 (syscall(s) 252)
4984 Again, in this case @value{GDBN} would not be able to display syscall's names.
4988 A call to @code{fork}.
4992 A call to @code{vfork}.
4994 @item load @r{[}@var{regexp}@r{]}
4995 @itemx unload @r{[}@var{regexp}@r{]}
4997 @kindex catch unload
4998 The loading or unloading of a shared library. If @var{regexp} is
4999 given, then the catchpoint will stop only if the regular expression
5000 matches one of the affected libraries.
5002 @item signal @r{[}@var{signal}@dots{} @r{|} @samp{all}@r{]}
5003 @kindex catch signal
5004 The delivery of a signal.
5006 With no arguments, this catchpoint will catch any signal that is not
5007 used internally by @value{GDBN}, specifically, all signals except
5008 @samp{SIGTRAP} and @samp{SIGINT}.
5010 With the argument @samp{all}, all signals, including those used by
5011 @value{GDBN}, will be caught. This argument cannot be used with other
5014 Otherwise, the arguments are a list of signal names as given to
5015 @code{handle} (@pxref{Signals}). Only signals specified in this list
5018 One reason that @code{catch signal} can be more useful than
5019 @code{handle} is that you can attach commands and conditions to the
5022 When a signal is caught by a catchpoint, the signal's @code{stop} and
5023 @code{print} settings, as specified by @code{handle}, are ignored.
5024 However, whether the signal is still delivered to the inferior depends
5025 on the @code{pass} setting; this can be changed in the catchpoint's
5030 @item tcatch @var{event}
5032 Set a catchpoint that is enabled only for one stop. The catchpoint is
5033 automatically deleted after the first time the event is caught.
5037 Use the @code{info break} command to list the current catchpoints.
5041 @subsection Deleting Breakpoints
5043 @cindex clearing breakpoints, watchpoints, catchpoints
5044 @cindex deleting breakpoints, watchpoints, catchpoints
5045 It is often necessary to eliminate a breakpoint, watchpoint, or
5046 catchpoint once it has done its job and you no longer want your program
5047 to stop there. This is called @dfn{deleting} the breakpoint. A
5048 breakpoint that has been deleted no longer exists; it is forgotten.
5050 With the @code{clear} command you can delete breakpoints according to
5051 where they are in your program. With the @code{delete} command you can
5052 delete individual breakpoints, watchpoints, or catchpoints by specifying
5053 their breakpoint numbers.
5055 It is not necessary to delete a breakpoint to proceed past it. @value{GDBN}
5056 automatically ignores breakpoints on the first instruction to be executed
5057 when you continue execution without changing the execution address.
5062 Delete any breakpoints at the next instruction to be executed in the
5063 selected stack frame (@pxref{Selection, ,Selecting a Frame}). When
5064 the innermost frame is selected, this is a good way to delete a
5065 breakpoint where your program just stopped.
5067 @item clear @var{location}
5068 Delete any breakpoints set at the specified @var{location}.
5069 @xref{Specify Location}, for the various forms of @var{location}; the
5070 most useful ones are listed below:
5073 @item clear @var{function}
5074 @itemx clear @var{filename}:@var{function}
5075 Delete any breakpoints set at entry to the named @var{function}.
5077 @item clear @var{linenum}
5078 @itemx clear @var{filename}:@var{linenum}
5079 Delete any breakpoints set at or within the code of the specified
5080 @var{linenum} of the specified @var{filename}.
5083 @cindex delete breakpoints
5085 @kindex d @r{(@code{delete})}
5086 @item delete @r{[}breakpoints@r{]} @r{[}@var{list}@dots{}@r{]}
5087 Delete the breakpoints, watchpoints, or catchpoints of the breakpoint
5088 list specified as argument. If no argument is specified, delete all
5089 breakpoints (@value{GDBN} asks confirmation, unless you have @code{set
5090 confirm off}). You can abbreviate this command as @code{d}.
5094 @subsection Disabling Breakpoints
5096 @cindex enable/disable a breakpoint
5097 Rather than deleting a breakpoint, watchpoint, or catchpoint, you might
5098 prefer to @dfn{disable} it. This makes the breakpoint inoperative as if
5099 it had been deleted, but remembers the information on the breakpoint so
5100 that you can @dfn{enable} it again later.
5102 You disable and enable breakpoints, watchpoints, and catchpoints with
5103 the @code{enable} and @code{disable} commands, optionally specifying
5104 one or more breakpoint numbers as arguments. Use @code{info break} to
5105 print a list of all breakpoints, watchpoints, and catchpoints if you
5106 do not know which numbers to use.
5108 Disabling and enabling a breakpoint that has multiple locations
5109 affects all of its locations.
5111 A breakpoint, watchpoint, or catchpoint can have any of several
5112 different states of enablement:
5116 Enabled. The breakpoint stops your program. A breakpoint set
5117 with the @code{break} command starts out in this state.
5119 Disabled. The breakpoint has no effect on your program.
5121 Enabled once. The breakpoint stops your program, but then becomes
5124 Enabled for a count. The breakpoint stops your program for the next
5125 N times, then becomes disabled.
5127 Enabled for deletion. The breakpoint stops your program, but
5128 immediately after it does so it is deleted permanently. A breakpoint
5129 set with the @code{tbreak} command starts out in this state.
5132 You can use the following commands to enable or disable breakpoints,
5133 watchpoints, and catchpoints:
5137 @kindex dis @r{(@code{disable})}
5138 @item disable @r{[}breakpoints@r{]} @r{[}@var{list}@dots{}@r{]}
5139 Disable the specified breakpoints---or all breakpoints, if none are
5140 listed. A disabled breakpoint has no effect but is not forgotten. All
5141 options such as ignore-counts, conditions and commands are remembered in
5142 case the breakpoint is enabled again later. You may abbreviate
5143 @code{disable} as @code{dis}.
5146 @item enable @r{[}breakpoints@r{]} @r{[}@var{list}@dots{}@r{]}
5147 Enable the specified breakpoints (or all defined breakpoints). They
5148 become effective once again in stopping your program.
5150 @item enable @r{[}breakpoints@r{]} once @var{list}@dots{}
5151 Enable the specified breakpoints temporarily. @value{GDBN} disables any
5152 of these breakpoints immediately after stopping your program.
5154 @item enable @r{[}breakpoints@r{]} count @var{count} @var{list}@dots{}
5155 Enable the specified breakpoints temporarily. @value{GDBN} records
5156 @var{count} with each of the specified breakpoints, and decrements a
5157 breakpoint's count when it is hit. When any count reaches 0,
5158 @value{GDBN} disables that breakpoint. If a breakpoint has an ignore
5159 count (@pxref{Conditions, ,Break Conditions}), that will be
5160 decremented to 0 before @var{count} is affected.
5162 @item enable @r{[}breakpoints@r{]} delete @var{list}@dots{}
5163 Enable the specified breakpoints to work once, then die. @value{GDBN}
5164 deletes any of these breakpoints as soon as your program stops there.
5165 Breakpoints set by the @code{tbreak} command start out in this state.
5168 @c FIXME: I think the following ``Except for [...] @code{tbreak}'' is
5169 @c confusing: tbreak is also initially enabled.
5170 Except for a breakpoint set with @code{tbreak} (@pxref{Set Breaks,
5171 ,Setting Breakpoints}), breakpoints that you set are initially enabled;
5172 subsequently, they become disabled or enabled only when you use one of
5173 the commands above. (The command @code{until} can set and delete a
5174 breakpoint of its own, but it does not change the state of your other
5175 breakpoints; see @ref{Continuing and Stepping, ,Continuing and
5179 @subsection Break Conditions
5180 @cindex conditional breakpoints
5181 @cindex breakpoint conditions
5183 @c FIXME what is scope of break condition expr? Context where wanted?
5184 @c in particular for a watchpoint?
5185 The simplest sort of breakpoint breaks every time your program reaches a
5186 specified place. You can also specify a @dfn{condition} for a
5187 breakpoint. A condition is just a Boolean expression in your
5188 programming language (@pxref{Expressions, ,Expressions}). A breakpoint with
5189 a condition evaluates the expression each time your program reaches it,
5190 and your program stops only if the condition is @emph{true}.
5192 This is the converse of using assertions for program validation; in that
5193 situation, you want to stop when the assertion is violated---that is,
5194 when the condition is false. In C, if you want to test an assertion expressed
5195 by the condition @var{assert}, you should set the condition
5196 @samp{! @var{assert}} on the appropriate breakpoint.
5198 Conditions are also accepted for watchpoints; you may not need them,
5199 since a watchpoint is inspecting the value of an expression anyhow---but
5200 it might be simpler, say, to just set a watchpoint on a variable name,
5201 and specify a condition that tests whether the new value is an interesting
5204 Break conditions can have side effects, and may even call functions in
5205 your program. This can be useful, for example, to activate functions
5206 that log program progress, or to use your own print functions to
5207 format special data structures. The effects are completely predictable
5208 unless there is another enabled breakpoint at the same address. (In
5209 that case, @value{GDBN} might see the other breakpoint first and stop your
5210 program without checking the condition of this one.) Note that
5211 breakpoint commands are usually more convenient and flexible than break
5213 purpose of performing side effects when a breakpoint is reached
5214 (@pxref{Break Commands, ,Breakpoint Command Lists}).
5216 Breakpoint conditions can also be evaluated on the target's side if
5217 the target supports it. Instead of evaluating the conditions locally,
5218 @value{GDBN} encodes the expression into an agent expression
5219 (@pxref{Agent Expressions}) suitable for execution on the target,
5220 independently of @value{GDBN}. Global variables become raw memory
5221 locations, locals become stack accesses, and so forth.
5223 In this case, @value{GDBN} will only be notified of a breakpoint trigger
5224 when its condition evaluates to true. This mechanism may provide faster
5225 response times depending on the performance characteristics of the target
5226 since it does not need to keep @value{GDBN} informed about
5227 every breakpoint trigger, even those with false conditions.
5229 Break conditions can be specified when a breakpoint is set, by using
5230 @samp{if} in the arguments to the @code{break} command. @xref{Set
5231 Breaks, ,Setting Breakpoints}. They can also be changed at any time
5232 with the @code{condition} command.
5234 You can also use the @code{if} keyword with the @code{watch} command.
5235 The @code{catch} command does not recognize the @code{if} keyword;
5236 @code{condition} is the only way to impose a further condition on a
5241 @item condition @var{bnum} @var{expression}
5242 Specify @var{expression} as the break condition for breakpoint,
5243 watchpoint, or catchpoint number @var{bnum}. After you set a condition,
5244 breakpoint @var{bnum} stops your program only if the value of
5245 @var{expression} is true (nonzero, in C). When you use
5246 @code{condition}, @value{GDBN} checks @var{expression} immediately for
5247 syntactic correctness, and to determine whether symbols in it have
5248 referents in the context of your breakpoint. If @var{expression} uses
5249 symbols not referenced in the context of the breakpoint, @value{GDBN}
5250 prints an error message:
5253 No symbol "foo" in current context.
5258 not actually evaluate @var{expression} at the time the @code{condition}
5259 command (or a command that sets a breakpoint with a condition, like
5260 @code{break if @dots{}}) is given, however. @xref{Expressions, ,Expressions}.
5262 @item condition @var{bnum}
5263 Remove the condition from breakpoint number @var{bnum}. It becomes
5264 an ordinary unconditional breakpoint.
5267 @cindex ignore count (of breakpoint)
5268 A special case of a breakpoint condition is to stop only when the
5269 breakpoint has been reached a certain number of times. This is so
5270 useful that there is a special way to do it, using the @dfn{ignore
5271 count} of the breakpoint. Every breakpoint has an ignore count, which
5272 is an integer. Most of the time, the ignore count is zero, and
5273 therefore has no effect. But if your program reaches a breakpoint whose
5274 ignore count is positive, then instead of stopping, it just decrements
5275 the ignore count by one and continues. As a result, if the ignore count
5276 value is @var{n}, the breakpoint does not stop the next @var{n} times
5277 your program reaches it.
5281 @item ignore @var{bnum} @var{count}
5282 Set the ignore count of breakpoint number @var{bnum} to @var{count}.
5283 The next @var{count} times the breakpoint is reached, your program's
5284 execution does not stop; other than to decrement the ignore count, @value{GDBN}
5287 To make the breakpoint stop the next time it is reached, specify
5290 When you use @code{continue} to resume execution of your program from a
5291 breakpoint, you can specify an ignore count directly as an argument to
5292 @code{continue}, rather than using @code{ignore}. @xref{Continuing and
5293 Stepping,,Continuing and Stepping}.
5295 If a breakpoint has a positive ignore count and a condition, the
5296 condition is not checked. Once the ignore count reaches zero,
5297 @value{GDBN} resumes checking the condition.
5299 You could achieve the effect of the ignore count with a condition such
5300 as @w{@samp{$foo-- <= 0}} using a debugger convenience variable that
5301 is decremented each time. @xref{Convenience Vars, ,Convenience
5305 Ignore counts apply to breakpoints, watchpoints, and catchpoints.
5308 @node Break Commands
5309 @subsection Breakpoint Command Lists
5311 @cindex breakpoint commands
5312 You can give any breakpoint (or watchpoint or catchpoint) a series of
5313 commands to execute when your program stops due to that breakpoint. For
5314 example, you might want to print the values of certain expressions, or
5315 enable other breakpoints.
5319 @kindex end@r{ (breakpoint commands)}
5320 @item commands @r{[}@var{list}@dots{}@r{]}
5321 @itemx @dots{} @var{command-list} @dots{}
5323 Specify a list of commands for the given breakpoints. The commands
5324 themselves appear on the following lines. Type a line containing just
5325 @code{end} to terminate the commands.
5327 To remove all commands from a breakpoint, type @code{commands} and
5328 follow it immediately with @code{end}; that is, give no commands.
5330 With no argument, @code{commands} refers to the last breakpoint,
5331 watchpoint, or catchpoint set (not to the breakpoint most recently
5332 encountered). If the most recent breakpoints were set with a single
5333 command, then the @code{commands} will apply to all the breakpoints
5334 set by that command. This applies to breakpoints set by
5335 @code{rbreak}, and also applies when a single @code{break} command
5336 creates multiple breakpoints (@pxref{Ambiguous Expressions,,Ambiguous
5340 Pressing @key{RET} as a means of repeating the last @value{GDBN} command is
5341 disabled within a @var{command-list}.
5343 You can use breakpoint commands to start your program up again. Simply
5344 use the @code{continue} command, or @code{step}, or any other command
5345 that resumes execution.
5347 Any other commands in the command list, after a command that resumes
5348 execution, are ignored. This is because any time you resume execution
5349 (even with a simple @code{next} or @code{step}), you may encounter
5350 another breakpoint---which could have its own command list, leading to
5351 ambiguities about which list to execute.
5354 If the first command you specify in a command list is @code{silent}, the
5355 usual message about stopping at a breakpoint is not printed. This may
5356 be desirable for breakpoints that are to print a specific message and
5357 then continue. If none of the remaining commands print anything, you
5358 see no sign that the breakpoint was reached. @code{silent} is
5359 meaningful only at the beginning of a breakpoint command list.
5361 The commands @code{echo}, @code{output}, and @code{printf} allow you to
5362 print precisely controlled output, and are often useful in silent
5363 breakpoints. @xref{Output, ,Commands for Controlled Output}.
5365 For example, here is how you could use breakpoint commands to print the
5366 value of @code{x} at entry to @code{foo} whenever @code{x} is positive.
5372 printf "x is %d\n",x
5377 One application for breakpoint commands is to compensate for one bug so
5378 you can test for another. Put a breakpoint just after the erroneous line
5379 of code, give it a condition to detect the case in which something
5380 erroneous has been done, and give it commands to assign correct values
5381 to any variables that need them. End with the @code{continue} command
5382 so that your program does not stop, and start with the @code{silent}
5383 command so that no output is produced. Here is an example:
5394 @node Dynamic Printf
5395 @subsection Dynamic Printf
5397 @cindex dynamic printf
5399 The dynamic printf command @code{dprintf} combines a breakpoint with
5400 formatted printing of your program's data to give you the effect of
5401 inserting @code{printf} calls into your program on-the-fly, without
5402 having to recompile it.
5404 In its most basic form, the output goes to the GDB console. However,
5405 you can set the variable @code{dprintf-style} for alternate handling.
5406 For instance, you can ask to format the output by calling your
5407 program's @code{printf} function. This has the advantage that the
5408 characters go to the program's output device, so they can recorded in
5409 redirects to files and so forth.
5411 If you are doing remote debugging with a stub or agent, you can also
5412 ask to have the printf handled by the remote agent. In addition to
5413 ensuring that the output goes to the remote program's device along
5414 with any other output the program might produce, you can also ask that
5415 the dprintf remain active even after disconnecting from the remote
5416 target. Using the stub/agent is also more efficient, as it can do
5417 everything without needing to communicate with @value{GDBN}.
5421 @item dprintf @var{location},@var{template},@var{expression}[,@var{expression}@dots{}]
5422 Whenever execution reaches @var{location}, print the values of one or
5423 more @var{expressions} under the control of the string @var{template}.
5424 To print several values, separate them with commas.
5426 @item set dprintf-style @var{style}
5427 Set the dprintf output to be handled in one of several different
5428 styles enumerated below. A change of style affects all existing
5429 dynamic printfs immediately. (If you need individual control over the
5430 print commands, simply define normal breakpoints with
5431 explicitly-supplied command lists.)
5435 @kindex dprintf-style gdb
5436 Handle the output using the @value{GDBN} @code{printf} command.
5439 @kindex dprintf-style call
5440 Handle the output by calling a function in your program (normally
5444 @kindex dprintf-style agent
5445 Have the remote debugging agent (such as @code{gdbserver}) handle
5446 the output itself. This style is only available for agents that
5447 support running commands on the target.
5450 @item set dprintf-function @var{function}
5451 Set the function to call if the dprintf style is @code{call}. By
5452 default its value is @code{printf}. You may set it to any expression.
5453 that @value{GDBN} can evaluate to a function, as per the @code{call}
5456 @item set dprintf-channel @var{channel}
5457 Set a ``channel'' for dprintf. If set to a non-empty value,
5458 @value{GDBN} will evaluate it as an expression and pass the result as
5459 a first argument to the @code{dprintf-function}, in the manner of
5460 @code{fprintf} and similar functions. Otherwise, the dprintf format
5461 string will be the first argument, in the manner of @code{printf}.
5463 As an example, if you wanted @code{dprintf} output to go to a logfile
5464 that is a standard I/O stream assigned to the variable @code{mylog},
5465 you could do the following:
5468 (gdb) set dprintf-style call
5469 (gdb) set dprintf-function fprintf
5470 (gdb) set dprintf-channel mylog
5471 (gdb) dprintf 25,"at line 25, glob=%d\n",glob
5472 Dprintf 1 at 0x123456: file main.c, line 25.
5474 1 dprintf keep y 0x00123456 in main at main.c:25
5475 call (void) fprintf (mylog,"at line 25, glob=%d\n",glob)
5480 Note that the @code{info break} displays the dynamic printf commands
5481 as normal breakpoint commands; you can thus easily see the effect of
5482 the variable settings.
5484 @item set disconnected-dprintf on
5485 @itemx set disconnected-dprintf off
5486 @kindex set disconnected-dprintf
5487 Choose whether @code{dprintf} commands should continue to run if
5488 @value{GDBN} has disconnected from the target. This only applies
5489 if the @code{dprintf-style} is @code{agent}.
5491 @item show disconnected-dprintf off
5492 @kindex show disconnected-dprintf
5493 Show the current choice for disconnected @code{dprintf}.
5497 @value{GDBN} does not check the validity of function and channel,
5498 relying on you to supply values that are meaningful for the contexts
5499 in which they are being used. For instance, the function and channel
5500 may be the values of local variables, but if that is the case, then
5501 all enabled dynamic prints must be at locations within the scope of
5502 those locals. If evaluation fails, @value{GDBN} will report an error.
5504 @node Save Breakpoints
5505 @subsection How to save breakpoints to a file
5507 To save breakpoint definitions to a file use the @w{@code{save
5508 breakpoints}} command.
5511 @kindex save breakpoints
5512 @cindex save breakpoints to a file for future sessions
5513 @item save breakpoints [@var{filename}]
5514 This command saves all current breakpoint definitions together with
5515 their commands and ignore counts, into a file @file{@var{filename}}
5516 suitable for use in a later debugging session. This includes all
5517 types of breakpoints (breakpoints, watchpoints, catchpoints,
5518 tracepoints). To read the saved breakpoint definitions, use the
5519 @code{source} command (@pxref{Command Files}). Note that watchpoints
5520 with expressions involving local variables may fail to be recreated
5521 because it may not be possible to access the context where the
5522 watchpoint is valid anymore. Because the saved breakpoint definitions
5523 are simply a sequence of @value{GDBN} commands that recreate the
5524 breakpoints, you can edit the file in your favorite editing program,
5525 and remove the breakpoint definitions you're not interested in, or
5526 that can no longer be recreated.
5529 @node Static Probe Points
5530 @subsection Static Probe Points
5532 @cindex static probe point, SystemTap
5533 @cindex static probe point, DTrace
5534 @value{GDBN} supports @dfn{SDT} probes in the code. @acronym{SDT} stands
5535 for Statically Defined Tracing, and the probes are designed to have a tiny
5536 runtime code and data footprint, and no dynamic relocations.
5538 Currently, the following types of probes are supported on
5539 ELF-compatible systems:
5543 @item @code{SystemTap} (@uref{http://sourceware.org/systemtap/})
5544 @acronym{SDT} probes@footnote{See
5545 @uref{http://sourceware.org/systemtap/wiki/AddingUserSpaceProbingToApps}
5546 for more information on how to add @code{SystemTap} @acronym{SDT}
5547 probes in your applications.}. @code{SystemTap} probes are usable
5548 from assembly, C and C@t{++} languages@footnote{See
5549 @uref{http://sourceware.org/systemtap/wiki/UserSpaceProbeImplementation}
5550 for a good reference on how the @acronym{SDT} probes are implemented.}.
5552 @item @code{DTrace} (@uref{http://oss.oracle.com/projects/DTrace})
5553 @acronym{USDT} probes. @code{DTrace} probes are usable from C and
5557 @cindex semaphores on static probe points
5558 Some @code{SystemTap} probes have an associated semaphore variable;
5559 for instance, this happens automatically if you defined your probe
5560 using a DTrace-style @file{.d} file. If your probe has a semaphore,
5561 @value{GDBN} will automatically enable it when you specify a
5562 breakpoint using the @samp{-probe-stap} notation. But, if you put a
5563 breakpoint at a probe's location by some other method (e.g.,
5564 @code{break file:line}), then @value{GDBN} will not automatically set
5565 the semaphore. @code{DTrace} probes do not support semaphores.
5567 You can examine the available static static probes using @code{info
5568 probes}, with optional arguments:
5572 @item info probes @r{[}@var{type}@r{]} @r{[}@var{provider} @r{[}@var{name} @r{[}@var{objfile}@r{]}@r{]}@r{]}
5573 If given, @var{type} is either @code{stap} for listing
5574 @code{SystemTap} probes or @code{dtrace} for listing @code{DTrace}
5575 probes. If omitted all probes are listed regardless of their types.
5577 If given, @var{provider} is a regular expression used to match against provider
5578 names when selecting which probes to list. If omitted, probes by all
5579 probes from all providers are listed.
5581 If given, @var{name} is a regular expression to match against probe names
5582 when selecting which probes to list. If omitted, probe names are not
5583 considered when deciding whether to display them.
5585 If given, @var{objfile} is a regular expression used to select which
5586 object files (executable or shared libraries) to examine. If not
5587 given, all object files are considered.
5589 @item info probes all
5590 List the available static probes, from all types.
5593 @cindex enabling and disabling probes
5594 Some probe points can be enabled and/or disabled. The effect of
5595 enabling or disabling a probe depends on the type of probe being
5596 handled. Some @code{DTrace} probes can be enabled or
5597 disabled, but @code{SystemTap} probes cannot be disabled.
5599 You can enable (or disable) one or more probes using the following
5600 commands, with optional arguments:
5603 @kindex enable probes
5604 @item enable probes @r{[}@var{provider} @r{[}@var{name} @r{[}@var{objfile}@r{]}@r{]}@r{]}
5605 If given, @var{provider} is a regular expression used to match against
5606 provider names when selecting which probes to enable. If omitted,
5607 all probes from all providers are enabled.
5609 If given, @var{name} is a regular expression to match against probe
5610 names when selecting which probes to enable. If omitted, probe names
5611 are not considered when deciding whether to enable them.
5613 If given, @var{objfile} is a regular expression used to select which
5614 object files (executable or shared libraries) to examine. If not
5615 given, all object files are considered.
5617 @kindex disable probes
5618 @item disable probes @r{[}@var{provider} @r{[}@var{name} @r{[}@var{objfile}@r{]}@r{]}@r{]}
5619 See the @code{enable probes} command above for a description of the
5620 optional arguments accepted by this command.
5623 @vindex $_probe_arg@r{, convenience variable}
5624 A probe may specify up to twelve arguments. These are available at the
5625 point at which the probe is defined---that is, when the current PC is
5626 at the probe's location. The arguments are available using the
5627 convenience variables (@pxref{Convenience Vars})
5628 @code{$_probe_arg0}@dots{}@code{$_probe_arg11}. In @code{SystemTap}
5629 probes each probe argument is an integer of the appropriate size;
5630 types are not preserved. In @code{DTrace} probes types are preserved
5631 provided that they are recognized as such by @value{GDBN}; otherwise
5632 the value of the probe argument will be a long integer. The
5633 convenience variable @code{$_probe_argc} holds the number of arguments
5634 at the current probe point.
5636 These variables are always available, but attempts to access them at
5637 any location other than a probe point will cause @value{GDBN} to give
5641 @c @ifclear BARETARGET
5642 @node Error in Breakpoints
5643 @subsection ``Cannot insert breakpoints''
5645 If you request too many active hardware-assisted breakpoints and
5646 watchpoints, you will see this error message:
5648 @c FIXME: the precise wording of this message may change; the relevant
5649 @c source change is not committed yet (Sep 3, 1999).
5651 Stopped; cannot insert breakpoints.
5652 You may have requested too many hardware breakpoints and watchpoints.
5656 This message is printed when you attempt to resume the program, since
5657 only then @value{GDBN} knows exactly how many hardware breakpoints and
5658 watchpoints it needs to insert.
5660 When this message is printed, you need to disable or remove some of the
5661 hardware-assisted breakpoints and watchpoints, and then continue.
5663 @node Breakpoint-related Warnings
5664 @subsection ``Breakpoint address adjusted...''
5665 @cindex breakpoint address adjusted
5667 Some processor architectures place constraints on the addresses at
5668 which breakpoints may be placed. For architectures thus constrained,
5669 @value{GDBN} will attempt to adjust the breakpoint's address to comply
5670 with the constraints dictated by the architecture.
5672 One example of such an architecture is the Fujitsu FR-V. The FR-V is
5673 a VLIW architecture in which a number of RISC-like instructions may be
5674 bundled together for parallel execution. The FR-V architecture
5675 constrains the location of a breakpoint instruction within such a
5676 bundle to the instruction with the lowest address. @value{GDBN}
5677 honors this constraint by adjusting a breakpoint's address to the
5678 first in the bundle.
5680 It is not uncommon for optimized code to have bundles which contain
5681 instructions from different source statements, thus it may happen that
5682 a breakpoint's address will be adjusted from one source statement to
5683 another. Since this adjustment may significantly alter @value{GDBN}'s
5684 breakpoint related behavior from what the user expects, a warning is
5685 printed when the breakpoint is first set and also when the breakpoint
5688 A warning like the one below is printed when setting a breakpoint
5689 that's been subject to address adjustment:
5692 warning: Breakpoint address adjusted from 0x00010414 to 0x00010410.
5695 Such warnings are printed both for user settable and @value{GDBN}'s
5696 internal breakpoints. If you see one of these warnings, you should
5697 verify that a breakpoint set at the adjusted address will have the
5698 desired affect. If not, the breakpoint in question may be removed and
5699 other breakpoints may be set which will have the desired behavior.
5700 E.g., it may be sufficient to place the breakpoint at a later
5701 instruction. A conditional breakpoint may also be useful in some
5702 cases to prevent the breakpoint from triggering too often.
5704 @value{GDBN} will also issue a warning when stopping at one of these
5705 adjusted breakpoints:
5708 warning: Breakpoint 1 address previously adjusted from 0x00010414
5712 When this warning is encountered, it may be too late to take remedial
5713 action except in cases where the breakpoint is hit earlier or more
5714 frequently than expected.
5716 @node Continuing and Stepping
5717 @section Continuing and Stepping
5721 @cindex resuming execution
5722 @dfn{Continuing} means resuming program execution until your program
5723 completes normally. In contrast, @dfn{stepping} means executing just
5724 one more ``step'' of your program, where ``step'' may mean either one
5725 line of source code, or one machine instruction (depending on what
5726 particular command you use). Either when continuing or when stepping,
5727 your program may stop even sooner, due to a breakpoint or a signal. (If
5728 it stops due to a signal, you may want to use @code{handle}, or use
5729 @samp{signal 0} to resume execution (@pxref{Signals, ,Signals}),
5730 or you may step into the signal's handler (@pxref{stepping and signal
5735 @kindex c @r{(@code{continue})}
5736 @kindex fg @r{(resume foreground execution)}
5737 @item continue @r{[}@var{ignore-count}@r{]}
5738 @itemx c @r{[}@var{ignore-count}@r{]}
5739 @itemx fg @r{[}@var{ignore-count}@r{]}
5740 Resume program execution, at the address where your program last stopped;
5741 any breakpoints set at that address are bypassed. The optional argument
5742 @var{ignore-count} allows you to specify a further number of times to
5743 ignore a breakpoint at this location; its effect is like that of
5744 @code{ignore} (@pxref{Conditions, ,Break Conditions}).
5746 The argument @var{ignore-count} is meaningful only when your program
5747 stopped due to a breakpoint. At other times, the argument to
5748 @code{continue} is ignored.
5750 The synonyms @code{c} and @code{fg} (for @dfn{foreground}, as the
5751 debugged program is deemed to be the foreground program) are provided
5752 purely for convenience, and have exactly the same behavior as
5756 To resume execution at a different place, you can use @code{return}
5757 (@pxref{Returning, ,Returning from a Function}) to go back to the
5758 calling function; or @code{jump} (@pxref{Jumping, ,Continuing at a
5759 Different Address}) to go to an arbitrary location in your program.
5761 A typical technique for using stepping is to set a breakpoint
5762 (@pxref{Breakpoints, ,Breakpoints; Watchpoints; and Catchpoints}) at the
5763 beginning of the function or the section of your program where a problem
5764 is believed to lie, run your program until it stops at that breakpoint,
5765 and then step through the suspect area, examining the variables that are
5766 interesting, until you see the problem happen.
5770 @kindex s @r{(@code{step})}
5772 Continue running your program until control reaches a different source
5773 line, then stop it and return control to @value{GDBN}. This command is
5774 abbreviated @code{s}.
5777 @c "without debugging information" is imprecise; actually "without line
5778 @c numbers in the debugging information". (gcc -g1 has debugging info but
5779 @c not line numbers). But it seems complex to try to make that
5780 @c distinction here.
5781 @emph{Warning:} If you use the @code{step} command while control is
5782 within a function that was compiled without debugging information,
5783 execution proceeds until control reaches a function that does have
5784 debugging information. Likewise, it will not step into a function which
5785 is compiled without debugging information. To step through functions
5786 without debugging information, use the @code{stepi} command, described
5790 The @code{step} command only stops at the first instruction of a source
5791 line. This prevents the multiple stops that could otherwise occur in
5792 @code{switch} statements, @code{for} loops, etc. @code{step} continues
5793 to stop if a function that has debugging information is called within
5794 the line. In other words, @code{step} @emph{steps inside} any functions
5795 called within the line.
5797 Also, the @code{step} command only enters a function if there is line
5798 number information for the function. Otherwise it acts like the
5799 @code{next} command. This avoids problems when using @code{cc -gl}
5800 on @acronym{MIPS} machines. Previously, @code{step} entered subroutines if there
5801 was any debugging information about the routine.
5803 @item step @var{count}
5804 Continue running as in @code{step}, but do so @var{count} times. If a
5805 breakpoint is reached, or a signal not related to stepping occurs before
5806 @var{count} steps, stepping stops right away.
5809 @kindex n @r{(@code{next})}
5810 @item next @r{[}@var{count}@r{]}
5811 Continue to the next source line in the current (innermost) stack frame.
5812 This is similar to @code{step}, but function calls that appear within
5813 the line of code are executed without stopping. Execution stops when
5814 control reaches a different line of code at the original stack level
5815 that was executing when you gave the @code{next} command. This command
5816 is abbreviated @code{n}.
5818 An argument @var{count} is a repeat count, as for @code{step}.
5821 @c FIX ME!! Do we delete this, or is there a way it fits in with
5822 @c the following paragraph? --- Vctoria
5824 @c @code{next} within a function that lacks debugging information acts like
5825 @c @code{step}, but any function calls appearing within the code of the
5826 @c function are executed without stopping.
5828 The @code{next} command only stops at the first instruction of a
5829 source line. This prevents multiple stops that could otherwise occur in
5830 @code{switch} statements, @code{for} loops, etc.
5832 @kindex set step-mode
5834 @cindex functions without line info, and stepping
5835 @cindex stepping into functions with no line info
5836 @itemx set step-mode on
5837 The @code{set step-mode on} command causes the @code{step} command to
5838 stop at the first instruction of a function which contains no debug line
5839 information rather than stepping over it.
5841 This is useful in cases where you may be interested in inspecting the
5842 machine instructions of a function which has no symbolic info and do not
5843 want @value{GDBN} to automatically skip over this function.
5845 @item set step-mode off
5846 Causes the @code{step} command to step over any functions which contains no
5847 debug information. This is the default.
5849 @item show step-mode
5850 Show whether @value{GDBN} will stop in or step over functions without
5851 source line debug information.
5854 @kindex fin @r{(@code{finish})}
5856 Continue running until just after function in the selected stack frame
5857 returns. Print the returned value (if any). This command can be
5858 abbreviated as @code{fin}.
5860 Contrast this with the @code{return} command (@pxref{Returning,
5861 ,Returning from a Function}).
5863 @kindex set print finish
5864 @kindex show print finish
5865 @item set print finish @r{[}on|off@r{]}
5866 @itemx show print finish
5867 By default the @code{finish} command will show the value that is
5868 returned by the function. This can be disabled using @code{set print
5869 finish off}. When disabled, the value is still entered into the value
5870 history (@pxref{Value History}), but not displayed.
5873 @kindex u @r{(@code{until})}
5874 @cindex run until specified location
5877 Continue running until a source line past the current line, in the
5878 current stack frame, is reached. This command is used to avoid single
5879 stepping through a loop more than once. It is like the @code{next}
5880 command, except that when @code{until} encounters a jump, it
5881 automatically continues execution until the program counter is greater
5882 than the address of the jump.
5884 This means that when you reach the end of a loop after single stepping
5885 though it, @code{until} makes your program continue execution until it
5886 exits the loop. In contrast, a @code{next} command at the end of a loop
5887 simply steps back to the beginning of the loop, which forces you to step
5888 through the next iteration.
5890 @code{until} always stops your program if it attempts to exit the current
5893 @code{until} may produce somewhat counterintuitive results if the order
5894 of machine code does not match the order of the source lines. For
5895 example, in the following excerpt from a debugging session, the @code{f}
5896 (@code{frame}) command shows that execution is stopped at line
5897 @code{206}; yet when we use @code{until}, we get to line @code{195}:
5901 #0 main (argc=4, argv=0xf7fffae8) at m4.c:206
5903 (@value{GDBP}) until
5904 195 for ( ; argc > 0; NEXTARG) @{
5907 This happened because, for execution efficiency, the compiler had
5908 generated code for the loop closure test at the end, rather than the
5909 start, of the loop---even though the test in a C @code{for}-loop is
5910 written before the body of the loop. The @code{until} command appeared
5911 to step back to the beginning of the loop when it advanced to this
5912 expression; however, it has not really gone to an earlier
5913 statement---not in terms of the actual machine code.
5915 @code{until} with no argument works by means of single
5916 instruction stepping, and hence is slower than @code{until} with an
5919 @item until @var{location}
5920 @itemx u @var{location}
5921 Continue running your program until either the specified @var{location} is
5922 reached, or the current stack frame returns. The location is any of
5923 the forms described in @ref{Specify Location}.
5924 This form of the command uses temporary breakpoints, and
5925 hence is quicker than @code{until} without an argument. The specified
5926 location is actually reached only if it is in the current frame. This
5927 implies that @code{until} can be used to skip over recursive function
5928 invocations. For instance in the code below, if the current location is
5929 line @code{96}, issuing @code{until 99} will execute the program up to
5930 line @code{99} in the same invocation of factorial, i.e., after the inner
5931 invocations have returned.
5934 94 int factorial (int value)
5936 96 if (value > 1) @{
5937 97 value *= factorial (value - 1);
5944 @kindex advance @var{location}
5945 @item advance @var{location}
5946 Continue running the program up to the given @var{location}. An argument is
5947 required, which should be of one of the forms described in
5948 @ref{Specify Location}.
5949 Execution will also stop upon exit from the current stack
5950 frame. This command is similar to @code{until}, but @code{advance} will
5951 not skip over recursive function calls, and the target location doesn't
5952 have to be in the same frame as the current one.
5956 @kindex si @r{(@code{stepi})}
5958 @itemx stepi @var{arg}
5960 Execute one machine instruction, then stop and return to the debugger.
5962 It is often useful to do @samp{display/i $pc} when stepping by machine
5963 instructions. This makes @value{GDBN} automatically display the next
5964 instruction to be executed, each time your program stops. @xref{Auto
5965 Display,, Automatic Display}.
5967 An argument is a repeat count, as in @code{step}.
5971 @kindex ni @r{(@code{nexti})}
5973 @itemx nexti @var{arg}
5975 Execute one machine instruction, but if it is a function call,
5976 proceed until the function returns.
5978 An argument is a repeat count, as in @code{next}.
5982 @anchor{range stepping}
5983 @cindex range stepping
5984 @cindex target-assisted range stepping
5985 By default, and if available, @value{GDBN} makes use of
5986 target-assisted @dfn{range stepping}. In other words, whenever you
5987 use a stepping command (e.g., @code{step}, @code{next}), @value{GDBN}
5988 tells the target to step the corresponding range of instruction
5989 addresses instead of issuing multiple single-steps. This speeds up
5990 line stepping, particularly for remote targets. Ideally, there should
5991 be no reason you would want to turn range stepping off. However, it's
5992 possible that a bug in the debug info, a bug in the remote stub (for
5993 remote targets), or even a bug in @value{GDBN} could make line
5994 stepping behave incorrectly when target-assisted range stepping is
5995 enabled. You can use the following command to turn off range stepping
5999 @kindex set range-stepping
6000 @kindex show range-stepping
6001 @item set range-stepping
6002 @itemx show range-stepping
6003 Control whether range stepping is enabled.
6005 If @code{on}, and the target supports it, @value{GDBN} tells the
6006 target to step a range of addresses itself, instead of issuing
6007 multiple single-steps. If @code{off}, @value{GDBN} always issues
6008 single-steps, even if range stepping is supported by the target. The
6009 default is @code{on}.
6013 @node Skipping Over Functions and Files
6014 @section Skipping Over Functions and Files
6015 @cindex skipping over functions and files
6017 The program you are debugging may contain some functions which are
6018 uninteresting to debug. The @code{skip} command lets you tell @value{GDBN} to
6019 skip a function, all functions in a file or a particular function in
6020 a particular file when stepping.
6022 For example, consider the following C function:
6033 Suppose you wish to step into the functions @code{foo} and @code{bar}, but you
6034 are not interested in stepping through @code{boring}. If you run @code{step}
6035 at line 103, you'll enter @code{boring()}, but if you run @code{next}, you'll
6036 step over both @code{foo} and @code{boring}!
6038 One solution is to @code{step} into @code{boring} and use the @code{finish}
6039 command to immediately exit it. But this can become tedious if @code{boring}
6040 is called from many places.
6042 A more flexible solution is to execute @kbd{skip boring}. This instructs
6043 @value{GDBN} never to step into @code{boring}. Now when you execute
6044 @code{step} at line 103, you'll step over @code{boring} and directly into
6047 Functions may be skipped by providing either a function name, linespec
6048 (@pxref{Specify Location}), regular expression that matches the function's
6049 name, file name or a @code{glob}-style pattern that matches the file name.
6051 On Posix systems the form of the regular expression is
6052 ``Extended Regular Expressions''. See for example @samp{man 7 regex}
6053 on @sc{gnu}/Linux systems. On non-Posix systems the form of the regular
6054 expression is whatever is provided by the @code{regcomp} function of
6055 the underlying system.
6056 See for example @samp{man 7 glob} on @sc{gnu}/Linux systems for a
6057 description of @code{glob}-style patterns.
6061 @item skip @r{[}@var{options}@r{]}
6062 The basic form of the @code{skip} command takes zero or more options
6063 that specify what to skip.
6064 The @var{options} argument is any useful combination of the following:
6067 @item -file @var{file}
6068 @itemx -fi @var{file}
6069 Functions in @var{file} will be skipped over when stepping.
6071 @item -gfile @var{file-glob-pattern}
6072 @itemx -gfi @var{file-glob-pattern}
6073 @cindex skipping over files via glob-style patterns
6074 Functions in files matching @var{file-glob-pattern} will be skipped
6078 (gdb) skip -gfi utils/*.c
6081 @item -function @var{linespec}
6082 @itemx -fu @var{linespec}
6083 Functions named by @var{linespec} or the function containing the line
6084 named by @var{linespec} will be skipped over when stepping.
6085 @xref{Specify Location}.
6087 @item -rfunction @var{regexp}
6088 @itemx -rfu @var{regexp}
6089 @cindex skipping over functions via regular expressions
6090 Functions whose name matches @var{regexp} will be skipped over when stepping.
6092 This form is useful for complex function names.
6093 For example, there is generally no need to step into C@t{++} @code{std::string}
6094 constructors or destructors. Plus with C@t{++} templates it can be hard to
6095 write out the full name of the function, and often it doesn't matter what
6096 the template arguments are. Specifying the function to be skipped as a
6097 regular expression makes this easier.
6100 (gdb) skip -rfu ^std::(allocator|basic_string)<.*>::~?\1 *\(
6103 If you want to skip every templated C@t{++} constructor and destructor
6104 in the @code{std} namespace you can do:
6107 (gdb) skip -rfu ^std::([a-zA-z0-9_]+)<.*>::~?\1 *\(
6111 If no options are specified, the function you're currently debugging
6114 @kindex skip function
6115 @item skip function @r{[}@var{linespec}@r{]}
6116 After running this command, the function named by @var{linespec} or the
6117 function containing the line named by @var{linespec} will be skipped over when
6118 stepping. @xref{Specify Location}.
6120 If you do not specify @var{linespec}, the function you're currently debugging
6123 (If you have a function called @code{file} that you want to skip, use
6124 @kbd{skip function file}.)
6127 @item skip file @r{[}@var{filename}@r{]}
6128 After running this command, any function whose source lives in @var{filename}
6129 will be skipped over when stepping.
6132 (gdb) skip file boring.c
6133 File boring.c will be skipped when stepping.
6136 If you do not specify @var{filename}, functions whose source lives in the file
6137 you're currently debugging will be skipped.
6140 Skips can be listed, deleted, disabled, and enabled, much like breakpoints.
6141 These are the commands for managing your list of skips:
6145 @item info skip @r{[}@var{range}@r{]}
6146 Print details about the specified skip(s). If @var{range} is not specified,
6147 print a table with details about all functions and files marked for skipping.
6148 @code{info skip} prints the following information about each skip:
6152 A number identifying this skip.
6153 @item Enabled or Disabled
6154 Enabled skips are marked with @samp{y}.
6155 Disabled skips are marked with @samp{n}.
6157 If the file name is a @samp{glob} pattern this is @samp{y}.
6158 Otherwise it is @samp{n}.
6160 The name or @samp{glob} pattern of the file to be skipped.
6161 If no file is specified this is @samp{<none>}.
6163 If the function name is a @samp{regular expression} this is @samp{y}.
6164 Otherwise it is @samp{n}.
6166 The name or regular expression of the function to skip.
6167 If no function is specified this is @samp{<none>}.
6171 @item skip delete @r{[}@var{range}@r{]}
6172 Delete the specified skip(s). If @var{range} is not specified, delete all
6176 @item skip enable @r{[}@var{range}@r{]}
6177 Enable the specified skip(s). If @var{range} is not specified, enable all
6180 @kindex skip disable
6181 @item skip disable @r{[}@var{range}@r{]}
6182 Disable the specified skip(s). If @var{range} is not specified, disable all
6185 @kindex set debug skip
6186 @item set debug skip @r{[}on|off@r{]}
6187 Set whether to print the debug output about skipping files and functions.
6189 @kindex show debug skip
6190 @item show debug skip
6191 Show whether the debug output about skipping files and functions is printed.
6199 A signal is an asynchronous event that can happen in a program. The
6200 operating system defines the possible kinds of signals, and gives each
6201 kind a name and a number. For example, in Unix @code{SIGINT} is the
6202 signal a program gets when you type an interrupt character (often @kbd{Ctrl-c});
6203 @code{SIGSEGV} is the signal a program gets from referencing a place in
6204 memory far away from all the areas in use; @code{SIGALRM} occurs when
6205 the alarm clock timer goes off (which happens only if your program has
6206 requested an alarm).
6208 @cindex fatal signals
6209 Some signals, including @code{SIGALRM}, are a normal part of the
6210 functioning of your program. Others, such as @code{SIGSEGV}, indicate
6211 errors; these signals are @dfn{fatal} (they kill your program immediately) if the
6212 program has not specified in advance some other way to handle the signal.
6213 @code{SIGINT} does not indicate an error in your program, but it is normally
6214 fatal so it can carry out the purpose of the interrupt: to kill the program.
6216 @value{GDBN} has the ability to detect any occurrence of a signal in your
6217 program. You can tell @value{GDBN} in advance what to do for each kind of
6220 @cindex handling signals
6221 Normally, @value{GDBN} is set up to let the non-erroneous signals like
6222 @code{SIGALRM} be silently passed to your program
6223 (so as not to interfere with their role in the program's functioning)
6224 but to stop your program immediately whenever an error signal happens.
6225 You can change these settings with the @code{handle} command.
6228 @kindex info signals
6232 Print a table of all the kinds of signals and how @value{GDBN} has been told to
6233 handle each one. You can use this to see the signal numbers of all
6234 the defined types of signals.
6236 @item info signals @var{sig}
6237 Similar, but print information only about the specified signal number.
6239 @code{info handle} is an alias for @code{info signals}.
6241 @item catch signal @r{[}@var{signal}@dots{} @r{|} @samp{all}@r{]}
6242 Set a catchpoint for the indicated signals. @xref{Set Catchpoints},
6243 for details about this command.
6246 @item handle @var{signal} @r{[}@var{keywords}@dots{}@r{]}
6247 Change the way @value{GDBN} handles signal @var{signal}. The @var{signal}
6248 can be the number of a signal or its name (with or without the
6249 @samp{SIG} at the beginning); a list of signal numbers of the form
6250 @samp{@var{low}-@var{high}}; or the word @samp{all}, meaning all the
6251 known signals. Optional arguments @var{keywords}, described below,
6252 say what change to make.
6256 The keywords allowed by the @code{handle} command can be abbreviated.
6257 Their full names are:
6261 @value{GDBN} should not stop your program when this signal happens. It may
6262 still print a message telling you that the signal has come in.
6265 @value{GDBN} should stop your program when this signal happens. This implies
6266 the @code{print} keyword as well.
6269 @value{GDBN} should print a message when this signal happens.
6272 @value{GDBN} should not mention the occurrence of the signal at all. This
6273 implies the @code{nostop} keyword as well.
6277 @value{GDBN} should allow your program to see this signal; your program
6278 can handle the signal, or else it may terminate if the signal is fatal
6279 and not handled. @code{pass} and @code{noignore} are synonyms.
6283 @value{GDBN} should not allow your program to see this signal.
6284 @code{nopass} and @code{ignore} are synonyms.
6288 When a signal stops your program, the signal is not visible to the
6290 continue. Your program sees the signal then, if @code{pass} is in
6291 effect for the signal in question @emph{at that time}. In other words,
6292 after @value{GDBN} reports a signal, you can use the @code{handle}
6293 command with @code{pass} or @code{nopass} to control whether your
6294 program sees that signal when you continue.
6296 The default is set to @code{nostop}, @code{noprint}, @code{pass} for
6297 non-erroneous signals such as @code{SIGALRM}, @code{SIGWINCH} and
6298 @code{SIGCHLD}, and to @code{stop}, @code{print}, @code{pass} for the
6301 You can also use the @code{signal} command to prevent your program from
6302 seeing a signal, or cause it to see a signal it normally would not see,
6303 or to give it any signal at any time. For example, if your program stopped
6304 due to some sort of memory reference error, you might store correct
6305 values into the erroneous variables and continue, hoping to see more
6306 execution; but your program would probably terminate immediately as
6307 a result of the fatal signal once it saw the signal. To prevent this,
6308 you can continue with @samp{signal 0}. @xref{Signaling, ,Giving your
6311 @cindex stepping and signal handlers
6312 @anchor{stepping and signal handlers}
6314 @value{GDBN} optimizes for stepping the mainline code. If a signal
6315 that has @code{handle nostop} and @code{handle pass} set arrives while
6316 a stepping command (e.g., @code{stepi}, @code{step}, @code{next}) is
6317 in progress, @value{GDBN} lets the signal handler run and then resumes
6318 stepping the mainline code once the signal handler returns. In other
6319 words, @value{GDBN} steps over the signal handler. This prevents
6320 signals that you've specified as not interesting (with @code{handle
6321 nostop}) from changing the focus of debugging unexpectedly. Note that
6322 the signal handler itself may still hit a breakpoint, stop for another
6323 signal that has @code{handle stop} in effect, or for any other event
6324 that normally results in stopping the stepping command sooner. Also
6325 note that @value{GDBN} still informs you that the program received a
6326 signal if @code{handle print} is set.
6328 @anchor{stepping into signal handlers}
6330 If you set @code{handle pass} for a signal, and your program sets up a
6331 handler for it, then issuing a stepping command, such as @code{step}
6332 or @code{stepi}, when your program is stopped due to the signal will
6333 step @emph{into} the signal handler (if the target supports that).
6335 Likewise, if you use the @code{queue-signal} command to queue a signal
6336 to be delivered to the current thread when execution of the thread
6337 resumes (@pxref{Signaling, ,Giving your Program a Signal}), then a
6338 stepping command will step into the signal handler.
6340 Here's an example, using @code{stepi} to step to the first instruction
6341 of @code{SIGUSR1}'s handler:
6344 (@value{GDBP}) handle SIGUSR1
6345 Signal Stop Print Pass to program Description
6346 SIGUSR1 Yes Yes Yes User defined signal 1
6350 Program received signal SIGUSR1, User defined signal 1.
6351 main () sigusr1.c:28
6354 sigusr1_handler () at sigusr1.c:9
6358 The same, but using @code{queue-signal} instead of waiting for the
6359 program to receive the signal first:
6364 (@value{GDBP}) queue-signal SIGUSR1
6366 sigusr1_handler () at sigusr1.c:9
6371 @cindex extra signal information
6372 @anchor{extra signal information}
6374 On some targets, @value{GDBN} can inspect extra signal information
6375 associated with the intercepted signal, before it is actually
6376 delivered to the program being debugged. This information is exported
6377 by the convenience variable @code{$_siginfo}, and consists of data
6378 that is passed by the kernel to the signal handler at the time of the
6379 receipt of a signal. The data type of the information itself is
6380 target dependent. You can see the data type using the @code{ptype
6381 $_siginfo} command. On Unix systems, it typically corresponds to the
6382 standard @code{siginfo_t} type, as defined in the @file{signal.h}
6385 Here's an example, on a @sc{gnu}/Linux system, printing the stray
6386 referenced address that raised a segmentation fault.
6390 (@value{GDBP}) continue
6391 Program received signal SIGSEGV, Segmentation fault.
6392 0x0000000000400766 in main ()
6394 (@value{GDBP}) ptype $_siginfo
6401 struct @{...@} _kill;
6402 struct @{...@} _timer;
6404 struct @{...@} _sigchld;
6405 struct @{...@} _sigfault;
6406 struct @{...@} _sigpoll;
6409 (@value{GDBP}) ptype $_siginfo._sifields._sigfault
6413 (@value{GDBP}) p $_siginfo._sifields._sigfault.si_addr
6414 $1 = (void *) 0x7ffff7ff7000
6418 Depending on target support, @code{$_siginfo} may also be writable.
6420 @cindex Intel MPX boundary violations
6421 @cindex boundary violations, Intel MPX
6422 On some targets, a @code{SIGSEGV} can be caused by a boundary
6423 violation, i.e., accessing an address outside of the allowed range.
6424 In those cases @value{GDBN} may displays additional information,
6425 depending on how @value{GDBN} has been told to handle the signal.
6426 With @code{handle stop SIGSEGV}, @value{GDBN} displays the violation
6427 kind: "Upper" or "Lower", the memory address accessed and the
6428 bounds, while with @code{handle nostop SIGSEGV} no additional
6429 information is displayed.
6431 The usual output of a segfault is:
6433 Program received signal SIGSEGV, Segmentation fault
6434 0x0000000000400d7c in upper () at i386-mpx-sigsegv.c:68
6435 68 value = *(p + len);
6438 While a bound violation is presented as:
6440 Program received signal SIGSEGV, Segmentation fault
6441 Upper bound violation while accessing address 0x7fffffffc3b3
6442 Bounds: [lower = 0x7fffffffc390, upper = 0x7fffffffc3a3]
6443 0x0000000000400d7c in upper () at i386-mpx-sigsegv.c:68
6444 68 value = *(p + len);
6448 @section Stopping and Starting Multi-thread Programs
6450 @cindex stopped threads
6451 @cindex threads, stopped
6453 @cindex continuing threads
6454 @cindex threads, continuing
6456 @value{GDBN} supports debugging programs with multiple threads
6457 (@pxref{Threads,, Debugging Programs with Multiple Threads}). There
6458 are two modes of controlling execution of your program within the
6459 debugger. In the default mode, referred to as @dfn{all-stop mode},
6460 when any thread in your program stops (for example, at a breakpoint
6461 or while being stepped), all other threads in the program are also stopped by
6462 @value{GDBN}. On some targets, @value{GDBN} also supports
6463 @dfn{non-stop mode}, in which other threads can continue to run freely while
6464 you examine the stopped thread in the debugger.
6467 * All-Stop Mode:: All threads stop when GDB takes control
6468 * Non-Stop Mode:: Other threads continue to execute
6469 * Background Execution:: Running your program asynchronously
6470 * Thread-Specific Breakpoints:: Controlling breakpoints
6471 * Interrupted System Calls:: GDB may interfere with system calls
6472 * Observer Mode:: GDB does not alter program behavior
6476 @subsection All-Stop Mode
6478 @cindex all-stop mode
6480 In all-stop mode, whenever your program stops under @value{GDBN} for any reason,
6481 @emph{all} threads of execution stop, not just the current thread. This
6482 allows you to examine the overall state of the program, including
6483 switching between threads, without worrying that things may change
6486 Conversely, whenever you restart the program, @emph{all} threads start
6487 executing. @emph{This is true even when single-stepping} with commands
6488 like @code{step} or @code{next}.
6490 In particular, @value{GDBN} cannot single-step all threads in lockstep.
6491 Since thread scheduling is up to your debugging target's operating
6492 system (not controlled by @value{GDBN}), other threads may
6493 execute more than one statement while the current thread completes a
6494 single step. Moreover, in general other threads stop in the middle of a
6495 statement, rather than at a clean statement boundary, when the program
6498 You might even find your program stopped in another thread after
6499 continuing or even single-stepping. This happens whenever some other
6500 thread runs into a breakpoint, a signal, or an exception before the
6501 first thread completes whatever you requested.
6503 @cindex automatic thread selection
6504 @cindex switching threads automatically
6505 @cindex threads, automatic switching
6506 Whenever @value{GDBN} stops your program, due to a breakpoint or a
6507 signal, it automatically selects the thread where that breakpoint or
6508 signal happened. @value{GDBN} alerts you to the context switch with a
6509 message such as @samp{[Switching to Thread @var{n}]} to identify the
6512 On some OSes, you can modify @value{GDBN}'s default behavior by
6513 locking the OS scheduler to allow only a single thread to run.
6516 @item set scheduler-locking @var{mode}
6517 @cindex scheduler locking mode
6518 @cindex lock scheduler
6519 Set the scheduler locking mode. It applies to normal execution,
6520 record mode, and replay mode. If it is @code{off}, then there is no
6521 locking and any thread may run at any time. If @code{on}, then only
6522 the current thread may run when the inferior is resumed. The
6523 @code{step} mode optimizes for single-stepping; it prevents other
6524 threads from preempting the current thread while you are stepping, so
6525 that the focus of debugging does not change unexpectedly. Other
6526 threads never get a chance to run when you step, and they are
6527 completely free to run when you use commands like @samp{continue},
6528 @samp{until}, or @samp{finish}. However, unless another thread hits a
6529 breakpoint during its timeslice, @value{GDBN} does not change the
6530 current thread away from the thread that you are debugging. The
6531 @code{replay} mode behaves like @code{off} in record mode and like
6532 @code{on} in replay mode.
6534 @item show scheduler-locking
6535 Display the current scheduler locking mode.
6538 @cindex resume threads of multiple processes simultaneously
6539 By default, when you issue one of the execution commands such as
6540 @code{continue}, @code{next} or @code{step}, @value{GDBN} allows only
6541 threads of the current inferior to run. For example, if @value{GDBN}
6542 is attached to two inferiors, each with two threads, the
6543 @code{continue} command resumes only the two threads of the current
6544 inferior. This is useful, for example, when you debug a program that
6545 forks and you want to hold the parent stopped (so that, for instance,
6546 it doesn't run to exit), while you debug the child. In other
6547 situations, you may not be interested in inspecting the current state
6548 of any of the processes @value{GDBN} is attached to, and you may want
6549 to resume them all until some breakpoint is hit. In the latter case,
6550 you can instruct @value{GDBN} to allow all threads of all the
6551 inferiors to run with the @w{@code{set schedule-multiple}} command.
6554 @kindex set schedule-multiple
6555 @item set schedule-multiple
6556 Set the mode for allowing threads of multiple processes to be resumed
6557 when an execution command is issued. When @code{on}, all threads of
6558 all processes are allowed to run. When @code{off}, only the threads
6559 of the current process are resumed. The default is @code{off}. The
6560 @code{scheduler-locking} mode takes precedence when set to @code{on},
6561 or while you are stepping and set to @code{step}.
6563 @item show schedule-multiple
6564 Display the current mode for resuming the execution of threads of
6569 @subsection Non-Stop Mode
6571 @cindex non-stop mode
6573 @c This section is really only a place-holder, and needs to be expanded
6574 @c with more details.
6576 For some multi-threaded targets, @value{GDBN} supports an optional
6577 mode of operation in which you can examine stopped program threads in
6578 the debugger while other threads continue to execute freely. This
6579 minimizes intrusion when debugging live systems, such as programs
6580 where some threads have real-time constraints or must continue to
6581 respond to external events. This is referred to as @dfn{non-stop} mode.
6583 In non-stop mode, when a thread stops to report a debugging event,
6584 @emph{only} that thread is stopped; @value{GDBN} does not stop other
6585 threads as well, in contrast to the all-stop mode behavior. Additionally,
6586 execution commands such as @code{continue} and @code{step} apply by default
6587 only to the current thread in non-stop mode, rather than all threads as
6588 in all-stop mode. This allows you to control threads explicitly in
6589 ways that are not possible in all-stop mode --- for example, stepping
6590 one thread while allowing others to run freely, stepping
6591 one thread while holding all others stopped, or stepping several threads
6592 independently and simultaneously.
6594 To enter non-stop mode, use this sequence of commands before you run
6595 or attach to your program:
6598 # If using the CLI, pagination breaks non-stop.
6601 # Finally, turn it on!
6605 You can use these commands to manipulate the non-stop mode setting:
6608 @kindex set non-stop
6609 @item set non-stop on
6610 Enable selection of non-stop mode.
6611 @item set non-stop off
6612 Disable selection of non-stop mode.
6613 @kindex show non-stop
6615 Show the current non-stop enablement setting.
6618 Note these commands only reflect whether non-stop mode is enabled,
6619 not whether the currently-executing program is being run in non-stop mode.
6620 In particular, the @code{set non-stop} preference is only consulted when
6621 @value{GDBN} starts or connects to the target program, and it is generally
6622 not possible to switch modes once debugging has started. Furthermore,
6623 since not all targets support non-stop mode, even when you have enabled
6624 non-stop mode, @value{GDBN} may still fall back to all-stop operation by
6627 In non-stop mode, all execution commands apply only to the current thread
6628 by default. That is, @code{continue} only continues one thread.
6629 To continue all threads, issue @code{continue -a} or @code{c -a}.
6631 You can use @value{GDBN}'s background execution commands
6632 (@pxref{Background Execution}) to run some threads in the background
6633 while you continue to examine or step others from @value{GDBN}.
6634 The MI execution commands (@pxref{GDB/MI Program Execution}) are
6635 always executed asynchronously in non-stop mode.
6637 Suspending execution is done with the @code{interrupt} command when
6638 running in the background, or @kbd{Ctrl-c} during foreground execution.
6639 In all-stop mode, this stops the whole process;
6640 but in non-stop mode the interrupt applies only to the current thread.
6641 To stop the whole program, use @code{interrupt -a}.
6643 Other execution commands do not currently support the @code{-a} option.
6645 In non-stop mode, when a thread stops, @value{GDBN} doesn't automatically make
6646 that thread current, as it does in all-stop mode. This is because the
6647 thread stop notifications are asynchronous with respect to @value{GDBN}'s
6648 command interpreter, and it would be confusing if @value{GDBN} unexpectedly
6649 changed to a different thread just as you entered a command to operate on the
6650 previously current thread.
6652 @node Background Execution
6653 @subsection Background Execution
6655 @cindex foreground execution
6656 @cindex background execution
6657 @cindex asynchronous execution
6658 @cindex execution, foreground, background and asynchronous
6660 @value{GDBN}'s execution commands have two variants: the normal
6661 foreground (synchronous) behavior, and a background
6662 (asynchronous) behavior. In foreground execution, @value{GDBN} waits for
6663 the program to report that some thread has stopped before prompting for
6664 another command. In background execution, @value{GDBN} immediately gives
6665 a command prompt so that you can issue other commands while your program runs.
6667 If the target doesn't support async mode, @value{GDBN} issues an error
6668 message if you attempt to use the background execution commands.
6670 @cindex @code{&}, background execution of commands
6671 To specify background execution, add a @code{&} to the command. For example,
6672 the background form of the @code{continue} command is @code{continue&}, or
6673 just @code{c&}. The execution commands that accept background execution
6679 @xref{Starting, , Starting your Program}.
6683 @xref{Attach, , Debugging an Already-running Process}.
6687 @xref{Continuing and Stepping, step}.
6691 @xref{Continuing and Stepping, stepi}.
6695 @xref{Continuing and Stepping, next}.
6699 @xref{Continuing and Stepping, nexti}.
6703 @xref{Continuing and Stepping, continue}.
6707 @xref{Continuing and Stepping, finish}.
6711 @xref{Continuing and Stepping, until}.
6715 Background execution is especially useful in conjunction with non-stop
6716 mode for debugging programs with multiple threads; see @ref{Non-Stop Mode}.
6717 However, you can also use these commands in the normal all-stop mode with
6718 the restriction that you cannot issue another execution command until the
6719 previous one finishes. Examples of commands that are valid in all-stop
6720 mode while the program is running include @code{help} and @code{info break}.
6722 You can interrupt your program while it is running in the background by
6723 using the @code{interrupt} command.
6730 Suspend execution of the running program. In all-stop mode,
6731 @code{interrupt} stops the whole process, but in non-stop mode, it stops
6732 only the current thread. To stop the whole program in non-stop mode,
6733 use @code{interrupt -a}.
6736 @node Thread-Specific Breakpoints
6737 @subsection Thread-Specific Breakpoints
6739 When your program has multiple threads (@pxref{Threads,, Debugging
6740 Programs with Multiple Threads}), you can choose whether to set
6741 breakpoints on all threads, or on a particular thread.
6744 @cindex breakpoints and threads
6745 @cindex thread breakpoints
6746 @kindex break @dots{} thread @var{thread-id}
6747 @item break @var{location} thread @var{thread-id}
6748 @itemx break @var{location} thread @var{thread-id} if @dots{}
6749 @var{location} specifies source lines; there are several ways of
6750 writing them (@pxref{Specify Location}), but the effect is always to
6751 specify some source line.
6753 Use the qualifier @samp{thread @var{thread-id}} with a breakpoint command
6754 to specify that you only want @value{GDBN} to stop the program when a
6755 particular thread reaches this breakpoint. The @var{thread-id} specifier
6756 is one of the thread identifiers assigned by @value{GDBN}, shown
6757 in the first column of the @samp{info threads} display.
6759 If you do not specify @samp{thread @var{thread-id}} when you set a
6760 breakpoint, the breakpoint applies to @emph{all} threads of your
6763 You can use the @code{thread} qualifier on conditional breakpoints as
6764 well; in this case, place @samp{thread @var{thread-id}} before or
6765 after the breakpoint condition, like this:
6768 (@value{GDBP}) break frik.c:13 thread 28 if bartab > lim
6773 Thread-specific breakpoints are automatically deleted when
6774 @value{GDBN} detects the corresponding thread is no longer in the
6775 thread list. For example:
6779 Thread-specific breakpoint 3 deleted - thread 28 no longer in the thread list.
6782 There are several ways for a thread to disappear, such as a regular
6783 thread exit, but also when you detach from the process with the
6784 @code{detach} command (@pxref{Attach, ,Debugging an Already-running
6785 Process}), or if @value{GDBN} loses the remote connection
6786 (@pxref{Remote Debugging}), etc. Note that with some targets,
6787 @value{GDBN} is only able to detect a thread has exited when the user
6788 explictly asks for the thread list with the @code{info threads}
6791 @node Interrupted System Calls
6792 @subsection Interrupted System Calls
6794 @cindex thread breakpoints and system calls
6795 @cindex system calls and thread breakpoints
6796 @cindex premature return from system calls
6797 There is an unfortunate side effect when using @value{GDBN} to debug
6798 multi-threaded programs. If one thread stops for a
6799 breakpoint, or for some other reason, and another thread is blocked in a
6800 system call, then the system call may return prematurely. This is a
6801 consequence of the interaction between multiple threads and the signals
6802 that @value{GDBN} uses to implement breakpoints and other events that
6805 To handle this problem, your program should check the return value of
6806 each system call and react appropriately. This is good programming
6809 For example, do not write code like this:
6815 The call to @code{sleep} will return early if a different thread stops
6816 at a breakpoint or for some other reason.
6818 Instead, write this:
6823 unslept = sleep (unslept);
6826 A system call is allowed to return early, so the system is still
6827 conforming to its specification. But @value{GDBN} does cause your
6828 multi-threaded program to behave differently than it would without
6831 Also, @value{GDBN} uses internal breakpoints in the thread library to
6832 monitor certain events such as thread creation and thread destruction.
6833 When such an event happens, a system call in another thread may return
6834 prematurely, even though your program does not appear to stop.
6837 @subsection Observer Mode
6839 If you want to build on non-stop mode and observe program behavior
6840 without any chance of disruption by @value{GDBN}, you can set
6841 variables to disable all of the debugger's attempts to modify state,
6842 whether by writing memory, inserting breakpoints, etc. These operate
6843 at a low level, intercepting operations from all commands.
6845 When all of these are set to @code{off}, then @value{GDBN} is said to
6846 be @dfn{observer mode}. As a convenience, the variable
6847 @code{observer} can be set to disable these, plus enable non-stop
6850 Note that @value{GDBN} will not prevent you from making nonsensical
6851 combinations of these settings. For instance, if you have enabled
6852 @code{may-insert-breakpoints} but disabled @code{may-write-memory},
6853 then breakpoints that work by writing trap instructions into the code
6854 stream will still not be able to be placed.
6859 @item set observer on
6860 @itemx set observer off
6861 When set to @code{on}, this disables all the permission variables
6862 below (except for @code{insert-fast-tracepoints}), plus enables
6863 non-stop debugging. Setting this to @code{off} switches back to
6864 normal debugging, though remaining in non-stop mode.
6867 Show whether observer mode is on or off.
6869 @kindex may-write-registers
6870 @item set may-write-registers on
6871 @itemx set may-write-registers off
6872 This controls whether @value{GDBN} will attempt to alter the values of
6873 registers, such as with assignment expressions in @code{print}, or the
6874 @code{jump} command. It defaults to @code{on}.
6876 @item show may-write-registers
6877 Show the current permission to write registers.
6879 @kindex may-write-memory
6880 @item set may-write-memory on
6881 @itemx set may-write-memory off
6882 This controls whether @value{GDBN} will attempt to alter the contents
6883 of memory, such as with assignment expressions in @code{print}. It
6884 defaults to @code{on}.
6886 @item show may-write-memory
6887 Show the current permission to write memory.
6889 @kindex may-insert-breakpoints
6890 @item set may-insert-breakpoints on
6891 @itemx set may-insert-breakpoints off
6892 This controls whether @value{GDBN} will attempt to insert breakpoints.
6893 This affects all breakpoints, including internal breakpoints defined
6894 by @value{GDBN}. It defaults to @code{on}.
6896 @item show may-insert-breakpoints
6897 Show the current permission to insert breakpoints.
6899 @kindex may-insert-tracepoints
6900 @item set may-insert-tracepoints on
6901 @itemx set may-insert-tracepoints off
6902 This controls whether @value{GDBN} will attempt to insert (regular)
6903 tracepoints at the beginning of a tracing experiment. It affects only
6904 non-fast tracepoints, fast tracepoints being under the control of
6905 @code{may-insert-fast-tracepoints}. It defaults to @code{on}.
6907 @item show may-insert-tracepoints
6908 Show the current permission to insert tracepoints.
6910 @kindex may-insert-fast-tracepoints
6911 @item set may-insert-fast-tracepoints on
6912 @itemx set may-insert-fast-tracepoints off
6913 This controls whether @value{GDBN} will attempt to insert fast
6914 tracepoints at the beginning of a tracing experiment. It affects only
6915 fast tracepoints, regular (non-fast) tracepoints being under the
6916 control of @code{may-insert-tracepoints}. It defaults to @code{on}.
6918 @item show may-insert-fast-tracepoints
6919 Show the current permission to insert fast tracepoints.
6921 @kindex may-interrupt
6922 @item set may-interrupt on
6923 @itemx set may-interrupt off
6924 This controls whether @value{GDBN} will attempt to interrupt or stop
6925 program execution. When this variable is @code{off}, the
6926 @code{interrupt} command will have no effect, nor will
6927 @kbd{Ctrl-c}. It defaults to @code{on}.
6929 @item show may-interrupt
6930 Show the current permission to interrupt or stop the program.
6934 @node Reverse Execution
6935 @chapter Running programs backward
6936 @cindex reverse execution
6937 @cindex running programs backward
6939 When you are debugging a program, it is not unusual to realize that
6940 you have gone too far, and some event of interest has already happened.
6941 If the target environment supports it, @value{GDBN} can allow you to
6942 ``rewind'' the program by running it backward.
6944 A target environment that supports reverse execution should be able
6945 to ``undo'' the changes in machine state that have taken place as the
6946 program was executing normally. Variables, registers etc.@: should
6947 revert to their previous values. Obviously this requires a great
6948 deal of sophistication on the part of the target environment; not
6949 all target environments can support reverse execution.
6951 When a program is executed in reverse, the instructions that
6952 have most recently been executed are ``un-executed'', in reverse
6953 order. The program counter runs backward, following the previous
6954 thread of execution in reverse. As each instruction is ``un-executed'',
6955 the values of memory and/or registers that were changed by that
6956 instruction are reverted to their previous states. After executing
6957 a piece of source code in reverse, all side effects of that code
6958 should be ``undone'', and all variables should be returned to their
6959 prior values@footnote{
6960 Note that some side effects are easier to undo than others. For instance,
6961 memory and registers are relatively easy, but device I/O is hard. Some
6962 targets may be able undo things like device I/O, and some may not.
6964 The contract between @value{GDBN} and the reverse executing target
6965 requires only that the target do something reasonable when
6966 @value{GDBN} tells it to execute backwards, and then report the
6967 results back to @value{GDBN}. Whatever the target reports back to
6968 @value{GDBN}, @value{GDBN} will report back to the user. @value{GDBN}
6969 assumes that the memory and registers that the target reports are in a
6970 consistant state, but @value{GDBN} accepts whatever it is given.
6973 On some platforms, @value{GDBN} has built-in support for reverse
6974 execution, activated with the @code{record} or @code{record btrace}
6975 commands. @xref{Process Record and Replay}. Some remote targets,
6976 typically full system emulators, support reverse execution directly
6977 without requiring any special command.
6979 If you are debugging in a target environment that supports
6980 reverse execution, @value{GDBN} provides the following commands.
6983 @kindex reverse-continue
6984 @kindex rc @r{(@code{reverse-continue})}
6985 @item reverse-continue @r{[}@var{ignore-count}@r{]}
6986 @itemx rc @r{[}@var{ignore-count}@r{]}
6987 Beginning at the point where your program last stopped, start executing
6988 in reverse. Reverse execution will stop for breakpoints and synchronous
6989 exceptions (signals), just like normal execution. Behavior of
6990 asynchronous signals depends on the target environment.
6992 @kindex reverse-step
6993 @kindex rs @r{(@code{step})}
6994 @item reverse-step @r{[}@var{count}@r{]}
6995 Run the program backward until control reaches the start of a
6996 different source line; then stop it, and return control to @value{GDBN}.
6998 Like the @code{step} command, @code{reverse-step} will only stop
6999 at the beginning of a source line. It ``un-executes'' the previously
7000 executed source line. If the previous source line included calls to
7001 debuggable functions, @code{reverse-step} will step (backward) into
7002 the called function, stopping at the beginning of the @emph{last}
7003 statement in the called function (typically a return statement).
7005 Also, as with the @code{step} command, if non-debuggable functions are
7006 called, @code{reverse-step} will run thru them backward without stopping.
7008 @kindex reverse-stepi
7009 @kindex rsi @r{(@code{reverse-stepi})}
7010 @item reverse-stepi @r{[}@var{count}@r{]}
7011 Reverse-execute one machine instruction. Note that the instruction
7012 to be reverse-executed is @emph{not} the one pointed to by the program
7013 counter, but the instruction executed prior to that one. For instance,
7014 if the last instruction was a jump, @code{reverse-stepi} will take you
7015 back from the destination of the jump to the jump instruction itself.
7017 @kindex reverse-next
7018 @kindex rn @r{(@code{reverse-next})}
7019 @item reverse-next @r{[}@var{count}@r{]}
7020 Run backward to the beginning of the previous line executed in
7021 the current (innermost) stack frame. If the line contains function
7022 calls, they will be ``un-executed'' without stopping. Starting from
7023 the first line of a function, @code{reverse-next} will take you back
7024 to the caller of that function, @emph{before} the function was called,
7025 just as the normal @code{next} command would take you from the last
7026 line of a function back to its return to its caller
7027 @footnote{Unless the code is too heavily optimized.}.
7029 @kindex reverse-nexti
7030 @kindex rni @r{(@code{reverse-nexti})}
7031 @item reverse-nexti @r{[}@var{count}@r{]}
7032 Like @code{nexti}, @code{reverse-nexti} executes a single instruction
7033 in reverse, except that called functions are ``un-executed'' atomically.
7034 That is, if the previously executed instruction was a return from
7035 another function, @code{reverse-nexti} will continue to execute
7036 in reverse until the call to that function (from the current stack
7039 @kindex reverse-finish
7040 @item reverse-finish
7041 Just as the @code{finish} command takes you to the point where the
7042 current function returns, @code{reverse-finish} takes you to the point
7043 where it was called. Instead of ending up at the end of the current
7044 function invocation, you end up at the beginning.
7046 @kindex set exec-direction
7047 @item set exec-direction
7048 Set the direction of target execution.
7049 @item set exec-direction reverse
7050 @cindex execute forward or backward in time
7051 @value{GDBN} will perform all execution commands in reverse, until the
7052 exec-direction mode is changed to ``forward''. Affected commands include
7053 @code{step, stepi, next, nexti, continue, and finish}. The @code{return}
7054 command cannot be used in reverse mode.
7055 @item set exec-direction forward
7056 @value{GDBN} will perform all execution commands in the normal fashion.
7057 This is the default.
7061 @node Process Record and Replay
7062 @chapter Recording Inferior's Execution and Replaying It
7063 @cindex process record and replay
7064 @cindex recording inferior's execution and replaying it
7066 On some platforms, @value{GDBN} provides a special @dfn{process record
7067 and replay} target that can record a log of the process execution, and
7068 replay it later with both forward and reverse execution commands.
7071 When this target is in use, if the execution log includes the record
7072 for the next instruction, @value{GDBN} will debug in @dfn{replay
7073 mode}. In the replay mode, the inferior does not really execute code
7074 instructions. Instead, all the events that normally happen during
7075 code execution are taken from the execution log. While code is not
7076 really executed in replay mode, the values of registers (including the
7077 program counter register) and the memory of the inferior are still
7078 changed as they normally would. Their contents are taken from the
7082 If the record for the next instruction is not in the execution log,
7083 @value{GDBN} will debug in @dfn{record mode}. In this mode, the
7084 inferior executes normally, and @value{GDBN} records the execution log
7087 The process record and replay target supports reverse execution
7088 (@pxref{Reverse Execution}), even if the platform on which the
7089 inferior runs does not. However, the reverse execution is limited in
7090 this case by the range of the instructions recorded in the execution
7091 log. In other words, reverse execution on platforms that don't
7092 support it directly can only be done in the replay mode.
7094 When debugging in the reverse direction, @value{GDBN} will work in
7095 replay mode as long as the execution log includes the record for the
7096 previous instruction; otherwise, it will work in record mode, if the
7097 platform supports reverse execution, or stop if not.
7099 Currently, process record and replay is supported on ARM, Aarch64,
7100 Moxie, PowerPC, PowerPC64, S/390, and x86 (i386/amd64) running
7101 GNU/Linux. Process record and replay can be used both when native
7102 debugging, and when remote debugging via @code{gdbserver}.
7104 For architecture environments that support process record and replay,
7105 @value{GDBN} provides the following commands:
7108 @kindex target record
7109 @kindex target record-full
7110 @kindex target record-btrace
7113 @kindex record btrace
7114 @kindex record btrace bts
7115 @kindex record btrace pt
7121 @kindex rec btrace bts
7122 @kindex rec btrace pt
7125 @item record @var{method}
7126 This command starts the process record and replay target. The
7127 recording method can be specified as parameter. Without a parameter
7128 the command uses the @code{full} recording method. The following
7129 recording methods are available:
7133 Full record/replay recording using @value{GDBN}'s software record and
7134 replay implementation. This method allows replaying and reverse
7137 @item btrace @var{format}
7138 Hardware-supported instruction recording, supported on Intel
7139 processors. This method does not record data. Further, the data is
7140 collected in a ring buffer so old data will be overwritten when the
7141 buffer is full. It allows limited reverse execution. Variables and
7142 registers are not available during reverse execution. In remote
7143 debugging, recording continues on disconnect. Recorded data can be
7144 inspected after reconnecting. The recording may be stopped using
7147 The recording format can be specified as parameter. Without a parameter
7148 the command chooses the recording format. The following recording
7149 formats are available:
7153 @cindex branch trace store
7154 Use the @dfn{Branch Trace Store} (@acronym{BTS}) recording format. In
7155 this format, the processor stores a from/to record for each executed
7156 branch in the btrace ring buffer.
7159 @cindex Intel Processor Trace
7160 Use the @dfn{Intel Processor Trace} recording format. In this
7161 format, the processor stores the execution trace in a compressed form
7162 that is afterwards decoded by @value{GDBN}.
7164 The trace can be recorded with very low overhead. The compressed
7165 trace format also allows small trace buffers to already contain a big
7166 number of instructions compared to @acronym{BTS}.
7168 Decoding the recorded execution trace, on the other hand, is more
7169 expensive than decoding @acronym{BTS} trace. This is mostly due to the
7170 increased number of instructions to process. You should increase the
7171 buffer-size with care.
7174 Not all recording formats may be available on all processors.
7177 The process record and replay target can only debug a process that is
7178 already running. Therefore, you need first to start the process with
7179 the @kbd{run} or @kbd{start} commands, and then start the recording
7180 with the @kbd{record @var{method}} command.
7182 @cindex displaced stepping, and process record and replay
7183 Displaced stepping (@pxref{Maintenance Commands,, displaced stepping})
7184 will be automatically disabled when process record and replay target
7185 is started. That's because the process record and replay target
7186 doesn't support displaced stepping.
7188 @cindex non-stop mode, and process record and replay
7189 @cindex asynchronous execution, and process record and replay
7190 If the inferior is in the non-stop mode (@pxref{Non-Stop Mode}) or in
7191 the asynchronous execution mode (@pxref{Background Execution}), not
7192 all recording methods are available. The @code{full} recording method
7193 does not support these two modes.
7198 Stop the process record and replay target. When process record and
7199 replay target stops, the entire execution log will be deleted and the
7200 inferior will either be terminated, or will remain in its final state.
7202 When you stop the process record and replay target in record mode (at
7203 the end of the execution log), the inferior will be stopped at the
7204 next instruction that would have been recorded. In other words, if
7205 you record for a while and then stop recording, the inferior process
7206 will be left in the same state as if the recording never happened.
7208 On the other hand, if the process record and replay target is stopped
7209 while in replay mode (that is, not at the end of the execution log,
7210 but at some earlier point), the inferior process will become ``live''
7211 at that earlier state, and it will then be possible to continue the
7212 usual ``live'' debugging of the process from that state.
7214 When the inferior process exits, or @value{GDBN} detaches from it,
7215 process record and replay target will automatically stop itself.
7219 Go to a specific location in the execution log. There are several
7220 ways to specify the location to go to:
7223 @item record goto begin
7224 @itemx record goto start
7225 Go to the beginning of the execution log.
7227 @item record goto end
7228 Go to the end of the execution log.
7230 @item record goto @var{n}
7231 Go to instruction number @var{n} in the execution log.
7235 @item record save @var{filename}
7236 Save the execution log to a file @file{@var{filename}}.
7237 Default filename is @file{gdb_record.@var{process_id}}, where
7238 @var{process_id} is the process ID of the inferior.
7240 This command may not be available for all recording methods.
7242 @kindex record restore
7243 @item record restore @var{filename}
7244 Restore the execution log from a file @file{@var{filename}}.
7245 File must have been created with @code{record save}.
7247 @kindex set record full
7248 @item set record full insn-number-max @var{limit}
7249 @itemx set record full insn-number-max unlimited
7250 Set the limit of instructions to be recorded for the @code{full}
7251 recording method. Default value is 200000.
7253 If @var{limit} is a positive number, then @value{GDBN} will start
7254 deleting instructions from the log once the number of the record
7255 instructions becomes greater than @var{limit}. For every new recorded
7256 instruction, @value{GDBN} will delete the earliest recorded
7257 instruction to keep the number of recorded instructions at the limit.
7258 (Since deleting recorded instructions loses information, @value{GDBN}
7259 lets you control what happens when the limit is reached, by means of
7260 the @code{stop-at-limit} option, described below.)
7262 If @var{limit} is @code{unlimited} or zero, @value{GDBN} will never
7263 delete recorded instructions from the execution log. The number of
7264 recorded instructions is limited only by the available memory.
7266 @kindex show record full
7267 @item show record full insn-number-max
7268 Show the limit of instructions to be recorded with the @code{full}
7271 @item set record full stop-at-limit
7272 Control the behavior of the @code{full} recording method when the
7273 number of recorded instructions reaches the limit. If ON (the
7274 default), @value{GDBN} will stop when the limit is reached for the
7275 first time and ask you whether you want to stop the inferior or
7276 continue running it and recording the execution log. If you decide
7277 to continue recording, each new recorded instruction will cause the
7278 oldest one to be deleted.
7280 If this option is OFF, @value{GDBN} will automatically delete the
7281 oldest record to make room for each new one, without asking.
7283 @item show record full stop-at-limit
7284 Show the current setting of @code{stop-at-limit}.
7286 @item set record full memory-query
7287 Control the behavior when @value{GDBN} is unable to record memory
7288 changes caused by an instruction for the @code{full} recording method.
7289 If ON, @value{GDBN} will query whether to stop the inferior in that
7292 If this option is OFF (the default), @value{GDBN} will automatically
7293 ignore the effect of such instructions on memory. Later, when
7294 @value{GDBN} replays this execution log, it will mark the log of this
7295 instruction as not accessible, and it will not affect the replay
7298 @item show record full memory-query
7299 Show the current setting of @code{memory-query}.
7301 @kindex set record btrace
7302 The @code{btrace} record target does not trace data. As a
7303 convenience, when replaying, @value{GDBN} reads read-only memory off
7304 the live program directly, assuming that the addresses of the
7305 read-only areas don't change. This for example makes it possible to
7306 disassemble code while replaying, but not to print variables.
7307 In some cases, being able to inspect variables might be useful.
7308 You can use the following command for that:
7310 @item set record btrace replay-memory-access
7311 Control the behavior of the @code{btrace} recording method when
7312 accessing memory during replay. If @code{read-only} (the default),
7313 @value{GDBN} will only allow accesses to read-only memory.
7314 If @code{read-write}, @value{GDBN} will allow accesses to read-only
7315 and to read-write memory. Beware that the accessed memory corresponds
7316 to the live target and not necessarily to the current replay
7319 @item set record btrace cpu @var{identifier}
7320 Set the processor to be used for enabling workarounds for processor
7321 errata when decoding the trace.
7323 Processor errata are defects in processor operation, caused by its
7324 design or manufacture. They can cause a trace not to match the
7325 specification. This, in turn, may cause trace decode to fail.
7326 @value{GDBN} can detect erroneous trace packets and correct them, thus
7327 avoiding the decoding failures. These corrections are known as
7328 @dfn{errata workarounds}, and are enabled based on the processor on
7329 which the trace was recorded.
7331 By default, @value{GDBN} attempts to detect the processor
7332 automatically, and apply the necessary workarounds for it. However,
7333 you may need to specify the processor if @value{GDBN} does not yet
7334 support it. This command allows you to do that, and also allows to
7335 disable the workarounds.
7337 The argument @var{identifier} identifies the @sc{cpu} and is of the
7338 form: @code{@var{vendor}:@var{procesor identifier}}. In addition,
7339 there are two special identifiers, @code{none} and @code{auto}
7342 The following vendor identifiers and corresponding processor
7343 identifiers are currently supported:
7345 @multitable @columnfractions .1 .9
7348 @tab @var{family}/@var{model}[/@var{stepping}]
7352 On GNU/Linux systems, the processor @var{family}, @var{model}, and
7353 @var{stepping} can be obtained from @code{/proc/cpuinfo}.
7355 If @var{identifier} is @code{auto}, enable errata workarounds for the
7356 processor on which the trace was recorded. If @var{identifier} is
7357 @code{none}, errata workarounds are disabled.
7359 For example, when using an old @value{GDBN} on a new system, decode
7360 may fail because @value{GDBN} does not support the new processor. It
7361 often suffices to specify an older processor that @value{GDBN}
7366 Active record target: record-btrace
7367 Recording format: Intel Processor Trace.
7369 Failed to configure the Intel Processor Trace decoder: unknown cpu.
7370 (gdb) set record btrace cpu intel:6/158
7372 Active record target: record-btrace
7373 Recording format: Intel Processor Trace.
7375 Recorded 84872 instructions in 3189 functions (0 gaps) for thread 1 (...).
7378 @kindex show record btrace
7379 @item show record btrace replay-memory-access
7380 Show the current setting of @code{replay-memory-access}.
7382 @item show record btrace cpu
7383 Show the processor to be used for enabling trace decode errata
7386 @kindex set record btrace bts
7387 @item set record btrace bts buffer-size @var{size}
7388 @itemx set record btrace bts buffer-size unlimited
7389 Set the requested ring buffer size for branch tracing in @acronym{BTS}
7390 format. Default is 64KB.
7392 If @var{size} is a positive number, then @value{GDBN} will try to
7393 allocate a buffer of at least @var{size} bytes for each new thread
7394 that uses the btrace recording method and the @acronym{BTS} format.
7395 The actually obtained buffer size may differ from the requested
7396 @var{size}. Use the @code{info record} command to see the actual
7397 buffer size for each thread that uses the btrace recording method and
7398 the @acronym{BTS} format.
7400 If @var{limit} is @code{unlimited} or zero, @value{GDBN} will try to
7401 allocate a buffer of 4MB.
7403 Bigger buffers mean longer traces. On the other hand, @value{GDBN} will
7404 also need longer to process the branch trace data before it can be used.
7406 @item show record btrace bts buffer-size @var{size}
7407 Show the current setting of the requested ring buffer size for branch
7408 tracing in @acronym{BTS} format.
7410 @kindex set record btrace pt
7411 @item set record btrace pt buffer-size @var{size}
7412 @itemx set record btrace pt buffer-size unlimited
7413 Set the requested ring buffer size for branch tracing in Intel
7414 Processor Trace format. Default is 16KB.
7416 If @var{size} is a positive number, then @value{GDBN} will try to
7417 allocate a buffer of at least @var{size} bytes for each new thread
7418 that uses the btrace recording method and the Intel Processor Trace
7419 format. The actually obtained buffer size may differ from the
7420 requested @var{size}. Use the @code{info record} command to see the
7421 actual buffer size for each thread.
7423 If @var{limit} is @code{unlimited} or zero, @value{GDBN} will try to
7424 allocate a buffer of 4MB.
7426 Bigger buffers mean longer traces. On the other hand, @value{GDBN} will
7427 also need longer to process the branch trace data before it can be used.
7429 @item show record btrace pt buffer-size @var{size}
7430 Show the current setting of the requested ring buffer size for branch
7431 tracing in Intel Processor Trace format.
7435 Show various statistics about the recording depending on the recording
7440 For the @code{full} recording method, it shows the state of process
7441 record and its in-memory execution log buffer, including:
7445 Whether in record mode or replay mode.
7447 Lowest recorded instruction number (counting from when the current execution log started recording instructions).
7449 Highest recorded instruction number.
7451 Current instruction about to be replayed (if in replay mode).
7453 Number of instructions contained in the execution log.
7455 Maximum number of instructions that may be contained in the execution log.
7459 For the @code{btrace} recording method, it shows:
7465 Number of instructions that have been recorded.
7467 Number of blocks of sequential control-flow formed by the recorded
7470 Whether in record mode or replay mode.
7473 For the @code{bts} recording format, it also shows:
7476 Size of the perf ring buffer.
7479 For the @code{pt} recording format, it also shows:
7482 Size of the perf ring buffer.
7486 @kindex record delete
7489 When record target runs in replay mode (``in the past''), delete the
7490 subsequent execution log and begin to record a new execution log starting
7491 from the current address. This means you will abandon the previously
7492 recorded ``future'' and begin recording a new ``future''.
7494 @kindex record instruction-history
7495 @kindex rec instruction-history
7496 @item record instruction-history
7497 Disassembles instructions from the recorded execution log. By
7498 default, ten instructions are disassembled. This can be changed using
7499 the @code{set record instruction-history-size} command. Instructions
7500 are printed in execution order.
7502 It can also print mixed source+disassembly if you specify the the
7503 @code{/m} or @code{/s} modifier, and print the raw instructions in hex
7504 as well as in symbolic form by specifying the @code{/r} modifier.
7506 The current position marker is printed for the instruction at the
7507 current program counter value. This instruction can appear multiple
7508 times in the trace and the current position marker will be printed
7509 every time. To omit the current position marker, specify the
7512 To better align the printed instructions when the trace contains
7513 instructions from more than one function, the function name may be
7514 omitted by specifying the @code{/f} modifier.
7516 Speculatively executed instructions are prefixed with @samp{?}. This
7517 feature is not available for all recording formats.
7519 There are several ways to specify what part of the execution log to
7523 @item record instruction-history @var{insn}
7524 Disassembles ten instructions starting from instruction number
7527 @item record instruction-history @var{insn}, +/-@var{n}
7528 Disassembles @var{n} instructions around instruction number
7529 @var{insn}. If @var{n} is preceded with @code{+}, disassembles
7530 @var{n} instructions after instruction number @var{insn}. If
7531 @var{n} is preceded with @code{-}, disassembles @var{n}
7532 instructions before instruction number @var{insn}.
7534 @item record instruction-history
7535 Disassembles ten more instructions after the last disassembly.
7537 @item record instruction-history -
7538 Disassembles ten more instructions before the last disassembly.
7540 @item record instruction-history @var{begin}, @var{end}
7541 Disassembles instructions beginning with instruction number
7542 @var{begin} until instruction number @var{end}. The instruction
7543 number @var{end} is included.
7546 This command may not be available for all recording methods.
7549 @item set record instruction-history-size @var{size}
7550 @itemx set record instruction-history-size unlimited
7551 Define how many instructions to disassemble in the @code{record
7552 instruction-history} command. The default value is 10.
7553 A @var{size} of @code{unlimited} means unlimited instructions.
7556 @item show record instruction-history-size
7557 Show how many instructions to disassemble in the @code{record
7558 instruction-history} command.
7560 @kindex record function-call-history
7561 @kindex rec function-call-history
7562 @item record function-call-history
7563 Prints the execution history at function granularity. It prints one
7564 line for each sequence of instructions that belong to the same
7565 function giving the name of that function, the source lines
7566 for this instruction sequence (if the @code{/l} modifier is
7567 specified), and the instructions numbers that form the sequence (if
7568 the @code{/i} modifier is specified). The function names are indented
7569 to reflect the call stack depth if the @code{/c} modifier is
7570 specified. The @code{/l}, @code{/i}, and @code{/c} modifiers can be
7574 (@value{GDBP}) @b{list 1, 10}
7585 (@value{GDBP}) @b{record function-call-history /ilc}
7586 1 bar inst 1,4 at foo.c:6,8
7587 2 foo inst 5,10 at foo.c:2,3
7588 3 bar inst 11,13 at foo.c:9,10
7591 By default, ten lines are printed. This can be changed using the
7592 @code{set record function-call-history-size} command. Functions are
7593 printed in execution order. There are several ways to specify what
7597 @item record function-call-history @var{func}
7598 Prints ten functions starting from function number @var{func}.
7600 @item record function-call-history @var{func}, +/-@var{n}
7601 Prints @var{n} functions around function number @var{func}. If
7602 @var{n} is preceded with @code{+}, prints @var{n} functions after
7603 function number @var{func}. If @var{n} is preceded with @code{-},
7604 prints @var{n} functions before function number @var{func}.
7606 @item record function-call-history
7607 Prints ten more functions after the last ten-line print.
7609 @item record function-call-history -
7610 Prints ten more functions before the last ten-line print.
7612 @item record function-call-history @var{begin}, @var{end}
7613 Prints functions beginning with function number @var{begin} until
7614 function number @var{end}. The function number @var{end} is included.
7617 This command may not be available for all recording methods.
7619 @item set record function-call-history-size @var{size}
7620 @itemx set record function-call-history-size unlimited
7621 Define how many lines to print in the
7622 @code{record function-call-history} command. The default value is 10.
7623 A size of @code{unlimited} means unlimited lines.
7625 @item show record function-call-history-size
7626 Show how many lines to print in the
7627 @code{record function-call-history} command.
7632 @chapter Examining the Stack
7634 When your program has stopped, the first thing you need to know is where it
7635 stopped and how it got there.
7638 Each time your program performs a function call, information about the call
7640 That information includes the location of the call in your program,
7641 the arguments of the call,
7642 and the local variables of the function being called.
7643 The information is saved in a block of data called a @dfn{stack frame}.
7644 The stack frames are allocated in a region of memory called the @dfn{call
7647 When your program stops, the @value{GDBN} commands for examining the
7648 stack allow you to see all of this information.
7650 @cindex selected frame
7651 One of the stack frames is @dfn{selected} by @value{GDBN} and many
7652 @value{GDBN} commands refer implicitly to the selected frame. In
7653 particular, whenever you ask @value{GDBN} for the value of a variable in
7654 your program, the value is found in the selected frame. There are
7655 special @value{GDBN} commands to select whichever frame you are
7656 interested in. @xref{Selection, ,Selecting a Frame}.
7658 When your program stops, @value{GDBN} automatically selects the
7659 currently executing frame and describes it briefly, similar to the
7660 @code{frame} command (@pxref{Frame Info, ,Information about a Frame}).
7663 * Frames:: Stack frames
7664 * Backtrace:: Backtraces
7665 * Selection:: Selecting a frame
7666 * Frame Info:: Information on a frame
7667 * Frame Apply:: Applying a command to several frames
7668 * Frame Filter Management:: Managing frame filters
7673 @section Stack Frames
7675 @cindex frame, definition
7677 The call stack is divided up into contiguous pieces called @dfn{stack
7678 frames}, or @dfn{frames} for short; each frame is the data associated
7679 with one call to one function. The frame contains the arguments given
7680 to the function, the function's local variables, and the address at
7681 which the function is executing.
7683 @cindex initial frame
7684 @cindex outermost frame
7685 @cindex innermost frame
7686 When your program is started, the stack has only one frame, that of the
7687 function @code{main}. This is called the @dfn{initial} frame or the
7688 @dfn{outermost} frame. Each time a function is called, a new frame is
7689 made. Each time a function returns, the frame for that function invocation
7690 is eliminated. If a function is recursive, there can be many frames for
7691 the same function. The frame for the function in which execution is
7692 actually occurring is called the @dfn{innermost} frame. This is the most
7693 recently created of all the stack frames that still exist.
7695 @cindex frame pointer
7696 Inside your program, stack frames are identified by their addresses. A
7697 stack frame consists of many bytes, each of which has its own address; each
7698 kind of computer has a convention for choosing one byte whose
7699 address serves as the address of the frame. Usually this address is kept
7700 in a register called the @dfn{frame pointer register}
7701 (@pxref{Registers, $fp}) while execution is going on in that frame.
7704 @cindex frame number
7705 @value{GDBN} labels each existing stack frame with a @dfn{level}, a
7706 number that is zero for the innermost frame, one for the frame that
7707 called it, and so on upward. These level numbers give you a way of
7708 designating stack frames in @value{GDBN} commands. The terms
7709 @dfn{frame number} and @dfn{frame level} can be used interchangeably to
7710 describe this number.
7712 @c The -fomit-frame-pointer below perennially causes hbox overflow
7713 @c underflow problems.
7714 @cindex frameless execution
7715 Some compilers provide a way to compile functions so that they operate
7716 without stack frames. (For example, the @value{NGCC} option
7718 @samp{-fomit-frame-pointer}
7720 generates functions without a frame.)
7721 This is occasionally done with heavily used library functions to save
7722 the frame setup time. @value{GDBN} has limited facilities for dealing
7723 with these function invocations. If the innermost function invocation
7724 has no stack frame, @value{GDBN} nevertheless regards it as though
7725 it had a separate frame, which is numbered zero as usual, allowing
7726 correct tracing of the function call chain. However, @value{GDBN} has
7727 no provision for frameless functions elsewhere in the stack.
7733 @cindex call stack traces
7734 A backtrace is a summary of how your program got where it is. It shows one
7735 line per frame, for many frames, starting with the currently executing
7736 frame (frame zero), followed by its caller (frame one), and on up the
7739 @anchor{backtrace-command}
7741 @kindex bt @r{(@code{backtrace})}
7742 To print a backtrace of the entire stack, use the @code{backtrace}
7743 command, or its alias @code{bt}. This command will print one line per
7744 frame for frames in the stack. By default, all stack frames are
7745 printed. You can stop the backtrace at any time by typing the system
7746 interrupt character, normally @kbd{Ctrl-c}.
7749 @item backtrace [@var{option}]@dots{} [@var{qualifier}]@dots{} [@var{count}]
7750 @itemx bt [@var{option}]@dots{} [@var{qualifier}]@dots{} [@var{count}]
7751 Print the backtrace of the entire stack.
7753 The optional @var{count} can be one of the following:
7758 Print only the innermost @var{n} frames, where @var{n} is a positive
7763 Print only the outermost @var{n} frames, where @var{n} is a positive
7771 Print the values of the local variables also. This can be combined
7772 with the optional @var{count} to limit the number of frames shown.
7775 Do not run Python frame filters on this backtrace. @xref{Frame
7776 Filter API}, for more information. Additionally use @ref{disable
7777 frame-filter all} to turn off all frame filters. This is only
7778 relevant when @value{GDBN} has been configured with @code{Python}
7782 A Python frame filter might decide to ``elide'' some frames. Normally
7783 such elided frames are still printed, but they are indented relative
7784 to the filtered frames that cause them to be elided. The @code{-hide}
7785 option causes elided frames to not be printed at all.
7788 The @code{backtrace} command also supports a number of options that
7789 allow overriding relevant global print settings as set by @code{set
7790 backtrace} and @code{set print} subcommands:
7793 @item -past-main [@code{on}|@code{off}]
7794 Set whether backtraces should continue past @code{main}. Related setting:
7795 @ref{set backtrace past-main}.
7797 @item -past-entry [@code{on}|@code{off}]
7798 Set whether backtraces should continue past the entry point of a program.
7799 Related setting: @ref{set backtrace past-entry}.
7801 @item -entry-values @code{no}|@code{only}|@code{preferred}|@code{if-needed}|@code{both}|@code{compact}|@code{default}
7802 Set printing of function arguments at function entry.
7803 Related setting: @ref{set print entry-values}.
7805 @item -frame-arguments @code{all}|@code{scalars}|@code{none}
7806 Set printing of non-scalar frame arguments.
7807 Related setting: @ref{set print frame-arguments}.
7809 @item -raw-frame-arguments [@code{on}|@code{off}]
7810 Set whether to print frame arguments in raw form.
7811 Related setting: @ref{set print raw-frame-arguments}.
7813 @item -frame-info @code{auto}|@code{source-line}|@code{location}|@code{source-and-location}|@code{location-and-address}|@code{short-location}
7814 Set printing of frame information.
7815 Related setting: @ref{set print frame-info}.
7818 The optional @var{qualifier} is maintained for backward compatibility.
7819 It can be one of the following:
7823 Equivalent to the @code{-full} option.
7826 Equivalent to the @code{-no-filters} option.
7829 Equivalent to the @code{-hide} option.
7836 The names @code{where} and @code{info stack} (abbreviated @code{info s})
7837 are additional aliases for @code{backtrace}.
7839 @cindex multiple threads, backtrace
7840 In a multi-threaded program, @value{GDBN} by default shows the
7841 backtrace only for the current thread. To display the backtrace for
7842 several or all of the threads, use the command @code{thread apply}
7843 (@pxref{Threads, thread apply}). For example, if you type @kbd{thread
7844 apply all backtrace}, @value{GDBN} will display the backtrace for all
7845 the threads; this is handy when you debug a core dump of a
7846 multi-threaded program.
7848 Each line in the backtrace shows the frame number and the function name.
7849 The program counter value is also shown---unless you use @code{set
7850 print address off}. The backtrace also shows the source file name and
7851 line number, as well as the arguments to the function. The program
7852 counter value is omitted if it is at the beginning of the code for that
7855 Here is an example of a backtrace. It was made with the command
7856 @samp{bt 3}, so it shows the innermost three frames.
7860 #0 m4_traceon (obs=0x24eb0, argc=1, argv=0x2b8c8)
7862 #1 0x6e38 in expand_macro (sym=0x2b600, data=...) at macro.c:242
7863 #2 0x6840 in expand_token (obs=0x0, t=177664, td=0xf7fffb08)
7865 (More stack frames follow...)
7870 The display for frame zero does not begin with a program counter
7871 value, indicating that your program has stopped at the beginning of the
7872 code for line @code{993} of @code{builtin.c}.
7875 The value of parameter @code{data} in frame 1 has been replaced by
7876 @code{@dots{}}. By default, @value{GDBN} prints the value of a parameter
7877 only if it is a scalar (integer, pointer, enumeration, etc). See command
7878 @kbd{set print frame-arguments} in @ref{Print Settings} for more details
7879 on how to configure the way function parameter values are printed.
7880 The command @kbd{set print frame-info} (@pxref{Print Settings}) controls
7881 what frame information is printed.
7883 @cindex optimized out, in backtrace
7884 @cindex function call arguments, optimized out
7885 If your program was compiled with optimizations, some compilers will
7886 optimize away arguments passed to functions if those arguments are
7887 never used after the call. Such optimizations generate code that
7888 passes arguments through registers, but doesn't store those arguments
7889 in the stack frame. @value{GDBN} has no way of displaying such
7890 arguments in stack frames other than the innermost one. Here's what
7891 such a backtrace might look like:
7895 #0 m4_traceon (obs=0x24eb0, argc=1, argv=0x2b8c8)
7897 #1 0x6e38 in expand_macro (sym=<optimized out>) at macro.c:242
7898 #2 0x6840 in expand_token (obs=0x0, t=<optimized out>, td=0xf7fffb08)
7900 (More stack frames follow...)
7905 The values of arguments that were not saved in their stack frames are
7906 shown as @samp{<optimized out>}.
7908 If you need to display the values of such optimized-out arguments,
7909 either deduce that from other variables whose values depend on the one
7910 you are interested in, or recompile without optimizations.
7912 @cindex backtrace beyond @code{main} function
7913 @cindex program entry point
7914 @cindex startup code, and backtrace
7915 Most programs have a standard user entry point---a place where system
7916 libraries and startup code transition into user code. For C this is
7917 @code{main}@footnote{
7918 Note that embedded programs (the so-called ``free-standing''
7919 environment) are not required to have a @code{main} function as the
7920 entry point. They could even have multiple entry points.}.
7921 When @value{GDBN} finds the entry function in a backtrace
7922 it will terminate the backtrace, to avoid tracing into highly
7923 system-specific (and generally uninteresting) code.
7925 If you need to examine the startup code, or limit the number of levels
7926 in a backtrace, you can change this behavior:
7929 @item set backtrace past-main
7930 @itemx set backtrace past-main on
7931 @anchor{set backtrace past-main}
7932 @kindex set backtrace
7933 Backtraces will continue past the user entry point.
7935 @item set backtrace past-main off
7936 Backtraces will stop when they encounter the user entry point. This is the
7939 @item show backtrace past-main
7940 @kindex show backtrace
7941 Display the current user entry point backtrace policy.
7943 @item set backtrace past-entry
7944 @itemx set backtrace past-entry on
7945 @anchor{set backtrace past-entry}
7946 Backtraces will continue past the internal entry point of an application.
7947 This entry point is encoded by the linker when the application is built,
7948 and is likely before the user entry point @code{main} (or equivalent) is called.
7950 @item set backtrace past-entry off
7951 Backtraces will stop when they encounter the internal entry point of an
7952 application. This is the default.
7954 @item show backtrace past-entry
7955 Display the current internal entry point backtrace policy.
7957 @item set backtrace limit @var{n}
7958 @itemx set backtrace limit 0
7959 @itemx set backtrace limit unlimited
7960 @anchor{set backtrace limit}
7961 @cindex backtrace limit
7962 Limit the backtrace to @var{n} levels. A value of @code{unlimited}
7963 or zero means unlimited levels.
7965 @item show backtrace limit
7966 Display the current limit on backtrace levels.
7969 You can control how file names are displayed.
7972 @item set filename-display
7973 @itemx set filename-display relative
7974 @cindex filename-display
7975 Display file names relative to the compilation directory. This is the default.
7977 @item set filename-display basename
7978 Display only basename of a filename.
7980 @item set filename-display absolute
7981 Display an absolute filename.
7983 @item show filename-display
7984 Show the current way to display filenames.
7988 @section Selecting a Frame
7990 Most commands for examining the stack and other data in your program work on
7991 whichever stack frame is selected at the moment. Here are the commands for
7992 selecting a stack frame; all of them finish by printing a brief description
7993 of the stack frame just selected.
7996 @kindex frame@r{, selecting}
7997 @kindex f @r{(@code{frame})}
7998 @item frame @r{[} @var{frame-selection-spec} @r{]}
7999 @item f @r{[} @var{frame-selection-spec} @r{]}
8000 The @command{frame} command allows different stack frames to be
8001 selected. The @var{frame-selection-spec} can be any of the following:
8006 @item level @var{num}
8007 Select frame level @var{num}. Recall that frame zero is the innermost
8008 (currently executing) frame, frame one is the frame that called the
8009 innermost one, and so on. The highest level frame is usually the one
8012 As this is the most common method of navigating the frame stack, the
8013 string @command{level} can be omitted. For example, the following two
8014 commands are equivalent:
8017 (@value{GDBP}) frame 3
8018 (@value{GDBP}) frame level 3
8021 @kindex frame address
8022 @item address @var{stack-address}
8023 Select the frame with stack address @var{stack-address}. The
8024 @var{stack-address} for a frame can be seen in the output of
8025 @command{info frame}, for example:
8029 Stack level 1, frame at 0x7fffffffda30:
8030 rip = 0x40066d in b (amd64-entry-value.cc:59); saved rip 0x4004c5
8031 tail call frame, caller of frame at 0x7fffffffda30
8032 source language c++.
8033 Arglist at unknown address.
8034 Locals at unknown address, Previous frame's sp is 0x7fffffffda30
8037 The @var{stack-address} for this frame is @code{0x7fffffffda30} as
8038 indicated by the line:
8041 Stack level 1, frame at 0x7fffffffda30:
8044 @kindex frame function
8045 @item function @var{function-name}
8046 Select the stack frame for function @var{function-name}. If there are
8047 multiple stack frames for function @var{function-name} then the inner
8048 most stack frame is selected.
8051 @item view @var{stack-address} @r{[} @var{pc-addr} @r{]}
8052 View a frame that is not part of @value{GDBN}'s backtrace. The frame
8053 viewed has stack address @var{stack-addr}, and optionally, a program
8054 counter address of @var{pc-addr}.
8056 This is useful mainly if the chaining of stack frames has been
8057 damaged by a bug, making it impossible for @value{GDBN} to assign
8058 numbers properly to all frames. In addition, this can be useful
8059 when your program has multiple stacks and switches between them.
8061 When viewing a frame outside the current backtrace using
8062 @command{frame view} then you can always return to the original
8063 stack using one of the previous stack frame selection instructions,
8064 for example @command{frame level 0}.
8070 Move @var{n} frames up the stack; @var{n} defaults to 1. For positive
8071 numbers @var{n}, this advances toward the outermost frame, to higher
8072 frame numbers, to frames that have existed longer.
8075 @kindex do @r{(@code{down})}
8077 Move @var{n} frames down the stack; @var{n} defaults to 1. For
8078 positive numbers @var{n}, this advances toward the innermost frame, to
8079 lower frame numbers, to frames that were created more recently.
8080 You may abbreviate @code{down} as @code{do}.
8083 All of these commands end by printing two lines of output describing the
8084 frame. The first line shows the frame number, the function name, the
8085 arguments, and the source file and line number of execution in that
8086 frame. The second line shows the text of that source line.
8094 #1 0x22f0 in main (argc=1, argv=0xf7fffbf4, env=0xf7fffbfc)
8096 10 read_input_file (argv[i]);
8100 After such a printout, the @code{list} command with no arguments
8101 prints ten lines centered on the point of execution in the frame.
8102 You can also edit the program at the point of execution with your favorite
8103 editing program by typing @code{edit}.
8104 @xref{List, ,Printing Source Lines},
8108 @kindex select-frame
8109 @item select-frame @r{[} @var{frame-selection-spec} @r{]}
8110 The @code{select-frame} command is a variant of @code{frame} that does
8111 not display the new frame after selecting it. This command is
8112 intended primarily for use in @value{GDBN} command scripts, where the
8113 output might be unnecessary and distracting. The
8114 @var{frame-selection-spec} is as for the @command{frame} command
8115 described in @ref{Selection, ,Selecting a Frame}.
8117 @kindex down-silently
8119 @item up-silently @var{n}
8120 @itemx down-silently @var{n}
8121 These two commands are variants of @code{up} and @code{down},
8122 respectively; they differ in that they do their work silently, without
8123 causing display of the new frame. They are intended primarily for use
8124 in @value{GDBN} command scripts, where the output might be unnecessary and
8129 @section Information About a Frame
8131 There are several other commands to print information about the selected
8137 When used without any argument, this command does not change which
8138 frame is selected, but prints a brief description of the currently
8139 selected stack frame. It can be abbreviated @code{f}. With an
8140 argument, this command is used to select a stack frame.
8141 @xref{Selection, ,Selecting a Frame}.
8144 @kindex info f @r{(@code{info frame})}
8147 This command prints a verbose description of the selected stack frame,
8152 the address of the frame
8154 the address of the next frame down (called by this frame)
8156 the address of the next frame up (caller of this frame)
8158 the language in which the source code corresponding to this frame is written
8160 the address of the frame's arguments
8162 the address of the frame's local variables
8164 the program counter saved in it (the address of execution in the caller frame)
8166 which registers were saved in the frame
8169 @noindent The verbose description is useful when
8170 something has gone wrong that has made the stack format fail to fit
8171 the usual conventions.
8173 @item info frame @r{[} @var{frame-selection-spec} @r{]}
8174 @itemx info f @r{[} @var{frame-selection-spec} @r{]}
8175 Print a verbose description of the frame selected by
8176 @var{frame-selection-spec}. The @var{frame-selection-spec} is the
8177 same as for the @command{frame} command (@pxref{Selection, ,Selecting
8178 a Frame}). The selected frame remains unchanged by this command.
8181 @item info args [-q]
8182 Print the arguments of the selected frame, each on a separate line.
8184 The optional flag @samp{-q}, which stands for @samp{quiet}, disables
8185 printing header information and messages explaining why no argument
8188 @item info args [-q] [-t @var{type_regexp}] [@var{regexp}]
8189 Like @kbd{info args}, but only print the arguments selected
8190 with the provided regexp(s).
8192 If @var{regexp} is provided, print only the arguments whose names
8193 match the regular expression @var{regexp}.
8195 If @var{type_regexp} is provided, print only the arguments whose
8196 types, as printed by the @code{whatis} command, match
8197 the regular expression @var{type_regexp}.
8198 If @var{type_regexp} contains space(s), it should be enclosed in
8199 quote characters. If needed, use backslash to escape the meaning
8200 of special characters or quotes.
8202 If both @var{regexp} and @var{type_regexp} are provided, an argument
8203 is printed only if its name matches @var{regexp} and its type matches
8206 @item info locals [-q]
8208 Print the local variables of the selected frame, each on a separate
8209 line. These are all variables (declared either static or automatic)
8210 accessible at the point of execution of the selected frame.
8212 The optional flag @samp{-q}, which stands for @samp{quiet}, disables
8213 printing header information and messages explaining why no local variables
8216 @item info locals [-q] [-t @var{type_regexp}] [@var{regexp}]
8217 Like @kbd{info locals}, but only print the local variables selected
8218 with the provided regexp(s).
8220 If @var{regexp} is provided, print only the local variables whose names
8221 match the regular expression @var{regexp}.
8223 If @var{type_regexp} is provided, print only the local variables whose
8224 types, as printed by the @code{whatis} command, match
8225 the regular expression @var{type_regexp}.
8226 If @var{type_regexp} contains space(s), it should be enclosed in
8227 quote characters. If needed, use backslash to escape the meaning
8228 of special characters or quotes.
8230 If both @var{regexp} and @var{type_regexp} are provided, a local variable
8231 is printed only if its name matches @var{regexp} and its type matches
8234 The command @kbd{info locals -q -t @var{type_regexp}} can usefully be
8235 combined with the commands @kbd{frame apply} and @kbd{thread apply}.
8236 For example, your program might use Resource Acquisition Is
8237 Initialization types (RAII) such as @code{lock_something_t}: each
8238 local variable of type @code{lock_something_t} automatically places a
8239 lock that is destroyed when the variable goes out of scope. You can
8240 then list all acquired locks in your program by doing
8242 thread apply all -s frame apply all -s info locals -q -t lock_something_t
8245 or the equivalent shorter form
8247 tfaas i lo -q -t lock_something_t
8253 @section Applying a Command to Several Frames.
8254 @anchor{frame apply}
8256 @cindex apply command to several frames
8258 @item frame apply [all | @var{count} | @var{-count} | level @var{level}@dots{}] [@var{option}]@dots{} @var{command}
8259 The @code{frame apply} command allows you to apply the named
8260 @var{command} to one or more frames.
8264 Specify @code{all} to apply @var{command} to all frames.
8267 Use @var{count} to apply @var{command} to the innermost @var{count}
8268 frames, where @var{count} is a positive number.
8271 Use @var{-count} to apply @var{command} to the outermost @var{count}
8272 frames, where @var{count} is a positive number.
8275 Use @code{level} to apply @var{command} to the set of frames identified
8276 by the @var{level} list. @var{level} is a frame level or a range of frame
8277 levels as @var{level1}-@var{level2}. The frame level is the number shown
8278 in the first field of the @samp{backtrace} command output.
8279 E.g., @samp{2-4 6-8 3} indicates to apply @var{command} for the frames
8280 at levels 2, 3, 4, 6, 7, 8, and then again on frame at level 3.
8284 Note that the frames on which @code{frame apply} applies a command are
8285 also influenced by the @code{set backtrace} settings such as @code{set
8286 backtrace past-main} and @code{set backtrace limit N}.
8287 @xref{Backtrace,,Backtraces}.
8289 The @code{frame apply} command also supports a number of options that
8290 allow overriding relevant @code{set backtrace} settings:
8293 @item -past-main [@code{on}|@code{off}]
8294 Whether backtraces should continue past @code{main}.
8295 Related setting: @ref{set backtrace past-main}.
8297 @item -past-entry [@code{on}|@code{off}]
8298 Whether backtraces should continue past the entry point of a program.
8299 Related setting: @ref{set backtrace past-entry}.
8302 By default, @value{GDBN} displays some frame information before the
8303 output produced by @var{command}, and an error raised during the
8304 execution of a @var{command} will abort @code{frame apply}. The
8305 following options can be used to fine-tune these behaviors:
8309 The flag @code{-c}, which stands for @samp{continue}, causes any
8310 errors in @var{command} to be displayed, and the execution of
8311 @code{frame apply} then continues.
8313 The flag @code{-s}, which stands for @samp{silent}, causes any errors
8314 or empty output produced by a @var{command} to be silently ignored.
8315 That is, the execution continues, but the frame information and errors
8318 The flag @code{-q} (@samp{quiet}) disables printing the frame
8322 The following example shows how the flags @code{-c} and @code{-s} are
8323 working when applying the command @code{p j} to all frames, where
8324 variable @code{j} can only be successfully printed in the outermost
8325 @code{#1 main} frame.
8329 (gdb) frame apply all p j
8330 #0 some_function (i=5) at fun.c:4
8331 No symbol "j" in current context.
8332 (gdb) frame apply all -c p j
8333 #0 some_function (i=5) at fun.c:4
8334 No symbol "j" in current context.
8335 #1 0x565555fb in main (argc=1, argv=0xffffd2c4) at fun.c:11
8337 (gdb) frame apply all -s p j
8338 #1 0x565555fb in main (argc=1, argv=0xffffd2c4) at fun.c:11
8344 By default, @samp{frame apply}, prints the frame location
8345 information before the command output:
8349 (gdb) frame apply all p $sp
8350 #0 some_function (i=5) at fun.c:4
8351 $4 = (void *) 0xffffd1e0
8352 #1 0x565555fb in main (argc=1, argv=0xffffd2c4) at fun.c:11
8353 $5 = (void *) 0xffffd1f0
8358 If the flag @code{-q} is given, no frame information is printed:
8361 (gdb) frame apply all -q p $sp
8362 $12 = (void *) 0xffffd1e0
8363 $13 = (void *) 0xffffd1f0
8373 @cindex apply a command to all frames (ignoring errors and empty output)
8374 @item faas @var{command}
8375 Shortcut for @code{frame apply all -s @var{command}}.
8376 Applies @var{command} on all frames, ignoring errors and empty output.
8378 It can for example be used to print a local variable or a function
8379 argument without knowing the frame where this variable or argument
8382 (@value{GDBP}) faas p some_local_var_i_do_not_remember_where_it_is
8385 The @code{faas} command accepts the same options as the @code{frame
8386 apply} command. @xref{frame apply}.
8388 Note that the command @code{tfaas @var{command}} applies @var{command}
8389 on all frames of all threads. See @xref{Threads,,Threads}.
8393 @node Frame Filter Management
8394 @section Management of Frame Filters.
8395 @cindex managing frame filters
8397 Frame filters are Python based utilities to manage and decorate the
8398 output of frames. @xref{Frame Filter API}, for further information.
8400 Managing frame filters is performed by several commands available
8401 within @value{GDBN}, detailed here.
8404 @kindex info frame-filter
8405 @item info frame-filter
8406 Print a list of installed frame filters from all dictionaries, showing
8407 their name, priority and enabled status.
8409 @kindex disable frame-filter
8410 @anchor{disable frame-filter all}
8411 @item disable frame-filter @var{filter-dictionary} @var{filter-name}
8412 Disable a frame filter in the dictionary matching
8413 @var{filter-dictionary} and @var{filter-name}. The
8414 @var{filter-dictionary} may be @code{all}, @code{global},
8415 @code{progspace}, or the name of the object file where the frame filter
8416 dictionary resides. When @code{all} is specified, all frame filters
8417 across all dictionaries are disabled. The @var{filter-name} is the name
8418 of the frame filter and is used when @code{all} is not the option for
8419 @var{filter-dictionary}. A disabled frame-filter is not deleted, it
8420 may be enabled again later.
8422 @kindex enable frame-filter
8423 @item enable frame-filter @var{filter-dictionary} @var{filter-name}
8424 Enable a frame filter in the dictionary matching
8425 @var{filter-dictionary} and @var{filter-name}. The
8426 @var{filter-dictionary} may be @code{all}, @code{global},
8427 @code{progspace} or the name of the object file where the frame filter
8428 dictionary resides. When @code{all} is specified, all frame filters across
8429 all dictionaries are enabled. The @var{filter-name} is the name of the frame
8430 filter and is used when @code{all} is not the option for
8431 @var{filter-dictionary}.
8436 (gdb) info frame-filter
8438 global frame-filters:
8439 Priority Enabled Name
8440 1000 No PrimaryFunctionFilter
8443 progspace /build/test frame-filters:
8444 Priority Enabled Name
8445 100 Yes ProgspaceFilter
8447 objfile /build/test frame-filters:
8448 Priority Enabled Name
8449 999 Yes BuildProgra Filter
8451 (gdb) disable frame-filter /build/test BuildProgramFilter
8452 (gdb) info frame-filter
8454 global frame-filters:
8455 Priority Enabled Name
8456 1000 No PrimaryFunctionFilter
8459 progspace /build/test frame-filters:
8460 Priority Enabled Name
8461 100 Yes ProgspaceFilter
8463 objfile /build/test frame-filters:
8464 Priority Enabled Name
8465 999 No BuildProgramFilter
8467 (gdb) enable frame-filter global PrimaryFunctionFilter
8468 (gdb) info frame-filter
8470 global frame-filters:
8471 Priority Enabled Name
8472 1000 Yes PrimaryFunctionFilter
8475 progspace /build/test frame-filters:
8476 Priority Enabled Name
8477 100 Yes ProgspaceFilter
8479 objfile /build/test frame-filters:
8480 Priority Enabled Name
8481 999 No BuildProgramFilter
8484 @kindex set frame-filter priority
8485 @item set frame-filter priority @var{filter-dictionary} @var{filter-name} @var{priority}
8486 Set the @var{priority} of a frame filter in the dictionary matching
8487 @var{filter-dictionary}, and the frame filter name matching
8488 @var{filter-name}. The @var{filter-dictionary} may be @code{global},
8489 @code{progspace} or the name of the object file where the frame filter
8490 dictionary resides. The @var{priority} is an integer.
8492 @kindex show frame-filter priority
8493 @item show frame-filter priority @var{filter-dictionary} @var{filter-name}
8494 Show the @var{priority} of a frame filter in the dictionary matching
8495 @var{filter-dictionary}, and the frame filter name matching
8496 @var{filter-name}. The @var{filter-dictionary} may be @code{global},
8497 @code{progspace} or the name of the object file where the frame filter
8503 (gdb) info frame-filter
8505 global frame-filters:
8506 Priority Enabled Name
8507 1000 Yes PrimaryFunctionFilter
8510 progspace /build/test frame-filters:
8511 Priority Enabled Name
8512 100 Yes ProgspaceFilter
8514 objfile /build/test frame-filters:
8515 Priority Enabled Name
8516 999 No BuildProgramFilter
8518 (gdb) set frame-filter priority global Reverse 50
8519 (gdb) info frame-filter
8521 global frame-filters:
8522 Priority Enabled Name
8523 1000 Yes PrimaryFunctionFilter
8526 progspace /build/test frame-filters:
8527 Priority Enabled Name
8528 100 Yes ProgspaceFilter
8530 objfile /build/test frame-filters:
8531 Priority Enabled Name
8532 999 No BuildProgramFilter
8537 @chapter Examining Source Files
8539 @value{GDBN} can print parts of your program's source, since the debugging
8540 information recorded in the program tells @value{GDBN} what source files were
8541 used to build it. When your program stops, @value{GDBN} spontaneously prints
8542 the line where it stopped. Likewise, when you select a stack frame
8543 (@pxref{Selection, ,Selecting a Frame}), @value{GDBN} prints the line where
8544 execution in that frame has stopped. You can print other portions of
8545 source files by explicit command.
8547 If you use @value{GDBN} through its @sc{gnu} Emacs interface, you may
8548 prefer to use Emacs facilities to view source; see @ref{Emacs, ,Using
8549 @value{GDBN} under @sc{gnu} Emacs}.
8552 * List:: Printing source lines
8553 * Specify Location:: How to specify code locations
8554 * Edit:: Editing source files
8555 * Search:: Searching source files
8556 * Source Path:: Specifying source directories
8557 * Machine Code:: Source and machine code
8561 @section Printing Source Lines
8564 @kindex l @r{(@code{list})}
8565 To print lines from a source file, use the @code{list} command
8566 (abbreviated @code{l}). By default, ten lines are printed.
8567 There are several ways to specify what part of the file you want to
8568 print; see @ref{Specify Location}, for the full list.
8570 Here are the forms of the @code{list} command most commonly used:
8573 @item list @var{linenum}
8574 Print lines centered around line number @var{linenum} in the
8575 current source file.
8577 @item list @var{function}
8578 Print lines centered around the beginning of function
8582 Print more lines. If the last lines printed were printed with a
8583 @code{list} command, this prints lines following the last lines
8584 printed; however, if the last line printed was a solitary line printed
8585 as part of displaying a stack frame (@pxref{Stack, ,Examining the
8586 Stack}), this prints lines centered around that line.
8589 Print lines just before the lines last printed.
8592 @cindex @code{list}, how many lines to display
8593 By default, @value{GDBN} prints ten source lines with any of these forms of
8594 the @code{list} command. You can change this using @code{set listsize}:
8597 @kindex set listsize
8598 @item set listsize @var{count}
8599 @itemx set listsize unlimited
8600 Make the @code{list} command display @var{count} source lines (unless
8601 the @code{list} argument explicitly specifies some other number).
8602 Setting @var{count} to @code{unlimited} or 0 means there's no limit.
8604 @kindex show listsize
8606 Display the number of lines that @code{list} prints.
8609 Repeating a @code{list} command with @key{RET} discards the argument,
8610 so it is equivalent to typing just @code{list}. This is more useful
8611 than listing the same lines again. An exception is made for an
8612 argument of @samp{-}; that argument is preserved in repetition so that
8613 each repetition moves up in the source file.
8615 In general, the @code{list} command expects you to supply zero, one or two
8616 @dfn{locations}. Locations specify source lines; there are several ways
8617 of writing them (@pxref{Specify Location}), but the effect is always
8618 to specify some source line.
8620 Here is a complete description of the possible arguments for @code{list}:
8623 @item list @var{location}
8624 Print lines centered around the line specified by @var{location}.
8626 @item list @var{first},@var{last}
8627 Print lines from @var{first} to @var{last}. Both arguments are
8628 locations. When a @code{list} command has two locations, and the
8629 source file of the second location is omitted, this refers to
8630 the same source file as the first location.
8632 @item list ,@var{last}
8633 Print lines ending with @var{last}.
8635 @item list @var{first},
8636 Print lines starting with @var{first}.
8639 Print lines just after the lines last printed.
8642 Print lines just before the lines last printed.
8645 As described in the preceding table.
8648 @node Specify Location
8649 @section Specifying a Location
8650 @cindex specifying location
8652 @cindex source location
8655 * Linespec Locations:: Linespec locations
8656 * Explicit Locations:: Explicit locations
8657 * Address Locations:: Address locations
8660 Several @value{GDBN} commands accept arguments that specify a location
8661 of your program's code. Since @value{GDBN} is a source-level
8662 debugger, a location usually specifies some line in the source code.
8663 Locations may be specified using three different formats:
8664 linespec locations, explicit locations, or address locations.
8666 @node Linespec Locations
8667 @subsection Linespec Locations
8668 @cindex linespec locations
8670 A @dfn{linespec} is a colon-separated list of source location parameters such
8671 as file name, function name, etc. Here are all the different ways of
8672 specifying a linespec:
8676 Specifies the line number @var{linenum} of the current source file.
8679 @itemx +@var{offset}
8680 Specifies the line @var{offset} lines before or after the @dfn{current
8681 line}. For the @code{list} command, the current line is the last one
8682 printed; for the breakpoint commands, this is the line at which
8683 execution stopped in the currently selected @dfn{stack frame}
8684 (@pxref{Frames, ,Frames}, for a description of stack frames.) When
8685 used as the second of the two linespecs in a @code{list} command,
8686 this specifies the line @var{offset} lines up or down from the first
8689 @item @var{filename}:@var{linenum}
8690 Specifies the line @var{linenum} in the source file @var{filename}.
8691 If @var{filename} is a relative file name, then it will match any
8692 source file name with the same trailing components. For example, if
8693 @var{filename} is @samp{gcc/expr.c}, then it will match source file
8694 name of @file{/build/trunk/gcc/expr.c}, but not
8695 @file{/build/trunk/libcpp/expr.c} or @file{/build/trunk/gcc/x-expr.c}.
8697 @item @var{function}
8698 Specifies the line that begins the body of the function @var{function}.
8699 For example, in C, this is the line with the open brace.
8701 By default, in C@t{++} and Ada, @var{function} is interpreted as
8702 specifying all functions named @var{function} in all scopes. For
8703 C@t{++}, this means in all namespaces and classes. For Ada, this
8704 means in all packages.
8706 For example, assuming a program with C@t{++} symbols named
8707 @code{A::B::func} and @code{B::func}, both commands @w{@kbd{break
8708 func}} and @w{@kbd{break B::func}} set a breakpoint on both symbols.
8710 Commands that accept a linespec let you override this with the
8711 @code{-qualified} option. For example, @w{@kbd{break -qualified
8712 func}} sets a breakpoint on a free-function named @code{func} ignoring
8713 any C@t{++} class methods and namespace functions called @code{func}.
8715 @xref{Explicit Locations}.
8717 @item @var{function}:@var{label}
8718 Specifies the line where @var{label} appears in @var{function}.
8720 @item @var{filename}:@var{function}
8721 Specifies the line that begins the body of the function @var{function}
8722 in the file @var{filename}. You only need the file name with a
8723 function name to avoid ambiguity when there are identically named
8724 functions in different source files.
8727 Specifies the line at which the label named @var{label} appears
8728 in the function corresponding to the currently selected stack frame.
8729 If there is no current selected stack frame (for instance, if the inferior
8730 is not running), then @value{GDBN} will not search for a label.
8732 @cindex breakpoint at static probe point
8733 @item -pstap|-probe-stap @r{[}@var{objfile}:@r{[}@var{provider}:@r{]}@r{]}@var{name}
8734 The @sc{gnu}/Linux tool @code{SystemTap} provides a way for
8735 applications to embed static probes. @xref{Static Probe Points}, for more
8736 information on finding and using static probes. This form of linespec
8737 specifies the location of such a static probe.
8739 If @var{objfile} is given, only probes coming from that shared library
8740 or executable matching @var{objfile} as a regular expression are considered.
8741 If @var{provider} is given, then only probes from that provider are considered.
8742 If several probes match the spec, @value{GDBN} will insert a breakpoint at
8743 each one of those probes.
8746 @node Explicit Locations
8747 @subsection Explicit Locations
8748 @cindex explicit locations
8750 @dfn{Explicit locations} allow the user to directly specify the source
8751 location's parameters using option-value pairs.
8753 Explicit locations are useful when several functions, labels, or
8754 file names have the same name (base name for files) in the program's
8755 sources. In these cases, explicit locations point to the source
8756 line you meant more accurately and unambiguously. Also, using
8757 explicit locations might be faster in large programs.
8759 For example, the linespec @samp{foo:bar} may refer to a function @code{bar}
8760 defined in the file named @file{foo} or the label @code{bar} in a function
8761 named @code{foo}. @value{GDBN} must search either the file system or
8762 the symbol table to know.
8764 The list of valid explicit location options is summarized in the
8768 @item -source @var{filename}
8769 The value specifies the source file name. To differentiate between
8770 files with the same base name, prepend as many directories as is necessary
8771 to uniquely identify the desired file, e.g., @file{foo/bar/baz.c}. Otherwise
8772 @value{GDBN} will use the first file it finds with the given base
8773 name. This option requires the use of either @code{-function} or @code{-line}.
8775 @item -function @var{function}
8776 The value specifies the name of a function. Operations
8777 on function locations unmodified by other options (such as @code{-label}
8778 or @code{-line}) refer to the line that begins the body of the function.
8779 In C, for example, this is the line with the open brace.
8781 By default, in C@t{++} and Ada, @var{function} is interpreted as
8782 specifying all functions named @var{function} in all scopes. For
8783 C@t{++}, this means in all namespaces and classes. For Ada, this
8784 means in all packages.
8786 For example, assuming a program with C@t{++} symbols named
8787 @code{A::B::func} and @code{B::func}, both commands @w{@kbd{break
8788 -function func}} and @w{@kbd{break -function B::func}} set a
8789 breakpoint on both symbols.
8791 You can use the @kbd{-qualified} flag to override this (see below).
8795 This flag makes @value{GDBN} interpret a function name specified with
8796 @kbd{-function} as a complete fully-qualified name.
8798 For example, assuming a C@t{++} program with symbols named
8799 @code{A::B::func} and @code{B::func}, the @w{@kbd{break -qualified
8800 -function B::func}} command sets a breakpoint on @code{B::func}, only.
8802 (Note: the @kbd{-qualified} option can precede a linespec as well
8803 (@pxref{Linespec Locations}), so the particular example above could be
8804 simplified as @w{@kbd{break -qualified B::func}}.)
8806 @item -label @var{label}
8807 The value specifies the name of a label. When the function
8808 name is not specified, the label is searched in the function of the currently
8809 selected stack frame.
8811 @item -line @var{number}
8812 The value specifies a line offset for the location. The offset may either
8813 be absolute (@code{-line 3}) or relative (@code{-line +3}), depending on
8814 the command. When specified without any other options, the line offset is
8815 relative to the current line.
8818 Explicit location options may be abbreviated by omitting any non-unique
8819 trailing characters from the option name, e.g., @w{@kbd{break -s main.c -li 3}}.
8821 @node Address Locations
8822 @subsection Address Locations
8823 @cindex address locations
8825 @dfn{Address locations} indicate a specific program address. They have
8826 the generalized form *@var{address}.
8828 For line-oriented commands, such as @code{list} and @code{edit}, this
8829 specifies a source line that contains @var{address}. For @code{break} and
8830 other breakpoint-oriented commands, this can be used to set breakpoints in
8831 parts of your program which do not have debugging information or
8834 Here @var{address} may be any expression valid in the current working
8835 language (@pxref{Languages, working language}) that specifies a code
8836 address. In addition, as a convenience, @value{GDBN} extends the
8837 semantics of expressions used in locations to cover several situations
8838 that frequently occur during debugging. Here are the various forms
8842 @item @var{expression}
8843 Any expression valid in the current working language.
8845 @item @var{funcaddr}
8846 An address of a function or procedure derived from its name. In C,
8847 C@t{++}, Objective-C, Fortran, minimal, and assembly, this is
8848 simply the function's name @var{function} (and actually a special case
8849 of a valid expression). In Pascal and Modula-2, this is
8850 @code{&@var{function}}. In Ada, this is @code{@var{function}'Address}
8851 (although the Pascal form also works).
8853 This form specifies the address of the function's first instruction,
8854 before the stack frame and arguments have been set up.
8856 @item '@var{filename}':@var{funcaddr}
8857 Like @var{funcaddr} above, but also specifies the name of the source
8858 file explicitly. This is useful if the name of the function does not
8859 specify the function unambiguously, e.g., if there are several
8860 functions with identical names in different source files.
8864 @section Editing Source Files
8865 @cindex editing source files
8868 @kindex e @r{(@code{edit})}
8869 To edit the lines in a source file, use the @code{edit} command.
8870 The editing program of your choice
8871 is invoked with the current line set to
8872 the active line in the program.
8873 Alternatively, there are several ways to specify what part of the file you
8874 want to print if you want to see other parts of the program:
8877 @item edit @var{location}
8878 Edit the source file specified by @code{location}. Editing starts at
8879 that @var{location}, e.g., at the specified source line of the
8880 specified file. @xref{Specify Location}, for all the possible forms
8881 of the @var{location} argument; here are the forms of the @code{edit}
8882 command most commonly used:
8885 @item edit @var{number}
8886 Edit the current source file with @var{number} as the active line number.
8888 @item edit @var{function}
8889 Edit the file containing @var{function} at the beginning of its definition.
8894 @subsection Choosing your Editor
8895 You can customize @value{GDBN} to use any editor you want
8897 The only restriction is that your editor (say @code{ex}), recognizes the
8898 following command-line syntax:
8900 ex +@var{number} file
8902 The optional numeric value +@var{number} specifies the number of the line in
8903 the file where to start editing.}.
8904 By default, it is @file{@value{EDITOR}}, but you can change this
8905 by setting the environment variable @code{EDITOR} before using
8906 @value{GDBN}. For example, to configure @value{GDBN} to use the
8907 @code{vi} editor, you could use these commands with the @code{sh} shell:
8913 or in the @code{csh} shell,
8915 setenv EDITOR /usr/bin/vi
8920 @section Searching Source Files
8921 @cindex searching source files
8923 There are two commands for searching through the current source file for a
8928 @kindex forward-search
8929 @kindex fo @r{(@code{forward-search})}
8930 @item forward-search @var{regexp}
8931 @itemx search @var{regexp}
8932 The command @samp{forward-search @var{regexp}} checks each line,
8933 starting with the one following the last line listed, for a match for
8934 @var{regexp}. It lists the line that is found. You can use the
8935 synonym @samp{search @var{regexp}} or abbreviate the command name as
8938 @kindex reverse-search
8939 @item reverse-search @var{regexp}
8940 The command @samp{reverse-search @var{regexp}} checks each line, starting
8941 with the one before the last line listed and going backward, for a match
8942 for @var{regexp}. It lists the line that is found. You can abbreviate
8943 this command as @code{rev}.
8947 @section Specifying Source Directories
8950 @cindex directories for source files
8951 Executable programs sometimes do not record the directories of the source
8952 files from which they were compiled, just the names. Even when they do,
8953 the directories could be moved between the compilation and your debugging
8954 session. @value{GDBN} has a list of directories to search for source files;
8955 this is called the @dfn{source path}. Each time @value{GDBN} wants a source file,
8956 it tries all the directories in the list, in the order they are present
8957 in the list, until it finds a file with the desired name.
8959 For example, suppose an executable references the file
8960 @file{/usr/src/foo-1.0/lib/foo.c}, does not record a compilation
8961 directory, and the @dfn{source path} is @file{/mnt/cross}.
8962 @value{GDBN} would look for the source file in the following
8967 @item @file{/usr/src/foo-1.0/lib/foo.c}
8968 @item @file{/mnt/cross/usr/src/foo-1.0/lib/foo.c}
8969 @item @file{/mnt/cross/foo.c}
8973 If the source file is not present at any of the above locations then
8974 an error is printed. @value{GDBN} does not look up the parts of the
8975 source file name, such as @file{/mnt/cross/src/foo-1.0/lib/foo.c}.
8976 Likewise, the subdirectories of the source path are not searched: if
8977 the source path is @file{/mnt/cross}, and the binary refers to
8978 @file{foo.c}, @value{GDBN} would not find it under
8979 @file{/mnt/cross/usr/src/foo-1.0/lib}.
8981 Plain file names, relative file names with leading directories, file
8982 names containing dots, etc.@: are all treated as described above,
8983 except that non-absolute file names are not looked up literally. If
8984 the @dfn{source path} is @file{/mnt/cross}, the source file is
8985 recorded as @file{../lib/foo.c}, and no compilation directory is
8986 recorded, then @value{GDBN} will search in the following locations:
8990 @item @file{/mnt/cross/../lib/foo.c}
8991 @item @file{/mnt/cross/foo.c}
8997 @vindex $cdir@r{, convenience variable}
8998 @vindex $cwd@r{, convenience variable}
8999 @cindex compilation directory
9000 @cindex current directory
9001 @cindex working directory
9002 @cindex directory, current
9003 @cindex directory, compilation
9004 The @dfn{source path} will always include two special entries
9005 @samp{$cdir} and @samp{$cwd}, these refer to the compilation directory
9006 (if one is recorded) and the current working directory respectively.
9008 @samp{$cdir} causes @value{GDBN} to search within the compilation
9009 directory, if one is recorded in the debug information. If no
9010 compilation directory is recorded in the debug information then
9011 @samp{$cdir} is ignored.
9013 @samp{$cwd} is not the same as @samp{.}---the former tracks the
9014 current working directory as it changes during your @value{GDBN}
9015 session, while the latter is immediately expanded to the current
9016 directory at the time you add an entry to the source path.
9018 If a compilation directory is recorded in the debug information, and
9019 @value{GDBN} has not found the source file after the first search
9020 using @dfn{source path}, then @value{GDBN} will combine the
9021 compilation directory and the filename, and then search for the source
9022 file again using the @dfn{source path}.
9024 For example, if the executable records the source file as
9025 @file{/usr/src/foo-1.0/lib/foo.c}, the compilation directory is
9026 recorded as @file{/project/build}, and the @dfn{source path} is
9027 @file{/mnt/cross:$cdir:$cwd} while the current working directory of
9028 the @value{GDBN} session is @file{/home/user}, then @value{GDBN} will
9029 search for the source file in the following loctions:
9033 @item @file{/usr/src/foo-1.0/lib/foo.c}
9034 @item @file{/mnt/cross/usr/src/foo-1.0/lib/foo.c}
9035 @item @file{/project/build/usr/src/foo-1.0/lib/foo.c}
9036 @item @file{/home/user/usr/src/foo-1.0/lib/foo.c}
9037 @item @file{/mnt/cross/project/build/usr/src/foo-1.0/lib/foo.c}
9038 @item @file{/project/build/project/build/usr/src/foo-1.0/lib/foo.c}
9039 @item @file{/home/user/project/build/usr/src/foo-1.0/lib/foo.c}
9040 @item @file{/mnt/cross/foo.c}
9041 @item @file{/project/build/foo.c}
9042 @item @file{/home/user/foo.c}
9046 If the file name in the previous example had been recorded in the
9047 executable as a relative path rather than an absolute path, then the
9048 first look up would not have occurred, but all of the remaining steps
9051 When searching for source files on MS-DOS and MS-Windows, where
9052 absolute paths start with a drive letter (e.g.
9053 @file{C:/project/foo.c}), @value{GDBN} will remove the drive letter
9054 from the file name before appending it to a search directory from
9055 @dfn{source path}; for instance if the executable references the
9056 source file @file{C:/project/foo.c} and @dfn{source path} is set to
9057 @file{D:/mnt/cross}, then @value{GDBN} will search in the following
9058 locations for the source file:
9062 @item @file{C:/project/foo.c}
9063 @item @file{D:/mnt/cross/project/foo.c}
9064 @item @file{D:/mnt/cross/foo.c}
9068 Note that the executable search path is @emph{not} used to locate the
9071 Whenever you reset or rearrange the source path, @value{GDBN} clears out
9072 any information it has cached about where source files are found and where
9073 each line is in the file.
9077 When you start @value{GDBN}, its source path includes only @samp{$cdir}
9078 and @samp{$cwd}, in that order.
9079 To add other directories, use the @code{directory} command.
9081 The search path is used to find both program source files and @value{GDBN}
9082 script files (read using the @samp{-command} option and @samp{source} command).
9084 In addition to the source path, @value{GDBN} provides a set of commands
9085 that manage a list of source path substitution rules. A @dfn{substitution
9086 rule} specifies how to rewrite source directories stored in the program's
9087 debug information in case the sources were moved to a different
9088 directory between compilation and debugging. A rule is made of
9089 two strings, the first specifying what needs to be rewritten in
9090 the path, and the second specifying how it should be rewritten.
9091 In @ref{set substitute-path}, we name these two parts @var{from} and
9092 @var{to} respectively. @value{GDBN} does a simple string replacement
9093 of @var{from} with @var{to} at the start of the directory part of the
9094 source file name, and uses that result instead of the original file
9095 name to look up the sources.
9097 Using the previous example, suppose the @file{foo-1.0} tree has been
9098 moved from @file{/usr/src} to @file{/mnt/cross}, then you can tell
9099 @value{GDBN} to replace @file{/usr/src} in all source path names with
9100 @file{/mnt/cross}. The first lookup will then be
9101 @file{/mnt/cross/foo-1.0/lib/foo.c} in place of the original location
9102 of @file{/usr/src/foo-1.0/lib/foo.c}. To define a source path
9103 substitution rule, use the @code{set substitute-path} command
9104 (@pxref{set substitute-path}).
9106 To avoid unexpected substitution results, a rule is applied only if the
9107 @var{from} part of the directory name ends at a directory separator.
9108 For instance, a rule substituting @file{/usr/source} into
9109 @file{/mnt/cross} will be applied to @file{/usr/source/foo-1.0} but
9110 not to @file{/usr/sourceware/foo-2.0}. And because the substitution
9111 is applied only at the beginning of the directory name, this rule will
9112 not be applied to @file{/root/usr/source/baz.c} either.
9114 In many cases, you can achieve the same result using the @code{directory}
9115 command. However, @code{set substitute-path} can be more efficient in
9116 the case where the sources are organized in a complex tree with multiple
9117 subdirectories. With the @code{directory} command, you need to add each
9118 subdirectory of your project. If you moved the entire tree while
9119 preserving its internal organization, then @code{set substitute-path}
9120 allows you to direct the debugger to all the sources with one single
9123 @code{set substitute-path} is also more than just a shortcut command.
9124 The source path is only used if the file at the original location no
9125 longer exists. On the other hand, @code{set substitute-path} modifies
9126 the debugger behavior to look at the rewritten location instead. So, if
9127 for any reason a source file that is not relevant to your executable is
9128 located at the original location, a substitution rule is the only
9129 method available to point @value{GDBN} at the new location.
9131 @cindex @samp{--with-relocated-sources}
9132 @cindex default source path substitution
9133 You can configure a default source path substitution rule by
9134 configuring @value{GDBN} with the
9135 @samp{--with-relocated-sources=@var{dir}} option. The @var{dir}
9136 should be the name of a directory under @value{GDBN}'s configured
9137 prefix (set with @samp{--prefix} or @samp{--exec-prefix}), and
9138 directory names in debug information under @var{dir} will be adjusted
9139 automatically if the installed @value{GDBN} is moved to a new
9140 location. This is useful if @value{GDBN}, libraries or executables
9141 with debug information and corresponding source code are being moved
9145 @item directory @var{dirname} @dots{}
9146 @item dir @var{dirname} @dots{}
9147 Add directory @var{dirname} to the front of the source path. Several
9148 directory names may be given to this command, separated by @samp{:}
9149 (@samp{;} on MS-DOS and MS-Windows, where @samp{:} usually appears as
9150 part of absolute file names) or
9151 whitespace. You may specify a directory that is already in the source
9152 path; this moves it forward, so @value{GDBN} searches it sooner.
9154 The special strings @samp{$cdir} (to refer to the compilation
9155 directory, if one is recorded), and @samp{$cwd} (to refer to the
9156 current working directory) can also be included in the list of
9157 directories @var{dirname}. Though these will already be in the source
9158 path they will be moved forward in the list so @value{GDBN} searches
9162 Reset the source path to its default value (@samp{$cdir:$cwd} on Unix systems). This requires confirmation.
9164 @c RET-repeat for @code{directory} is explicitly disabled, but since
9165 @c repeating it would be a no-op we do not say that. (thanks to RMS)
9167 @item set directories @var{path-list}
9168 @kindex set directories
9169 Set the source path to @var{path-list}.
9170 @samp{$cdir:$cwd} are added if missing.
9172 @item show directories
9173 @kindex show directories
9174 Print the source path: show which directories it contains.
9176 @anchor{set substitute-path}
9177 @item set substitute-path @var{from} @var{to}
9178 @kindex set substitute-path
9179 Define a source path substitution rule, and add it at the end of the
9180 current list of existing substitution rules. If a rule with the same
9181 @var{from} was already defined, then the old rule is also deleted.
9183 For example, if the file @file{/foo/bar/baz.c} was moved to
9184 @file{/mnt/cross/baz.c}, then the command
9187 (@value{GDBP}) set substitute-path /foo/bar /mnt/cross
9191 will tell @value{GDBN} to replace @samp{/foo/bar} with
9192 @samp{/mnt/cross}, which will allow @value{GDBN} to find the file
9193 @file{baz.c} even though it was moved.
9195 In the case when more than one substitution rule have been defined,
9196 the rules are evaluated one by one in the order where they have been
9197 defined. The first one matching, if any, is selected to perform
9200 For instance, if we had entered the following commands:
9203 (@value{GDBP}) set substitute-path /usr/src/include /mnt/include
9204 (@value{GDBP}) set substitute-path /usr/src /mnt/src
9208 @value{GDBN} would then rewrite @file{/usr/src/include/defs.h} into
9209 @file{/mnt/include/defs.h} by using the first rule. However, it would
9210 use the second rule to rewrite @file{/usr/src/lib/foo.c} into
9211 @file{/mnt/src/lib/foo.c}.
9214 @item unset substitute-path [path]
9215 @kindex unset substitute-path
9216 If a path is specified, search the current list of substitution rules
9217 for a rule that would rewrite that path. Delete that rule if found.
9218 A warning is emitted by the debugger if no rule could be found.
9220 If no path is specified, then all substitution rules are deleted.
9222 @item show substitute-path [path]
9223 @kindex show substitute-path
9224 If a path is specified, then print the source path substitution rule
9225 which would rewrite that path, if any.
9227 If no path is specified, then print all existing source path substitution
9232 If your source path is cluttered with directories that are no longer of
9233 interest, @value{GDBN} may sometimes cause confusion by finding the wrong
9234 versions of source. You can correct the situation as follows:
9238 Use @code{directory} with no argument to reset the source path to its default value.
9241 Use @code{directory} with suitable arguments to reinstall the
9242 directories you want in the source path. You can add all the
9243 directories in one command.
9247 @section Source and Machine Code
9248 @cindex source line and its code address
9250 You can use the command @code{info line} to map source lines to program
9251 addresses (and vice versa), and the command @code{disassemble} to display
9252 a range of addresses as machine instructions. You can use the command
9253 @code{set disassemble-next-line} to set whether to disassemble next
9254 source line when execution stops. When run under @sc{gnu} Emacs
9255 mode, the @code{info line} command causes the arrow to point to the
9256 line specified. Also, @code{info line} prints addresses in symbolic form as
9262 @itemx info line @var{location}
9263 Print the starting and ending addresses of the compiled code for
9264 source line @var{location}. You can specify source lines in any of
9265 the ways documented in @ref{Specify Location}. With no @var{location}
9266 information about the current source line is printed.
9269 For example, we can use @code{info line} to discover the location of
9270 the object code for the first line of function
9271 @code{m4_changequote}:
9274 (@value{GDBP}) info line m4_changequote
9275 Line 895 of "builtin.c" starts at pc 0x634c <m4_changequote> and \
9276 ends at 0x6350 <m4_changequote+4>.
9280 @cindex code address and its source line
9281 We can also inquire (using @code{*@var{addr}} as the form for
9282 @var{location}) what source line covers a particular address:
9284 (@value{GDBP}) info line *0x63ff
9285 Line 926 of "builtin.c" starts at pc 0x63e4 <m4_changequote+152> and \
9286 ends at 0x6404 <m4_changequote+184>.
9289 @cindex @code{$_} and @code{info line}
9290 @cindex @code{x} command, default address
9291 @kindex x@r{(examine), and} info line
9292 After @code{info line}, the default address for the @code{x} command
9293 is changed to the starting address of the line, so that @samp{x/i} is
9294 sufficient to begin examining the machine code (@pxref{Memory,
9295 ,Examining Memory}). Also, this address is saved as the value of the
9296 convenience variable @code{$_} (@pxref{Convenience Vars, ,Convenience
9299 @cindex info line, repeated calls
9300 After @code{info line}, using @code{info line} again without
9301 specifying a location will display information about the next source
9306 @cindex assembly instructions
9307 @cindex instructions, assembly
9308 @cindex machine instructions
9309 @cindex listing machine instructions
9311 @itemx disassemble /m
9312 @itemx disassemble /s
9313 @itemx disassemble /r
9314 This specialized command dumps a range of memory as machine
9315 instructions. It can also print mixed source+disassembly by specifying
9316 the @code{/m} or @code{/s} modifier and print the raw instructions in hex
9317 as well as in symbolic form by specifying the @code{/r} modifier.
9318 The default memory range is the function surrounding the
9319 program counter of the selected frame. A single argument to this
9320 command is a program counter value; @value{GDBN} dumps the function
9321 surrounding this value. When two arguments are given, they should
9322 be separated by a comma, possibly surrounded by whitespace. The
9323 arguments specify a range of addresses to dump, in one of two forms:
9326 @item @var{start},@var{end}
9327 the addresses from @var{start} (inclusive) to @var{end} (exclusive)
9328 @item @var{start},+@var{length}
9329 the addresses from @var{start} (inclusive) to
9330 @code{@var{start}+@var{length}} (exclusive).
9334 When 2 arguments are specified, the name of the function is also
9335 printed (since there could be several functions in the given range).
9337 The argument(s) can be any expression yielding a numeric value, such as
9338 @samp{0x32c4}, @samp{&main+10} or @samp{$pc - 8}.
9340 If the range of memory being disassembled contains current program counter,
9341 the instruction at that location is shown with a @code{=>} marker.
9344 The following example shows the disassembly of a range of addresses of
9345 HP PA-RISC 2.0 code:
9348 (@value{GDBP}) disas 0x32c4, 0x32e4
9349 Dump of assembler code from 0x32c4 to 0x32e4:
9350 0x32c4 <main+204>: addil 0,dp
9351 0x32c8 <main+208>: ldw 0x22c(sr0,r1),r26
9352 0x32cc <main+212>: ldil 0x3000,r31
9353 0x32d0 <main+216>: ble 0x3f8(sr4,r31)
9354 0x32d4 <main+220>: ldo 0(r31),rp
9355 0x32d8 <main+224>: addil -0x800,dp
9356 0x32dc <main+228>: ldo 0x588(r1),r26
9357 0x32e0 <main+232>: ldil 0x3000,r31
9358 End of assembler dump.
9361 Here is an example showing mixed source+assembly for Intel x86
9362 with @code{/m} or @code{/s}, when the program is stopped just after
9363 function prologue in a non-optimized function with no inline code.
9366 (@value{GDBP}) disas /m main
9367 Dump of assembler code for function main:
9369 0x08048330 <+0>: push %ebp
9370 0x08048331 <+1>: mov %esp,%ebp
9371 0x08048333 <+3>: sub $0x8,%esp
9372 0x08048336 <+6>: and $0xfffffff0,%esp
9373 0x08048339 <+9>: sub $0x10,%esp
9375 6 printf ("Hello.\n");
9376 => 0x0804833c <+12>: movl $0x8048440,(%esp)
9377 0x08048343 <+19>: call 0x8048284 <puts@@plt>
9381 0x08048348 <+24>: mov $0x0,%eax
9382 0x0804834d <+29>: leave
9383 0x0804834e <+30>: ret
9385 End of assembler dump.
9388 The @code{/m} option is deprecated as its output is not useful when
9389 there is either inlined code or re-ordered code.
9390 The @code{/s} option is the preferred choice.
9391 Here is an example for AMD x86-64 showing the difference between
9392 @code{/m} output and @code{/s} output.
9393 This example has one inline function defined in a header file,
9394 and the code is compiled with @samp{-O2} optimization.
9395 Note how the @code{/m} output is missing the disassembly of
9396 several instructions that are present in the @code{/s} output.
9426 (@value{GDBP}) disas /m main
9427 Dump of assembler code for function main:
9431 0x0000000000400400 <+0>: mov 0x200c2e(%rip),%eax # 0x601034 <y>
9432 0x0000000000400417 <+23>: mov %eax,0x200c13(%rip) # 0x601030 <x>
9436 0x000000000040041d <+29>: xor %eax,%eax
9437 0x000000000040041f <+31>: retq
9438 0x0000000000400420 <+32>: add %eax,%eax
9439 0x0000000000400422 <+34>: jmp 0x400417 <main+23>
9441 End of assembler dump.
9442 (@value{GDBP}) disas /s main
9443 Dump of assembler code for function main:
9447 0x0000000000400400 <+0>: mov 0x200c2e(%rip),%eax # 0x601034 <y>
9451 0x0000000000400406 <+6>: test %eax,%eax
9452 0x0000000000400408 <+8>: js 0x400420 <main+32>
9457 0x000000000040040a <+10>: lea 0xa(%rax),%edx
9458 0x000000000040040d <+13>: test %eax,%eax
9459 0x000000000040040f <+15>: mov $0x1,%eax
9460 0x0000000000400414 <+20>: cmovne %edx,%eax
9464 0x0000000000400417 <+23>: mov %eax,0x200c13(%rip) # 0x601030 <x>
9468 0x000000000040041d <+29>: xor %eax,%eax
9469 0x000000000040041f <+31>: retq
9473 0x0000000000400420 <+32>: add %eax,%eax
9474 0x0000000000400422 <+34>: jmp 0x400417 <main+23>
9475 End of assembler dump.
9478 Here is another example showing raw instructions in hex for AMD x86-64,
9481 (gdb) disas /r 0x400281,+10
9482 Dump of assembler code from 0x400281 to 0x40028b:
9483 0x0000000000400281: 38 36 cmp %dh,(%rsi)
9484 0x0000000000400283: 2d 36 34 2e 73 sub $0x732e3436,%eax
9485 0x0000000000400288: 6f outsl %ds:(%rsi),(%dx)
9486 0x0000000000400289: 2e 32 00 xor %cs:(%rax),%al
9487 End of assembler dump.
9490 Addresses cannot be specified as a location (@pxref{Specify Location}).
9491 So, for example, if you want to disassemble function @code{bar}
9492 in file @file{foo.c}, you must type @samp{disassemble 'foo.c'::bar}
9493 and not @samp{disassemble foo.c:bar}.
9495 Some architectures have more than one commonly-used set of instruction
9496 mnemonics or other syntax.
9498 For programs that were dynamically linked and use shared libraries,
9499 instructions that call functions or branch to locations in the shared
9500 libraries might show a seemingly bogus location---it's actually a
9501 location of the relocation table. On some architectures, @value{GDBN}
9502 might be able to resolve these to actual function names.
9505 @kindex set disassembler-options
9506 @cindex disassembler options
9507 @item set disassembler-options @var{option1}[,@var{option2}@dots{}]
9508 This command controls the passing of target specific information to
9509 the disassembler. For a list of valid options, please refer to the
9510 @code{-M}/@code{--disassembler-options} section of the @samp{objdump}
9511 manual and/or the output of @kbd{objdump --help}
9512 (@pxref{objdump,,objdump,binutils,The GNU Binary Utilities}).
9513 The default value is the empty string.
9515 If it is necessary to specify more than one disassembler option, then
9516 multiple options can be placed together into a comma separated list.
9517 Currently this command is only supported on targets ARM, MIPS, PowerPC
9520 @kindex show disassembler-options
9521 @item show disassembler-options
9522 Show the current setting of the disassembler options.
9526 @kindex set disassembly-flavor
9527 @cindex Intel disassembly flavor
9528 @cindex AT&T disassembly flavor
9529 @item set disassembly-flavor @var{instruction-set}
9530 Select the instruction set to use when disassembling the
9531 program via the @code{disassemble} or @code{x/i} commands.
9533 Currently this command is only defined for the Intel x86 family. You
9534 can set @var{instruction-set} to either @code{intel} or @code{att}.
9535 The default is @code{att}, the AT&T flavor used by default by Unix
9536 assemblers for x86-based targets.
9538 @kindex show disassembly-flavor
9539 @item show disassembly-flavor
9540 Show the current setting of the disassembly flavor.
9544 @kindex set disassemble-next-line
9545 @kindex show disassemble-next-line
9546 @item set disassemble-next-line
9547 @itemx show disassemble-next-line
9548 Control whether or not @value{GDBN} will disassemble the next source
9549 line or instruction when execution stops. If ON, @value{GDBN} will
9550 display disassembly of the next source line when execution of the
9551 program being debugged stops. This is @emph{in addition} to
9552 displaying the source line itself, which @value{GDBN} always does if
9553 possible. If the next source line cannot be displayed for some reason
9554 (e.g., if @value{GDBN} cannot find the source file, or there's no line
9555 info in the debug info), @value{GDBN} will display disassembly of the
9556 next @emph{instruction} instead of showing the next source line. If
9557 AUTO, @value{GDBN} will display disassembly of next instruction only
9558 if the source line cannot be displayed. This setting causes
9559 @value{GDBN} to display some feedback when you step through a function
9560 with no line info or whose source file is unavailable. The default is
9561 OFF, which means never display the disassembly of the next line or
9567 @chapter Examining Data
9569 @cindex printing data
9570 @cindex examining data
9573 The usual way to examine data in your program is with the @code{print}
9574 command (abbreviated @code{p}), or its synonym @code{inspect}. It
9575 evaluates and prints the value of an expression of the language your
9576 program is written in (@pxref{Languages, ,Using @value{GDBN} with
9577 Different Languages}). It may also print the expression using a
9578 Python-based pretty-printer (@pxref{Pretty Printing}).
9581 @item print [[@var{options}] --] @var{expr}
9582 @itemx print [[@var{options}] --] /@var{f} @var{expr}
9583 @var{expr} is an expression (in the source language). By default the
9584 value of @var{expr} is printed in a format appropriate to its data type;
9585 you can choose a different format by specifying @samp{/@var{f}}, where
9586 @var{f} is a letter specifying the format; see @ref{Output Formats,,Output
9589 @anchor{print options}
9590 The @code{print} command supports a number of options that allow
9591 overriding relevant global print settings as set by @code{set print}
9595 @item -address [@code{on}|@code{off}]
9596 Set printing of addresses.
9597 Related setting: @ref{set print address}.
9599 @item -array [@code{on}|@code{off}]
9600 Pretty formatting of arrays.
9601 Related setting: @ref{set print array}.
9603 @item -array-indexes [@code{on}|@code{off}]
9604 Set printing of array indexes.
9605 Related setting: @ref{set print array-indexes}.
9607 @item -elements @var{number-of-elements}|@code{unlimited}
9608 Set limit on string chars or array elements to print. The value
9609 @code{unlimited} causes there to be no limit. Related setting:
9610 @ref{set print elements}.
9612 @item -max-depth @var{depth}|@code{unlimited}
9613 Set the threshold after which nested structures are replaced with
9614 ellipsis. Related setting: @ref{set print max-depth}.
9616 @item -null-stop [@code{on}|@code{off}]
9617 Set printing of char arrays to stop at first null char. Related
9618 setting: @ref{set print null-stop}.
9620 @item -object [@code{on}|@code{off}]
9621 Set printing C@t{++} virtual function tables. Related setting:
9622 @ref{set print object}.
9624 @item -pretty [@code{on}|@code{off}]
9625 Set pretty formatting of structures. Related setting: @ref{set print
9628 @item -repeats @var{number-of-repeats}|@code{unlimited}
9629 Set threshold for repeated print elements. @code{unlimited} causes
9630 all elements to be individually printed. Related setting: @ref{set
9633 @item -static-members [@code{on}|@code{off}]
9634 Set printing C@t{++} static members. Related setting: @ref{set print
9637 @item -symbol [@code{on}|@code{off}]
9638 Set printing of symbol names when printing pointers. Related setting:
9639 @ref{set print symbol}.
9641 @item -union [@code{on}|@code{off}]
9642 Set printing of unions interior to structures. Related setting:
9643 @ref{set print union}.
9645 @item -vtbl [@code{on}|@code{off}]
9646 Set printing of C++ virtual function tables. Related setting:
9647 @ref{set print vtbl}.
9650 Because the @code{print} command accepts arbitrary expressions which
9651 may look like options (including abbreviations), if you specify any
9652 command option, then you must use a double dash (@code{--}) to mark
9653 the end of option processing.
9655 For example, this prints the value of the @code{-r} expression:
9658 (@value{GDBP}) print -r
9661 While this repeats the last value in the value history (see below)
9662 with the @code{-raw} option in effect:
9665 (@value{GDBP}) print -r --
9668 Here is an example including both on option and an expression:
9672 (@value{GDBP}) print -pretty -- *myptr
9684 @item print [@var{options}]
9685 @itemx print [@var{options}] /@var{f}
9686 @cindex reprint the last value
9687 If you omit @var{expr}, @value{GDBN} displays the last value again (from the
9688 @dfn{value history}; @pxref{Value History, ,Value History}). This allows you to
9689 conveniently inspect the same value in an alternative format.
9692 A more low-level way of examining data is with the @code{x} command.
9693 It examines data in memory at a specified address and prints it in a
9694 specified format. @xref{Memory, ,Examining Memory}.
9696 If you are interested in information about types, or about how the
9697 fields of a struct or a class are declared, use the @code{ptype @var{exp}}
9698 command rather than @code{print}. @xref{Symbols, ,Examining the Symbol
9701 @cindex exploring hierarchical data structures
9703 Another way of examining values of expressions and type information is
9704 through the Python extension command @code{explore} (available only if
9705 the @value{GDBN} build is configured with @code{--with-python}). It
9706 offers an interactive way to start at the highest level (or, the most
9707 abstract level) of the data type of an expression (or, the data type
9708 itself) and explore all the way down to leaf scalar values/fields
9709 embedded in the higher level data types.
9712 @item explore @var{arg}
9713 @var{arg} is either an expression (in the source language), or a type
9714 visible in the current context of the program being debugged.
9717 The working of the @code{explore} command can be illustrated with an
9718 example. If a data type @code{struct ComplexStruct} is defined in your
9728 struct ComplexStruct
9730 struct SimpleStruct *ss_p;
9736 followed by variable declarations as
9739 struct SimpleStruct ss = @{ 10, 1.11 @};
9740 struct ComplexStruct cs = @{ &ss, @{ 0, 1, 2, 3, 4, 5, 6, 7, 8, 9 @} @};
9744 then, the value of the variable @code{cs} can be explored using the
9745 @code{explore} command as follows.
9749 The value of `cs' is a struct/class of type `struct ComplexStruct' with
9750 the following fields:
9752 ss_p = <Enter 0 to explore this field of type `struct SimpleStruct *'>
9753 arr = <Enter 1 to explore this field of type `int [10]'>
9755 Enter the field number of choice:
9759 Since the fields of @code{cs} are not scalar values, you are being
9760 prompted to chose the field you want to explore. Let's say you choose
9761 the field @code{ss_p} by entering @code{0}. Then, since this field is a
9762 pointer, you will be asked if it is pointing to a single value. From
9763 the declaration of @code{cs} above, it is indeed pointing to a single
9764 value, hence you enter @code{y}. If you enter @code{n}, then you will
9765 be asked if it were pointing to an array of values, in which case this
9766 field will be explored as if it were an array.
9769 `cs.ss_p' is a pointer to a value of type `struct SimpleStruct'
9770 Continue exploring it as a pointer to a single value [y/n]: y
9771 The value of `*(cs.ss_p)' is a struct/class of type `struct
9772 SimpleStruct' with the following fields:
9774 i = 10 .. (Value of type `int')
9775 d = 1.1100000000000001 .. (Value of type `double')
9777 Press enter to return to parent value:
9781 If the field @code{arr} of @code{cs} was chosen for exploration by
9782 entering @code{1} earlier, then since it is as array, you will be
9783 prompted to enter the index of the element in the array that you want
9787 `cs.arr' is an array of `int'.
9788 Enter the index of the element you want to explore in `cs.arr': 5
9790 `(cs.arr)[5]' is a scalar value of type `int'.
9794 Press enter to return to parent value:
9797 In general, at any stage of exploration, you can go deeper towards the
9798 leaf values by responding to the prompts appropriately, or hit the
9799 return key to return to the enclosing data structure (the @i{higher}
9800 level data structure).
9802 Similar to exploring values, you can use the @code{explore} command to
9803 explore types. Instead of specifying a value (which is typically a
9804 variable name or an expression valid in the current context of the
9805 program being debugged), you specify a type name. If you consider the
9806 same example as above, your can explore the type
9807 @code{struct ComplexStruct} by passing the argument
9808 @code{struct ComplexStruct} to the @code{explore} command.
9811 (gdb) explore struct ComplexStruct
9815 By responding to the prompts appropriately in the subsequent interactive
9816 session, you can explore the type @code{struct ComplexStruct} in a
9817 manner similar to how the value @code{cs} was explored in the above
9820 The @code{explore} command also has two sub-commands,
9821 @code{explore value} and @code{explore type}. The former sub-command is
9822 a way to explicitly specify that value exploration of the argument is
9823 being invoked, while the latter is a way to explicitly specify that type
9824 exploration of the argument is being invoked.
9827 @item explore value @var{expr}
9828 @cindex explore value
9829 This sub-command of @code{explore} explores the value of the
9830 expression @var{expr} (if @var{expr} is an expression valid in the
9831 current context of the program being debugged). The behavior of this
9832 command is identical to that of the behavior of the @code{explore}
9833 command being passed the argument @var{expr}.
9835 @item explore type @var{arg}
9836 @cindex explore type
9837 This sub-command of @code{explore} explores the type of @var{arg} (if
9838 @var{arg} is a type visible in the current context of program being
9839 debugged), or the type of the value/expression @var{arg} (if @var{arg}
9840 is an expression valid in the current context of the program being
9841 debugged). If @var{arg} is a type, then the behavior of this command is
9842 identical to that of the @code{explore} command being passed the
9843 argument @var{arg}. If @var{arg} is an expression, then the behavior of
9844 this command will be identical to that of the @code{explore} command
9845 being passed the type of @var{arg} as the argument.
9849 * Expressions:: Expressions
9850 * Ambiguous Expressions:: Ambiguous Expressions
9851 * Variables:: Program variables
9852 * Arrays:: Artificial arrays
9853 * Output Formats:: Output formats
9854 * Memory:: Examining memory
9855 * Auto Display:: Automatic display
9856 * Print Settings:: Print settings
9857 * Pretty Printing:: Python pretty printing
9858 * Value History:: Value history
9859 * Convenience Vars:: Convenience variables
9860 * Convenience Funs:: Convenience functions
9861 * Registers:: Registers
9862 * Floating Point Hardware:: Floating point hardware
9863 * Vector Unit:: Vector Unit
9864 * OS Information:: Auxiliary data provided by operating system
9865 * Memory Region Attributes:: Memory region attributes
9866 * Dump/Restore Files:: Copy between memory and a file
9867 * Core File Generation:: Cause a program dump its core
9868 * Character Sets:: Debugging programs that use a different
9869 character set than GDB does
9870 * Caching Target Data:: Data caching for targets
9871 * Searching Memory:: Searching memory for a sequence of bytes
9872 * Value Sizes:: Managing memory allocated for values
9876 @section Expressions
9879 @code{print} and many other @value{GDBN} commands accept an expression and
9880 compute its value. Any kind of constant, variable or operator defined
9881 by the programming language you are using is valid in an expression in
9882 @value{GDBN}. This includes conditional expressions, function calls,
9883 casts, and string constants. It also includes preprocessor macros, if
9884 you compiled your program to include this information; see
9887 @cindex arrays in expressions
9888 @value{GDBN} supports array constants in expressions input by
9889 the user. The syntax is @{@var{element}, @var{element}@dots{}@}. For example,
9890 you can use the command @code{print @{1, 2, 3@}} to create an array
9891 of three integers. If you pass an array to a function or assign it
9892 to a program variable, @value{GDBN} copies the array to memory that
9893 is @code{malloc}ed in the target program.
9895 Because C is so widespread, most of the expressions shown in examples in
9896 this manual are in C. @xref{Languages, , Using @value{GDBN} with Different
9897 Languages}, for information on how to use expressions in other
9900 In this section, we discuss operators that you can use in @value{GDBN}
9901 expressions regardless of your programming language.
9903 @cindex casts, in expressions
9904 Casts are supported in all languages, not just in C, because it is so
9905 useful to cast a number into a pointer in order to examine a structure
9906 at that address in memory.
9907 @c FIXME: casts supported---Mod2 true?
9909 @value{GDBN} supports these operators, in addition to those common
9910 to programming languages:
9914 @samp{@@} is a binary operator for treating parts of memory as arrays.
9915 @xref{Arrays, ,Artificial Arrays}, for more information.
9918 @samp{::} allows you to specify a variable in terms of the file or
9919 function where it is defined. @xref{Variables, ,Program Variables}.
9921 @cindex @{@var{type}@}
9922 @cindex type casting memory
9923 @cindex memory, viewing as typed object
9924 @cindex casts, to view memory
9925 @item @{@var{type}@} @var{addr}
9926 Refers to an object of type @var{type} stored at address @var{addr} in
9927 memory. The address @var{addr} may be any expression whose value is
9928 an integer or pointer (but parentheses are required around binary
9929 operators, just as in a cast). This construct is allowed regardless
9930 of what kind of data is normally supposed to reside at @var{addr}.
9933 @node Ambiguous Expressions
9934 @section Ambiguous Expressions
9935 @cindex ambiguous expressions
9937 Expressions can sometimes contain some ambiguous elements. For instance,
9938 some programming languages (notably Ada, C@t{++} and Objective-C) permit
9939 a single function name to be defined several times, for application in
9940 different contexts. This is called @dfn{overloading}. Another example
9941 involving Ada is generics. A @dfn{generic package} is similar to C@t{++}
9942 templates and is typically instantiated several times, resulting in
9943 the same function name being defined in different contexts.
9945 In some cases and depending on the language, it is possible to adjust
9946 the expression to remove the ambiguity. For instance in C@t{++}, you
9947 can specify the signature of the function you want to break on, as in
9948 @kbd{break @var{function}(@var{types})}. In Ada, using the fully
9949 qualified name of your function often makes the expression unambiguous
9952 When an ambiguity that needs to be resolved is detected, the debugger
9953 has the capability to display a menu of numbered choices for each
9954 possibility, and then waits for the selection with the prompt @samp{>}.
9955 The first option is always @samp{[0] cancel}, and typing @kbd{0 @key{RET}}
9956 aborts the current command. If the command in which the expression was
9957 used allows more than one choice to be selected, the next option in the
9958 menu is @samp{[1] all}, and typing @kbd{1 @key{RET}} selects all possible
9961 For example, the following session excerpt shows an attempt to set a
9962 breakpoint at the overloaded symbol @code{String::after}.
9963 We choose three particular definitions of that function name:
9965 @c FIXME! This is likely to change to show arg type lists, at least
9968 (@value{GDBP}) b String::after
9971 [2] file:String.cc; line number:867
9972 [3] file:String.cc; line number:860
9973 [4] file:String.cc; line number:875
9974 [5] file:String.cc; line number:853
9975 [6] file:String.cc; line number:846
9976 [7] file:String.cc; line number:735
9978 Breakpoint 1 at 0xb26c: file String.cc, line 867.
9979 Breakpoint 2 at 0xb344: file String.cc, line 875.
9980 Breakpoint 3 at 0xafcc: file String.cc, line 846.
9981 Multiple breakpoints were set.
9982 Use the "delete" command to delete unwanted
9989 @kindex set multiple-symbols
9990 @item set multiple-symbols @var{mode}
9991 @cindex multiple-symbols menu
9993 This option allows you to adjust the debugger behavior when an expression
9996 By default, @var{mode} is set to @code{all}. If the command with which
9997 the expression is used allows more than one choice, then @value{GDBN}
9998 automatically selects all possible choices. For instance, inserting
9999 a breakpoint on a function using an ambiguous name results in a breakpoint
10000 inserted on each possible match. However, if a unique choice must be made,
10001 then @value{GDBN} uses the menu to help you disambiguate the expression.
10002 For instance, printing the address of an overloaded function will result
10003 in the use of the menu.
10005 When @var{mode} is set to @code{ask}, the debugger always uses the menu
10006 when an ambiguity is detected.
10008 Finally, when @var{mode} is set to @code{cancel}, the debugger reports
10009 an error due to the ambiguity and the command is aborted.
10011 @kindex show multiple-symbols
10012 @item show multiple-symbols
10013 Show the current value of the @code{multiple-symbols} setting.
10017 @section Program Variables
10019 The most common kind of expression to use is the name of a variable
10022 Variables in expressions are understood in the selected stack frame
10023 (@pxref{Selection, ,Selecting a Frame}); they must be either:
10027 global (or file-static)
10034 visible according to the scope rules of the
10035 programming language from the point of execution in that frame
10038 @noindent This means that in the function
10053 you can examine and use the variable @code{a} whenever your program is
10054 executing within the function @code{foo}, but you can only use or
10055 examine the variable @code{b} while your program is executing inside
10056 the block where @code{b} is declared.
10058 @cindex variable name conflict
10059 There is an exception: you can refer to a variable or function whose
10060 scope is a single source file even if the current execution point is not
10061 in this file. But it is possible to have more than one such variable or
10062 function with the same name (in different source files). If that
10063 happens, referring to that name has unpredictable effects. If you wish,
10064 you can specify a static variable in a particular function or file by
10065 using the colon-colon (@code{::}) notation:
10067 @cindex colon-colon, context for variables/functions
10069 @c info cannot cope with a :: index entry, but why deprive hard copy readers?
10070 @cindex @code{::}, context for variables/functions
10073 @var{file}::@var{variable}
10074 @var{function}::@var{variable}
10078 Here @var{file} or @var{function} is the name of the context for the
10079 static @var{variable}. In the case of file names, you can use quotes to
10080 make sure @value{GDBN} parses the file name as a single word---for example,
10081 to print a global value of @code{x} defined in @file{f2.c}:
10084 (@value{GDBP}) p 'f2.c'::x
10087 The @code{::} notation is normally used for referring to
10088 static variables, since you typically disambiguate uses of local variables
10089 in functions by selecting the appropriate frame and using the
10090 simple name of the variable. However, you may also use this notation
10091 to refer to local variables in frames enclosing the selected frame:
10100 process (a); /* Stop here */
10111 For example, if there is a breakpoint at the commented line,
10112 here is what you might see
10113 when the program stops after executing the call @code{bar(0)}:
10118 (@value{GDBP}) p bar::a
10120 (@value{GDBP}) up 2
10121 #2 0x080483d0 in foo (a=5) at foobar.c:12
10124 (@value{GDBP}) p bar::a
10128 @cindex C@t{++} scope resolution
10129 These uses of @samp{::} are very rarely in conflict with the very
10130 similar use of the same notation in C@t{++}. When they are in
10131 conflict, the C@t{++} meaning takes precedence; however, this can be
10132 overridden by quoting the file or function name with single quotes.
10134 For example, suppose the program is stopped in a method of a class
10135 that has a field named @code{includefile}, and there is also an
10136 include file named @file{includefile} that defines a variable,
10137 @code{some_global}.
10140 (@value{GDBP}) p includefile
10142 (@value{GDBP}) p includefile::some_global
10143 A syntax error in expression, near `'.
10144 (@value{GDBP}) p 'includefile'::some_global
10148 @cindex wrong values
10149 @cindex variable values, wrong
10150 @cindex function entry/exit, wrong values of variables
10151 @cindex optimized code, wrong values of variables
10153 @emph{Warning:} Occasionally, a local variable may appear to have the
10154 wrong value at certain points in a function---just after entry to a new
10155 scope, and just before exit.
10157 You may see this problem when you are stepping by machine instructions.
10158 This is because, on most machines, it takes more than one instruction to
10159 set up a stack frame (including local variable definitions); if you are
10160 stepping by machine instructions, variables may appear to have the wrong
10161 values until the stack frame is completely built. On exit, it usually
10162 also takes more than one machine instruction to destroy a stack frame;
10163 after you begin stepping through that group of instructions, local
10164 variable definitions may be gone.
10166 This may also happen when the compiler does significant optimizations.
10167 To be sure of always seeing accurate values, turn off all optimization
10170 @cindex ``No symbol "foo" in current context''
10171 Another possible effect of compiler optimizations is to optimize
10172 unused variables out of existence, or assign variables to registers (as
10173 opposed to memory addresses). Depending on the support for such cases
10174 offered by the debug info format used by the compiler, @value{GDBN}
10175 might not be able to display values for such local variables. If that
10176 happens, @value{GDBN} will print a message like this:
10179 No symbol "foo" in current context.
10182 To solve such problems, either recompile without optimizations, or use a
10183 different debug info format, if the compiler supports several such
10184 formats. @xref{Compilation}, for more information on choosing compiler
10185 options. @xref{C, ,C and C@t{++}}, for more information about debug
10186 info formats that are best suited to C@t{++} programs.
10188 If you ask to print an object whose contents are unknown to
10189 @value{GDBN}, e.g., because its data type is not completely specified
10190 by the debug information, @value{GDBN} will say @samp{<incomplete
10191 type>}. @xref{Symbols, incomplete type}, for more about this.
10193 @cindex no debug info variables
10194 If you try to examine or use the value of a (global) variable for
10195 which @value{GDBN} has no type information, e.g., because the program
10196 includes no debug information, @value{GDBN} displays an error message.
10197 @xref{Symbols, unknown type}, for more about unknown types. If you
10198 cast the variable to its declared type, @value{GDBN} gets the
10199 variable's value using the cast-to type as the variable's type. For
10200 example, in a C program:
10203 (@value{GDBP}) p var
10204 'var' has unknown type; cast it to its declared type
10205 (@value{GDBP}) p (float) var
10209 If you append @kbd{@@entry} string to a function parameter name you get its
10210 value at the time the function got called. If the value is not available an
10211 error message is printed. Entry values are available only with some compilers.
10212 Entry values are normally also printed at the function parameter list according
10213 to @ref{set print entry-values}.
10216 Breakpoint 1, d (i=30) at gdb.base/entry-value.c:29
10222 (gdb) print i@@entry
10226 Strings are identified as arrays of @code{char} values without specified
10227 signedness. Arrays of either @code{signed char} or @code{unsigned char} get
10228 printed as arrays of 1 byte sized integers. @code{-fsigned-char} or
10229 @code{-funsigned-char} @value{NGCC} options have no effect as @value{GDBN}
10230 defines literal string type @code{"char"} as @code{char} without a sign.
10235 signed char var1[] = "A";
10238 You get during debugging
10243 $2 = @{65 'A', 0 '\0'@}
10247 @section Artificial Arrays
10249 @cindex artificial array
10251 @kindex @@@r{, referencing memory as an array}
10252 It is often useful to print out several successive objects of the
10253 same type in memory; a section of an array, or an array of
10254 dynamically determined size for which only a pointer exists in the
10257 You can do this by referring to a contiguous span of memory as an
10258 @dfn{artificial array}, using the binary operator @samp{@@}. The left
10259 operand of @samp{@@} should be the first element of the desired array
10260 and be an individual object. The right operand should be the desired length
10261 of the array. The result is an array value whose elements are all of
10262 the type of the left argument. The first element is actually the left
10263 argument; the second element comes from bytes of memory immediately
10264 following those that hold the first element, and so on. Here is an
10265 example. If a program says
10268 int *array = (int *) malloc (len * sizeof (int));
10272 you can print the contents of @code{array} with
10278 The left operand of @samp{@@} must reside in memory. Array values made
10279 with @samp{@@} in this way behave just like other arrays in terms of
10280 subscripting, and are coerced to pointers when used in expressions.
10281 Artificial arrays most often appear in expressions via the value history
10282 (@pxref{Value History, ,Value History}), after printing one out.
10284 Another way to create an artificial array is to use a cast.
10285 This re-interprets a value as if it were an array.
10286 The value need not be in memory:
10288 (@value{GDBP}) p/x (short[2])0x12345678
10289 $1 = @{0x1234, 0x5678@}
10292 As a convenience, if you leave the array length out (as in
10293 @samp{(@var{type}[])@var{value}}) @value{GDBN} calculates the size to fill
10294 the value (as @samp{sizeof(@var{value})/sizeof(@var{type})}:
10296 (@value{GDBP}) p/x (short[])0x12345678
10297 $2 = @{0x1234, 0x5678@}
10300 Sometimes the artificial array mechanism is not quite enough; in
10301 moderately complex data structures, the elements of interest may not
10302 actually be adjacent---for example, if you are interested in the values
10303 of pointers in an array. One useful work-around in this situation is
10304 to use a convenience variable (@pxref{Convenience Vars, ,Convenience
10305 Variables}) as a counter in an expression that prints the first
10306 interesting value, and then repeat that expression via @key{RET}. For
10307 instance, suppose you have an array @code{dtab} of pointers to
10308 structures, and you are interested in the values of a field @code{fv}
10309 in each structure. Here is an example of what you might type:
10319 @node Output Formats
10320 @section Output Formats
10322 @cindex formatted output
10323 @cindex output formats
10324 By default, @value{GDBN} prints a value according to its data type. Sometimes
10325 this is not what you want. For example, you might want to print a number
10326 in hex, or a pointer in decimal. Or you might want to view data in memory
10327 at a certain address as a character string or as an instruction. To do
10328 these things, specify an @dfn{output format} when you print a value.
10330 The simplest use of output formats is to say how to print a value
10331 already computed. This is done by starting the arguments of the
10332 @code{print} command with a slash and a format letter. The format
10333 letters supported are:
10337 Regard the bits of the value as an integer, and print the integer in
10341 Print as integer in signed decimal.
10344 Print as integer in unsigned decimal.
10347 Print as integer in octal.
10350 Print as integer in binary. The letter @samp{t} stands for ``two''.
10351 @footnote{@samp{b} cannot be used because these format letters are also
10352 used with the @code{x} command, where @samp{b} stands for ``byte'';
10353 see @ref{Memory,,Examining Memory}.}
10356 @cindex unknown address, locating
10357 @cindex locate address
10358 Print as an address, both absolute in hexadecimal and as an offset from
10359 the nearest preceding symbol. You can use this format used to discover
10360 where (in what function) an unknown address is located:
10363 (@value{GDBP}) p/a 0x54320
10364 $3 = 0x54320 <_initialize_vx+396>
10368 The command @code{info symbol 0x54320} yields similar results.
10369 @xref{Symbols, info symbol}.
10372 Regard as an integer and print it as a character constant. This
10373 prints both the numerical value and its character representation. The
10374 character representation is replaced with the octal escape @samp{\nnn}
10375 for characters outside the 7-bit @sc{ascii} range.
10377 Without this format, @value{GDBN} displays @code{char},
10378 @w{@code{unsigned char}}, and @w{@code{signed char}} data as character
10379 constants. Single-byte members of vectors are displayed as integer
10383 Regard the bits of the value as a floating point number and print
10384 using typical floating point syntax.
10387 @cindex printing strings
10388 @cindex printing byte arrays
10389 Regard as a string, if possible. With this format, pointers to single-byte
10390 data are displayed as null-terminated strings and arrays of single-byte data
10391 are displayed as fixed-length strings. Other values are displayed in their
10394 Without this format, @value{GDBN} displays pointers to and arrays of
10395 @code{char}, @w{@code{unsigned char}}, and @w{@code{signed char}} as
10396 strings. Single-byte members of a vector are displayed as an integer
10400 Like @samp{x} formatting, the value is treated as an integer and
10401 printed as hexadecimal, but leading zeros are printed to pad the value
10402 to the size of the integer type.
10405 @cindex raw printing
10406 Print using the @samp{raw} formatting. By default, @value{GDBN} will
10407 use a Python-based pretty-printer, if one is available (@pxref{Pretty
10408 Printing}). This typically results in a higher-level display of the
10409 value's contents. The @samp{r} format bypasses any Python
10410 pretty-printer which might exist.
10413 For example, to print the program counter in hex (@pxref{Registers}), type
10420 Note that no space is required before the slash; this is because command
10421 names in @value{GDBN} cannot contain a slash.
10423 To reprint the last value in the value history with a different format,
10424 you can use the @code{print} command with just a format and no
10425 expression. For example, @samp{p/x} reprints the last value in hex.
10428 @section Examining Memory
10430 You can use the command @code{x} (for ``examine'') to examine memory in
10431 any of several formats, independently of your program's data types.
10433 @cindex examining memory
10435 @kindex x @r{(examine memory)}
10436 @item x/@var{nfu} @var{addr}
10437 @itemx x @var{addr}
10439 Use the @code{x} command to examine memory.
10442 @var{n}, @var{f}, and @var{u} are all optional parameters that specify how
10443 much memory to display and how to format it; @var{addr} is an
10444 expression giving the address where you want to start displaying memory.
10445 If you use defaults for @var{nfu}, you need not type the slash @samp{/}.
10446 Several commands set convenient defaults for @var{addr}.
10449 @item @var{n}, the repeat count
10450 The repeat count is a decimal integer; the default is 1. It specifies
10451 how much memory (counting by units @var{u}) to display. If a negative
10452 number is specified, memory is examined backward from @var{addr}.
10453 @c This really is **decimal**; unaffected by 'set radix' as of GDB
10456 @item @var{f}, the display format
10457 The display format is one of the formats used by @code{print}
10458 (@samp{x}, @samp{d}, @samp{u}, @samp{o}, @samp{t}, @samp{a}, @samp{c},
10459 @samp{f}, @samp{s}), and in addition @samp{i} (for machine instructions).
10460 The default is @samp{x} (hexadecimal) initially. The default changes
10461 each time you use either @code{x} or @code{print}.
10463 @item @var{u}, the unit size
10464 The unit size is any of
10470 Halfwords (two bytes).
10472 Words (four bytes). This is the initial default.
10474 Giant words (eight bytes).
10477 Each time you specify a unit size with @code{x}, that size becomes the
10478 default unit the next time you use @code{x}. For the @samp{i} format,
10479 the unit size is ignored and is normally not written. For the @samp{s} format,
10480 the unit size defaults to @samp{b}, unless it is explicitly given.
10481 Use @kbd{x /hs} to display 16-bit char strings and @kbd{x /ws} to display
10482 32-bit strings. The next use of @kbd{x /s} will again display 8-bit strings.
10483 Note that the results depend on the programming language of the
10484 current compilation unit. If the language is C, the @samp{s}
10485 modifier will use the UTF-16 encoding while @samp{w} will use
10486 UTF-32. The encoding is set by the programming language and cannot
10489 @item @var{addr}, starting display address
10490 @var{addr} is the address where you want @value{GDBN} to begin displaying
10491 memory. The expression need not have a pointer value (though it may);
10492 it is always interpreted as an integer address of a byte of memory.
10493 @xref{Expressions, ,Expressions}, for more information on expressions. The default for
10494 @var{addr} is usually just after the last address examined---but several
10495 other commands also set the default address: @code{info breakpoints} (to
10496 the address of the last breakpoint listed), @code{info line} (to the
10497 starting address of a line), and @code{print} (if you use it to display
10498 a value from memory).
10501 For example, @samp{x/3uh 0x54320} is a request to display three halfwords
10502 (@code{h}) of memory, formatted as unsigned decimal integers (@samp{u}),
10503 starting at address @code{0x54320}. @samp{x/4xw $sp} prints the four
10504 words (@samp{w}) of memory above the stack pointer (here, @samp{$sp};
10505 @pxref{Registers, ,Registers}) in hexadecimal (@samp{x}).
10507 You can also specify a negative repeat count to examine memory backward
10508 from the given address. For example, @samp{x/-3uh 0x54320} prints three
10509 halfwords (@code{h}) at @code{0x54314}, @code{0x54328}, and @code{0x5431c}.
10511 Since the letters indicating unit sizes are all distinct from the
10512 letters specifying output formats, you do not have to remember whether
10513 unit size or format comes first; either order works. The output
10514 specifications @samp{4xw} and @samp{4wx} mean exactly the same thing.
10515 (However, the count @var{n} must come first; @samp{wx4} does not work.)
10517 Even though the unit size @var{u} is ignored for the formats @samp{s}
10518 and @samp{i}, you might still want to use a count @var{n}; for example,
10519 @samp{3i} specifies that you want to see three machine instructions,
10520 including any operands. For convenience, especially when used with
10521 the @code{display} command, the @samp{i} format also prints branch delay
10522 slot instructions, if any, beyond the count specified, which immediately
10523 follow the last instruction that is within the count. The command
10524 @code{disassemble} gives an alternative way of inspecting machine
10525 instructions; see @ref{Machine Code,,Source and Machine Code}.
10527 If a negative repeat count is specified for the formats @samp{s} or @samp{i},
10528 the command displays null-terminated strings or instructions before the given
10529 address as many as the absolute value of the given number. For the @samp{i}
10530 format, we use line number information in the debug info to accurately locate
10531 instruction boundaries while disassembling backward. If line info is not
10532 available, the command stops examining memory with an error message.
10534 All the defaults for the arguments to @code{x} are designed to make it
10535 easy to continue scanning memory with minimal specifications each time
10536 you use @code{x}. For example, after you have inspected three machine
10537 instructions with @samp{x/3i @var{addr}}, you can inspect the next seven
10538 with just @samp{x/7}. If you use @key{RET} to repeat the @code{x} command,
10539 the repeat count @var{n} is used again; the other arguments default as
10540 for successive uses of @code{x}.
10542 When examining machine instructions, the instruction at current program
10543 counter is shown with a @code{=>} marker. For example:
10546 (@value{GDBP}) x/5i $pc-6
10547 0x804837f <main+11>: mov %esp,%ebp
10548 0x8048381 <main+13>: push %ecx
10549 0x8048382 <main+14>: sub $0x4,%esp
10550 => 0x8048385 <main+17>: movl $0x8048460,(%esp)
10551 0x804838c <main+24>: call 0x80482d4 <puts@@plt>
10554 @cindex @code{$_}, @code{$__}, and value history
10555 The addresses and contents printed by the @code{x} command are not saved
10556 in the value history because there is often too much of them and they
10557 would get in the way. Instead, @value{GDBN} makes these values available for
10558 subsequent use in expressions as values of the convenience variables
10559 @code{$_} and @code{$__}. After an @code{x} command, the last address
10560 examined is available for use in expressions in the convenience variable
10561 @code{$_}. The contents of that address, as examined, are available in
10562 the convenience variable @code{$__}.
10564 If the @code{x} command has a repeat count, the address and contents saved
10565 are from the last memory unit printed; this is not the same as the last
10566 address printed if several units were printed on the last line of output.
10568 @anchor{addressable memory unit}
10569 @cindex addressable memory unit
10570 Most targets have an addressable memory unit size of 8 bits. This means
10571 that to each memory address are associated 8 bits of data. Some
10572 targets, however, have other addressable memory unit sizes.
10573 Within @value{GDBN} and this document, the term
10574 @dfn{addressable memory unit} (or @dfn{memory unit} for short) is used
10575 when explicitly referring to a chunk of data of that size. The word
10576 @dfn{byte} is used to refer to a chunk of data of 8 bits, regardless of
10577 the addressable memory unit size of the target. For most systems,
10578 addressable memory unit is a synonym of byte.
10580 @cindex remote memory comparison
10581 @cindex target memory comparison
10582 @cindex verify remote memory image
10583 @cindex verify target memory image
10584 When you are debugging a program running on a remote target machine
10585 (@pxref{Remote Debugging}), you may wish to verify the program's image
10586 in the remote machine's memory against the executable file you
10587 downloaded to the target. Or, on any target, you may want to check
10588 whether the program has corrupted its own read-only sections. The
10589 @code{compare-sections} command is provided for such situations.
10592 @kindex compare-sections
10593 @item compare-sections @r{[}@var{section-name}@r{|}@code{-r}@r{]}
10594 Compare the data of a loadable section @var{section-name} in the
10595 executable file of the program being debugged with the same section in
10596 the target machine's memory, and report any mismatches. With no
10597 arguments, compares all loadable sections. With an argument of
10598 @code{-r}, compares all loadable read-only sections.
10600 Note: for remote targets, this command can be accelerated if the
10601 target supports computing the CRC checksum of a block of memory
10602 (@pxref{qCRC packet}).
10606 @section Automatic Display
10607 @cindex automatic display
10608 @cindex display of expressions
10610 If you find that you want to print the value of an expression frequently
10611 (to see how it changes), you might want to add it to the @dfn{automatic
10612 display list} so that @value{GDBN} prints its value each time your program stops.
10613 Each expression added to the list is given a number to identify it;
10614 to remove an expression from the list, you specify that number.
10615 The automatic display looks like this:
10619 3: bar[5] = (struct hack *) 0x3804
10623 This display shows item numbers, expressions and their current values. As with
10624 displays you request manually using @code{x} or @code{print}, you can
10625 specify the output format you prefer; in fact, @code{display} decides
10626 whether to use @code{print} or @code{x} depending your format
10627 specification---it uses @code{x} if you specify either the @samp{i}
10628 or @samp{s} format, or a unit size; otherwise it uses @code{print}.
10632 @item display @var{expr}
10633 Add the expression @var{expr} to the list of expressions to display
10634 each time your program stops. @xref{Expressions, ,Expressions}.
10636 @code{display} does not repeat if you press @key{RET} again after using it.
10638 @item display/@var{fmt} @var{expr}
10639 For @var{fmt} specifying only a display format and not a size or
10640 count, add the expression @var{expr} to the auto-display list but
10641 arrange to display it each time in the specified format @var{fmt}.
10642 @xref{Output Formats,,Output Formats}.
10644 @item display/@var{fmt} @var{addr}
10645 For @var{fmt} @samp{i} or @samp{s}, or including a unit-size or a
10646 number of units, add the expression @var{addr} as a memory address to
10647 be examined each time your program stops. Examining means in effect
10648 doing @samp{x/@var{fmt} @var{addr}}. @xref{Memory, ,Examining Memory}.
10651 For example, @samp{display/i $pc} can be helpful, to see the machine
10652 instruction about to be executed each time execution stops (@samp{$pc}
10653 is a common name for the program counter; @pxref{Registers, ,Registers}).
10656 @kindex delete display
10658 @item undisplay @var{dnums}@dots{}
10659 @itemx delete display @var{dnums}@dots{}
10660 Remove items from the list of expressions to display. Specify the
10661 numbers of the displays that you want affected with the command
10662 argument @var{dnums}. It can be a single display number, one of the
10663 numbers shown in the first field of the @samp{info display} display;
10664 or it could be a range of display numbers, as in @code{2-4}.
10666 @code{undisplay} does not repeat if you press @key{RET} after using it.
10667 (Otherwise you would just get the error @samp{No display number @dots{}}.)
10669 @kindex disable display
10670 @item disable display @var{dnums}@dots{}
10671 Disable the display of item numbers @var{dnums}. A disabled display
10672 item is not printed automatically, but is not forgotten. It may be
10673 enabled again later. Specify the numbers of the displays that you
10674 want affected with the command argument @var{dnums}. It can be a
10675 single display number, one of the numbers shown in the first field of
10676 the @samp{info display} display; or it could be a range of display
10677 numbers, as in @code{2-4}.
10679 @kindex enable display
10680 @item enable display @var{dnums}@dots{}
10681 Enable display of item numbers @var{dnums}. It becomes effective once
10682 again in auto display of its expression, until you specify otherwise.
10683 Specify the numbers of the displays that you want affected with the
10684 command argument @var{dnums}. It can be a single display number, one
10685 of the numbers shown in the first field of the @samp{info display}
10686 display; or it could be a range of display numbers, as in @code{2-4}.
10689 Display the current values of the expressions on the list, just as is
10690 done when your program stops.
10692 @kindex info display
10694 Print the list of expressions previously set up to display
10695 automatically, each one with its item number, but without showing the
10696 values. This includes disabled expressions, which are marked as such.
10697 It also includes expressions which would not be displayed right now
10698 because they refer to automatic variables not currently available.
10701 @cindex display disabled out of scope
10702 If a display expression refers to local variables, then it does not make
10703 sense outside the lexical context for which it was set up. Such an
10704 expression is disabled when execution enters a context where one of its
10705 variables is not defined. For example, if you give the command
10706 @code{display last_char} while inside a function with an argument
10707 @code{last_char}, @value{GDBN} displays this argument while your program
10708 continues to stop inside that function. When it stops elsewhere---where
10709 there is no variable @code{last_char}---the display is disabled
10710 automatically. The next time your program stops where @code{last_char}
10711 is meaningful, you can enable the display expression once again.
10713 @node Print Settings
10714 @section Print Settings
10716 @cindex format options
10717 @cindex print settings
10718 @value{GDBN} provides the following ways to control how arrays, structures,
10719 and symbols are printed.
10722 These settings are useful for debugging programs in any language:
10726 @anchor{set print address}
10727 @item set print address
10728 @itemx set print address on
10729 @cindex print/don't print memory addresses
10730 @value{GDBN} prints memory addresses showing the location of stack
10731 traces, structure values, pointer values, breakpoints, and so forth,
10732 even when it also displays the contents of those addresses. The default
10733 is @code{on}. For example, this is what a stack frame display looks like with
10734 @code{set print address on}:
10739 #0 set_quotes (lq=0x34c78 "<<", rq=0x34c88 ">>")
10741 530 if (lquote != def_lquote)
10745 @item set print address off
10746 Do not print addresses when displaying their contents. For example,
10747 this is the same stack frame displayed with @code{set print address off}:
10751 (@value{GDBP}) set print addr off
10753 #0 set_quotes (lq="<<", rq=">>") at input.c:530
10754 530 if (lquote != def_lquote)
10758 You can use @samp{set print address off} to eliminate all machine
10759 dependent displays from the @value{GDBN} interface. For example, with
10760 @code{print address off}, you should get the same text for backtraces on
10761 all machines---whether or not they involve pointer arguments.
10764 @item show print address
10765 Show whether or not addresses are to be printed.
10768 When @value{GDBN} prints a symbolic address, it normally prints the
10769 closest earlier symbol plus an offset. If that symbol does not uniquely
10770 identify the address (for example, it is a name whose scope is a single
10771 source file), you may need to clarify. One way to do this is with
10772 @code{info line}, for example @samp{info line *0x4537}. Alternately,
10773 you can set @value{GDBN} to print the source file and line number when
10774 it prints a symbolic address:
10777 @item set print symbol-filename on
10778 @cindex source file and line of a symbol
10779 @cindex symbol, source file and line
10780 Tell @value{GDBN} to print the source file name and line number of a
10781 symbol in the symbolic form of an address.
10783 @item set print symbol-filename off
10784 Do not print source file name and line number of a symbol. This is the
10787 @item show print symbol-filename
10788 Show whether or not @value{GDBN} will print the source file name and
10789 line number of a symbol in the symbolic form of an address.
10792 Another situation where it is helpful to show symbol filenames and line
10793 numbers is when disassembling code; @value{GDBN} shows you the line
10794 number and source file that corresponds to each instruction.
10796 Also, you may wish to see the symbolic form only if the address being
10797 printed is reasonably close to the closest earlier symbol:
10800 @item set print max-symbolic-offset @var{max-offset}
10801 @itemx set print max-symbolic-offset unlimited
10802 @cindex maximum value for offset of closest symbol
10803 Tell @value{GDBN} to only display the symbolic form of an address if the
10804 offset between the closest earlier symbol and the address is less than
10805 @var{max-offset}. The default is @code{unlimited}, which tells @value{GDBN}
10806 to always print the symbolic form of an address if any symbol precedes
10807 it. Zero is equivalent to @code{unlimited}.
10809 @item show print max-symbolic-offset
10810 Ask how large the maximum offset is that @value{GDBN} prints in a
10814 @cindex wild pointer, interpreting
10815 @cindex pointer, finding referent
10816 If you have a pointer and you are not sure where it points, try
10817 @samp{set print symbol-filename on}. Then you can determine the name
10818 and source file location of the variable where it points, using
10819 @samp{p/a @var{pointer}}. This interprets the address in symbolic form.
10820 For example, here @value{GDBN} shows that a variable @code{ptt} points
10821 at another variable @code{t}, defined in @file{hi2.c}:
10824 (@value{GDBP}) set print symbol-filename on
10825 (@value{GDBP}) p/a ptt
10826 $4 = 0xe008 <t in hi2.c>
10830 @emph{Warning:} For pointers that point to a local variable, @samp{p/a}
10831 does not show the symbol name and filename of the referent, even with
10832 the appropriate @code{set print} options turned on.
10835 You can also enable @samp{/a}-like formatting all the time using
10836 @samp{set print symbol on}:
10838 @anchor{set print symbol}
10840 @item set print symbol on
10841 Tell @value{GDBN} to print the symbol corresponding to an address, if
10844 @item set print symbol off
10845 Tell @value{GDBN} not to print the symbol corresponding to an
10846 address. In this mode, @value{GDBN} will still print the symbol
10847 corresponding to pointers to functions. This is the default.
10849 @item show print symbol
10850 Show whether @value{GDBN} will display the symbol corresponding to an
10854 Other settings control how different kinds of objects are printed:
10857 @anchor{set print array}
10858 @item set print array
10859 @itemx set print array on
10860 @cindex pretty print arrays
10861 Pretty print arrays. This format is more convenient to read,
10862 but uses more space. The default is off.
10864 @item set print array off
10865 Return to compressed format for arrays.
10867 @item show print array
10868 Show whether compressed or pretty format is selected for displaying
10871 @cindex print array indexes
10872 @anchor{set print array-indexes}
10873 @item set print array-indexes
10874 @itemx set print array-indexes on
10875 Print the index of each element when displaying arrays. May be more
10876 convenient to locate a given element in the array or quickly find the
10877 index of a given element in that printed array. The default is off.
10879 @item set print array-indexes off
10880 Stop printing element indexes when displaying arrays.
10882 @item show print array-indexes
10883 Show whether the index of each element is printed when displaying
10886 @anchor{set print elements}
10887 @item set print elements @var{number-of-elements}
10888 @itemx set print elements unlimited
10889 @cindex number of array elements to print
10890 @cindex limit on number of printed array elements
10891 Set a limit on how many elements of an array @value{GDBN} will print.
10892 If @value{GDBN} is printing a large array, it stops printing after it has
10893 printed the number of elements set by the @code{set print elements} command.
10894 This limit also applies to the display of strings.
10895 When @value{GDBN} starts, this limit is set to 200.
10896 Setting @var{number-of-elements} to @code{unlimited} or zero means
10897 that the number of elements to print is unlimited.
10899 @item show print elements
10900 Display the number of elements of a large array that @value{GDBN} will print.
10901 If the number is 0, then the printing is unlimited.
10903 @anchor{set print frame-arguments}
10904 @item set print frame-arguments @var{value}
10905 @kindex set print frame-arguments
10906 @cindex printing frame argument values
10907 @cindex print all frame argument values
10908 @cindex print frame argument values for scalars only
10909 @cindex do not print frame arguments
10910 This command allows to control how the values of arguments are printed
10911 when the debugger prints a frame (@pxref{Frames}). The possible
10916 The values of all arguments are printed.
10919 Print the value of an argument only if it is a scalar. The value of more
10920 complex arguments such as arrays, structures, unions, etc, is replaced
10921 by @code{@dots{}}. This is the default. Here is an example where
10922 only scalar arguments are shown:
10925 #1 0x08048361 in call_me (i=3, s=@dots{}, ss=0xbf8d508c, u=@dots{}, e=green)
10930 None of the argument values are printed. Instead, the value of each argument
10931 is replaced by @code{@dots{}}. In this case, the example above now becomes:
10934 #1 0x08048361 in call_me (i=@dots{}, s=@dots{}, ss=@dots{}, u=@dots{}, e=@dots{})
10939 Only the presence of arguments is indicated by @code{@dots{}}.
10940 The @code{@dots{}} are not printed for function without any arguments.
10941 None of the argument names and values are printed.
10942 In this case, the example above now becomes:
10945 #1 0x08048361 in call_me (@dots{}) at frame-args.c:23
10950 By default, only scalar arguments are printed. This command can be used
10951 to configure the debugger to print the value of all arguments, regardless
10952 of their type. However, it is often advantageous to not print the value
10953 of more complex parameters. For instance, it reduces the amount of
10954 information printed in each frame, making the backtrace more readable.
10955 Also, it improves performance when displaying Ada frames, because
10956 the computation of large arguments can sometimes be CPU-intensive,
10957 especially in large applications. Setting @code{print frame-arguments}
10958 to @code{scalars} (the default), @code{none} or @code{presence} avoids
10959 this computation, thus speeding up the display of each Ada frame.
10961 @item show print frame-arguments
10962 Show how the value of arguments should be displayed when printing a frame.
10964 @anchor{set print raw-frame-arguments}
10965 @item set print raw-frame-arguments on
10966 Print frame arguments in raw, non pretty-printed, form.
10968 @item set print raw-frame-arguments off
10969 Print frame arguments in pretty-printed form, if there is a pretty-printer
10970 for the value (@pxref{Pretty Printing}),
10971 otherwise print the value in raw form.
10972 This is the default.
10974 @item show print raw-frame-arguments
10975 Show whether to print frame arguments in raw form.
10977 @anchor{set print entry-values}
10978 @item set print entry-values @var{value}
10979 @kindex set print entry-values
10980 Set printing of frame argument values at function entry. In some cases
10981 @value{GDBN} can determine the value of function argument which was passed by
10982 the function caller, even if the value was modified inside the called function
10983 and therefore is different. With optimized code, the current value could be
10984 unavailable, but the entry value may still be known.
10986 The default value is @code{default} (see below for its description). Older
10987 @value{GDBN} behaved as with the setting @code{no}. Compilers not supporting
10988 this feature will behave in the @code{default} setting the same way as with the
10991 This functionality is currently supported only by DWARF 2 debugging format and
10992 the compiler has to produce @samp{DW_TAG_call_site} tags. With
10993 @value{NGCC}, you need to specify @option{-O -g} during compilation, to get
10996 The @var{value} parameter can be one of the following:
11000 Print only actual parameter values, never print values from function entry
11004 #0 different (val=6)
11005 #0 lost (val=<optimized out>)
11007 #0 invalid (val=<optimized out>)
11011 Print only parameter values from function entry point. The actual parameter
11012 values are never printed.
11014 #0 equal (val@@entry=5)
11015 #0 different (val@@entry=5)
11016 #0 lost (val@@entry=5)
11017 #0 born (val@@entry=<optimized out>)
11018 #0 invalid (val@@entry=<optimized out>)
11022 Print only parameter values from function entry point. If value from function
11023 entry point is not known while the actual value is known, print the actual
11024 value for such parameter.
11026 #0 equal (val@@entry=5)
11027 #0 different (val@@entry=5)
11028 #0 lost (val@@entry=5)
11030 #0 invalid (val@@entry=<optimized out>)
11034 Print actual parameter values. If actual parameter value is not known while
11035 value from function entry point is known, print the entry point value for such
11039 #0 different (val=6)
11040 #0 lost (val@@entry=5)
11042 #0 invalid (val=<optimized out>)
11046 Always print both the actual parameter value and its value from function entry
11047 point, even if values of one or both are not available due to compiler
11050 #0 equal (val=5, val@@entry=5)
11051 #0 different (val=6, val@@entry=5)
11052 #0 lost (val=<optimized out>, val@@entry=5)
11053 #0 born (val=10, val@@entry=<optimized out>)
11054 #0 invalid (val=<optimized out>, val@@entry=<optimized out>)
11058 Print the actual parameter value if it is known and also its value from
11059 function entry point if it is known. If neither is known, print for the actual
11060 value @code{<optimized out>}. If not in MI mode (@pxref{GDB/MI}) and if both
11061 values are known and identical, print the shortened
11062 @code{param=param@@entry=VALUE} notation.
11064 #0 equal (val=val@@entry=5)
11065 #0 different (val=6, val@@entry=5)
11066 #0 lost (val@@entry=5)
11068 #0 invalid (val=<optimized out>)
11072 Always print the actual parameter value. Print also its value from function
11073 entry point, but only if it is known. If not in MI mode (@pxref{GDB/MI}) and
11074 if both values are known and identical, print the shortened
11075 @code{param=param@@entry=VALUE} notation.
11077 #0 equal (val=val@@entry=5)
11078 #0 different (val=6, val@@entry=5)
11079 #0 lost (val=<optimized out>, val@@entry=5)
11081 #0 invalid (val=<optimized out>)
11085 For analysis messages on possible failures of frame argument values at function
11086 entry resolution see @ref{set debug entry-values}.
11088 @item show print entry-values
11089 Show the method being used for printing of frame argument values at function
11092 @anchor{set print frame-info}
11093 @item set print frame-info @var{value}
11094 @kindex set print frame-info
11095 @cindex printing frame information
11096 @cindex frame information, printing
11097 This command allows to control the information printed when
11098 the debugger prints a frame. See @ref{Frames}, @ref{Backtrace},
11099 for a general explanation about frames and frame information.
11100 Note that some other settings (such as @code{set print frame-arguments}
11101 and @code{set print address}) are also influencing if and how some frame
11102 information is displayed. In particular, the frame program counter is never
11103 printed if @code{set print address} is off.
11105 The possible values for @code{set print frame-info} are:
11107 @item short-location
11108 Print the frame level, the program counter (if not at the
11109 beginning of the location source line), the function, the function
11112 Same as @code{short-location} but also print the source file and source line
11114 @item location-and-address
11115 Same as @code{location} but print the program counter even if located at the
11116 beginning of the location source line.
11118 Print the program counter (if not at the beginning of the location
11119 source line), the line number and the source line.
11120 @item source-and-location
11121 Print what @code{location} and @code{source-line} are printing.
11123 The information printed for a frame is decided automatically
11124 by the @value{GDBN} command that prints a frame.
11125 For example, @code{frame} prints the information printed by
11126 @code{source-and-location} while @code{stepi} will switch between
11127 @code{source-line} and @code{source-and-location} depending on the program
11129 The default value is @code{auto}.
11132 @anchor{set print repeats}
11133 @item set print repeats @var{number-of-repeats}
11134 @itemx set print repeats unlimited
11135 @cindex repeated array elements
11136 Set the threshold for suppressing display of repeated array
11137 elements. When the number of consecutive identical elements of an
11138 array exceeds the threshold, @value{GDBN} prints the string
11139 @code{"<repeats @var{n} times>"}, where @var{n} is the number of
11140 identical repetitions, instead of displaying the identical elements
11141 themselves. Setting the threshold to @code{unlimited} or zero will
11142 cause all elements to be individually printed. The default threshold
11145 @item show print repeats
11146 Display the current threshold for printing repeated identical
11149 @anchor{set print max-depth}
11150 @item set print max-depth @var{depth}
11151 @item set print max-depth unlimited
11152 @cindex printing nested structures
11153 Set the threshold after which nested structures are replaced with
11154 ellipsis, this can make visualising deeply nested structures easier.
11156 For example, given this C code
11159 typedef struct s1 @{ int a; @} s1;
11160 typedef struct s2 @{ s1 b; @} s2;
11161 typedef struct s3 @{ s2 c; @} s3;
11162 typedef struct s4 @{ s3 d; @} s4;
11164 s4 var = @{ @{ @{ @{ 3 @} @} @} @};
11167 The following table shows how different values of @var{depth} will
11168 effect how @code{var} is printed by @value{GDBN}:
11170 @multitable @columnfractions .3 .7
11171 @headitem @var{depth} setting @tab Result of @samp{p var}
11173 @tab @code{$1 = @{d = @{c = @{b = @{a = 3@}@}@}@}}
11175 @tab @code{$1 = @{...@}}
11177 @tab @code{$1 = @{d = @{...@}@}}
11179 @tab @code{$1 = @{d = @{c = @{...@}@}@}}
11181 @tab @code{$1 = @{d = @{c = @{b = @{...@}@}@}@}}
11183 @tab @code{$1 = @{d = @{c = @{b = @{a = 3@}@}@}@}}
11186 To see the contents of structures that have been hidden the user can
11187 either increase the print max-depth, or they can print the elements of
11188 the structure that are visible, for example
11191 (gdb) set print max-depth 2
11193 $1 = @{d = @{c = @{...@}@}@}
11195 $2 = @{c = @{b = @{...@}@}@}
11197 $3 = @{b = @{a = 3@}@}
11200 The pattern used to replace nested structures varies based on
11201 language, for most languages @code{@{...@}} is used, but Fortran uses
11204 @item show print max-depth
11205 Display the current threshold after which nested structures are
11206 replaces with ellipsis.
11208 @anchor{set print null-stop}
11209 @item set print null-stop
11210 @cindex @sc{null} elements in arrays
11211 Cause @value{GDBN} to stop printing the characters of an array when the first
11212 @sc{null} is encountered. This is useful when large arrays actually
11213 contain only short strings.
11214 The default is off.
11216 @item show print null-stop
11217 Show whether @value{GDBN} stops printing an array on the first
11218 @sc{null} character.
11220 @anchor{set print pretty}
11221 @item set print pretty on
11222 @cindex print structures in indented form
11223 @cindex indentation in structure display
11224 Cause @value{GDBN} to print structures in an indented format with one member
11225 per line, like this:
11240 @item set print pretty off
11241 Cause @value{GDBN} to print structures in a compact format, like this:
11245 $1 = @{next = 0x0, flags = @{sweet = 1, sour = 1@}, \
11246 meat = 0x54 "Pork"@}
11251 This is the default format.
11253 @item show print pretty
11254 Show which format @value{GDBN} is using to print structures.
11256 @item set print sevenbit-strings on
11257 @cindex eight-bit characters in strings
11258 @cindex octal escapes in strings
11259 Print using only seven-bit characters; if this option is set,
11260 @value{GDBN} displays any eight-bit characters (in strings or
11261 character values) using the notation @code{\}@var{nnn}. This setting is
11262 best if you are working in English (@sc{ascii}) and you use the
11263 high-order bit of characters as a marker or ``meta'' bit.
11265 @item set print sevenbit-strings off
11266 Print full eight-bit characters. This allows the use of more
11267 international character sets, and is the default.
11269 @item show print sevenbit-strings
11270 Show whether or not @value{GDBN} is printing only seven-bit characters.
11272 @anchor{set print union}
11273 @item set print union on
11274 @cindex unions in structures, printing
11275 Tell @value{GDBN} to print unions which are contained in structures
11276 and other unions. This is the default setting.
11278 @item set print union off
11279 Tell @value{GDBN} not to print unions which are contained in
11280 structures and other unions. @value{GDBN} will print @code{"@{...@}"}
11283 @item show print union
11284 Ask @value{GDBN} whether or not it will print unions which are contained in
11285 structures and other unions.
11287 For example, given the declarations
11290 typedef enum @{Tree, Bug@} Species;
11291 typedef enum @{Big_tree, Acorn, Seedling@} Tree_forms;
11292 typedef enum @{Caterpillar, Cocoon, Butterfly@}
11303 struct thing foo = @{Tree, @{Acorn@}@};
11307 with @code{set print union on} in effect @samp{p foo} would print
11310 $1 = @{it = Tree, form = @{tree = Acorn, bug = Cocoon@}@}
11314 and with @code{set print union off} in effect it would print
11317 $1 = @{it = Tree, form = @{...@}@}
11321 @code{set print union} affects programs written in C-like languages
11327 These settings are of interest when debugging C@t{++} programs:
11330 @cindex demangling C@t{++} names
11331 @item set print demangle
11332 @itemx set print demangle on
11333 Print C@t{++} names in their source form rather than in the encoded
11334 (``mangled'') form passed to the assembler and linker for type-safe
11335 linkage. The default is on.
11337 @item show print demangle
11338 Show whether C@t{++} names are printed in mangled or demangled form.
11340 @item set print asm-demangle
11341 @itemx set print asm-demangle on
11342 Print C@t{++} names in their source form rather than their mangled form, even
11343 in assembler code printouts such as instruction disassemblies.
11344 The default is off.
11346 @item show print asm-demangle
11347 Show whether C@t{++} names in assembly listings are printed in mangled
11350 @cindex C@t{++} symbol decoding style
11351 @cindex symbol decoding style, C@t{++}
11352 @kindex set demangle-style
11353 @item set demangle-style @var{style}
11354 Choose among several encoding schemes used by different compilers to represent
11355 C@t{++} names. If you omit @var{style}, you will see a list of possible
11356 formats. The default value is @var{auto}, which lets @value{GDBN} choose a
11357 decoding style by inspecting your program.
11359 @item show demangle-style
11360 Display the encoding style currently in use for decoding C@t{++} symbols.
11362 @anchor{set print object}
11363 @item set print object
11364 @itemx set print object on
11365 @cindex derived type of an object, printing
11366 @cindex display derived types
11367 When displaying a pointer to an object, identify the @emph{actual}
11368 (derived) type of the object rather than the @emph{declared} type, using
11369 the virtual function table. Note that the virtual function table is
11370 required---this feature can only work for objects that have run-time
11371 type identification; a single virtual method in the object's declared
11372 type is sufficient. Note that this setting is also taken into account when
11373 working with variable objects via MI (@pxref{GDB/MI}).
11375 @item set print object off
11376 Display only the declared type of objects, without reference to the
11377 virtual function table. This is the default setting.
11379 @item show print object
11380 Show whether actual, or declared, object types are displayed.
11382 @anchor{set print static-members}
11383 @item set print static-members
11384 @itemx set print static-members on
11385 @cindex static members of C@t{++} objects
11386 Print static members when displaying a C@t{++} object. The default is on.
11388 @item set print static-members off
11389 Do not print static members when displaying a C@t{++} object.
11391 @item show print static-members
11392 Show whether C@t{++} static members are printed or not.
11394 @item set print pascal_static-members
11395 @itemx set print pascal_static-members on
11396 @cindex static members of Pascal objects
11397 @cindex Pascal objects, static members display
11398 Print static members when displaying a Pascal object. The default is on.
11400 @item set print pascal_static-members off
11401 Do not print static members when displaying a Pascal object.
11403 @item show print pascal_static-members
11404 Show whether Pascal static members are printed or not.
11406 @c These don't work with HP ANSI C++ yet.
11407 @anchor{set print vtbl}
11408 @item set print vtbl
11409 @itemx set print vtbl on
11410 @cindex pretty print C@t{++} virtual function tables
11411 @cindex virtual functions (C@t{++}) display
11412 @cindex VTBL display
11413 Pretty print C@t{++} virtual function tables. The default is off.
11414 (The @code{vtbl} commands do not work on programs compiled with the HP
11415 ANSI C@t{++} compiler (@code{aCC}).)
11417 @item set print vtbl off
11418 Do not pretty print C@t{++} virtual function tables.
11420 @item show print vtbl
11421 Show whether C@t{++} virtual function tables are pretty printed, or not.
11424 @node Pretty Printing
11425 @section Pretty Printing
11427 @value{GDBN} provides a mechanism to allow pretty-printing of values using
11428 Python code. It greatly simplifies the display of complex objects. This
11429 mechanism works for both MI and the CLI.
11432 * Pretty-Printer Introduction:: Introduction to pretty-printers
11433 * Pretty-Printer Example:: An example pretty-printer
11434 * Pretty-Printer Commands:: Pretty-printer commands
11437 @node Pretty-Printer Introduction
11438 @subsection Pretty-Printer Introduction
11440 When @value{GDBN} prints a value, it first sees if there is a pretty-printer
11441 registered for the value. If there is then @value{GDBN} invokes the
11442 pretty-printer to print the value. Otherwise the value is printed normally.
11444 Pretty-printers are normally named. This makes them easy to manage.
11445 The @samp{info pretty-printer} command will list all the installed
11446 pretty-printers with their names.
11447 If a pretty-printer can handle multiple data types, then its
11448 @dfn{subprinters} are the printers for the individual data types.
11449 Each such subprinter has its own name.
11450 The format of the name is @var{printer-name};@var{subprinter-name}.
11452 Pretty-printers are installed by @dfn{registering} them with @value{GDBN}.
11453 Typically they are automatically loaded and registered when the corresponding
11454 debug information is loaded, thus making them available without having to
11455 do anything special.
11457 There are three places where a pretty-printer can be registered.
11461 Pretty-printers registered globally are available when debugging
11465 Pretty-printers registered with a program space are available only
11466 when debugging that program.
11467 @xref{Progspaces In Python}, for more details on program spaces in Python.
11470 Pretty-printers registered with an objfile are loaded and unloaded
11471 with the corresponding objfile (e.g., shared library).
11472 @xref{Objfiles In Python}, for more details on objfiles in Python.
11475 @xref{Selecting Pretty-Printers}, for further information on how
11476 pretty-printers are selected,
11478 @xref{Writing a Pretty-Printer}, for implementing pretty printers
11481 @node Pretty-Printer Example
11482 @subsection Pretty-Printer Example
11484 Here is how a C@t{++} @code{std::string} looks without a pretty-printer:
11487 (@value{GDBP}) print s
11489 static npos = 4294967295,
11491 <std::allocator<char>> = @{
11492 <__gnu_cxx::new_allocator<char>> = @{
11493 <No data fields>@}, <No data fields>
11495 members of std::basic_string<char, std::char_traits<char>,
11496 std::allocator<char> >::_Alloc_hider:
11497 _M_p = 0x804a014 "abcd"
11502 With a pretty-printer for @code{std::string} only the contents are printed:
11505 (@value{GDBP}) print s
11509 @node Pretty-Printer Commands
11510 @subsection Pretty-Printer Commands
11511 @cindex pretty-printer commands
11514 @kindex info pretty-printer
11515 @item info pretty-printer [@var{object-regexp} [@var{name-regexp}]]
11516 Print the list of installed pretty-printers.
11517 This includes disabled pretty-printers, which are marked as such.
11519 @var{object-regexp} is a regular expression matching the objects
11520 whose pretty-printers to list.
11521 Objects can be @code{global}, the program space's file
11522 (@pxref{Progspaces In Python}),
11523 and the object files within that program space (@pxref{Objfiles In Python}).
11524 @xref{Selecting Pretty-Printers}, for details on how @value{GDBN}
11525 looks up a printer from these three objects.
11527 @var{name-regexp} is a regular expression matching the name of the printers
11530 @kindex disable pretty-printer
11531 @item disable pretty-printer [@var{object-regexp} [@var{name-regexp}]]
11532 Disable pretty-printers matching @var{object-regexp} and @var{name-regexp}.
11533 A disabled pretty-printer is not forgotten, it may be enabled again later.
11535 @kindex enable pretty-printer
11536 @item enable pretty-printer [@var{object-regexp} [@var{name-regexp}]]
11537 Enable pretty-printers matching @var{object-regexp} and @var{name-regexp}.
11542 Suppose we have three pretty-printers installed: one from library1.so
11543 named @code{foo} that prints objects of type @code{foo}, and
11544 another from library2.so named @code{bar} that prints two types of objects,
11545 @code{bar1} and @code{bar2}.
11548 (gdb) info pretty-printer
11555 (gdb) info pretty-printer library2
11560 (gdb) disable pretty-printer library1
11562 2 of 3 printers enabled
11563 (gdb) info pretty-printer
11570 (gdb) disable pretty-printer library2 bar;bar1
11572 1 of 3 printers enabled
11573 (gdb) info pretty-printer library2
11580 (gdb) disable pretty-printer library2 bar
11582 0 of 3 printers enabled
11583 (gdb) info pretty-printer library2
11592 Note that for @code{bar} the entire printer can be disabled,
11593 as can each individual subprinter.
11595 @node Value History
11596 @section Value History
11598 @cindex value history
11599 @cindex history of values printed by @value{GDBN}
11600 Values printed by the @code{print} command are saved in the @value{GDBN}
11601 @dfn{value history}. This allows you to refer to them in other expressions.
11602 Values are kept until the symbol table is re-read or discarded
11603 (for example with the @code{file} or @code{symbol-file} commands).
11604 When the symbol table changes, the value history is discarded,
11605 since the values may contain pointers back to the types defined in the
11610 @cindex history number
11611 The values printed are given @dfn{history numbers} by which you can
11612 refer to them. These are successive integers starting with one.
11613 @code{print} shows you the history number assigned to a value by
11614 printing @samp{$@var{num} = } before the value; here @var{num} is the
11617 To refer to any previous value, use @samp{$} followed by the value's
11618 history number. The way @code{print} labels its output is designed to
11619 remind you of this. Just @code{$} refers to the most recent value in
11620 the history, and @code{$$} refers to the value before that.
11621 @code{$$@var{n}} refers to the @var{n}th value from the end; @code{$$2}
11622 is the value just prior to @code{$$}, @code{$$1} is equivalent to
11623 @code{$$}, and @code{$$0} is equivalent to @code{$}.
11625 For example, suppose you have just printed a pointer to a structure and
11626 want to see the contents of the structure. It suffices to type
11632 If you have a chain of structures where the component @code{next} points
11633 to the next one, you can print the contents of the next one with this:
11640 You can print successive links in the chain by repeating this
11641 command---which you can do by just typing @key{RET}.
11643 Note that the history records values, not expressions. If the value of
11644 @code{x} is 4 and you type these commands:
11652 then the value recorded in the value history by the @code{print} command
11653 remains 4 even though the value of @code{x} has changed.
11656 @kindex show values
11658 Print the last ten values in the value history, with their item numbers.
11659 This is like @samp{p@ $$9} repeated ten times, except that @code{show
11660 values} does not change the history.
11662 @item show values @var{n}
11663 Print ten history values centered on history item number @var{n}.
11665 @item show values +
11666 Print ten history values just after the values last printed. If no more
11667 values are available, @code{show values +} produces no display.
11670 Pressing @key{RET} to repeat @code{show values @var{n}} has exactly the
11671 same effect as @samp{show values +}.
11673 @node Convenience Vars
11674 @section Convenience Variables
11676 @cindex convenience variables
11677 @cindex user-defined variables
11678 @value{GDBN} provides @dfn{convenience variables} that you can use within
11679 @value{GDBN} to hold on to a value and refer to it later. These variables
11680 exist entirely within @value{GDBN}; they are not part of your program, and
11681 setting a convenience variable has no direct effect on further execution
11682 of your program. That is why you can use them freely.
11684 Convenience variables are prefixed with @samp{$}. Any name preceded by
11685 @samp{$} can be used for a convenience variable, unless it is one of
11686 the predefined machine-specific register names (@pxref{Registers, ,Registers}).
11687 (Value history references, in contrast, are @emph{numbers} preceded
11688 by @samp{$}. @xref{Value History, ,Value History}.)
11690 You can save a value in a convenience variable with an assignment
11691 expression, just as you would set a variable in your program.
11695 set $foo = *object_ptr
11699 would save in @code{$foo} the value contained in the object pointed to by
11702 Using a convenience variable for the first time creates it, but its
11703 value is @code{void} until you assign a new value. You can alter the
11704 value with another assignment at any time.
11706 Convenience variables have no fixed types. You can assign a convenience
11707 variable any type of value, including structures and arrays, even if
11708 that variable already has a value of a different type. The convenience
11709 variable, when used as an expression, has the type of its current value.
11712 @kindex show convenience
11713 @cindex show all user variables and functions
11714 @item show convenience
11715 Print a list of convenience variables used so far, and their values,
11716 as well as a list of the convenience functions.
11717 Abbreviated @code{show conv}.
11719 @kindex init-if-undefined
11720 @cindex convenience variables, initializing
11721 @item init-if-undefined $@var{variable} = @var{expression}
11722 Set a convenience variable if it has not already been set. This is useful
11723 for user-defined commands that keep some state. It is similar, in concept,
11724 to using local static variables with initializers in C (except that
11725 convenience variables are global). It can also be used to allow users to
11726 override default values used in a command script.
11728 If the variable is already defined then the expression is not evaluated so
11729 any side-effects do not occur.
11732 One of the ways to use a convenience variable is as a counter to be
11733 incremented or a pointer to be advanced. For example, to print
11734 a field from successive elements of an array of structures:
11738 print bar[$i++]->contents
11742 Repeat that command by typing @key{RET}.
11744 Some convenience variables are created automatically by @value{GDBN} and given
11745 values likely to be useful.
11748 @vindex $_@r{, convenience variable}
11750 The variable @code{$_} is automatically set by the @code{x} command to
11751 the last address examined (@pxref{Memory, ,Examining Memory}). Other
11752 commands which provide a default address for @code{x} to examine also
11753 set @code{$_} to that address; these commands include @code{info line}
11754 and @code{info breakpoint}. The type of @code{$_} is @code{void *}
11755 except when set by the @code{x} command, in which case it is a pointer
11756 to the type of @code{$__}.
11758 @vindex $__@r{, convenience variable}
11760 The variable @code{$__} is automatically set by the @code{x} command
11761 to the value found in the last address examined. Its type is chosen
11762 to match the format in which the data was printed.
11765 @vindex $_exitcode@r{, convenience variable}
11766 When the program being debugged terminates normally, @value{GDBN}
11767 automatically sets this variable to the exit code of the program, and
11768 resets @code{$_exitsignal} to @code{void}.
11771 @vindex $_exitsignal@r{, convenience variable}
11772 When the program being debugged dies due to an uncaught signal,
11773 @value{GDBN} automatically sets this variable to that signal's number,
11774 and resets @code{$_exitcode} to @code{void}.
11776 To distinguish between whether the program being debugged has exited
11777 (i.e., @code{$_exitcode} is not @code{void}) or signalled (i.e.,
11778 @code{$_exitsignal} is not @code{void}), the convenience function
11779 @code{$_isvoid} can be used (@pxref{Convenience Funs,, Convenience
11780 Functions}). For example, considering the following source code:
11783 #include <signal.h>
11786 main (int argc, char *argv[])
11793 A valid way of telling whether the program being debugged has exited
11794 or signalled would be:
11797 (@value{GDBP}) define has_exited_or_signalled
11798 Type commands for definition of ``has_exited_or_signalled''.
11799 End with a line saying just ``end''.
11800 >if $_isvoid ($_exitsignal)
11801 >echo The program has exited\n
11803 >echo The program has signalled\n
11809 Program terminated with signal SIGALRM, Alarm clock.
11810 The program no longer exists.
11811 (@value{GDBP}) has_exited_or_signalled
11812 The program has signalled
11815 As can be seen, @value{GDBN} correctly informs that the program being
11816 debugged has signalled, since it calls @code{raise} and raises a
11817 @code{SIGALRM} signal. If the program being debugged had not called
11818 @code{raise}, then @value{GDBN} would report a normal exit:
11821 (@value{GDBP}) has_exited_or_signalled
11822 The program has exited
11826 The variable @code{$_exception} is set to the exception object being
11827 thrown at an exception-related catchpoint. @xref{Set Catchpoints}.
11830 @itemx $_probe_arg0@dots{}$_probe_arg11
11831 Arguments to a static probe. @xref{Static Probe Points}.
11834 @vindex $_sdata@r{, inspect, convenience variable}
11835 The variable @code{$_sdata} contains extra collected static tracepoint
11836 data. @xref{Tracepoint Actions,,Tracepoint Action Lists}. Note that
11837 @code{$_sdata} could be empty, if not inspecting a trace buffer, or
11838 if extra static tracepoint data has not been collected.
11841 @vindex $_siginfo@r{, convenience variable}
11842 The variable @code{$_siginfo} contains extra signal information
11843 (@pxref{extra signal information}). Note that @code{$_siginfo}
11844 could be empty, if the application has not yet received any signals.
11845 For example, it will be empty before you execute the @code{run} command.
11848 @vindex $_tlb@r{, convenience variable}
11849 The variable @code{$_tlb} is automatically set when debugging
11850 applications running on MS-Windows in native mode or connected to
11851 gdbserver that supports the @code{qGetTIBAddr} request.
11852 @xref{General Query Packets}.
11853 This variable contains the address of the thread information block.
11856 The number of the current inferior. @xref{Inferiors and
11857 Programs, ,Debugging Multiple Inferiors and Programs}.
11860 The thread number of the current thread. @xref{thread numbers}.
11863 The global number of the current thread. @xref{global thread numbers}.
11867 @vindex $_gdb_major@r{, convenience variable}
11868 @vindex $_gdb_minor@r{, convenience variable}
11869 The major and minor version numbers of the running @value{GDBN}.
11870 Development snapshots and pretest versions have their minor version
11871 incremented by one; thus, @value{GDBN} pretest 9.11.90 will produce
11872 the value 12 for @code{$_gdb_minor}. These variables allow you to
11873 write scripts that work with different versions of @value{GDBN}
11874 without errors caused by features unavailable in some of those
11877 @item $_shell_exitcode
11878 @itemx $_shell_exitsignal
11879 @vindex $_shell_exitcode@r{, convenience variable}
11880 @vindex $_shell_exitsignal@r{, convenience variable}
11881 @cindex shell command, exit code
11882 @cindex shell command, exit signal
11883 @cindex exit status of shell commands
11884 @value{GDBN} commands such as @code{shell} and @code{|} are launching
11885 shell commands. When a launched command terminates, @value{GDBN}
11886 automatically maintains the variables @code{$_shell_exitcode}
11887 and @code{$_shell_exitsignal} according to the exit status of the last
11888 launched command. These variables are set and used similarly to
11889 the variables @code{$_exitcode} and @code{$_exitsignal}.
11893 @node Convenience Funs
11894 @section Convenience Functions
11896 @cindex convenience functions
11897 @value{GDBN} also supplies some @dfn{convenience functions}. These
11898 have a syntax similar to convenience variables. A convenience
11899 function can be used in an expression just like an ordinary function;
11900 however, a convenience function is implemented internally to
11903 These functions do not require @value{GDBN} to be configured with
11904 @code{Python} support, which means that they are always available.
11908 @item $_isvoid (@var{expr})
11909 @findex $_isvoid@r{, convenience function}
11910 Return one if the expression @var{expr} is @code{void}. Otherwise it
11913 A @code{void} expression is an expression where the type of the result
11914 is @code{void}. For example, you can examine a convenience variable
11915 (see @ref{Convenience Vars,, Convenience Variables}) to check whether
11919 (@value{GDBP}) print $_exitcode
11921 (@value{GDBP}) print $_isvoid ($_exitcode)
11924 Starting program: ./a.out
11925 [Inferior 1 (process 29572) exited normally]
11926 (@value{GDBP}) print $_exitcode
11928 (@value{GDBP}) print $_isvoid ($_exitcode)
11932 In the example above, we used @code{$_isvoid} to check whether
11933 @code{$_exitcode} is @code{void} before and after the execution of the
11934 program being debugged. Before the execution there is no exit code to
11935 be examined, therefore @code{$_exitcode} is @code{void}. After the
11936 execution the program being debugged returned zero, therefore
11937 @code{$_exitcode} is zero, which means that it is not @code{void}
11940 The @code{void} expression can also be a call of a function from the
11941 program being debugged. For example, given the following function:
11950 The result of calling it inside @value{GDBN} is @code{void}:
11953 (@value{GDBP}) print foo ()
11955 (@value{GDBP}) print $_isvoid (foo ())
11957 (@value{GDBP}) set $v = foo ()
11958 (@value{GDBP}) print $v
11960 (@value{GDBP}) print $_isvoid ($v)
11966 These functions require @value{GDBN} to be configured with
11967 @code{Python} support.
11971 @item $_memeq(@var{buf1}, @var{buf2}, @var{length})
11972 @findex $_memeq@r{, convenience function}
11973 Returns one if the @var{length} bytes at the addresses given by
11974 @var{buf1} and @var{buf2} are equal.
11975 Otherwise it returns zero.
11977 @item $_regex(@var{str}, @var{regex})
11978 @findex $_regex@r{, convenience function}
11979 Returns one if the string @var{str} matches the regular expression
11980 @var{regex}. Otherwise it returns zero.
11981 The syntax of the regular expression is that specified by @code{Python}'s
11982 regular expression support.
11984 @item $_streq(@var{str1}, @var{str2})
11985 @findex $_streq@r{, convenience function}
11986 Returns one if the strings @var{str1} and @var{str2} are equal.
11987 Otherwise it returns zero.
11989 @item $_strlen(@var{str})
11990 @findex $_strlen@r{, convenience function}
11991 Returns the length of string @var{str}.
11993 @item $_caller_is(@var{name}@r{[}, @var{number_of_frames}@r{]})
11994 @findex $_caller_is@r{, convenience function}
11995 Returns one if the calling function's name is equal to @var{name}.
11996 Otherwise it returns zero.
11998 If the optional argument @var{number_of_frames} is provided,
11999 it is the number of frames up in the stack to look.
12007 at testsuite/gdb.python/py-caller-is.c:21
12008 #1 0x00000000004005a0 in middle_func ()
12009 at testsuite/gdb.python/py-caller-is.c:27
12010 #2 0x00000000004005ab in top_func ()
12011 at testsuite/gdb.python/py-caller-is.c:33
12012 #3 0x00000000004005b6 in main ()
12013 at testsuite/gdb.python/py-caller-is.c:39
12014 (gdb) print $_caller_is ("middle_func")
12016 (gdb) print $_caller_is ("top_func", 2)
12020 @item $_caller_matches(@var{regexp}@r{[}, @var{number_of_frames}@r{]})
12021 @findex $_caller_matches@r{, convenience function}
12022 Returns one if the calling function's name matches the regular expression
12023 @var{regexp}. Otherwise it returns zero.
12025 If the optional argument @var{number_of_frames} is provided,
12026 it is the number of frames up in the stack to look.
12029 @item $_any_caller_is(@var{name}@r{[}, @var{number_of_frames}@r{]})
12030 @findex $_any_caller_is@r{, convenience function}
12031 Returns one if any calling function's name is equal to @var{name}.
12032 Otherwise it returns zero.
12034 If the optional argument @var{number_of_frames} is provided,
12035 it is the number of frames up in the stack to look.
12038 This function differs from @code{$_caller_is} in that this function
12039 checks all stack frames from the immediate caller to the frame specified
12040 by @var{number_of_frames}, whereas @code{$_caller_is} only checks the
12041 frame specified by @var{number_of_frames}.
12043 @item $_any_caller_matches(@var{regexp}@r{[}, @var{number_of_frames}@r{]})
12044 @findex $_any_caller_matches@r{, convenience function}
12045 Returns one if any calling function's name matches the regular expression
12046 @var{regexp}. Otherwise it returns zero.
12048 If the optional argument @var{number_of_frames} is provided,
12049 it is the number of frames up in the stack to look.
12052 This function differs from @code{$_caller_matches} in that this function
12053 checks all stack frames from the immediate caller to the frame specified
12054 by @var{number_of_frames}, whereas @code{$_caller_matches} only checks the
12055 frame specified by @var{number_of_frames}.
12057 @item $_as_string(@var{value})
12058 @findex $_as_string@r{, convenience function}
12059 Return the string representation of @var{value}.
12061 This function is useful to obtain the textual label (enumerator) of an
12062 enumeration value. For example, assuming the variable @var{node} is of
12063 an enumerated type:
12066 (gdb) printf "Visiting node of type %s\n", $_as_string(node)
12067 Visiting node of type NODE_INTEGER
12070 @item $_cimag(@var{value})
12071 @itemx $_creal(@var{value})
12072 @findex $_cimag@r{, convenience function}
12073 @findex $_creal@r{, convenience function}
12074 Return the imaginary (@code{$_cimag}) or real (@code{$_creal}) part of
12075 the complex number @var{value}.
12077 The type of the imaginary or real part depends on the type of the
12078 complex number, e.g., using @code{$_cimag} on a @code{float complex}
12079 will return an imaginary part of type @code{float}.
12083 @value{GDBN} provides the ability to list and get help on
12084 convenience functions.
12087 @item help function
12088 @kindex help function
12089 @cindex show all convenience functions
12090 Print a list of all convenience functions.
12097 You can refer to machine register contents, in expressions, as variables
12098 with names starting with @samp{$}. The names of registers are different
12099 for each machine; use @code{info registers} to see the names used on
12103 @kindex info registers
12104 @item info registers
12105 Print the names and values of all registers except floating-point
12106 and vector registers (in the selected stack frame).
12108 @kindex info all-registers
12109 @cindex floating point registers
12110 @item info all-registers
12111 Print the names and values of all registers, including floating-point
12112 and vector registers (in the selected stack frame).
12114 @item info registers @var{reggroup} @dots{}
12115 Print the name and value of the registers in each of the specified
12116 @var{reggroup}s. The @var{reggoup} can be any of those returned by
12117 @code{maint print reggroups} (@pxref{Maintenance Commands}).
12119 @item info registers @var{regname} @dots{}
12120 Print the @dfn{relativized} value of each specified register @var{regname}.
12121 As discussed in detail below, register values are normally relative to
12122 the selected stack frame. The @var{regname} may be any register name valid on
12123 the machine you are using, with or without the initial @samp{$}.
12126 @anchor{standard registers}
12127 @cindex stack pointer register
12128 @cindex program counter register
12129 @cindex process status register
12130 @cindex frame pointer register
12131 @cindex standard registers
12132 @value{GDBN} has four ``standard'' register names that are available (in
12133 expressions) on most machines---whenever they do not conflict with an
12134 architecture's canonical mnemonics for registers. The register names
12135 @code{$pc} and @code{$sp} are used for the program counter register and
12136 the stack pointer. @code{$fp} is used for a register that contains a
12137 pointer to the current stack frame, and @code{$ps} is used for a
12138 register that contains the processor status. For example,
12139 you could print the program counter in hex with
12146 or print the instruction to be executed next with
12153 or add four to the stack pointer@footnote{This is a way of removing
12154 one word from the stack, on machines where stacks grow downward in
12155 memory (most machines, nowadays). This assumes that the innermost
12156 stack frame is selected; setting @code{$sp} is not allowed when other
12157 stack frames are selected. To pop entire frames off the stack,
12158 regardless of machine architecture, use @code{return};
12159 see @ref{Returning, ,Returning from a Function}.} with
12165 Whenever possible, these four standard register names are available on
12166 your machine even though the machine has different canonical mnemonics,
12167 so long as there is no conflict. The @code{info registers} command
12168 shows the canonical names. For example, on the SPARC, @code{info
12169 registers} displays the processor status register as @code{$psr} but you
12170 can also refer to it as @code{$ps}; and on x86-based machines @code{$ps}
12171 is an alias for the @sc{eflags} register.
12173 @value{GDBN} always considers the contents of an ordinary register as an
12174 integer when the register is examined in this way. Some machines have
12175 special registers which can hold nothing but floating point; these
12176 registers are considered to have floating point values. There is no way
12177 to refer to the contents of an ordinary register as floating point value
12178 (although you can @emph{print} it as a floating point value with
12179 @samp{print/f $@var{regname}}).
12181 Some registers have distinct ``raw'' and ``virtual'' data formats. This
12182 means that the data format in which the register contents are saved by
12183 the operating system is not the same one that your program normally
12184 sees. For example, the registers of the 68881 floating point
12185 coprocessor are always saved in ``extended'' (raw) format, but all C
12186 programs expect to work with ``double'' (virtual) format. In such
12187 cases, @value{GDBN} normally works with the virtual format only (the format
12188 that makes sense for your program), but the @code{info registers} command
12189 prints the data in both formats.
12191 @cindex SSE registers (x86)
12192 @cindex MMX registers (x86)
12193 Some machines have special registers whose contents can be interpreted
12194 in several different ways. For example, modern x86-based machines
12195 have SSE and MMX registers that can hold several values packed
12196 together in several different formats. @value{GDBN} refers to such
12197 registers in @code{struct} notation:
12200 (@value{GDBP}) print $xmm1
12202 v4_float = @{0, 3.43859137e-038, 1.54142831e-044, 1.821688e-044@},
12203 v2_double = @{9.92129282474342e-303, 2.7585945287983262e-313@},
12204 v16_int8 = "\000\000\000\000\3706;\001\v\000\000\000\r\000\000",
12205 v8_int16 = @{0, 0, 14072, 315, 11, 0, 13, 0@},
12206 v4_int32 = @{0, 20657912, 11, 13@},
12207 v2_int64 = @{88725056443645952, 55834574859@},
12208 uint128 = 0x0000000d0000000b013b36f800000000
12213 To set values of such registers, you need to tell @value{GDBN} which
12214 view of the register you wish to change, as if you were assigning
12215 value to a @code{struct} member:
12218 (@value{GDBP}) set $xmm1.uint128 = 0x000000000000000000000000FFFFFFFF
12221 Normally, register values are relative to the selected stack frame
12222 (@pxref{Selection, ,Selecting a Frame}). This means that you get the
12223 value that the register would contain if all stack frames farther in
12224 were exited and their saved registers restored. In order to see the
12225 true contents of hardware registers, you must select the innermost
12226 frame (with @samp{frame 0}).
12228 @cindex caller-saved registers
12229 @cindex call-clobbered registers
12230 @cindex volatile registers
12231 @cindex <not saved> values
12232 Usually ABIs reserve some registers as not needed to be saved by the
12233 callee (a.k.a.: ``caller-saved'', ``call-clobbered'' or ``volatile''
12234 registers). It may therefore not be possible for @value{GDBN} to know
12235 the value a register had before the call (in other words, in the outer
12236 frame), if the register value has since been changed by the callee.
12237 @value{GDBN} tries to deduce where the inner frame saved
12238 (``callee-saved'') registers, from the debug info, unwind info, or the
12239 machine code generated by your compiler. If some register is not
12240 saved, and @value{GDBN} knows the register is ``caller-saved'' (via
12241 its own knowledge of the ABI, or because the debug/unwind info
12242 explicitly says the register's value is undefined), @value{GDBN}
12243 displays @w{@samp{<not saved>}} as the register's value. With targets
12244 that @value{GDBN} has no knowledge of the register saving convention,
12245 if a register was not saved by the callee, then its value and location
12246 in the outer frame are assumed to be the same of the inner frame.
12247 This is usually harmless, because if the register is call-clobbered,
12248 the caller either does not care what is in the register after the
12249 call, or has code to restore the value that it does care about. Note,
12250 however, that if you change such a register in the outer frame, you
12251 may also be affecting the inner frame. Also, the more ``outer'' the
12252 frame is you're looking at, the more likely a call-clobbered
12253 register's value is to be wrong, in the sense that it doesn't actually
12254 represent the value the register had just before the call.
12256 @node Floating Point Hardware
12257 @section Floating Point Hardware
12258 @cindex floating point
12260 Depending on the configuration, @value{GDBN} may be able to give
12261 you more information about the status of the floating point hardware.
12266 Display hardware-dependent information about the floating
12267 point unit. The exact contents and layout vary depending on the
12268 floating point chip. Currently, @samp{info float} is supported on
12269 the ARM and x86 machines.
12273 @section Vector Unit
12274 @cindex vector unit
12276 Depending on the configuration, @value{GDBN} may be able to give you
12277 more information about the status of the vector unit.
12280 @kindex info vector
12282 Display information about the vector unit. The exact contents and
12283 layout vary depending on the hardware.
12286 @node OS Information
12287 @section Operating System Auxiliary Information
12288 @cindex OS information
12290 @value{GDBN} provides interfaces to useful OS facilities that can help
12291 you debug your program.
12293 @cindex auxiliary vector
12294 @cindex vector, auxiliary
12295 Some operating systems supply an @dfn{auxiliary vector} to programs at
12296 startup. This is akin to the arguments and environment that you
12297 specify for a program, but contains a system-dependent variety of
12298 binary values that tell system libraries important details about the
12299 hardware, operating system, and process. Each value's purpose is
12300 identified by an integer tag; the meanings are well-known but system-specific.
12301 Depending on the configuration and operating system facilities,
12302 @value{GDBN} may be able to show you this information. For remote
12303 targets, this functionality may further depend on the remote stub's
12304 support of the @samp{qXfer:auxv:read} packet, see
12305 @ref{qXfer auxiliary vector read}.
12310 Display the auxiliary vector of the inferior, which can be either a
12311 live process or a core dump file. @value{GDBN} prints each tag value
12312 numerically, and also shows names and text descriptions for recognized
12313 tags. Some values in the vector are numbers, some bit masks, and some
12314 pointers to strings or other data. @value{GDBN} displays each value in the
12315 most appropriate form for a recognized tag, and in hexadecimal for
12316 an unrecognized tag.
12319 On some targets, @value{GDBN} can access operating system-specific
12320 information and show it to you. The types of information available
12321 will differ depending on the type of operating system running on the
12322 target. The mechanism used to fetch the data is described in
12323 @ref{Operating System Information}. For remote targets, this
12324 functionality depends on the remote stub's support of the
12325 @samp{qXfer:osdata:read} packet, see @ref{qXfer osdata read}.
12329 @item info os @var{infotype}
12331 Display OS information of the requested type.
12333 On @sc{gnu}/Linux, the following values of @var{infotype} are valid:
12335 @anchor{linux info os infotypes}
12337 @kindex info os cpus
12339 Display the list of all CPUs/cores. For each CPU/core, @value{GDBN} prints
12340 the available fields from /proc/cpuinfo. For each supported architecture
12341 different fields are available. Two common entries are processor which gives
12342 CPU number and bogomips; a system constant that is calculated during
12343 kernel initialization.
12345 @kindex info os files
12347 Display the list of open file descriptors on the target. For each
12348 file descriptor, @value{GDBN} prints the identifier of the process
12349 owning the descriptor, the command of the owning process, the value
12350 of the descriptor, and the target of the descriptor.
12352 @kindex info os modules
12354 Display the list of all loaded kernel modules on the target. For each
12355 module, @value{GDBN} prints the module name, the size of the module in
12356 bytes, the number of times the module is used, the dependencies of the
12357 module, the status of the module, and the address of the loaded module
12360 @kindex info os msg
12362 Display the list of all System V message queues on the target. For each
12363 message queue, @value{GDBN} prints the message queue key, the message
12364 queue identifier, the access permissions, the current number of bytes
12365 on the queue, the current number of messages on the queue, the processes
12366 that last sent and received a message on the queue, the user and group
12367 of the owner and creator of the message queue, the times at which a
12368 message was last sent and received on the queue, and the time at which
12369 the message queue was last changed.
12371 @kindex info os processes
12373 Display the list of processes on the target. For each process,
12374 @value{GDBN} prints the process identifier, the name of the user, the
12375 command corresponding to the process, and the list of processor cores
12376 that the process is currently running on. (To understand what these
12377 properties mean, for this and the following info types, please consult
12378 the general @sc{gnu}/Linux documentation.)
12380 @kindex info os procgroups
12382 Display the list of process groups on the target. For each process,
12383 @value{GDBN} prints the identifier of the process group that it belongs
12384 to, the command corresponding to the process group leader, the process
12385 identifier, and the command line of the process. The list is sorted
12386 first by the process group identifier, then by the process identifier,
12387 so that processes belonging to the same process group are grouped together
12388 and the process group leader is listed first.
12390 @kindex info os semaphores
12392 Display the list of all System V semaphore sets on the target. For each
12393 semaphore set, @value{GDBN} prints the semaphore set key, the semaphore
12394 set identifier, the access permissions, the number of semaphores in the
12395 set, the user and group of the owner and creator of the semaphore set,
12396 and the times at which the semaphore set was operated upon and changed.
12398 @kindex info os shm
12400 Display the list of all System V shared-memory regions on the target.
12401 For each shared-memory region, @value{GDBN} prints the region key,
12402 the shared-memory identifier, the access permissions, the size of the
12403 region, the process that created the region, the process that last
12404 attached to or detached from the region, the current number of live
12405 attaches to the region, and the times at which the region was last
12406 attached to, detach from, and changed.
12408 @kindex info os sockets
12410 Display the list of Internet-domain sockets on the target. For each
12411 socket, @value{GDBN} prints the address and port of the local and
12412 remote endpoints, the current state of the connection, the creator of
12413 the socket, the IP address family of the socket, and the type of the
12416 @kindex info os threads
12418 Display the list of threads running on the target. For each thread,
12419 @value{GDBN} prints the identifier of the process that the thread
12420 belongs to, the command of the process, the thread identifier, and the
12421 processor core that it is currently running on. The main thread of a
12422 process is not listed.
12426 If @var{infotype} is omitted, then list the possible values for
12427 @var{infotype} and the kind of OS information available for each
12428 @var{infotype}. If the target does not return a list of possible
12429 types, this command will report an error.
12432 @node Memory Region Attributes
12433 @section Memory Region Attributes
12434 @cindex memory region attributes
12436 @dfn{Memory region attributes} allow you to describe special handling
12437 required by regions of your target's memory. @value{GDBN} uses
12438 attributes to determine whether to allow certain types of memory
12439 accesses; whether to use specific width accesses; and whether to cache
12440 target memory. By default the description of memory regions is
12441 fetched from the target (if the current target supports this), but the
12442 user can override the fetched regions.
12444 Defined memory regions can be individually enabled and disabled. When a
12445 memory region is disabled, @value{GDBN} uses the default attributes when
12446 accessing memory in that region. Similarly, if no memory regions have
12447 been defined, @value{GDBN} uses the default attributes when accessing
12450 When a memory region is defined, it is given a number to identify it;
12451 to enable, disable, or remove a memory region, you specify that number.
12455 @item mem @var{lower} @var{upper} @var{attributes}@dots{}
12456 Define a memory region bounded by @var{lower} and @var{upper} with
12457 attributes @var{attributes}@dots{}, and add it to the list of regions
12458 monitored by @value{GDBN}. Note that @var{upper} == 0 is a special
12459 case: it is treated as the target's maximum memory address.
12460 (0xffff on 16 bit targets, 0xffffffff on 32 bit targets, etc.)
12463 Discard any user changes to the memory regions and use target-supplied
12464 regions, if available, or no regions if the target does not support.
12467 @item delete mem @var{nums}@dots{}
12468 Remove memory regions @var{nums}@dots{} from the list of regions
12469 monitored by @value{GDBN}.
12471 @kindex disable mem
12472 @item disable mem @var{nums}@dots{}
12473 Disable monitoring of memory regions @var{nums}@dots{}.
12474 A disabled memory region is not forgotten.
12475 It may be enabled again later.
12478 @item enable mem @var{nums}@dots{}
12479 Enable monitoring of memory regions @var{nums}@dots{}.
12483 Print a table of all defined memory regions, with the following columns
12487 @item Memory Region Number
12488 @item Enabled or Disabled.
12489 Enabled memory regions are marked with @samp{y}.
12490 Disabled memory regions are marked with @samp{n}.
12493 The address defining the inclusive lower bound of the memory region.
12496 The address defining the exclusive upper bound of the memory region.
12499 The list of attributes set for this memory region.
12504 @subsection Attributes
12506 @subsubsection Memory Access Mode
12507 The access mode attributes set whether @value{GDBN} may make read or
12508 write accesses to a memory region.
12510 While these attributes prevent @value{GDBN} from performing invalid
12511 memory accesses, they do nothing to prevent the target system, I/O DMA,
12512 etc.@: from accessing memory.
12516 Memory is read only.
12518 Memory is write only.
12520 Memory is read/write. This is the default.
12523 @subsubsection Memory Access Size
12524 The access size attribute tells @value{GDBN} to use specific sized
12525 accesses in the memory region. Often memory mapped device registers
12526 require specific sized accesses. If no access size attribute is
12527 specified, @value{GDBN} may use accesses of any size.
12531 Use 8 bit memory accesses.
12533 Use 16 bit memory accesses.
12535 Use 32 bit memory accesses.
12537 Use 64 bit memory accesses.
12540 @c @subsubsection Hardware/Software Breakpoints
12541 @c The hardware/software breakpoint attributes set whether @value{GDBN}
12542 @c will use hardware or software breakpoints for the internal breakpoints
12543 @c used by the step, next, finish, until, etc. commands.
12547 @c Always use hardware breakpoints
12548 @c @item swbreak (default)
12551 @subsubsection Data Cache
12552 The data cache attributes set whether @value{GDBN} will cache target
12553 memory. While this generally improves performance by reducing debug
12554 protocol overhead, it can lead to incorrect results because @value{GDBN}
12555 does not know about volatile variables or memory mapped device
12560 Enable @value{GDBN} to cache target memory.
12562 Disable @value{GDBN} from caching target memory. This is the default.
12565 @subsection Memory Access Checking
12566 @value{GDBN} can be instructed to refuse accesses to memory that is
12567 not explicitly described. This can be useful if accessing such
12568 regions has undesired effects for a specific target, or to provide
12569 better error checking. The following commands control this behaviour.
12572 @kindex set mem inaccessible-by-default
12573 @item set mem inaccessible-by-default [on|off]
12574 If @code{on} is specified, make @value{GDBN} treat memory not
12575 explicitly described by the memory ranges as non-existent and refuse accesses
12576 to such memory. The checks are only performed if there's at least one
12577 memory range defined. If @code{off} is specified, make @value{GDBN}
12578 treat the memory not explicitly described by the memory ranges as RAM.
12579 The default value is @code{on}.
12580 @kindex show mem inaccessible-by-default
12581 @item show mem inaccessible-by-default
12582 Show the current handling of accesses to unknown memory.
12586 @c @subsubsection Memory Write Verification
12587 @c The memory write verification attributes set whether @value{GDBN}
12588 @c will re-reads data after each write to verify the write was successful.
12592 @c @item noverify (default)
12595 @node Dump/Restore Files
12596 @section Copy Between Memory and a File
12597 @cindex dump/restore files
12598 @cindex append data to a file
12599 @cindex dump data to a file
12600 @cindex restore data from a file
12602 You can use the commands @code{dump}, @code{append}, and
12603 @code{restore} to copy data between target memory and a file. The
12604 @code{dump} and @code{append} commands write data to a file, and the
12605 @code{restore} command reads data from a file back into the inferior's
12606 memory. Files may be in binary, Motorola S-record, Intel hex,
12607 Tektronix Hex, or Verilog Hex format; however, @value{GDBN} can only
12608 append to binary files, and cannot read from Verilog Hex files.
12613 @item dump @r{[}@var{format}@r{]} memory @var{filename} @var{start_addr} @var{end_addr}
12614 @itemx dump @r{[}@var{format}@r{]} value @var{filename} @var{expr}
12615 Dump the contents of memory from @var{start_addr} to @var{end_addr},
12616 or the value of @var{expr}, to @var{filename} in the given format.
12618 The @var{format} parameter may be any one of:
12625 Motorola S-record format.
12627 Tektronix Hex format.
12629 Verilog Hex format.
12632 @value{GDBN} uses the same definitions of these formats as the
12633 @sc{gnu} binary utilities, like @samp{objdump} and @samp{objcopy}. If
12634 @var{format} is omitted, @value{GDBN} dumps the data in raw binary
12638 @item append @r{[}binary@r{]} memory @var{filename} @var{start_addr} @var{end_addr}
12639 @itemx append @r{[}binary@r{]} value @var{filename} @var{expr}
12640 Append the contents of memory from @var{start_addr} to @var{end_addr},
12641 or the value of @var{expr}, to the file @var{filename}, in raw binary form.
12642 (@value{GDBN} can only append data to files in raw binary form.)
12645 @item restore @var{filename} @r{[}binary@r{]} @var{bias} @var{start} @var{end}
12646 Restore the contents of file @var{filename} into memory. The
12647 @code{restore} command can automatically recognize any known @sc{bfd}
12648 file format, except for raw binary. To restore a raw binary file you
12649 must specify the optional keyword @code{binary} after the filename.
12651 If @var{bias} is non-zero, its value will be added to the addresses
12652 contained in the file. Binary files always start at address zero, so
12653 they will be restored at address @var{bias}. Other bfd files have
12654 a built-in location; they will be restored at offset @var{bias}
12655 from that location.
12657 If @var{start} and/or @var{end} are non-zero, then only data between
12658 file offset @var{start} and file offset @var{end} will be restored.
12659 These offsets are relative to the addresses in the file, before
12660 the @var{bias} argument is applied.
12664 @node Core File Generation
12665 @section How to Produce a Core File from Your Program
12666 @cindex dump core from inferior
12668 A @dfn{core file} or @dfn{core dump} is a file that records the memory
12669 image of a running process and its process status (register values
12670 etc.). Its primary use is post-mortem debugging of a program that
12671 crashed while it ran outside a debugger. A program that crashes
12672 automatically produces a core file, unless this feature is disabled by
12673 the user. @xref{Files}, for information on invoking @value{GDBN} in
12674 the post-mortem debugging mode.
12676 Occasionally, you may wish to produce a core file of the program you
12677 are debugging in order to preserve a snapshot of its state.
12678 @value{GDBN} has a special command for that.
12682 @kindex generate-core-file
12683 @item generate-core-file [@var{file}]
12684 @itemx gcore [@var{file}]
12685 Produce a core dump of the inferior process. The optional argument
12686 @var{file} specifies the file name where to put the core dump. If not
12687 specified, the file name defaults to @file{core.@var{pid}}, where
12688 @var{pid} is the inferior process ID.
12690 Note that this command is implemented only for some systems (as of
12691 this writing, @sc{gnu}/Linux, FreeBSD, Solaris, and S390).
12693 On @sc{gnu}/Linux, this command can take into account the value of the
12694 file @file{/proc/@var{pid}/coredump_filter} when generating the core
12695 dump (@pxref{set use-coredump-filter}), and by default honors the
12696 @code{VM_DONTDUMP} flag for mappings where it is present in the file
12697 @file{/proc/@var{pid}/smaps} (@pxref{set dump-excluded-mappings}).
12699 @kindex set use-coredump-filter
12700 @anchor{set use-coredump-filter}
12701 @item set use-coredump-filter on
12702 @itemx set use-coredump-filter off
12703 Enable or disable the use of the file
12704 @file{/proc/@var{pid}/coredump_filter} when generating core dump
12705 files. This file is used by the Linux kernel to decide what types of
12706 memory mappings will be dumped or ignored when generating a core dump
12707 file. @var{pid} is the process ID of a currently running process.
12709 To make use of this feature, you have to write in the
12710 @file{/proc/@var{pid}/coredump_filter} file a value, in hexadecimal,
12711 which is a bit mask representing the memory mapping types. If a bit
12712 is set in the bit mask, then the memory mappings of the corresponding
12713 types will be dumped; otherwise, they will be ignored. This
12714 configuration is inherited by child processes. For more information
12715 about the bits that can be set in the
12716 @file{/proc/@var{pid}/coredump_filter} file, please refer to the
12717 manpage of @code{core(5)}.
12719 By default, this option is @code{on}. If this option is turned
12720 @code{off}, @value{GDBN} does not read the @file{coredump_filter} file
12721 and instead uses the same default value as the Linux kernel in order
12722 to decide which pages will be dumped in the core dump file. This
12723 value is currently @code{0x33}, which means that bits @code{0}
12724 (anonymous private mappings), @code{1} (anonymous shared mappings),
12725 @code{4} (ELF headers) and @code{5} (private huge pages) are active.
12726 This will cause these memory mappings to be dumped automatically.
12728 @kindex set dump-excluded-mappings
12729 @anchor{set dump-excluded-mappings}
12730 @item set dump-excluded-mappings on
12731 @itemx set dump-excluded-mappings off
12732 If @code{on} is specified, @value{GDBN} will dump memory mappings
12733 marked with the @code{VM_DONTDUMP} flag. This flag is represented in
12734 the file @file{/proc/@var{pid}/smaps} with the acronym @code{dd}.
12736 The default value is @code{off}.
12739 @node Character Sets
12740 @section Character Sets
12741 @cindex character sets
12743 @cindex translating between character sets
12744 @cindex host character set
12745 @cindex target character set
12747 If the program you are debugging uses a different character set to
12748 represent characters and strings than the one @value{GDBN} uses itself,
12749 @value{GDBN} can automatically translate between the character sets for
12750 you. The character set @value{GDBN} uses we call the @dfn{host
12751 character set}; the one the inferior program uses we call the
12752 @dfn{target character set}.
12754 For example, if you are running @value{GDBN} on a @sc{gnu}/Linux system, which
12755 uses the ISO Latin 1 character set, but you are using @value{GDBN}'s
12756 remote protocol (@pxref{Remote Debugging}) to debug a program
12757 running on an IBM mainframe, which uses the @sc{ebcdic} character set,
12758 then the host character set is Latin-1, and the target character set is
12759 @sc{ebcdic}. If you give @value{GDBN} the command @code{set
12760 target-charset EBCDIC-US}, then @value{GDBN} translates between
12761 @sc{ebcdic} and Latin 1 as you print character or string values, or use
12762 character and string literals in expressions.
12764 @value{GDBN} has no way to automatically recognize which character set
12765 the inferior program uses; you must tell it, using the @code{set
12766 target-charset} command, described below.
12768 Here are the commands for controlling @value{GDBN}'s character set
12772 @item set target-charset @var{charset}
12773 @kindex set target-charset
12774 Set the current target character set to @var{charset}. To display the
12775 list of supported target character sets, type
12776 @kbd{@w{set target-charset @key{TAB}@key{TAB}}}.
12778 @item set host-charset @var{charset}
12779 @kindex set host-charset
12780 Set the current host character set to @var{charset}.
12782 By default, @value{GDBN} uses a host character set appropriate to the
12783 system it is running on; you can override that default using the
12784 @code{set host-charset} command. On some systems, @value{GDBN} cannot
12785 automatically determine the appropriate host character set. In this
12786 case, @value{GDBN} uses @samp{UTF-8}.
12788 @value{GDBN} can only use certain character sets as its host character
12789 set. If you type @kbd{@w{set host-charset @key{TAB}@key{TAB}}},
12790 @value{GDBN} will list the host character sets it supports.
12792 @item set charset @var{charset}
12793 @kindex set charset
12794 Set the current host and target character sets to @var{charset}. As
12795 above, if you type @kbd{@w{set charset @key{TAB}@key{TAB}}},
12796 @value{GDBN} will list the names of the character sets that can be used
12797 for both host and target.
12800 @kindex show charset
12801 Show the names of the current host and target character sets.
12803 @item show host-charset
12804 @kindex show host-charset
12805 Show the name of the current host character set.
12807 @item show target-charset
12808 @kindex show target-charset
12809 Show the name of the current target character set.
12811 @item set target-wide-charset @var{charset}
12812 @kindex set target-wide-charset
12813 Set the current target's wide character set to @var{charset}. This is
12814 the character set used by the target's @code{wchar_t} type. To
12815 display the list of supported wide character sets, type
12816 @kbd{@w{set target-wide-charset @key{TAB}@key{TAB}}}.
12818 @item show target-wide-charset
12819 @kindex show target-wide-charset
12820 Show the name of the current target's wide character set.
12823 Here is an example of @value{GDBN}'s character set support in action.
12824 Assume that the following source code has been placed in the file
12825 @file{charset-test.c}:
12831 = @{72, 101, 108, 108, 111, 44, 32, 119,
12832 111, 114, 108, 100, 33, 10, 0@};
12833 char ibm1047_hello[]
12834 = @{200, 133, 147, 147, 150, 107, 64, 166,
12835 150, 153, 147, 132, 90, 37, 0@};
12839 printf ("Hello, world!\n");
12843 In this program, @code{ascii_hello} and @code{ibm1047_hello} are arrays
12844 containing the string @samp{Hello, world!} followed by a newline,
12845 encoded in the @sc{ascii} and @sc{ibm1047} character sets.
12847 We compile the program, and invoke the debugger on it:
12850 $ gcc -g charset-test.c -o charset-test
12851 $ gdb -nw charset-test
12852 GNU gdb 2001-12-19-cvs
12853 Copyright 2001 Free Software Foundation, Inc.
12858 We can use the @code{show charset} command to see what character sets
12859 @value{GDBN} is currently using to interpret and display characters and
12863 (@value{GDBP}) show charset
12864 The current host and target character set is `ISO-8859-1'.
12868 For the sake of printing this manual, let's use @sc{ascii} as our
12869 initial character set:
12871 (@value{GDBP}) set charset ASCII
12872 (@value{GDBP}) show charset
12873 The current host and target character set is `ASCII'.
12877 Let's assume that @sc{ascii} is indeed the correct character set for our
12878 host system --- in other words, let's assume that if @value{GDBN} prints
12879 characters using the @sc{ascii} character set, our terminal will display
12880 them properly. Since our current target character set is also
12881 @sc{ascii}, the contents of @code{ascii_hello} print legibly:
12884 (@value{GDBP}) print ascii_hello
12885 $1 = 0x401698 "Hello, world!\n"
12886 (@value{GDBP}) print ascii_hello[0]
12891 @value{GDBN} uses the target character set for character and string
12892 literals you use in expressions:
12895 (@value{GDBP}) print '+'
12900 The @sc{ascii} character set uses the number 43 to encode the @samp{+}
12903 @value{GDBN} relies on the user to tell it which character set the
12904 target program uses. If we print @code{ibm1047_hello} while our target
12905 character set is still @sc{ascii}, we get jibberish:
12908 (@value{GDBP}) print ibm1047_hello
12909 $4 = 0x4016a8 "\310\205\223\223\226k@@\246\226\231\223\204Z%"
12910 (@value{GDBP}) print ibm1047_hello[0]
12915 If we invoke the @code{set target-charset} followed by @key{TAB}@key{TAB},
12916 @value{GDBN} tells us the character sets it supports:
12919 (@value{GDBP}) set target-charset
12920 ASCII EBCDIC-US IBM1047 ISO-8859-1
12921 (@value{GDBP}) set target-charset
12924 We can select @sc{ibm1047} as our target character set, and examine the
12925 program's strings again. Now the @sc{ascii} string is wrong, but
12926 @value{GDBN} translates the contents of @code{ibm1047_hello} from the
12927 target character set, @sc{ibm1047}, to the host character set,
12928 @sc{ascii}, and they display correctly:
12931 (@value{GDBP}) set target-charset IBM1047
12932 (@value{GDBP}) show charset
12933 The current host character set is `ASCII'.
12934 The current target character set is `IBM1047'.
12935 (@value{GDBP}) print ascii_hello
12936 $6 = 0x401698 "\110\145%%?\054\040\167?\162%\144\041\012"
12937 (@value{GDBP}) print ascii_hello[0]
12939 (@value{GDBP}) print ibm1047_hello
12940 $8 = 0x4016a8 "Hello, world!\n"
12941 (@value{GDBP}) print ibm1047_hello[0]
12946 As above, @value{GDBN} uses the target character set for character and
12947 string literals you use in expressions:
12950 (@value{GDBP}) print '+'
12955 The @sc{ibm1047} character set uses the number 78 to encode the @samp{+}
12958 @node Caching Target Data
12959 @section Caching Data of Targets
12960 @cindex caching data of targets
12962 @value{GDBN} caches data exchanged between the debugger and a target.
12963 Each cache is associated with the address space of the inferior.
12964 @xref{Inferiors and Programs}, about inferior and address space.
12965 Such caching generally improves performance in remote debugging
12966 (@pxref{Remote Debugging}), because it reduces the overhead of the
12967 remote protocol by bundling memory reads and writes into large chunks.
12968 Unfortunately, simply caching everything would lead to incorrect results,
12969 since @value{GDBN} does not necessarily know anything about volatile
12970 values, memory-mapped I/O addresses, etc. Furthermore, in non-stop mode
12971 (@pxref{Non-Stop Mode}) memory can be changed @emph{while} a gdb command
12973 Therefore, by default, @value{GDBN} only caches data
12974 known to be on the stack@footnote{In non-stop mode, it is moderately
12975 rare for a running thread to modify the stack of a stopped thread
12976 in a way that would interfere with a backtrace, and caching of
12977 stack reads provides a significant speed up of remote backtraces.} or
12978 in the code segment.
12979 Other regions of memory can be explicitly marked as
12980 cacheable; @pxref{Memory Region Attributes}.
12983 @kindex set remotecache
12984 @item set remotecache on
12985 @itemx set remotecache off
12986 This option no longer does anything; it exists for compatibility
12989 @kindex show remotecache
12990 @item show remotecache
12991 Show the current state of the obsolete remotecache flag.
12993 @kindex set stack-cache
12994 @item set stack-cache on
12995 @itemx set stack-cache off
12996 Enable or disable caching of stack accesses. When @code{on}, use
12997 caching. By default, this option is @code{on}.
12999 @kindex show stack-cache
13000 @item show stack-cache
13001 Show the current state of data caching for memory accesses.
13003 @kindex set code-cache
13004 @item set code-cache on
13005 @itemx set code-cache off
13006 Enable or disable caching of code segment accesses. When @code{on},
13007 use caching. By default, this option is @code{on}. This improves
13008 performance of disassembly in remote debugging.
13010 @kindex show code-cache
13011 @item show code-cache
13012 Show the current state of target memory cache for code segment
13015 @kindex info dcache
13016 @item info dcache @r{[}line@r{]}
13017 Print the information about the performance of data cache of the
13018 current inferior's address space. The information displayed
13019 includes the dcache width and depth, and for each cache line, its
13020 number, address, and how many times it was referenced. This
13021 command is useful for debugging the data cache operation.
13023 If a line number is specified, the contents of that line will be
13026 @item set dcache size @var{size}
13027 @cindex dcache size
13028 @kindex set dcache size
13029 Set maximum number of entries in dcache (dcache depth above).
13031 @item set dcache line-size @var{line-size}
13032 @cindex dcache line-size
13033 @kindex set dcache line-size
13034 Set number of bytes each dcache entry caches (dcache width above).
13035 Must be a power of 2.
13037 @item show dcache size
13038 @kindex show dcache size
13039 Show maximum number of dcache entries. @xref{Caching Target Data, info dcache}.
13041 @item show dcache line-size
13042 @kindex show dcache line-size
13043 Show default size of dcache lines.
13047 @node Searching Memory
13048 @section Search Memory
13049 @cindex searching memory
13051 Memory can be searched for a particular sequence of bytes with the
13052 @code{find} command.
13056 @item find @r{[}/@var{sn}@r{]} @var{start_addr}, +@var{len}, @var{val1} @r{[}, @var{val2}, @dots{}@r{]}
13057 @itemx find @r{[}/@var{sn}@r{]} @var{start_addr}, @var{end_addr}, @var{val1} @r{[}, @var{val2}, @dots{}@r{]}
13058 Search memory for the sequence of bytes specified by @var{val1}, @var{val2},
13059 etc. The search begins at address @var{start_addr} and continues for either
13060 @var{len} bytes or through to @var{end_addr} inclusive.
13063 @var{s} and @var{n} are optional parameters.
13064 They may be specified in either order, apart or together.
13067 @item @var{s}, search query size
13068 The size of each search query value.
13074 halfwords (two bytes)
13078 giant words (eight bytes)
13081 All values are interpreted in the current language.
13082 This means, for example, that if the current source language is C/C@t{++}
13083 then searching for the string ``hello'' includes the trailing '\0'.
13084 The null terminator can be removed from searching by using casts,
13085 e.g.: @samp{@{char[5]@}"hello"}.
13087 If the value size is not specified, it is taken from the
13088 value's type in the current language.
13089 This is useful when one wants to specify the search
13090 pattern as a mixture of types.
13091 Note that this means, for example, that in the case of C-like languages
13092 a search for an untyped 0x42 will search for @samp{(int) 0x42}
13093 which is typically four bytes.
13095 @item @var{n}, maximum number of finds
13096 The maximum number of matches to print. The default is to print all finds.
13099 You can use strings as search values. Quote them with double-quotes
13101 The string value is copied into the search pattern byte by byte,
13102 regardless of the endianness of the target and the size specification.
13104 The address of each match found is printed as well as a count of the
13105 number of matches found.
13107 The address of the last value found is stored in convenience variable
13109 A count of the number of matches is stored in @samp{$numfound}.
13111 For example, if stopped at the @code{printf} in this function:
13117 static char hello[] = "hello-hello";
13118 static struct @{ char c; short s; int i; @}
13119 __attribute__ ((packed)) mixed
13120 = @{ 'c', 0x1234, 0x87654321 @};
13121 printf ("%s\n", hello);
13126 you get during debugging:
13129 (gdb) find &hello[0], +sizeof(hello), "hello"
13130 0x804956d <hello.1620+6>
13132 (gdb) find &hello[0], +sizeof(hello), 'h', 'e', 'l', 'l', 'o'
13133 0x8049567 <hello.1620>
13134 0x804956d <hello.1620+6>
13136 (gdb) find &hello[0], +sizeof(hello), @{char[5]@}"hello"
13137 0x8049567 <hello.1620>
13138 0x804956d <hello.1620+6>
13140 (gdb) find /b1 &hello[0], +sizeof(hello), 'h', 0x65, 'l'
13141 0x8049567 <hello.1620>
13143 (gdb) find &mixed, +sizeof(mixed), (char) 'c', (short) 0x1234, (int) 0x87654321
13144 0x8049560 <mixed.1625>
13146 (gdb) print $numfound
13149 $2 = (void *) 0x8049560
13153 @section Value Sizes
13155 Whenever @value{GDBN} prints a value memory will be allocated within
13156 @value{GDBN} to hold the contents of the value. It is possible in
13157 some languages with dynamic typing systems, that an invalid program
13158 may indicate a value that is incorrectly large, this in turn may cause
13159 @value{GDBN} to try and allocate an overly large ammount of memory.
13162 @kindex set max-value-size
13163 @item set max-value-size @var{bytes}
13164 @itemx set max-value-size unlimited
13165 Set the maximum size of memory that @value{GDBN} will allocate for the
13166 contents of a value to @var{bytes}, trying to display a value that
13167 requires more memory than that will result in an error.
13169 Setting this variable does not effect values that have already been
13170 allocated within @value{GDBN}, only future allocations.
13172 There's a minimum size that @code{max-value-size} can be set to in
13173 order that @value{GDBN} can still operate correctly, this minimum is
13174 currently 16 bytes.
13176 The limit applies to the results of some subexpressions as well as to
13177 complete expressions. For example, an expression denoting a simple
13178 integer component, such as @code{x.y.z}, may fail if the size of
13179 @var{x.y} is dynamic and exceeds @var{bytes}. On the other hand,
13180 @value{GDBN} is sometimes clever; the expression @code{A[i]}, where
13181 @var{A} is an array variable with non-constant size, will generally
13182 succeed regardless of the bounds on @var{A}, as long as the component
13183 size is less than @var{bytes}.
13185 The default value of @code{max-value-size} is currently 64k.
13187 @kindex show max-value-size
13188 @item show max-value-size
13189 Show the maximum size of memory, in bytes, that @value{GDBN} will
13190 allocate for the contents of a value.
13193 @node Optimized Code
13194 @chapter Debugging Optimized Code
13195 @cindex optimized code, debugging
13196 @cindex debugging optimized code
13198 Almost all compilers support optimization. With optimization
13199 disabled, the compiler generates assembly code that corresponds
13200 directly to your source code, in a simplistic way. As the compiler
13201 applies more powerful optimizations, the generated assembly code
13202 diverges from your original source code. With help from debugging
13203 information generated by the compiler, @value{GDBN} can map from
13204 the running program back to constructs from your original source.
13206 @value{GDBN} is more accurate with optimization disabled. If you
13207 can recompile without optimization, it is easier to follow the
13208 progress of your program during debugging. But, there are many cases
13209 where you may need to debug an optimized version.
13211 When you debug a program compiled with @samp{-g -O}, remember that the
13212 optimizer has rearranged your code; the debugger shows you what is
13213 really there. Do not be too surprised when the execution path does not
13214 exactly match your source file! An extreme example: if you define a
13215 variable, but never use it, @value{GDBN} never sees that
13216 variable---because the compiler optimizes it out of existence.
13218 Some things do not work as well with @samp{-g -O} as with just
13219 @samp{-g}, particularly on machines with instruction scheduling. If in
13220 doubt, recompile with @samp{-g} alone, and if this fixes the problem,
13221 please report it to us as a bug (including a test case!).
13222 @xref{Variables}, for more information about debugging optimized code.
13225 * Inline Functions:: How @value{GDBN} presents inlining
13226 * Tail Call Frames:: @value{GDBN} analysis of jumps to functions
13229 @node Inline Functions
13230 @section Inline Functions
13231 @cindex inline functions, debugging
13233 @dfn{Inlining} is an optimization that inserts a copy of the function
13234 body directly at each call site, instead of jumping to a shared
13235 routine. @value{GDBN} displays inlined functions just like
13236 non-inlined functions. They appear in backtraces. You can view their
13237 arguments and local variables, step into them with @code{step}, skip
13238 them with @code{next}, and escape from them with @code{finish}.
13239 You can check whether a function was inlined by using the
13240 @code{info frame} command.
13242 For @value{GDBN} to support inlined functions, the compiler must
13243 record information about inlining in the debug information ---
13244 @value{NGCC} using the @sc{dwarf 2} format does this, and several
13245 other compilers do also. @value{GDBN} only supports inlined functions
13246 when using @sc{dwarf 2}. Versions of @value{NGCC} before 4.1
13247 do not emit two required attributes (@samp{DW_AT_call_file} and
13248 @samp{DW_AT_call_line}); @value{GDBN} does not display inlined
13249 function calls with earlier versions of @value{NGCC}. It instead
13250 displays the arguments and local variables of inlined functions as
13251 local variables in the caller.
13253 The body of an inlined function is directly included at its call site;
13254 unlike a non-inlined function, there are no instructions devoted to
13255 the call. @value{GDBN} still pretends that the call site and the
13256 start of the inlined function are different instructions. Stepping to
13257 the call site shows the call site, and then stepping again shows
13258 the first line of the inlined function, even though no additional
13259 instructions are executed.
13261 This makes source-level debugging much clearer; you can see both the
13262 context of the call and then the effect of the call. Only stepping by
13263 a single instruction using @code{stepi} or @code{nexti} does not do
13264 this; single instruction steps always show the inlined body.
13266 There are some ways that @value{GDBN} does not pretend that inlined
13267 function calls are the same as normal calls:
13271 Setting breakpoints at the call site of an inlined function may not
13272 work, because the call site does not contain any code. @value{GDBN}
13273 may incorrectly move the breakpoint to the next line of the enclosing
13274 function, after the call. This limitation will be removed in a future
13275 version of @value{GDBN}; until then, set a breakpoint on an earlier line
13276 or inside the inlined function instead.
13279 @value{GDBN} cannot locate the return value of inlined calls after
13280 using the @code{finish} command. This is a limitation of compiler-generated
13281 debugging information; after @code{finish}, you can step to the next line
13282 and print a variable where your program stored the return value.
13286 @node Tail Call Frames
13287 @section Tail Call Frames
13288 @cindex tail call frames, debugging
13290 Function @code{B} can call function @code{C} in its very last statement. In
13291 unoptimized compilation the call of @code{C} is immediately followed by return
13292 instruction at the end of @code{B} code. Optimizing compiler may replace the
13293 call and return in function @code{B} into one jump to function @code{C}
13294 instead. Such use of a jump instruction is called @dfn{tail call}.
13296 During execution of function @code{C}, there will be no indication in the
13297 function call stack frames that it was tail-called from @code{B}. If function
13298 @code{A} regularly calls function @code{B} which tail-calls function @code{C},
13299 then @value{GDBN} will see @code{A} as the caller of @code{C}. However, in
13300 some cases @value{GDBN} can determine that @code{C} was tail-called from
13301 @code{B}, and it will then create fictitious call frame for that, with the
13302 return address set up as if @code{B} called @code{C} normally.
13304 This functionality is currently supported only by DWARF 2 debugging format and
13305 the compiler has to produce @samp{DW_TAG_call_site} tags. With
13306 @value{NGCC}, you need to specify @option{-O -g} during compilation, to get
13309 @kbd{info frame} command (@pxref{Frame Info}) will indicate the tail call frame
13310 kind by text @code{tail call frame} such as in this sample @value{GDBN} output:
13314 0x40066b <b(int, double)+11>: jmp 0x400640 <c(int, double)>
13316 Stack level 1, frame at 0x7fffffffda30:
13317 rip = 0x40066d in b (amd64-entry-value.cc:59); saved rip 0x4004c5
13318 tail call frame, caller of frame at 0x7fffffffda30
13319 source language c++.
13320 Arglist at unknown address.
13321 Locals at unknown address, Previous frame's sp is 0x7fffffffda30
13324 The detection of all the possible code path executions can find them ambiguous.
13325 There is no execution history stored (possible @ref{Reverse Execution} is never
13326 used for this purpose) and the last known caller could have reached the known
13327 callee by multiple different jump sequences. In such case @value{GDBN} still
13328 tries to show at least all the unambiguous top tail callers and all the
13329 unambiguous bottom tail calees, if any.
13332 @anchor{set debug entry-values}
13333 @item set debug entry-values
13334 @kindex set debug entry-values
13335 When set to on, enables printing of analysis messages for both frame argument
13336 values at function entry and tail calls. It will show all the possible valid
13337 tail calls code paths it has considered. It will also print the intersection
13338 of them with the final unambiguous (possibly partial or even empty) code path
13341 @item show debug entry-values
13342 @kindex show debug entry-values
13343 Show the current state of analysis messages printing for both frame argument
13344 values at function entry and tail calls.
13347 The analysis messages for tail calls can for example show why the virtual tail
13348 call frame for function @code{c} has not been recognized (due to the indirect
13349 reference by variable @code{x}):
13352 static void __attribute__((noinline, noclone)) c (void);
13353 void (*x) (void) = c;
13354 static void __attribute__((noinline, noclone)) a (void) @{ x++; @}
13355 static void __attribute__((noinline, noclone)) c (void) @{ a (); @}
13356 int main (void) @{ x (); return 0; @}
13358 Breakpoint 1, DW_OP_entry_value resolving cannot find
13359 DW_TAG_call_site 0x40039a in main
13361 3 static void __attribute__((noinline, noclone)) a (void) @{ x++; @}
13364 #1 0x000000000040039a in main () at t.c:5
13367 Another possibility is an ambiguous virtual tail call frames resolution:
13371 static void __attribute__((noinline, noclone)) f (void) @{ i++; @}
13372 static void __attribute__((noinline, noclone)) e (void) @{ f (); @}
13373 static void __attribute__((noinline, noclone)) d (void) @{ f (); @}
13374 static void __attribute__((noinline, noclone)) c (void) @{ d (); @}
13375 static void __attribute__((noinline, noclone)) b (void)
13376 @{ if (i) c (); else e (); @}
13377 static void __attribute__((noinline, noclone)) a (void) @{ b (); @}
13378 int main (void) @{ a (); return 0; @}
13380 tailcall: initial: 0x4004d2(a) 0x4004ce(b) 0x4004b2(c) 0x4004a2(d)
13381 tailcall: compare: 0x4004d2(a) 0x4004cc(b) 0x400492(e)
13382 tailcall: reduced: 0x4004d2(a) |
13385 #1 0x00000000004004d2 in a () at t.c:8
13386 #2 0x0000000000400395 in main () at t.c:9
13389 @set CALLSEQ1A @code{main@value{ARROW}a@value{ARROW}b@value{ARROW}c@value{ARROW}d@value{ARROW}f}
13390 @set CALLSEQ2A @code{main@value{ARROW}a@value{ARROW}b@value{ARROW}e@value{ARROW}f}
13392 @c Convert CALLSEQ#A to CALLSEQ#B depending on HAVE_MAKEINFO_CLICK.
13393 @ifset HAVE_MAKEINFO_CLICK
13394 @set ARROW @click{}
13395 @set CALLSEQ1B @clicksequence{@value{CALLSEQ1A}}
13396 @set CALLSEQ2B @clicksequence{@value{CALLSEQ2A}}
13398 @ifclear HAVE_MAKEINFO_CLICK
13400 @set CALLSEQ1B @value{CALLSEQ1A}
13401 @set CALLSEQ2B @value{CALLSEQ2A}
13404 Frames #0 and #2 are real, #1 is a virtual tail call frame.
13405 The code can have possible execution paths @value{CALLSEQ1B} or
13406 @value{CALLSEQ2B}, @value{GDBN} cannot find which one from the inferior state.
13408 @code{initial:} state shows some random possible calling sequence @value{GDBN}
13409 has found. It then finds another possible calling sequcen - that one is
13410 prefixed by @code{compare:}. The non-ambiguous intersection of these two is
13411 printed as the @code{reduced:} calling sequence. That one could have many
13412 futher @code{compare:} and @code{reduced:} statements as long as there remain
13413 any non-ambiguous sequence entries.
13415 For the frame of function @code{b} in both cases there are different possible
13416 @code{$pc} values (@code{0x4004cc} or @code{0x4004ce}), therefore this frame is
13417 also ambigous. The only non-ambiguous frame is the one for function @code{a},
13418 therefore this one is displayed to the user while the ambiguous frames are
13421 There can be also reasons why printing of frame argument values at function
13426 static void __attribute__((noinline, noclone)) c (int i) @{ v++; @}
13427 static void __attribute__((noinline, noclone)) a (int i);
13428 static void __attribute__((noinline, noclone)) b (int i) @{ a (i); @}
13429 static void __attribute__((noinline, noclone)) a (int i)
13430 @{ if (i) b (i - 1); else c (0); @}
13431 int main (void) @{ a (5); return 0; @}
13434 #0 c (i=i@@entry=0) at t.c:2
13435 #1 0x0000000000400428 in a (DW_OP_entry_value resolving has found
13436 function "a" at 0x400420 can call itself via tail calls
13437 i=<optimized out>) at t.c:6
13438 #2 0x000000000040036e in main () at t.c:7
13441 @value{GDBN} cannot find out from the inferior state if and how many times did
13442 function @code{a} call itself (via function @code{b}) as these calls would be
13443 tail calls. Such tail calls would modify thue @code{i} variable, therefore
13444 @value{GDBN} cannot be sure the value it knows would be right - @value{GDBN}
13445 prints @code{<optimized out>} instead.
13448 @chapter C Preprocessor Macros
13450 Some languages, such as C and C@t{++}, provide a way to define and invoke
13451 ``preprocessor macros'' which expand into strings of tokens.
13452 @value{GDBN} can evaluate expressions containing macro invocations, show
13453 the result of macro expansion, and show a macro's definition, including
13454 where it was defined.
13456 You may need to compile your program specially to provide @value{GDBN}
13457 with information about preprocessor macros. Most compilers do not
13458 include macros in their debugging information, even when you compile
13459 with the @option{-g} flag. @xref{Compilation}.
13461 A program may define a macro at one point, remove that definition later,
13462 and then provide a different definition after that. Thus, at different
13463 points in the program, a macro may have different definitions, or have
13464 no definition at all. If there is a current stack frame, @value{GDBN}
13465 uses the macros in scope at that frame's source code line. Otherwise,
13466 @value{GDBN} uses the macros in scope at the current listing location;
13469 Whenever @value{GDBN} evaluates an expression, it always expands any
13470 macro invocations present in the expression. @value{GDBN} also provides
13471 the following commands for working with macros explicitly.
13475 @kindex macro expand
13476 @cindex macro expansion, showing the results of preprocessor
13477 @cindex preprocessor macro expansion, showing the results of
13478 @cindex expanding preprocessor macros
13479 @item macro expand @var{expression}
13480 @itemx macro exp @var{expression}
13481 Show the results of expanding all preprocessor macro invocations in
13482 @var{expression}. Since @value{GDBN} simply expands macros, but does
13483 not parse the result, @var{expression} need not be a valid expression;
13484 it can be any string of tokens.
13487 @item macro expand-once @var{expression}
13488 @itemx macro exp1 @var{expression}
13489 @cindex expand macro once
13490 @i{(This command is not yet implemented.)} Show the results of
13491 expanding those preprocessor macro invocations that appear explicitly in
13492 @var{expression}. Macro invocations appearing in that expansion are
13493 left unchanged. This command allows you to see the effect of a
13494 particular macro more clearly, without being confused by further
13495 expansions. Since @value{GDBN} simply expands macros, but does not
13496 parse the result, @var{expression} need not be a valid expression; it
13497 can be any string of tokens.
13500 @cindex macro definition, showing
13501 @cindex definition of a macro, showing
13502 @cindex macros, from debug info
13503 @item info macro [-a|-all] [--] @var{macro}
13504 Show the current definition or all definitions of the named @var{macro},
13505 and describe the source location or compiler command-line where that
13506 definition was established. The optional double dash is to signify the end of
13507 argument processing and the beginning of @var{macro} for non C-like macros where
13508 the macro may begin with a hyphen.
13510 @kindex info macros
13511 @item info macros @var{location}
13512 Show all macro definitions that are in effect at the location specified
13513 by @var{location}, and describe the source location or compiler
13514 command-line where those definitions were established.
13516 @kindex macro define
13517 @cindex user-defined macros
13518 @cindex defining macros interactively
13519 @cindex macros, user-defined
13520 @item macro define @var{macro} @var{replacement-list}
13521 @itemx macro define @var{macro}(@var{arglist}) @var{replacement-list}
13522 Introduce a definition for a preprocessor macro named @var{macro},
13523 invocations of which are replaced by the tokens given in
13524 @var{replacement-list}. The first form of this command defines an
13525 ``object-like'' macro, which takes no arguments; the second form
13526 defines a ``function-like'' macro, which takes the arguments given in
13529 A definition introduced by this command is in scope in every
13530 expression evaluated in @value{GDBN}, until it is removed with the
13531 @code{macro undef} command, described below. The definition overrides
13532 all definitions for @var{macro} present in the program being debugged,
13533 as well as any previous user-supplied definition.
13535 @kindex macro undef
13536 @item macro undef @var{macro}
13537 Remove any user-supplied definition for the macro named @var{macro}.
13538 This command only affects definitions provided with the @code{macro
13539 define} command, described above; it cannot remove definitions present
13540 in the program being debugged.
13544 List all the macros defined using the @code{macro define} command.
13547 @cindex macros, example of debugging with
13548 Here is a transcript showing the above commands in action. First, we
13549 show our source files:
13554 #include "sample.h"
13557 #define ADD(x) (M + x)
13562 printf ("Hello, world!\n");
13564 printf ("We're so creative.\n");
13566 printf ("Goodbye, world!\n");
13573 Now, we compile the program using the @sc{gnu} C compiler,
13574 @value{NGCC}. We pass the @option{-gdwarf-2}@footnote{This is the
13575 minimum. Recent versions of @value{NGCC} support @option{-gdwarf-3}
13576 and @option{-gdwarf-4}; we recommend always choosing the most recent
13577 version of DWARF.} @emph{and} @option{-g3} flags to ensure the compiler
13578 includes information about preprocessor macros in the debugging
13582 $ gcc -gdwarf-2 -g3 sample.c -o sample
13586 Now, we start @value{GDBN} on our sample program:
13590 GNU gdb 2002-05-06-cvs
13591 Copyright 2002 Free Software Foundation, Inc.
13592 GDB is free software, @dots{}
13596 We can expand macros and examine their definitions, even when the
13597 program is not running. @value{GDBN} uses the current listing position
13598 to decide which macro definitions are in scope:
13601 (@value{GDBP}) list main
13604 5 #define ADD(x) (M + x)
13609 10 printf ("Hello, world!\n");
13611 12 printf ("We're so creative.\n");
13612 (@value{GDBP}) info macro ADD
13613 Defined at /home/jimb/gdb/macros/play/sample.c:5
13614 #define ADD(x) (M + x)
13615 (@value{GDBP}) info macro Q
13616 Defined at /home/jimb/gdb/macros/play/sample.h:1
13617 included at /home/jimb/gdb/macros/play/sample.c:2
13619 (@value{GDBP}) macro expand ADD(1)
13620 expands to: (42 + 1)
13621 (@value{GDBP}) macro expand-once ADD(1)
13622 expands to: once (M + 1)
13626 In the example above, note that @code{macro expand-once} expands only
13627 the macro invocation explicit in the original text --- the invocation of
13628 @code{ADD} --- but does not expand the invocation of the macro @code{M},
13629 which was introduced by @code{ADD}.
13631 Once the program is running, @value{GDBN} uses the macro definitions in
13632 force at the source line of the current stack frame:
13635 (@value{GDBP}) break main
13636 Breakpoint 1 at 0x8048370: file sample.c, line 10.
13638 Starting program: /home/jimb/gdb/macros/play/sample
13640 Breakpoint 1, main () at sample.c:10
13641 10 printf ("Hello, world!\n");
13645 At line 10, the definition of the macro @code{N} at line 9 is in force:
13648 (@value{GDBP}) info macro N
13649 Defined at /home/jimb/gdb/macros/play/sample.c:9
13651 (@value{GDBP}) macro expand N Q M
13652 expands to: 28 < 42
13653 (@value{GDBP}) print N Q M
13658 As we step over directives that remove @code{N}'s definition, and then
13659 give it a new definition, @value{GDBN} finds the definition (or lack
13660 thereof) in force at each point:
13663 (@value{GDBP}) next
13665 12 printf ("We're so creative.\n");
13666 (@value{GDBP}) info macro N
13667 The symbol `N' has no definition as a C/C++ preprocessor macro
13668 at /home/jimb/gdb/macros/play/sample.c:12
13669 (@value{GDBP}) next
13671 14 printf ("Goodbye, world!\n");
13672 (@value{GDBP}) info macro N
13673 Defined at /home/jimb/gdb/macros/play/sample.c:13
13675 (@value{GDBP}) macro expand N Q M
13676 expands to: 1729 < 42
13677 (@value{GDBP}) print N Q M
13682 In addition to source files, macros can be defined on the compilation command
13683 line using the @option{-D@var{name}=@var{value}} syntax. For macros defined in
13684 such a way, @value{GDBN} displays the location of their definition as line zero
13685 of the source file submitted to the compiler.
13688 (@value{GDBP}) info macro __STDC__
13689 Defined at /home/jimb/gdb/macros/play/sample.c:0
13696 @chapter Tracepoints
13697 @c This chapter is based on the documentation written by Michael
13698 @c Snyder, David Taylor, Jim Blandy, and Elena Zannoni.
13700 @cindex tracepoints
13701 In some applications, it is not feasible for the debugger to interrupt
13702 the program's execution long enough for the developer to learn
13703 anything helpful about its behavior. If the program's correctness
13704 depends on its real-time behavior, delays introduced by a debugger
13705 might cause the program to change its behavior drastically, or perhaps
13706 fail, even when the code itself is correct. It is useful to be able
13707 to observe the program's behavior without interrupting it.
13709 Using @value{GDBN}'s @code{trace} and @code{collect} commands, you can
13710 specify locations in the program, called @dfn{tracepoints}, and
13711 arbitrary expressions to evaluate when those tracepoints are reached.
13712 Later, using the @code{tfind} command, you can examine the values
13713 those expressions had when the program hit the tracepoints. The
13714 expressions may also denote objects in memory---structures or arrays,
13715 for example---whose values @value{GDBN} should record; while visiting
13716 a particular tracepoint, you may inspect those objects as if they were
13717 in memory at that moment. However, because @value{GDBN} records these
13718 values without interacting with you, it can do so quickly and
13719 unobtrusively, hopefully not disturbing the program's behavior.
13721 The tracepoint facility is currently available only for remote
13722 targets. @xref{Targets}. In addition, your remote target must know
13723 how to collect trace data. This functionality is implemented in the
13724 remote stub; however, none of the stubs distributed with @value{GDBN}
13725 support tracepoints as of this writing. The format of the remote
13726 packets used to implement tracepoints are described in @ref{Tracepoint
13729 It is also possible to get trace data from a file, in a manner reminiscent
13730 of corefiles; you specify the filename, and use @code{tfind} to search
13731 through the file. @xref{Trace Files}, for more details.
13733 This chapter describes the tracepoint commands and features.
13736 * Set Tracepoints::
13737 * Analyze Collected Data::
13738 * Tracepoint Variables::
13742 @node Set Tracepoints
13743 @section Commands to Set Tracepoints
13745 Before running such a @dfn{trace experiment}, an arbitrary number of
13746 tracepoints can be set. A tracepoint is actually a special type of
13747 breakpoint (@pxref{Set Breaks}), so you can manipulate it using
13748 standard breakpoint commands. For instance, as with breakpoints,
13749 tracepoint numbers are successive integers starting from one, and many
13750 of the commands associated with tracepoints take the tracepoint number
13751 as their argument, to identify which tracepoint to work on.
13753 For each tracepoint, you can specify, in advance, some arbitrary set
13754 of data that you want the target to collect in the trace buffer when
13755 it hits that tracepoint. The collected data can include registers,
13756 local variables, or global data. Later, you can use @value{GDBN}
13757 commands to examine the values these data had at the time the
13758 tracepoint was hit.
13760 Tracepoints do not support every breakpoint feature. Ignore counts on
13761 tracepoints have no effect, and tracepoints cannot run @value{GDBN}
13762 commands when they are hit. Tracepoints may not be thread-specific
13765 @cindex fast tracepoints
13766 Some targets may support @dfn{fast tracepoints}, which are inserted in
13767 a different way (such as with a jump instead of a trap), that is
13768 faster but possibly restricted in where they may be installed.
13770 @cindex static tracepoints
13771 @cindex markers, static tracepoints
13772 @cindex probing markers, static tracepoints
13773 Regular and fast tracepoints are dynamic tracing facilities, meaning
13774 that they can be used to insert tracepoints at (almost) any location
13775 in the target. Some targets may also support controlling @dfn{static
13776 tracepoints} from @value{GDBN}. With static tracing, a set of
13777 instrumentation points, also known as @dfn{markers}, are embedded in
13778 the target program, and can be activated or deactivated by name or
13779 address. These are usually placed at locations which facilitate
13780 investigating what the target is actually doing. @value{GDBN}'s
13781 support for static tracing includes being able to list instrumentation
13782 points, and attach them with @value{GDBN} defined high level
13783 tracepoints that expose the whole range of convenience of
13784 @value{GDBN}'s tracepoints support. Namely, support for collecting
13785 registers values and values of global or local (to the instrumentation
13786 point) variables; tracepoint conditions and trace state variables.
13787 The act of installing a @value{GDBN} static tracepoint on an
13788 instrumentation point, or marker, is referred to as @dfn{probing} a
13789 static tracepoint marker.
13791 @code{gdbserver} supports tracepoints on some target systems.
13792 @xref{Server,,Tracepoints support in @code{gdbserver}}.
13794 This section describes commands to set tracepoints and associated
13795 conditions and actions.
13798 * Create and Delete Tracepoints::
13799 * Enable and Disable Tracepoints::
13800 * Tracepoint Passcounts::
13801 * Tracepoint Conditions::
13802 * Trace State Variables::
13803 * Tracepoint Actions::
13804 * Listing Tracepoints::
13805 * Listing Static Tracepoint Markers::
13806 * Starting and Stopping Trace Experiments::
13807 * Tracepoint Restrictions::
13810 @node Create and Delete Tracepoints
13811 @subsection Create and Delete Tracepoints
13814 @cindex set tracepoint
13816 @item trace @var{location}
13817 The @code{trace} command is very similar to the @code{break} command.
13818 Its argument @var{location} can be any valid location.
13819 @xref{Specify Location}. The @code{trace} command defines a tracepoint,
13820 which is a point in the target program where the debugger will briefly stop,
13821 collect some data, and then allow the program to continue. Setting a tracepoint
13822 or changing its actions takes effect immediately if the remote stub
13823 supports the @samp{InstallInTrace} feature (@pxref{install tracepoint
13825 If remote stub doesn't support the @samp{InstallInTrace} feature, all
13826 these changes don't take effect until the next @code{tstart}
13827 command, and once a trace experiment is running, further changes will
13828 not have any effect until the next trace experiment starts. In addition,
13829 @value{GDBN} supports @dfn{pending tracepoints}---tracepoints whose
13830 address is not yet resolved. (This is similar to pending breakpoints.)
13831 Pending tracepoints are not downloaded to the target and not installed
13832 until they are resolved. The resolution of pending tracepoints requires
13833 @value{GDBN} support---when debugging with the remote target, and
13834 @value{GDBN} disconnects from the remote stub (@pxref{disconnected
13835 tracing}), pending tracepoints can not be resolved (and downloaded to
13836 the remote stub) while @value{GDBN} is disconnected.
13838 Here are some examples of using the @code{trace} command:
13841 (@value{GDBP}) @b{trace foo.c:121} // a source file and line number
13843 (@value{GDBP}) @b{trace +2} // 2 lines forward
13845 (@value{GDBP}) @b{trace my_function} // first source line of function
13847 (@value{GDBP}) @b{trace *my_function} // EXACT start address of function
13849 (@value{GDBP}) @b{trace *0x2117c4} // an address
13853 You can abbreviate @code{trace} as @code{tr}.
13855 @item trace @var{location} if @var{cond}
13856 Set a tracepoint with condition @var{cond}; evaluate the expression
13857 @var{cond} each time the tracepoint is reached, and collect data only
13858 if the value is nonzero---that is, if @var{cond} evaluates as true.
13859 @xref{Tracepoint Conditions, ,Tracepoint Conditions}, for more
13860 information on tracepoint conditions.
13862 @item ftrace @var{location} [ if @var{cond} ]
13863 @cindex set fast tracepoint
13864 @cindex fast tracepoints, setting
13866 The @code{ftrace} command sets a fast tracepoint. For targets that
13867 support them, fast tracepoints will use a more efficient but possibly
13868 less general technique to trigger data collection, such as a jump
13869 instruction instead of a trap, or some sort of hardware support. It
13870 may not be possible to create a fast tracepoint at the desired
13871 location, in which case the command will exit with an explanatory
13874 @value{GDBN} handles arguments to @code{ftrace} exactly as for
13877 On 32-bit x86-architecture systems, fast tracepoints normally need to
13878 be placed at an instruction that is 5 bytes or longer, but can be
13879 placed at 4-byte instructions if the low 64K of memory of the target
13880 program is available to install trampolines. Some Unix-type systems,
13881 such as @sc{gnu}/Linux, exclude low addresses from the program's
13882 address space; but for instance with the Linux kernel it is possible
13883 to let @value{GDBN} use this area by doing a @command{sysctl} command
13884 to set the @code{mmap_min_addr} kernel parameter, as in
13887 sudo sysctl -w vm.mmap_min_addr=32768
13891 which sets the low address to 32K, which leaves plenty of room for
13892 trampolines. The minimum address should be set to a page boundary.
13894 @item strace @var{location} [ if @var{cond} ]
13895 @cindex set static tracepoint
13896 @cindex static tracepoints, setting
13897 @cindex probe static tracepoint marker
13899 The @code{strace} command sets a static tracepoint. For targets that
13900 support it, setting a static tracepoint probes a static
13901 instrumentation point, or marker, found at @var{location}. It may not
13902 be possible to set a static tracepoint at the desired location, in
13903 which case the command will exit with an explanatory message.
13905 @value{GDBN} handles arguments to @code{strace} exactly as for
13906 @code{trace}, with the addition that the user can also specify
13907 @code{-m @var{marker}} as @var{location}. This probes the marker
13908 identified by the @var{marker} string identifier. This identifier
13909 depends on the static tracepoint backend library your program is
13910 using. You can find all the marker identifiers in the @samp{ID} field
13911 of the @code{info static-tracepoint-markers} command output.
13912 @xref{Listing Static Tracepoint Markers,,Listing Static Tracepoint
13913 Markers}. For example, in the following small program using the UST
13919 trace_mark(ust, bar33, "str %s", "FOOBAZ");
13924 the marker id is composed of joining the first two arguments to the
13925 @code{trace_mark} call with a slash, which translates to:
13928 (@value{GDBP}) info static-tracepoint-markers
13929 Cnt Enb ID Address What
13930 1 n ust/bar33 0x0000000000400ddc in main at stexample.c:22
13936 so you may probe the marker above with:
13939 (@value{GDBP}) strace -m ust/bar33
13942 Static tracepoints accept an extra collect action --- @code{collect
13943 $_sdata}. This collects arbitrary user data passed in the probe point
13944 call to the tracing library. In the UST example above, you'll see
13945 that the third argument to @code{trace_mark} is a printf-like format
13946 string. The user data is then the result of running that formating
13947 string against the following arguments. Note that @code{info
13948 static-tracepoint-markers} command output lists that format string in
13949 the @samp{Data:} field.
13951 You can inspect this data when analyzing the trace buffer, by printing
13952 the $_sdata variable like any other variable available to
13953 @value{GDBN}. @xref{Tracepoint Actions,,Tracepoint Action Lists}.
13956 @cindex last tracepoint number
13957 @cindex recent tracepoint number
13958 @cindex tracepoint number
13959 The convenience variable @code{$tpnum} records the tracepoint number
13960 of the most recently set tracepoint.
13962 @kindex delete tracepoint
13963 @cindex tracepoint deletion
13964 @item delete tracepoint @r{[}@var{num}@r{]}
13965 Permanently delete one or more tracepoints. With no argument, the
13966 default is to delete all tracepoints. Note that the regular
13967 @code{delete} command can remove tracepoints also.
13972 (@value{GDBP}) @b{delete trace 1 2 3} // remove three tracepoints
13974 (@value{GDBP}) @b{delete trace} // remove all tracepoints
13978 You can abbreviate this command as @code{del tr}.
13981 @node Enable and Disable Tracepoints
13982 @subsection Enable and Disable Tracepoints
13984 These commands are deprecated; they are equivalent to plain @code{disable} and @code{enable}.
13987 @kindex disable tracepoint
13988 @item disable tracepoint @r{[}@var{num}@r{]}
13989 Disable tracepoint @var{num}, or all tracepoints if no argument
13990 @var{num} is given. A disabled tracepoint will have no effect during
13991 a trace experiment, but it is not forgotten. You can re-enable
13992 a disabled tracepoint using the @code{enable tracepoint} command.
13993 If the command is issued during a trace experiment and the debug target
13994 has support for disabling tracepoints during a trace experiment, then the
13995 change will be effective immediately. Otherwise, it will be applied to the
13996 next trace experiment.
13998 @kindex enable tracepoint
13999 @item enable tracepoint @r{[}@var{num}@r{]}
14000 Enable tracepoint @var{num}, or all tracepoints. If this command is
14001 issued during a trace experiment and the debug target supports enabling
14002 tracepoints during a trace experiment, then the enabled tracepoints will
14003 become effective immediately. Otherwise, they will become effective the
14004 next time a trace experiment is run.
14007 @node Tracepoint Passcounts
14008 @subsection Tracepoint Passcounts
14012 @cindex tracepoint pass count
14013 @item passcount @r{[}@var{n} @r{[}@var{num}@r{]]}
14014 Set the @dfn{passcount} of a tracepoint. The passcount is a way to
14015 automatically stop a trace experiment. If a tracepoint's passcount is
14016 @var{n}, then the trace experiment will be automatically stopped on
14017 the @var{n}'th time that tracepoint is hit. If the tracepoint number
14018 @var{num} is not specified, the @code{passcount} command sets the
14019 passcount of the most recently defined tracepoint. If no passcount is
14020 given, the trace experiment will run until stopped explicitly by the
14026 (@value{GDBP}) @b{passcount 5 2} // Stop on the 5th execution of
14027 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// tracepoint 2}
14029 (@value{GDBP}) @b{passcount 12} // Stop on the 12th execution of the
14030 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// most recently defined tracepoint.}
14031 (@value{GDBP}) @b{trace foo}
14032 (@value{GDBP}) @b{pass 3}
14033 (@value{GDBP}) @b{trace bar}
14034 (@value{GDBP}) @b{pass 2}
14035 (@value{GDBP}) @b{trace baz}
14036 (@value{GDBP}) @b{pass 1} // Stop tracing when foo has been
14037 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// executed 3 times OR when bar has}
14038 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// been executed 2 times}
14039 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// OR when baz has been executed 1 time.}
14043 @node Tracepoint Conditions
14044 @subsection Tracepoint Conditions
14045 @cindex conditional tracepoints
14046 @cindex tracepoint conditions
14048 The simplest sort of tracepoint collects data every time your program
14049 reaches a specified place. You can also specify a @dfn{condition} for
14050 a tracepoint. A condition is just a Boolean expression in your
14051 programming language (@pxref{Expressions, ,Expressions}). A
14052 tracepoint with a condition evaluates the expression each time your
14053 program reaches it, and data collection happens only if the condition
14056 Tracepoint conditions can be specified when a tracepoint is set, by
14057 using @samp{if} in the arguments to the @code{trace} command.
14058 @xref{Create and Delete Tracepoints, ,Setting Tracepoints}. They can
14059 also be set or changed at any time with the @code{condition} command,
14060 just as with breakpoints.
14062 Unlike breakpoint conditions, @value{GDBN} does not actually evaluate
14063 the conditional expression itself. Instead, @value{GDBN} encodes the
14064 expression into an agent expression (@pxref{Agent Expressions})
14065 suitable for execution on the target, independently of @value{GDBN}.
14066 Global variables become raw memory locations, locals become stack
14067 accesses, and so forth.
14069 For instance, suppose you have a function that is usually called
14070 frequently, but should not be called after an error has occurred. You
14071 could use the following tracepoint command to collect data about calls
14072 of that function that happen while the error code is propagating
14073 through the program; an unconditional tracepoint could end up
14074 collecting thousands of useless trace frames that you would have to
14078 (@value{GDBP}) @kbd{trace normal_operation if errcode > 0}
14081 @node Trace State Variables
14082 @subsection Trace State Variables
14083 @cindex trace state variables
14085 A @dfn{trace state variable} is a special type of variable that is
14086 created and managed by target-side code. The syntax is the same as
14087 that for GDB's convenience variables (a string prefixed with ``$''),
14088 but they are stored on the target. They must be created explicitly,
14089 using a @code{tvariable} command. They are always 64-bit signed
14092 Trace state variables are remembered by @value{GDBN}, and downloaded
14093 to the target along with tracepoint information when the trace
14094 experiment starts. There are no intrinsic limits on the number of
14095 trace state variables, beyond memory limitations of the target.
14097 @cindex convenience variables, and trace state variables
14098 Although trace state variables are managed by the target, you can use
14099 them in print commands and expressions as if they were convenience
14100 variables; @value{GDBN} will get the current value from the target
14101 while the trace experiment is running. Trace state variables share
14102 the same namespace as other ``$'' variables, which means that you
14103 cannot have trace state variables with names like @code{$23} or
14104 @code{$pc}, nor can you have a trace state variable and a convenience
14105 variable with the same name.
14109 @item tvariable $@var{name} [ = @var{expression} ]
14111 The @code{tvariable} command creates a new trace state variable named
14112 @code{$@var{name}}, and optionally gives it an initial value of
14113 @var{expression}. The @var{expression} is evaluated when this command is
14114 entered; the result will be converted to an integer if possible,
14115 otherwise @value{GDBN} will report an error. A subsequent
14116 @code{tvariable} command specifying the same name does not create a
14117 variable, but instead assigns the supplied initial value to the
14118 existing variable of that name, overwriting any previous initial
14119 value. The default initial value is 0.
14121 @item info tvariables
14122 @kindex info tvariables
14123 List all the trace state variables along with their initial values.
14124 Their current values may also be displayed, if the trace experiment is
14127 @item delete tvariable @r{[} $@var{name} @dots{} @r{]}
14128 @kindex delete tvariable
14129 Delete the given trace state variables, or all of them if no arguments
14134 @node Tracepoint Actions
14135 @subsection Tracepoint Action Lists
14139 @cindex tracepoint actions
14140 @item actions @r{[}@var{num}@r{]}
14141 This command will prompt for a list of actions to be taken when the
14142 tracepoint is hit. If the tracepoint number @var{num} is not
14143 specified, this command sets the actions for the one that was most
14144 recently defined (so that you can define a tracepoint and then say
14145 @code{actions} without bothering about its number). You specify the
14146 actions themselves on the following lines, one action at a time, and
14147 terminate the actions list with a line containing just @code{end}. So
14148 far, the only defined actions are @code{collect}, @code{teval}, and
14149 @code{while-stepping}.
14151 @code{actions} is actually equivalent to @code{commands} (@pxref{Break
14152 Commands, ,Breakpoint Command Lists}), except that only the defined
14153 actions are allowed; any other @value{GDBN} command is rejected.
14155 @cindex remove actions from a tracepoint
14156 To remove all actions from a tracepoint, type @samp{actions @var{num}}
14157 and follow it immediately with @samp{end}.
14160 (@value{GDBP}) @b{collect @var{data}} // collect some data
14162 (@value{GDBP}) @b{while-stepping 5} // single-step 5 times, collect data
14164 (@value{GDBP}) @b{end} // signals the end of actions.
14167 In the following example, the action list begins with @code{collect}
14168 commands indicating the things to be collected when the tracepoint is
14169 hit. Then, in order to single-step and collect additional data
14170 following the tracepoint, a @code{while-stepping} command is used,
14171 followed by the list of things to be collected after each step in a
14172 sequence of single steps. The @code{while-stepping} command is
14173 terminated by its own separate @code{end} command. Lastly, the action
14174 list is terminated by an @code{end} command.
14177 (@value{GDBP}) @b{trace foo}
14178 (@value{GDBP}) @b{actions}
14179 Enter actions for tracepoint 1, one per line:
14182 > while-stepping 12
14183 > collect $pc, arr[i]
14188 @kindex collect @r{(tracepoints)}
14189 @item collect@r{[}/@var{mods}@r{]} @var{expr1}, @var{expr2}, @dots{}
14190 Collect values of the given expressions when the tracepoint is hit.
14191 This command accepts a comma-separated list of any valid expressions.
14192 In addition to global, static, or local variables, the following
14193 special arguments are supported:
14197 Collect all registers.
14200 Collect all function arguments.
14203 Collect all local variables.
14206 Collect the return address. This is helpful if you want to see more
14209 @emph{Note:} The return address location can not always be reliably
14210 determined up front, and the wrong address / registers may end up
14211 collected instead. On some architectures the reliability is higher
14212 for tracepoints at function entry, while on others it's the opposite.
14213 When this happens, backtracing will stop because the return address is
14214 found unavailable (unless another collect rule happened to match it).
14217 Collects the number of arguments from the static probe at which the
14218 tracepoint is located.
14219 @xref{Static Probe Points}.
14221 @item $_probe_arg@var{n}
14222 @var{n} is an integer between 0 and 11. Collects the @var{n}th argument
14223 from the static probe at which the tracepoint is located.
14224 @xref{Static Probe Points}.
14227 @vindex $_sdata@r{, collect}
14228 Collect static tracepoint marker specific data. Only available for
14229 static tracepoints. @xref{Tracepoint Actions,,Tracepoint Action
14230 Lists}. On the UST static tracepoints library backend, an
14231 instrumentation point resembles a @code{printf} function call. The
14232 tracing library is able to collect user specified data formatted to a
14233 character string using the format provided by the programmer that
14234 instrumented the program. Other backends have similar mechanisms.
14235 Here's an example of a UST marker call:
14238 const char master_name[] = "$your_name";
14239 trace_mark(channel1, marker1, "hello %s", master_name)
14242 In this case, collecting @code{$_sdata} collects the string
14243 @samp{hello $yourname}. When analyzing the trace buffer, you can
14244 inspect @samp{$_sdata} like any other variable available to
14248 You can give several consecutive @code{collect} commands, each one
14249 with a single argument, or one @code{collect} command with several
14250 arguments separated by commas; the effect is the same.
14252 The optional @var{mods} changes the usual handling of the arguments.
14253 @code{s} requests that pointers to chars be handled as strings, in
14254 particular collecting the contents of the memory being pointed at, up
14255 to the first zero. The upper bound is by default the value of the
14256 @code{print elements} variable; if @code{s} is followed by a decimal
14257 number, that is the upper bound instead. So for instance
14258 @samp{collect/s25 mystr} collects as many as 25 characters at
14261 The command @code{info scope} (@pxref{Symbols, info scope}) is
14262 particularly useful for figuring out what data to collect.
14264 @kindex teval @r{(tracepoints)}
14265 @item teval @var{expr1}, @var{expr2}, @dots{}
14266 Evaluate the given expressions when the tracepoint is hit. This
14267 command accepts a comma-separated list of expressions. The results
14268 are discarded, so this is mainly useful for assigning values to trace
14269 state variables (@pxref{Trace State Variables}) without adding those
14270 values to the trace buffer, as would be the case if the @code{collect}
14273 @kindex while-stepping @r{(tracepoints)}
14274 @item while-stepping @var{n}
14275 Perform @var{n} single-step instruction traces after the tracepoint,
14276 collecting new data after each step. The @code{while-stepping}
14277 command is followed by the list of what to collect while stepping
14278 (followed by its own @code{end} command):
14281 > while-stepping 12
14282 > collect $regs, myglobal
14288 Note that @code{$pc} is not automatically collected by
14289 @code{while-stepping}; you need to explicitly collect that register if
14290 you need it. You may abbreviate @code{while-stepping} as @code{ws} or
14293 @item set default-collect @var{expr1}, @var{expr2}, @dots{}
14294 @kindex set default-collect
14295 @cindex default collection action
14296 This variable is a list of expressions to collect at each tracepoint
14297 hit. It is effectively an additional @code{collect} action prepended
14298 to every tracepoint action list. The expressions are parsed
14299 individually for each tracepoint, so for instance a variable named
14300 @code{xyz} may be interpreted as a global for one tracepoint, and a
14301 local for another, as appropriate to the tracepoint's location.
14303 @item show default-collect
14304 @kindex show default-collect
14305 Show the list of expressions that are collected by default at each
14310 @node Listing Tracepoints
14311 @subsection Listing Tracepoints
14314 @kindex info tracepoints @r{[}@var{n}@dots{}@r{]}
14315 @kindex info tp @r{[}@var{n}@dots{}@r{]}
14316 @cindex information about tracepoints
14317 @item info tracepoints @r{[}@var{num}@dots{}@r{]}
14318 Display information about the tracepoint @var{num}. If you don't
14319 specify a tracepoint number, displays information about all the
14320 tracepoints defined so far. The format is similar to that used for
14321 @code{info breakpoints}; in fact, @code{info tracepoints} is the same
14322 command, simply restricting itself to tracepoints.
14324 A tracepoint's listing may include additional information specific to
14329 its passcount as given by the @code{passcount @var{n}} command
14332 the state about installed on target of each location
14336 (@value{GDBP}) @b{info trace}
14337 Num Type Disp Enb Address What
14338 1 tracepoint keep y 0x0804ab57 in foo() at main.cxx:7
14340 collect globfoo, $regs
14345 2 tracepoint keep y <MULTIPLE>
14347 2.1 y 0x0804859c in func4 at change-loc.h:35
14348 installed on target
14349 2.2 y 0xb7ffc480 in func4 at change-loc.h:35
14350 installed on target
14351 2.3 y <PENDING> set_tracepoint
14352 3 tracepoint keep y 0x080485b1 in foo at change-loc.c:29
14353 not installed on target
14358 This command can be abbreviated @code{info tp}.
14361 @node Listing Static Tracepoint Markers
14362 @subsection Listing Static Tracepoint Markers
14365 @kindex info static-tracepoint-markers
14366 @cindex information about static tracepoint markers
14367 @item info static-tracepoint-markers
14368 Display information about all static tracepoint markers defined in the
14371 For each marker, the following columns are printed:
14375 An incrementing counter, output to help readability. This is not a
14378 The marker ID, as reported by the target.
14379 @item Enabled or Disabled
14380 Probed markers are tagged with @samp{y}. @samp{n} identifies marks
14381 that are not enabled.
14383 Where the marker is in your program, as a memory address.
14385 Where the marker is in the source for your program, as a file and line
14386 number. If the debug information included in the program does not
14387 allow @value{GDBN} to locate the source of the marker, this column
14388 will be left blank.
14392 In addition, the following information may be printed for each marker:
14396 User data passed to the tracing library by the marker call. In the
14397 UST backend, this is the format string passed as argument to the
14399 @item Static tracepoints probing the marker
14400 The list of static tracepoints attached to the marker.
14404 (@value{GDBP}) info static-tracepoint-markers
14405 Cnt ID Enb Address What
14406 1 ust/bar2 y 0x0000000000400e1a in main at stexample.c:25
14407 Data: number1 %d number2 %d
14408 Probed by static tracepoints: #2
14409 2 ust/bar33 n 0x0000000000400c87 in main at stexample.c:24
14415 @node Starting and Stopping Trace Experiments
14416 @subsection Starting and Stopping Trace Experiments
14419 @kindex tstart [ @var{notes} ]
14420 @cindex start a new trace experiment
14421 @cindex collected data discarded
14423 This command starts the trace experiment, and begins collecting data.
14424 It has the side effect of discarding all the data collected in the
14425 trace buffer during the previous trace experiment. If any arguments
14426 are supplied, they are taken as a note and stored with the trace
14427 experiment's state. The notes may be arbitrary text, and are
14428 especially useful with disconnected tracing in a multi-user context;
14429 the notes can explain what the trace is doing, supply user contact
14430 information, and so forth.
14432 @kindex tstop [ @var{notes} ]
14433 @cindex stop a running trace experiment
14435 This command stops the trace experiment. If any arguments are
14436 supplied, they are recorded with the experiment as a note. This is
14437 useful if you are stopping a trace started by someone else, for
14438 instance if the trace is interfering with the system's behavior and
14439 needs to be stopped quickly.
14441 @strong{Note}: a trace experiment and data collection may stop
14442 automatically if any tracepoint's passcount is reached
14443 (@pxref{Tracepoint Passcounts}), or if the trace buffer becomes full.
14446 @cindex status of trace data collection
14447 @cindex trace experiment, status of
14449 This command displays the status of the current trace data
14453 Here is an example of the commands we described so far:
14456 (@value{GDBP}) @b{trace gdb_c_test}
14457 (@value{GDBP}) @b{actions}
14458 Enter actions for tracepoint #1, one per line.
14459 > collect $regs,$locals,$args
14460 > while-stepping 11
14464 (@value{GDBP}) @b{tstart}
14465 [time passes @dots{}]
14466 (@value{GDBP}) @b{tstop}
14469 @anchor{disconnected tracing}
14470 @cindex disconnected tracing
14471 You can choose to continue running the trace experiment even if
14472 @value{GDBN} disconnects from the target, voluntarily or
14473 involuntarily. For commands such as @code{detach}, the debugger will
14474 ask what you want to do with the trace. But for unexpected
14475 terminations (@value{GDBN} crash, network outage), it would be
14476 unfortunate to lose hard-won trace data, so the variable
14477 @code{disconnected-tracing} lets you decide whether the trace should
14478 continue running without @value{GDBN}.
14481 @item set disconnected-tracing on
14482 @itemx set disconnected-tracing off
14483 @kindex set disconnected-tracing
14484 Choose whether a tracing run should continue to run if @value{GDBN}
14485 has disconnected from the target. Note that @code{detach} or
14486 @code{quit} will ask you directly what to do about a running trace no
14487 matter what this variable's setting, so the variable is mainly useful
14488 for handling unexpected situations, such as loss of the network.
14490 @item show disconnected-tracing
14491 @kindex show disconnected-tracing
14492 Show the current choice for disconnected tracing.
14496 When you reconnect to the target, the trace experiment may or may not
14497 still be running; it might have filled the trace buffer in the
14498 meantime, or stopped for one of the other reasons. If it is running,
14499 it will continue after reconnection.
14501 Upon reconnection, the target will upload information about the
14502 tracepoints in effect. @value{GDBN} will then compare that
14503 information to the set of tracepoints currently defined, and attempt
14504 to match them up, allowing for the possibility that the numbers may
14505 have changed due to creation and deletion in the meantime. If one of
14506 the target's tracepoints does not match any in @value{GDBN}, the
14507 debugger will create a new tracepoint, so that you have a number with
14508 which to specify that tracepoint. This matching-up process is
14509 necessarily heuristic, and it may result in useless tracepoints being
14510 created; you may simply delete them if they are of no use.
14512 @cindex circular trace buffer
14513 If your target agent supports a @dfn{circular trace buffer}, then you
14514 can run a trace experiment indefinitely without filling the trace
14515 buffer; when space runs out, the agent deletes already-collected trace
14516 frames, oldest first, until there is enough room to continue
14517 collecting. This is especially useful if your tracepoints are being
14518 hit too often, and your trace gets terminated prematurely because the
14519 buffer is full. To ask for a circular trace buffer, simply set
14520 @samp{circular-trace-buffer} to on. You can set this at any time,
14521 including during tracing; if the agent can do it, it will change
14522 buffer handling on the fly, otherwise it will not take effect until
14526 @item set circular-trace-buffer on
14527 @itemx set circular-trace-buffer off
14528 @kindex set circular-trace-buffer
14529 Choose whether a tracing run should use a linear or circular buffer
14530 for trace data. A linear buffer will not lose any trace data, but may
14531 fill up prematurely, while a circular buffer will discard old trace
14532 data, but it will have always room for the latest tracepoint hits.
14534 @item show circular-trace-buffer
14535 @kindex show circular-trace-buffer
14536 Show the current choice for the trace buffer. Note that this may not
14537 match the agent's current buffer handling, nor is it guaranteed to
14538 match the setting that might have been in effect during a past run,
14539 for instance if you are looking at frames from a trace file.
14544 @item set trace-buffer-size @var{n}
14545 @itemx set trace-buffer-size unlimited
14546 @kindex set trace-buffer-size
14547 Request that the target use a trace buffer of @var{n} bytes. Not all
14548 targets will honor the request; they may have a compiled-in size for
14549 the trace buffer, or some other limitation. Set to a value of
14550 @code{unlimited} or @code{-1} to let the target use whatever size it
14551 likes. This is also the default.
14553 @item show trace-buffer-size
14554 @kindex show trace-buffer-size
14555 Show the current requested size for the trace buffer. Note that this
14556 will only match the actual size if the target supports size-setting,
14557 and was able to handle the requested size. For instance, if the
14558 target can only change buffer size between runs, this variable will
14559 not reflect the change until the next run starts. Use @code{tstatus}
14560 to get a report of the actual buffer size.
14564 @item set trace-user @var{text}
14565 @kindex set trace-user
14567 @item show trace-user
14568 @kindex show trace-user
14570 @item set trace-notes @var{text}
14571 @kindex set trace-notes
14572 Set the trace run's notes.
14574 @item show trace-notes
14575 @kindex show trace-notes
14576 Show the trace run's notes.
14578 @item set trace-stop-notes @var{text}
14579 @kindex set trace-stop-notes
14580 Set the trace run's stop notes. The handling of the note is as for
14581 @code{tstop} arguments; the set command is convenient way to fix a
14582 stop note that is mistaken or incomplete.
14584 @item show trace-stop-notes
14585 @kindex show trace-stop-notes
14586 Show the trace run's stop notes.
14590 @node Tracepoint Restrictions
14591 @subsection Tracepoint Restrictions
14593 @cindex tracepoint restrictions
14594 There are a number of restrictions on the use of tracepoints. As
14595 described above, tracepoint data gathering occurs on the target
14596 without interaction from @value{GDBN}. Thus the full capabilities of
14597 the debugger are not available during data gathering, and then at data
14598 examination time, you will be limited by only having what was
14599 collected. The following items describe some common problems, but it
14600 is not exhaustive, and you may run into additional difficulties not
14606 Tracepoint expressions are intended to gather objects (lvalues). Thus
14607 the full flexibility of GDB's expression evaluator is not available.
14608 You cannot call functions, cast objects to aggregate types, access
14609 convenience variables or modify values (except by assignment to trace
14610 state variables). Some language features may implicitly call
14611 functions (for instance Objective-C fields with accessors), and therefore
14612 cannot be collected either.
14615 Collection of local variables, either individually or in bulk with
14616 @code{$locals} or @code{$args}, during @code{while-stepping} may
14617 behave erratically. The stepping action may enter a new scope (for
14618 instance by stepping into a function), or the location of the variable
14619 may change (for instance it is loaded into a register). The
14620 tracepoint data recorded uses the location information for the
14621 variables that is correct for the tracepoint location. When the
14622 tracepoint is created, it is not possible, in general, to determine
14623 where the steps of a @code{while-stepping} sequence will advance the
14624 program---particularly if a conditional branch is stepped.
14627 Collection of an incompletely-initialized or partially-destroyed object
14628 may result in something that @value{GDBN} cannot display, or displays
14629 in a misleading way.
14632 When @value{GDBN} displays a pointer to character it automatically
14633 dereferences the pointer to also display characters of the string
14634 being pointed to. However, collecting the pointer during tracing does
14635 not automatically collect the string. You need to explicitly
14636 dereference the pointer and provide size information if you want to
14637 collect not only the pointer, but the memory pointed to. For example,
14638 @code{*ptr@@50} can be used to collect the 50 element array pointed to
14642 It is not possible to collect a complete stack backtrace at a
14643 tracepoint. Instead, you may collect the registers and a few hundred
14644 bytes from the stack pointer with something like @code{*(unsigned char *)$esp@@300}
14645 (adjust to use the name of the actual stack pointer register on your
14646 target architecture, and the amount of stack you wish to capture).
14647 Then the @code{backtrace} command will show a partial backtrace when
14648 using a trace frame. The number of stack frames that can be examined
14649 depends on the sizes of the frames in the collected stack. Note that
14650 if you ask for a block so large that it goes past the bottom of the
14651 stack, the target agent may report an error trying to read from an
14655 If you do not collect registers at a tracepoint, @value{GDBN} can
14656 infer that the value of @code{$pc} must be the same as the address of
14657 the tracepoint and use that when you are looking at a trace frame
14658 for that tracepoint. However, this cannot work if the tracepoint has
14659 multiple locations (for instance if it was set in a function that was
14660 inlined), or if it has a @code{while-stepping} loop. In those cases
14661 @value{GDBN} will warn you that it can't infer @code{$pc}, and default
14666 @node Analyze Collected Data
14667 @section Using the Collected Data
14669 After the tracepoint experiment ends, you use @value{GDBN} commands
14670 for examining the trace data. The basic idea is that each tracepoint
14671 collects a trace @dfn{snapshot} every time it is hit and another
14672 snapshot every time it single-steps. All these snapshots are
14673 consecutively numbered from zero and go into a buffer, and you can
14674 examine them later. The way you examine them is to @dfn{focus} on a
14675 specific trace snapshot. When the remote stub is focused on a trace
14676 snapshot, it will respond to all @value{GDBN} requests for memory and
14677 registers by reading from the buffer which belongs to that snapshot,
14678 rather than from @emph{real} memory or registers of the program being
14679 debugged. This means that @strong{all} @value{GDBN} commands
14680 (@code{print}, @code{info registers}, @code{backtrace}, etc.) will
14681 behave as if we were currently debugging the program state as it was
14682 when the tracepoint occurred. Any requests for data that are not in
14683 the buffer will fail.
14686 * tfind:: How to select a trace snapshot
14687 * tdump:: How to display all data for a snapshot
14688 * save tracepoints:: How to save tracepoints for a future run
14692 @subsection @code{tfind @var{n}}
14695 @cindex select trace snapshot
14696 @cindex find trace snapshot
14697 The basic command for selecting a trace snapshot from the buffer is
14698 @code{tfind @var{n}}, which finds trace snapshot number @var{n},
14699 counting from zero. If no argument @var{n} is given, the next
14700 snapshot is selected.
14702 Here are the various forms of using the @code{tfind} command.
14706 Find the first snapshot in the buffer. This is a synonym for
14707 @code{tfind 0} (since 0 is the number of the first snapshot).
14710 Stop debugging trace snapshots, resume @emph{live} debugging.
14713 Same as @samp{tfind none}.
14716 No argument means find the next trace snapshot or find the first
14717 one if no trace snapshot is selected.
14720 Find the previous trace snapshot before the current one. This permits
14721 retracing earlier steps.
14723 @item tfind tracepoint @var{num}
14724 Find the next snapshot associated with tracepoint @var{num}. Search
14725 proceeds forward from the last examined trace snapshot. If no
14726 argument @var{num} is given, it means find the next snapshot collected
14727 for the same tracepoint as the current snapshot.
14729 @item tfind pc @var{addr}
14730 Find the next snapshot associated with the value @var{addr} of the
14731 program counter. Search proceeds forward from the last examined trace
14732 snapshot. If no argument @var{addr} is given, it means find the next
14733 snapshot with the same value of PC as the current snapshot.
14735 @item tfind outside @var{addr1}, @var{addr2}
14736 Find the next snapshot whose PC is outside the given range of
14737 addresses (exclusive).
14739 @item tfind range @var{addr1}, @var{addr2}
14740 Find the next snapshot whose PC is between @var{addr1} and
14741 @var{addr2} (inclusive).
14743 @item tfind line @r{[}@var{file}:@r{]}@var{n}
14744 Find the next snapshot associated with the source line @var{n}. If
14745 the optional argument @var{file} is given, refer to line @var{n} in
14746 that source file. Search proceeds forward from the last examined
14747 trace snapshot. If no argument @var{n} is given, it means find the
14748 next line other than the one currently being examined; thus saying
14749 @code{tfind line} repeatedly can appear to have the same effect as
14750 stepping from line to line in a @emph{live} debugging session.
14753 The default arguments for the @code{tfind} commands are specifically
14754 designed to make it easy to scan through the trace buffer. For
14755 instance, @code{tfind} with no argument selects the next trace
14756 snapshot, and @code{tfind -} with no argument selects the previous
14757 trace snapshot. So, by giving one @code{tfind} command, and then
14758 simply hitting @key{RET} repeatedly you can examine all the trace
14759 snapshots in order. Or, by saying @code{tfind -} and then hitting
14760 @key{RET} repeatedly you can examine the snapshots in reverse order.
14761 The @code{tfind line} command with no argument selects the snapshot
14762 for the next source line executed. The @code{tfind pc} command with
14763 no argument selects the next snapshot with the same program counter
14764 (PC) as the current frame. The @code{tfind tracepoint} command with
14765 no argument selects the next trace snapshot collected by the same
14766 tracepoint as the current one.
14768 In addition to letting you scan through the trace buffer manually,
14769 these commands make it easy to construct @value{GDBN} scripts that
14770 scan through the trace buffer and print out whatever collected data
14771 you are interested in. Thus, if we want to examine the PC, FP, and SP
14772 registers from each trace frame in the buffer, we can say this:
14775 (@value{GDBP}) @b{tfind start}
14776 (@value{GDBP}) @b{while ($trace_frame != -1)}
14777 > printf "Frame %d, PC = %08X, SP = %08X, FP = %08X\n", \
14778 $trace_frame, $pc, $sp, $fp
14782 Frame 0, PC = 0020DC64, SP = 0030BF3C, FP = 0030BF44
14783 Frame 1, PC = 0020DC6C, SP = 0030BF38, FP = 0030BF44
14784 Frame 2, PC = 0020DC70, SP = 0030BF34, FP = 0030BF44
14785 Frame 3, PC = 0020DC74, SP = 0030BF30, FP = 0030BF44
14786 Frame 4, PC = 0020DC78, SP = 0030BF2C, FP = 0030BF44
14787 Frame 5, PC = 0020DC7C, SP = 0030BF28, FP = 0030BF44
14788 Frame 6, PC = 0020DC80, SP = 0030BF24, FP = 0030BF44
14789 Frame 7, PC = 0020DC84, SP = 0030BF20, FP = 0030BF44
14790 Frame 8, PC = 0020DC88, SP = 0030BF1C, FP = 0030BF44
14791 Frame 9, PC = 0020DC8E, SP = 0030BF18, FP = 0030BF44
14792 Frame 10, PC = 00203F6C, SP = 0030BE3C, FP = 0030BF14
14795 Or, if we want to examine the variable @code{X} at each source line in
14799 (@value{GDBP}) @b{tfind start}
14800 (@value{GDBP}) @b{while ($trace_frame != -1)}
14801 > printf "Frame %d, X == %d\n", $trace_frame, X
14811 @subsection @code{tdump}
14813 @cindex dump all data collected at tracepoint
14814 @cindex tracepoint data, display
14816 This command takes no arguments. It prints all the data collected at
14817 the current trace snapshot.
14820 (@value{GDBP}) @b{trace 444}
14821 (@value{GDBP}) @b{actions}
14822 Enter actions for tracepoint #2, one per line:
14823 > collect $regs, $locals, $args, gdb_long_test
14826 (@value{GDBP}) @b{tstart}
14828 (@value{GDBP}) @b{tfind line 444}
14829 #0 gdb_test (p1=0x11, p2=0x22, p3=0x33, p4=0x44, p5=0x55, p6=0x66)
14831 444 printp( "%s: arguments = 0x%X 0x%X 0x%X 0x%X 0x%X 0x%X\n", )
14833 (@value{GDBP}) @b{tdump}
14834 Data collected at tracepoint 2, trace frame 1:
14835 d0 0xc4aa0085 -995491707
14839 d4 0x71aea3d 119204413
14842 d7 0x380035 3670069
14843 a0 0x19e24a 1696330
14844 a1 0x3000668 50333288
14846 a3 0x322000 3284992
14847 a4 0x3000698 50333336
14848 a5 0x1ad3cc 1758156
14849 fp 0x30bf3c 0x30bf3c
14850 sp 0x30bf34 0x30bf34
14852 pc 0x20b2c8 0x20b2c8
14856 p = 0x20e5b4 "gdb-test"
14863 gdb_long_test = 17 '\021'
14868 @code{tdump} works by scanning the tracepoint's current collection
14869 actions and printing the value of each expression listed. So
14870 @code{tdump} can fail, if after a run, you change the tracepoint's
14871 actions to mention variables that were not collected during the run.
14873 Also, for tracepoints with @code{while-stepping} loops, @code{tdump}
14874 uses the collected value of @code{$pc} to distinguish between trace
14875 frames that were collected at the tracepoint hit, and frames that were
14876 collected while stepping. This allows it to correctly choose whether
14877 to display the basic list of collections, or the collections from the
14878 body of the while-stepping loop. However, if @code{$pc} was not collected,
14879 then @code{tdump} will always attempt to dump using the basic collection
14880 list, and may fail if a while-stepping frame does not include all the
14881 same data that is collected at the tracepoint hit.
14882 @c This is getting pretty arcane, example would be good.
14884 @node save tracepoints
14885 @subsection @code{save tracepoints @var{filename}}
14886 @kindex save tracepoints
14887 @kindex save-tracepoints
14888 @cindex save tracepoints for future sessions
14890 This command saves all current tracepoint definitions together with
14891 their actions and passcounts, into a file @file{@var{filename}}
14892 suitable for use in a later debugging session. To read the saved
14893 tracepoint definitions, use the @code{source} command (@pxref{Command
14894 Files}). The @w{@code{save-tracepoints}} command is a deprecated
14895 alias for @w{@code{save tracepoints}}
14897 @node Tracepoint Variables
14898 @section Convenience Variables for Tracepoints
14899 @cindex tracepoint variables
14900 @cindex convenience variables for tracepoints
14903 @vindex $trace_frame
14904 @item (int) $trace_frame
14905 The current trace snapshot (a.k.a.@: @dfn{frame}) number, or -1 if no
14906 snapshot is selected.
14908 @vindex $tracepoint
14909 @item (int) $tracepoint
14910 The tracepoint for the current trace snapshot.
14912 @vindex $trace_line
14913 @item (int) $trace_line
14914 The line number for the current trace snapshot.
14916 @vindex $trace_file
14917 @item (char []) $trace_file
14918 The source file for the current trace snapshot.
14920 @vindex $trace_func
14921 @item (char []) $trace_func
14922 The name of the function containing @code{$tracepoint}.
14925 Note: @code{$trace_file} is not suitable for use in @code{printf},
14926 use @code{output} instead.
14928 Here's a simple example of using these convenience variables for
14929 stepping through all the trace snapshots and printing some of their
14930 data. Note that these are not the same as trace state variables,
14931 which are managed by the target.
14934 (@value{GDBP}) @b{tfind start}
14936 (@value{GDBP}) @b{while $trace_frame != -1}
14937 > output $trace_file
14938 > printf ", line %d (tracepoint #%d)\n", $trace_line, $tracepoint
14944 @section Using Trace Files
14945 @cindex trace files
14947 In some situations, the target running a trace experiment may no
14948 longer be available; perhaps it crashed, or the hardware was needed
14949 for a different activity. To handle these cases, you can arrange to
14950 dump the trace data into a file, and later use that file as a source
14951 of trace data, via the @code{target tfile} command.
14956 @item tsave [ -r ] @var{filename}
14957 @itemx tsave [-ctf] @var{dirname}
14958 Save the trace data to @var{filename}. By default, this command
14959 assumes that @var{filename} refers to the host filesystem, so if
14960 necessary @value{GDBN} will copy raw trace data up from the target and
14961 then save it. If the target supports it, you can also supply the
14962 optional argument @code{-r} (``remote'') to direct the target to save
14963 the data directly into @var{filename} in its own filesystem, which may be
14964 more efficient if the trace buffer is very large. (Note, however, that
14965 @code{target tfile} can only read from files accessible to the host.)
14966 By default, this command will save trace frame in tfile format.
14967 You can supply the optional argument @code{-ctf} to save data in CTF
14968 format. The @dfn{Common Trace Format} (CTF) is proposed as a trace format
14969 that can be shared by multiple debugging and tracing tools. Please go to
14970 @indicateurl{http://www.efficios.com/ctf} to get more information.
14972 @kindex target tfile
14976 @item target tfile @var{filename}
14977 @itemx target ctf @var{dirname}
14978 Use the file named @var{filename} or directory named @var{dirname} as
14979 a source of trace data. Commands that examine data work as they do with
14980 a live target, but it is not possible to run any new trace experiments.
14981 @code{tstatus} will report the state of the trace run at the moment
14982 the data was saved, as well as the current trace frame you are examining.
14983 Both @var{filename} and @var{dirname} must be on a filesystem accessible to
14987 (@value{GDBP}) target ctf ctf.ctf
14988 (@value{GDBP}) tfind
14989 Found trace frame 0, tracepoint 2
14990 39 ++a; /* set tracepoint 1 here */
14991 (@value{GDBP}) tdump
14992 Data collected at tracepoint 2, trace frame 0:
14996 c = @{"123", "456", "789", "123", "456", "789"@}
14997 d = @{@{@{a = 1, b = 2@}, @{a = 3, b = 4@}@}, @{@{a = 5, b = 6@}, @{a = 7, b = 8@}@}@}
15005 @chapter Debugging Programs That Use Overlays
15008 If your program is too large to fit completely in your target system's
15009 memory, you can sometimes use @dfn{overlays} to work around this
15010 problem. @value{GDBN} provides some support for debugging programs that
15014 * How Overlays Work:: A general explanation of overlays.
15015 * Overlay Commands:: Managing overlays in @value{GDBN}.
15016 * Automatic Overlay Debugging:: @value{GDBN} can find out which overlays are
15017 mapped by asking the inferior.
15018 * Overlay Sample Program:: A sample program using overlays.
15021 @node How Overlays Work
15022 @section How Overlays Work
15023 @cindex mapped overlays
15024 @cindex unmapped overlays
15025 @cindex load address, overlay's
15026 @cindex mapped address
15027 @cindex overlay area
15029 Suppose you have a computer whose instruction address space is only 64
15030 kilobytes long, but which has much more memory which can be accessed by
15031 other means: special instructions, segment registers, or memory
15032 management hardware, for example. Suppose further that you want to
15033 adapt a program which is larger than 64 kilobytes to run on this system.
15035 One solution is to identify modules of your program which are relatively
15036 independent, and need not call each other directly; call these modules
15037 @dfn{overlays}. Separate the overlays from the main program, and place
15038 their machine code in the larger memory. Place your main program in
15039 instruction memory, but leave at least enough space there to hold the
15040 largest overlay as well.
15042 Now, to call a function located in an overlay, you must first copy that
15043 overlay's machine code from the large memory into the space set aside
15044 for it in the instruction memory, and then jump to its entry point
15047 @c NB: In the below the mapped area's size is greater or equal to the
15048 @c size of all overlays. This is intentional to remind the developer
15049 @c that overlays don't necessarily need to be the same size.
15053 Data Instruction Larger
15054 Address Space Address Space Address Space
15055 +-----------+ +-----------+ +-----------+
15057 +-----------+ +-----------+ +-----------+<-- overlay 1
15058 | program | | main | .----| overlay 1 | load address
15059 | variables | | program | | +-----------+
15060 | and heap | | | | | |
15061 +-----------+ | | | +-----------+<-- overlay 2
15062 | | +-----------+ | | | load address
15063 +-----------+ | | | .-| overlay 2 |
15065 mapped --->+-----------+ | | +-----------+
15066 address | | | | | |
15067 | overlay | <-' | | |
15068 | area | <---' +-----------+<-- overlay 3
15069 | | <---. | | load address
15070 +-----------+ `--| overlay 3 |
15077 @anchor{A code overlay}A code overlay
15081 The diagram (@pxref{A code overlay}) shows a system with separate data
15082 and instruction address spaces. To map an overlay, the program copies
15083 its code from the larger address space to the instruction address space.
15084 Since the overlays shown here all use the same mapped address, only one
15085 may be mapped at a time. For a system with a single address space for
15086 data and instructions, the diagram would be similar, except that the
15087 program variables and heap would share an address space with the main
15088 program and the overlay area.
15090 An overlay loaded into instruction memory and ready for use is called a
15091 @dfn{mapped} overlay; its @dfn{mapped address} is its address in the
15092 instruction memory. An overlay not present (or only partially present)
15093 in instruction memory is called @dfn{unmapped}; its @dfn{load address}
15094 is its address in the larger memory. The mapped address is also called
15095 the @dfn{virtual memory address}, or @dfn{VMA}; the load address is also
15096 called the @dfn{load memory address}, or @dfn{LMA}.
15098 Unfortunately, overlays are not a completely transparent way to adapt a
15099 program to limited instruction memory. They introduce a new set of
15100 global constraints you must keep in mind as you design your program:
15105 Before calling or returning to a function in an overlay, your program
15106 must make sure that overlay is actually mapped. Otherwise, the call or
15107 return will transfer control to the right address, but in the wrong
15108 overlay, and your program will probably crash.
15111 If the process of mapping an overlay is expensive on your system, you
15112 will need to choose your overlays carefully to minimize their effect on
15113 your program's performance.
15116 The executable file you load onto your system must contain each
15117 overlay's instructions, appearing at the overlay's load address, not its
15118 mapped address. However, each overlay's instructions must be relocated
15119 and its symbols defined as if the overlay were at its mapped address.
15120 You can use GNU linker scripts to specify different load and relocation
15121 addresses for pieces of your program; see @ref{Overlay Description,,,
15122 ld.info, Using ld: the GNU linker}.
15125 The procedure for loading executable files onto your system must be able
15126 to load their contents into the larger address space as well as the
15127 instruction and data spaces.
15131 The overlay system described above is rather simple, and could be
15132 improved in many ways:
15137 If your system has suitable bank switch registers or memory management
15138 hardware, you could use those facilities to make an overlay's load area
15139 contents simply appear at their mapped address in instruction space.
15140 This would probably be faster than copying the overlay to its mapped
15141 area in the usual way.
15144 If your overlays are small enough, you could set aside more than one
15145 overlay area, and have more than one overlay mapped at a time.
15148 You can use overlays to manage data, as well as instructions. In
15149 general, data overlays are even less transparent to your design than
15150 code overlays: whereas code overlays only require care when you call or
15151 return to functions, data overlays require care every time you access
15152 the data. Also, if you change the contents of a data overlay, you
15153 must copy its contents back out to its load address before you can copy a
15154 different data overlay into the same mapped area.
15159 @node Overlay Commands
15160 @section Overlay Commands
15162 To use @value{GDBN}'s overlay support, each overlay in your program must
15163 correspond to a separate section of the executable file. The section's
15164 virtual memory address and load memory address must be the overlay's
15165 mapped and load addresses. Identifying overlays with sections allows
15166 @value{GDBN} to determine the appropriate address of a function or
15167 variable, depending on whether the overlay is mapped or not.
15169 @value{GDBN}'s overlay commands all start with the word @code{overlay};
15170 you can abbreviate this as @code{ov} or @code{ovly}. The commands are:
15175 Disable @value{GDBN}'s overlay support. When overlay support is
15176 disabled, @value{GDBN} assumes that all functions and variables are
15177 always present at their mapped addresses. By default, @value{GDBN}'s
15178 overlay support is disabled.
15180 @item overlay manual
15181 @cindex manual overlay debugging
15182 Enable @dfn{manual} overlay debugging. In this mode, @value{GDBN}
15183 relies on you to tell it which overlays are mapped, and which are not,
15184 using the @code{overlay map-overlay} and @code{overlay unmap-overlay}
15185 commands described below.
15187 @item overlay map-overlay @var{overlay}
15188 @itemx overlay map @var{overlay}
15189 @cindex map an overlay
15190 Tell @value{GDBN} that @var{overlay} is now mapped; @var{overlay} must
15191 be the name of the object file section containing the overlay. When an
15192 overlay is mapped, @value{GDBN} assumes it can find the overlay's
15193 functions and variables at their mapped addresses. @value{GDBN} assumes
15194 that any other overlays whose mapped ranges overlap that of
15195 @var{overlay} are now unmapped.
15197 @item overlay unmap-overlay @var{overlay}
15198 @itemx overlay unmap @var{overlay}
15199 @cindex unmap an overlay
15200 Tell @value{GDBN} that @var{overlay} is no longer mapped; @var{overlay}
15201 must be the name of the object file section containing the overlay.
15202 When an overlay is unmapped, @value{GDBN} assumes it can find the
15203 overlay's functions and variables at their load addresses.
15206 Enable @dfn{automatic} overlay debugging. In this mode, @value{GDBN}
15207 consults a data structure the overlay manager maintains in the inferior
15208 to see which overlays are mapped. For details, see @ref{Automatic
15209 Overlay Debugging}.
15211 @item overlay load-target
15212 @itemx overlay load
15213 @cindex reloading the overlay table
15214 Re-read the overlay table from the inferior. Normally, @value{GDBN}
15215 re-reads the table @value{GDBN} automatically each time the inferior
15216 stops, so this command should only be necessary if you have changed the
15217 overlay mapping yourself using @value{GDBN}. This command is only
15218 useful when using automatic overlay debugging.
15220 @item overlay list-overlays
15221 @itemx overlay list
15222 @cindex listing mapped overlays
15223 Display a list of the overlays currently mapped, along with their mapped
15224 addresses, load addresses, and sizes.
15228 Normally, when @value{GDBN} prints a code address, it includes the name
15229 of the function the address falls in:
15232 (@value{GDBP}) print main
15233 $3 = @{int ()@} 0x11a0 <main>
15236 When overlay debugging is enabled, @value{GDBN} recognizes code in
15237 unmapped overlays, and prints the names of unmapped functions with
15238 asterisks around them. For example, if @code{foo} is a function in an
15239 unmapped overlay, @value{GDBN} prints it this way:
15242 (@value{GDBP}) overlay list
15243 No sections are mapped.
15244 (@value{GDBP}) print foo
15245 $5 = @{int (int)@} 0x100000 <*foo*>
15248 When @code{foo}'s overlay is mapped, @value{GDBN} prints the function's
15252 (@value{GDBP}) overlay list
15253 Section .ov.foo.text, loaded at 0x100000 - 0x100034,
15254 mapped at 0x1016 - 0x104a
15255 (@value{GDBP}) print foo
15256 $6 = @{int (int)@} 0x1016 <foo>
15259 When overlay debugging is enabled, @value{GDBN} can find the correct
15260 address for functions and variables in an overlay, whether or not the
15261 overlay is mapped. This allows most @value{GDBN} commands, like
15262 @code{break} and @code{disassemble}, to work normally, even on unmapped
15263 code. However, @value{GDBN}'s breakpoint support has some limitations:
15267 @cindex breakpoints in overlays
15268 @cindex overlays, setting breakpoints in
15269 You can set breakpoints in functions in unmapped overlays, as long as
15270 @value{GDBN} can write to the overlay at its load address.
15272 @value{GDBN} can not set hardware or simulator-based breakpoints in
15273 unmapped overlays. However, if you set a breakpoint at the end of your
15274 overlay manager (and tell @value{GDBN} which overlays are now mapped, if
15275 you are using manual overlay management), @value{GDBN} will re-set its
15276 breakpoints properly.
15280 @node Automatic Overlay Debugging
15281 @section Automatic Overlay Debugging
15282 @cindex automatic overlay debugging
15284 @value{GDBN} can automatically track which overlays are mapped and which
15285 are not, given some simple co-operation from the overlay manager in the
15286 inferior. If you enable automatic overlay debugging with the
15287 @code{overlay auto} command (@pxref{Overlay Commands}), @value{GDBN}
15288 looks in the inferior's memory for certain variables describing the
15289 current state of the overlays.
15291 Here are the variables your overlay manager must define to support
15292 @value{GDBN}'s automatic overlay debugging:
15296 @item @code{_ovly_table}:
15297 This variable must be an array of the following structures:
15302 /* The overlay's mapped address. */
15305 /* The size of the overlay, in bytes. */
15306 unsigned long size;
15308 /* The overlay's load address. */
15311 /* Non-zero if the overlay is currently mapped;
15313 unsigned long mapped;
15317 @item @code{_novlys}:
15318 This variable must be a four-byte signed integer, holding the total
15319 number of elements in @code{_ovly_table}.
15323 To decide whether a particular overlay is mapped or not, @value{GDBN}
15324 looks for an entry in @w{@code{_ovly_table}} whose @code{vma} and
15325 @code{lma} members equal the VMA and LMA of the overlay's section in the
15326 executable file. When @value{GDBN} finds a matching entry, it consults
15327 the entry's @code{mapped} member to determine whether the overlay is
15330 In addition, your overlay manager may define a function called
15331 @code{_ovly_debug_event}. If this function is defined, @value{GDBN}
15332 will silently set a breakpoint there. If the overlay manager then
15333 calls this function whenever it has changed the overlay table, this
15334 will enable @value{GDBN} to accurately keep track of which overlays
15335 are in program memory, and update any breakpoints that may be set
15336 in overlays. This will allow breakpoints to work even if the
15337 overlays are kept in ROM or other non-writable memory while they
15338 are not being executed.
15340 @node Overlay Sample Program
15341 @section Overlay Sample Program
15342 @cindex overlay example program
15344 When linking a program which uses overlays, you must place the overlays
15345 at their load addresses, while relocating them to run at their mapped
15346 addresses. To do this, you must write a linker script (@pxref{Overlay
15347 Description,,, ld.info, Using ld: the GNU linker}). Unfortunately,
15348 since linker scripts are specific to a particular host system, target
15349 architecture, and target memory layout, this manual cannot provide
15350 portable sample code demonstrating @value{GDBN}'s overlay support.
15352 However, the @value{GDBN} source distribution does contain an overlaid
15353 program, with linker scripts for a few systems, as part of its test
15354 suite. The program consists of the following files from
15355 @file{gdb/testsuite/gdb.base}:
15359 The main program file.
15361 A simple overlay manager, used by @file{overlays.c}.
15366 Overlay modules, loaded and used by @file{overlays.c}.
15369 Linker scripts for linking the test program on the @code{d10v-elf}
15370 and @code{m32r-elf} targets.
15373 You can build the test program using the @code{d10v-elf} GCC
15374 cross-compiler like this:
15377 $ d10v-elf-gcc -g -c overlays.c
15378 $ d10v-elf-gcc -g -c ovlymgr.c
15379 $ d10v-elf-gcc -g -c foo.c
15380 $ d10v-elf-gcc -g -c bar.c
15381 $ d10v-elf-gcc -g -c baz.c
15382 $ d10v-elf-gcc -g -c grbx.c
15383 $ d10v-elf-gcc -g overlays.o ovlymgr.o foo.o bar.o \
15384 baz.o grbx.o -Wl,-Td10v.ld -o overlays
15387 The build process is identical for any other architecture, except that
15388 you must substitute the appropriate compiler and linker script for the
15389 target system for @code{d10v-elf-gcc} and @code{d10v.ld}.
15393 @chapter Using @value{GDBN} with Different Languages
15396 Although programming languages generally have common aspects, they are
15397 rarely expressed in the same manner. For instance, in ANSI C,
15398 dereferencing a pointer @code{p} is accomplished by @code{*p}, but in
15399 Modula-2, it is accomplished by @code{p^}. Values can also be
15400 represented (and displayed) differently. Hex numbers in C appear as
15401 @samp{0x1ae}, while in Modula-2 they appear as @samp{1AEH}.
15403 @cindex working language
15404 Language-specific information is built into @value{GDBN} for some languages,
15405 allowing you to express operations like the above in your program's
15406 native language, and allowing @value{GDBN} to output values in a manner
15407 consistent with the syntax of your program's native language. The
15408 language you use to build expressions is called the @dfn{working
15412 * Setting:: Switching between source languages
15413 * Show:: Displaying the language
15414 * Checks:: Type and range checks
15415 * Supported Languages:: Supported languages
15416 * Unsupported Languages:: Unsupported languages
15420 @section Switching Between Source Languages
15422 There are two ways to control the working language---either have @value{GDBN}
15423 set it automatically, or select it manually yourself. You can use the
15424 @code{set language} command for either purpose. On startup, @value{GDBN}
15425 defaults to setting the language automatically. The working language is
15426 used to determine how expressions you type are interpreted, how values
15429 In addition to the working language, every source file that
15430 @value{GDBN} knows about has its own working language. For some object
15431 file formats, the compiler might indicate which language a particular
15432 source file is in. However, most of the time @value{GDBN} infers the
15433 language from the name of the file. The language of a source file
15434 controls whether C@t{++} names are demangled---this way @code{backtrace} can
15435 show each frame appropriately for its own language. There is no way to
15436 set the language of a source file from within @value{GDBN}, but you can
15437 set the language associated with a filename extension. @xref{Show, ,
15438 Displaying the Language}.
15440 This is most commonly a problem when you use a program, such
15441 as @code{cfront} or @code{f2c}, that generates C but is written in
15442 another language. In that case, make the
15443 program use @code{#line} directives in its C output; that way
15444 @value{GDBN} will know the correct language of the source code of the original
15445 program, and will display that source code, not the generated C code.
15448 * Filenames:: Filename extensions and languages.
15449 * Manually:: Setting the working language manually
15450 * Automatically:: Having @value{GDBN} infer the source language
15454 @subsection List of Filename Extensions and Languages
15456 If a source file name ends in one of the following extensions, then
15457 @value{GDBN} infers that its language is the one indicated.
15475 C@t{++} source file
15481 Objective-C source file
15485 Fortran source file
15488 Modula-2 source file
15492 Assembler source file. This actually behaves almost like C, but
15493 @value{GDBN} does not skip over function prologues when stepping.
15496 In addition, you may set the language associated with a filename
15497 extension. @xref{Show, , Displaying the Language}.
15500 @subsection Setting the Working Language
15502 If you allow @value{GDBN} to set the language automatically,
15503 expressions are interpreted the same way in your debugging session and
15506 @kindex set language
15507 If you wish, you may set the language manually. To do this, issue the
15508 command @samp{set language @var{lang}}, where @var{lang} is the name of
15509 a language, such as
15510 @code{c} or @code{modula-2}.
15511 For a list of the supported languages, type @samp{set language}.
15513 Setting the language manually prevents @value{GDBN} from updating the working
15514 language automatically. This can lead to confusion if you try
15515 to debug a program when the working language is not the same as the
15516 source language, when an expression is acceptable to both
15517 languages---but means different things. For instance, if the current
15518 source file were written in C, and @value{GDBN} was parsing Modula-2, a
15526 might not have the effect you intended. In C, this means to add
15527 @code{b} and @code{c} and place the result in @code{a}. The result
15528 printed would be the value of @code{a}. In Modula-2, this means to compare
15529 @code{a} to the result of @code{b+c}, yielding a @code{BOOLEAN} value.
15531 @node Automatically
15532 @subsection Having @value{GDBN} Infer the Source Language
15534 To have @value{GDBN} set the working language automatically, use
15535 @samp{set language local} or @samp{set language auto}. @value{GDBN}
15536 then infers the working language. That is, when your program stops in a
15537 frame (usually by encountering a breakpoint), @value{GDBN} sets the
15538 working language to the language recorded for the function in that
15539 frame. If the language for a frame is unknown (that is, if the function
15540 or block corresponding to the frame was defined in a source file that
15541 does not have a recognized extension), the current working language is
15542 not changed, and @value{GDBN} issues a warning.
15544 This may not seem necessary for most programs, which are written
15545 entirely in one source language. However, program modules and libraries
15546 written in one source language can be used by a main program written in
15547 a different source language. Using @samp{set language auto} in this
15548 case frees you from having to set the working language manually.
15551 @section Displaying the Language
15553 The following commands help you find out which language is the
15554 working language, and also what language source files were written in.
15557 @item show language
15558 @anchor{show language}
15559 @kindex show language
15560 Display the current working language. This is the
15561 language you can use with commands such as @code{print} to
15562 build and compute expressions that may involve variables in your program.
15565 @kindex info frame@r{, show the source language}
15566 Display the source language for this frame. This language becomes the
15567 working language if you use an identifier from this frame.
15568 @xref{Frame Info, ,Information about a Frame}, to identify the other
15569 information listed here.
15572 @kindex info source@r{, show the source language}
15573 Display the source language of this source file.
15574 @xref{Symbols, ,Examining the Symbol Table}, to identify the other
15575 information listed here.
15578 In unusual circumstances, you may have source files with extensions
15579 not in the standard list. You can then set the extension associated
15580 with a language explicitly:
15583 @item set extension-language @var{ext} @var{language}
15584 @kindex set extension-language
15585 Tell @value{GDBN} that source files with extension @var{ext} are to be
15586 assumed as written in the source language @var{language}.
15588 @item info extensions
15589 @kindex info extensions
15590 List all the filename extensions and the associated languages.
15594 @section Type and Range Checking
15596 Some languages are designed to guard you against making seemingly common
15597 errors through a series of compile- and run-time checks. These include
15598 checking the type of arguments to functions and operators and making
15599 sure mathematical overflows are caught at run time. Checks such as
15600 these help to ensure a program's correctness once it has been compiled
15601 by eliminating type mismatches and providing active checks for range
15602 errors when your program is running.
15604 By default @value{GDBN} checks for these errors according to the
15605 rules of the current source language. Although @value{GDBN} does not check
15606 the statements in your program, it can check expressions entered directly
15607 into @value{GDBN} for evaluation via the @code{print} command, for example.
15610 * Type Checking:: An overview of type checking
15611 * Range Checking:: An overview of range checking
15614 @cindex type checking
15615 @cindex checks, type
15616 @node Type Checking
15617 @subsection An Overview of Type Checking
15619 Some languages, such as C and C@t{++}, are strongly typed, meaning that the
15620 arguments to operators and functions have to be of the correct type,
15621 otherwise an error occurs. These checks prevent type mismatch
15622 errors from ever causing any run-time problems. For example,
15625 int klass::my_method(char *b) @{ return b ? 1 : 2; @}
15627 (@value{GDBP}) print obj.my_method (0)
15630 (@value{GDBP}) print obj.my_method (0x1234)
15631 Cannot resolve method klass::my_method to any overloaded instance
15634 The second example fails because in C@t{++} the integer constant
15635 @samp{0x1234} is not type-compatible with the pointer parameter type.
15637 For the expressions you use in @value{GDBN} commands, you can tell
15638 @value{GDBN} to not enforce strict type checking or
15639 to treat any mismatches as errors and abandon the expression;
15640 When type checking is disabled, @value{GDBN} successfully evaluates
15641 expressions like the second example above.
15643 Even if type checking is off, there may be other reasons
15644 related to type that prevent @value{GDBN} from evaluating an expression.
15645 For instance, @value{GDBN} does not know how to add an @code{int} and
15646 a @code{struct foo}. These particular type errors have nothing to do
15647 with the language in use and usually arise from expressions which make
15648 little sense to evaluate anyway.
15650 @value{GDBN} provides some additional commands for controlling type checking:
15652 @kindex set check type
15653 @kindex show check type
15655 @item set check type on
15656 @itemx set check type off
15657 Set strict type checking on or off. If any type mismatches occur in
15658 evaluating an expression while type checking is on, @value{GDBN} prints a
15659 message and aborts evaluation of the expression.
15661 @item show check type
15662 Show the current setting of type checking and whether @value{GDBN}
15663 is enforcing strict type checking rules.
15666 @cindex range checking
15667 @cindex checks, range
15668 @node Range Checking
15669 @subsection An Overview of Range Checking
15671 In some languages (such as Modula-2), it is an error to exceed the
15672 bounds of a type; this is enforced with run-time checks. Such range
15673 checking is meant to ensure program correctness by making sure
15674 computations do not overflow, or indices on an array element access do
15675 not exceed the bounds of the array.
15677 For expressions you use in @value{GDBN} commands, you can tell
15678 @value{GDBN} to treat range errors in one of three ways: ignore them,
15679 always treat them as errors and abandon the expression, or issue
15680 warnings but evaluate the expression anyway.
15682 A range error can result from numerical overflow, from exceeding an
15683 array index bound, or when you type a constant that is not a member
15684 of any type. Some languages, however, do not treat overflows as an
15685 error. In many implementations of C, mathematical overflow causes the
15686 result to ``wrap around'' to lower values---for example, if @var{m} is
15687 the largest integer value, and @var{s} is the smallest, then
15690 @var{m} + 1 @result{} @var{s}
15693 This, too, is specific to individual languages, and in some cases
15694 specific to individual compilers or machines. @xref{Supported Languages, ,
15695 Supported Languages}, for further details on specific languages.
15697 @value{GDBN} provides some additional commands for controlling the range checker:
15699 @kindex set check range
15700 @kindex show check range
15702 @item set check range auto
15703 Set range checking on or off based on the current working language.
15704 @xref{Supported Languages, ,Supported Languages}, for the default settings for
15707 @item set check range on
15708 @itemx set check range off
15709 Set range checking on or off, overriding the default setting for the
15710 current working language. A warning is issued if the setting does not
15711 match the language default. If a range error occurs and range checking is on,
15712 then a message is printed and evaluation of the expression is aborted.
15714 @item set check range warn
15715 Output messages when the @value{GDBN} range checker detects a range error,
15716 but attempt to evaluate the expression anyway. Evaluating the
15717 expression may still be impossible for other reasons, such as accessing
15718 memory that the process does not own (a typical example from many Unix
15722 Show the current setting of the range checker, and whether or not it is
15723 being set automatically by @value{GDBN}.
15726 @node Supported Languages
15727 @section Supported Languages
15729 @value{GDBN} supports C, C@t{++}, D, Go, Objective-C, Fortran,
15730 OpenCL C, Pascal, Rust, assembly, Modula-2, and Ada.
15731 @c This is false ...
15732 Some @value{GDBN} features may be used in expressions regardless of the
15733 language you use: the @value{GDBN} @code{@@} and @code{::} operators,
15734 and the @samp{@{type@}addr} construct (@pxref{Expressions,
15735 ,Expressions}) can be used with the constructs of any supported
15738 The following sections detail to what degree each source language is
15739 supported by @value{GDBN}. These sections are not meant to be language
15740 tutorials or references, but serve only as a reference guide to what the
15741 @value{GDBN} expression parser accepts, and what input and output
15742 formats should look like for different languages. There are many good
15743 books written on each of these languages; please look to these for a
15744 language reference or tutorial.
15747 * C:: C and C@t{++}
15750 * Objective-C:: Objective-C
15751 * OpenCL C:: OpenCL C
15752 * Fortran:: Fortran
15755 * Modula-2:: Modula-2
15760 @subsection C and C@t{++}
15762 @cindex C and C@t{++}
15763 @cindex expressions in C or C@t{++}
15765 Since C and C@t{++} are so closely related, many features of @value{GDBN} apply
15766 to both languages. Whenever this is the case, we discuss those languages
15770 @cindex @code{g++}, @sc{gnu} C@t{++} compiler
15771 @cindex @sc{gnu} C@t{++}
15772 The C@t{++} debugging facilities are jointly implemented by the C@t{++}
15773 compiler and @value{GDBN}. Therefore, to debug your C@t{++} code
15774 effectively, you must compile your C@t{++} programs with a supported
15775 C@t{++} compiler, such as @sc{gnu} @code{g++}, or the HP ANSI C@t{++}
15776 compiler (@code{aCC}).
15779 * C Operators:: C and C@t{++} operators
15780 * C Constants:: C and C@t{++} constants
15781 * C Plus Plus Expressions:: C@t{++} expressions
15782 * C Defaults:: Default settings for C and C@t{++}
15783 * C Checks:: C and C@t{++} type and range checks
15784 * Debugging C:: @value{GDBN} and C
15785 * Debugging C Plus Plus:: @value{GDBN} features for C@t{++}
15786 * Decimal Floating Point:: Numbers in Decimal Floating Point format
15790 @subsubsection C and C@t{++} Operators
15792 @cindex C and C@t{++} operators
15794 Operators must be defined on values of specific types. For instance,
15795 @code{+} is defined on numbers, but not on structures. Operators are
15796 often defined on groups of types.
15798 For the purposes of C and C@t{++}, the following definitions hold:
15803 @emph{Integral types} include @code{int} with any of its storage-class
15804 specifiers; @code{char}; @code{enum}; and, for C@t{++}, @code{bool}.
15807 @emph{Floating-point types} include @code{float}, @code{double}, and
15808 @code{long double} (if supported by the target platform).
15811 @emph{Pointer types} include all types defined as @code{(@var{type} *)}.
15814 @emph{Scalar types} include all of the above.
15819 The following operators are supported. They are listed here
15820 in order of increasing precedence:
15824 The comma or sequencing operator. Expressions in a comma-separated list
15825 are evaluated from left to right, with the result of the entire
15826 expression being the last expression evaluated.
15829 Assignment. The value of an assignment expression is the value
15830 assigned. Defined on scalar types.
15833 Used in an expression of the form @w{@code{@var{a} @var{op}= @var{b}}},
15834 and translated to @w{@code{@var{a} = @var{a op b}}}.
15835 @w{@code{@var{op}=}} and @code{=} have the same precedence. The operator
15836 @var{op} is any one of the operators @code{|}, @code{^}, @code{&},
15837 @code{<<}, @code{>>}, @code{+}, @code{-}, @code{*}, @code{/}, @code{%}.
15840 The ternary operator. @code{@var{a} ? @var{b} : @var{c}} can be thought
15841 of as: if @var{a} then @var{b} else @var{c}. The argument @var{a}
15842 should be of an integral type.
15845 Logical @sc{or}. Defined on integral types.
15848 Logical @sc{and}. Defined on integral types.
15851 Bitwise @sc{or}. Defined on integral types.
15854 Bitwise exclusive-@sc{or}. Defined on integral types.
15857 Bitwise @sc{and}. Defined on integral types.
15860 Equality and inequality. Defined on scalar types. The value of these
15861 expressions is 0 for false and non-zero for true.
15863 @item <@r{, }>@r{, }<=@r{, }>=
15864 Less than, greater than, less than or equal, greater than or equal.
15865 Defined on scalar types. The value of these expressions is 0 for false
15866 and non-zero for true.
15869 left shift, and right shift. Defined on integral types.
15872 The @value{GDBN} ``artificial array'' operator (@pxref{Expressions, ,Expressions}).
15875 Addition and subtraction. Defined on integral types, floating-point types and
15878 @item *@r{, }/@r{, }%
15879 Multiplication, division, and modulus. Multiplication and division are
15880 defined on integral and floating-point types. Modulus is defined on
15884 Increment and decrement. When appearing before a variable, the
15885 operation is performed before the variable is used in an expression;
15886 when appearing after it, the variable's value is used before the
15887 operation takes place.
15890 Pointer dereferencing. Defined on pointer types. Same precedence as
15894 Address operator. Defined on variables. Same precedence as @code{++}.
15896 For debugging C@t{++}, @value{GDBN} implements a use of @samp{&} beyond what is
15897 allowed in the C@t{++} language itself: you can use @samp{&(&@var{ref})}
15898 to examine the address
15899 where a C@t{++} reference variable (declared with @samp{&@var{ref}}) is
15903 Negative. Defined on integral and floating-point types. Same
15904 precedence as @code{++}.
15907 Logical negation. Defined on integral types. Same precedence as
15911 Bitwise complement operator. Defined on integral types. Same precedence as
15916 Structure member, and pointer-to-structure member. For convenience,
15917 @value{GDBN} regards the two as equivalent, choosing whether to dereference a
15918 pointer based on the stored type information.
15919 Defined on @code{struct} and @code{union} data.
15922 Dereferences of pointers to members.
15925 Array indexing. @code{@var{a}[@var{i}]} is defined as
15926 @code{*(@var{a}+@var{i})}. Same precedence as @code{->}.
15929 Function parameter list. Same precedence as @code{->}.
15932 C@t{++} scope resolution operator. Defined on @code{struct}, @code{union},
15933 and @code{class} types.
15936 Doubled colons also represent the @value{GDBN} scope operator
15937 (@pxref{Expressions, ,Expressions}). Same precedence as @code{::},
15941 If an operator is redefined in the user code, @value{GDBN} usually
15942 attempts to invoke the redefined version instead of using the operator's
15943 predefined meaning.
15946 @subsubsection C and C@t{++} Constants
15948 @cindex C and C@t{++} constants
15950 @value{GDBN} allows you to express the constants of C and C@t{++} in the
15955 Integer constants are a sequence of digits. Octal constants are
15956 specified by a leading @samp{0} (i.e.@: zero), and hexadecimal constants
15957 by a leading @samp{0x} or @samp{0X}. Constants may also end with a letter
15958 @samp{l}, specifying that the constant should be treated as a
15962 Floating point constants are a sequence of digits, followed by a decimal
15963 point, followed by a sequence of digits, and optionally followed by an
15964 exponent. An exponent is of the form:
15965 @samp{@w{e@r{[[}+@r{]|}-@r{]}@var{nnn}}}, where @var{nnn} is another
15966 sequence of digits. The @samp{+} is optional for positive exponents.
15967 A floating-point constant may also end with a letter @samp{f} or
15968 @samp{F}, specifying that the constant should be treated as being of
15969 the @code{float} (as opposed to the default @code{double}) type; or with
15970 a letter @samp{l} or @samp{L}, which specifies a @code{long double}
15974 Enumerated constants consist of enumerated identifiers, or their
15975 integral equivalents.
15978 Character constants are a single character surrounded by single quotes
15979 (@code{'}), or a number---the ordinal value of the corresponding character
15980 (usually its @sc{ascii} value). Within quotes, the single character may
15981 be represented by a letter or by @dfn{escape sequences}, which are of
15982 the form @samp{\@var{nnn}}, where @var{nnn} is the octal representation
15983 of the character's ordinal value; or of the form @samp{\@var{x}}, where
15984 @samp{@var{x}} is a predefined special character---for example,
15985 @samp{\n} for newline.
15987 Wide character constants can be written by prefixing a character
15988 constant with @samp{L}, as in C. For example, @samp{L'x'} is the wide
15989 form of @samp{x}. The target wide character set is used when
15990 computing the value of this constant (@pxref{Character Sets}).
15993 String constants are a sequence of character constants surrounded by
15994 double quotes (@code{"}). Any valid character constant (as described
15995 above) may appear. Double quotes within the string must be preceded by
15996 a backslash, so for instance @samp{"a\"b'c"} is a string of five
15999 Wide string constants can be written by prefixing a string constant
16000 with @samp{L}, as in C. The target wide character set is used when
16001 computing the value of this constant (@pxref{Character Sets}).
16004 Pointer constants are an integral value. You can also write pointers
16005 to constants using the C operator @samp{&}.
16008 Array constants are comma-separated lists surrounded by braces @samp{@{}
16009 and @samp{@}}; for example, @samp{@{1,2,3@}} is a three-element array of
16010 integers, @samp{@{@{1,2@}, @{3,4@}, @{5,6@}@}} is a three-by-two array,
16011 and @samp{@{&"hi", &"there", &"fred"@}} is a three-element array of pointers.
16014 @node C Plus Plus Expressions
16015 @subsubsection C@t{++} Expressions
16017 @cindex expressions in C@t{++}
16018 @value{GDBN} expression handling can interpret most C@t{++} expressions.
16020 @cindex debugging C@t{++} programs
16021 @cindex C@t{++} compilers
16022 @cindex debug formats and C@t{++}
16023 @cindex @value{NGCC} and C@t{++}
16025 @emph{Warning:} @value{GDBN} can only debug C@t{++} code if you use
16026 the proper compiler and the proper debug format. Currently,
16027 @value{GDBN} works best when debugging C@t{++} code that is compiled
16028 with the most recent version of @value{NGCC} possible. The DWARF
16029 debugging format is preferred; @value{NGCC} defaults to this on most
16030 popular platforms. Other compilers and/or debug formats are likely to
16031 work badly or not at all when using @value{GDBN} to debug C@t{++}
16032 code. @xref{Compilation}.
16037 @cindex member functions
16039 Member function calls are allowed; you can use expressions like
16042 count = aml->GetOriginal(x, y)
16045 @vindex this@r{, inside C@t{++} member functions}
16046 @cindex namespace in C@t{++}
16048 While a member function is active (in the selected stack frame), your
16049 expressions have the same namespace available as the member function;
16050 that is, @value{GDBN} allows implicit references to the class instance
16051 pointer @code{this} following the same rules as C@t{++}. @code{using}
16052 declarations in the current scope are also respected by @value{GDBN}.
16054 @cindex call overloaded functions
16055 @cindex overloaded functions, calling
16056 @cindex type conversions in C@t{++}
16058 You can call overloaded functions; @value{GDBN} resolves the function
16059 call to the right definition, with some restrictions. @value{GDBN} does not
16060 perform overload resolution involving user-defined type conversions,
16061 calls to constructors, or instantiations of templates that do not exist
16062 in the program. It also cannot handle ellipsis argument lists or
16065 It does perform integral conversions and promotions, floating-point
16066 promotions, arithmetic conversions, pointer conversions, conversions of
16067 class objects to base classes, and standard conversions such as those of
16068 functions or arrays to pointers; it requires an exact match on the
16069 number of function arguments.
16071 Overload resolution is always performed, unless you have specified
16072 @code{set overload-resolution off}. @xref{Debugging C Plus Plus,
16073 ,@value{GDBN} Features for C@t{++}}.
16075 You must specify @code{set overload-resolution off} in order to use an
16076 explicit function signature to call an overloaded function, as in
16078 p 'foo(char,int)'('x', 13)
16081 The @value{GDBN} command-completion facility can simplify this;
16082 see @ref{Completion, ,Command Completion}.
16084 @cindex reference declarations
16086 @value{GDBN} understands variables declared as C@t{++} lvalue or rvalue
16087 references; you can use them in expressions just as you do in C@t{++}
16088 source---they are automatically dereferenced.
16090 In the parameter list shown when @value{GDBN} displays a frame, the values of
16091 reference variables are not displayed (unlike other variables); this
16092 avoids clutter, since references are often used for large structures.
16093 The @emph{address} of a reference variable is always shown, unless
16094 you have specified @samp{set print address off}.
16097 @value{GDBN} supports the C@t{++} name resolution operator @code{::}---your
16098 expressions can use it just as expressions in your program do. Since
16099 one scope may be defined in another, you can use @code{::} repeatedly if
16100 necessary, for example in an expression like
16101 @samp{@var{scope1}::@var{scope2}::@var{name}}. @value{GDBN} also allows
16102 resolving name scope by reference to source files, in both C and C@t{++}
16103 debugging (@pxref{Variables, ,Program Variables}).
16106 @value{GDBN} performs argument-dependent lookup, following the C@t{++}
16111 @subsubsection C and C@t{++} Defaults
16113 @cindex C and C@t{++} defaults
16115 If you allow @value{GDBN} to set range checking automatically, it
16116 defaults to @code{off} whenever the working language changes to
16117 C or C@t{++}. This happens regardless of whether you or @value{GDBN}
16118 selects the working language.
16120 If you allow @value{GDBN} to set the language automatically, it
16121 recognizes source files whose names end with @file{.c}, @file{.C}, or
16122 @file{.cc}, etc, and when @value{GDBN} enters code compiled from one of
16123 these files, it sets the working language to C or C@t{++}.
16124 @xref{Automatically, ,Having @value{GDBN} Infer the Source Language},
16125 for further details.
16128 @subsubsection C and C@t{++} Type and Range Checks
16130 @cindex C and C@t{++} checks
16132 By default, when @value{GDBN} parses C or C@t{++} expressions, strict type
16133 checking is used. However, if you turn type checking off, @value{GDBN}
16134 will allow certain non-standard conversions, such as promoting integer
16135 constants to pointers.
16137 Range checking, if turned on, is done on mathematical operations. Array
16138 indices are not checked, since they are often used to index a pointer
16139 that is not itself an array.
16142 @subsubsection @value{GDBN} and C
16144 The @code{set print union} and @code{show print union} commands apply to
16145 the @code{union} type. When set to @samp{on}, any @code{union} that is
16146 inside a @code{struct} or @code{class} is also printed. Otherwise, it
16147 appears as @samp{@{...@}}.
16149 The @code{@@} operator aids in the debugging of dynamic arrays, formed
16150 with pointers and a memory allocation function. @xref{Expressions,
16153 @node Debugging C Plus Plus
16154 @subsubsection @value{GDBN} Features for C@t{++}
16156 @cindex commands for C@t{++}
16158 Some @value{GDBN} commands are particularly useful with C@t{++}, and some are
16159 designed specifically for use with C@t{++}. Here is a summary:
16162 @cindex break in overloaded functions
16163 @item @r{breakpoint menus}
16164 When you want a breakpoint in a function whose name is overloaded,
16165 @value{GDBN} has the capability to display a menu of possible breakpoint
16166 locations to help you specify which function definition you want.
16167 @xref{Ambiguous Expressions,,Ambiguous Expressions}.
16169 @cindex overloading in C@t{++}
16170 @item rbreak @var{regex}
16171 Setting breakpoints using regular expressions is helpful for setting
16172 breakpoints on overloaded functions that are not members of any special
16174 @xref{Set Breaks, ,Setting Breakpoints}.
16176 @cindex C@t{++} exception handling
16178 @itemx catch rethrow
16180 Debug C@t{++} exception handling using these commands. @xref{Set
16181 Catchpoints, , Setting Catchpoints}.
16183 @cindex inheritance
16184 @item ptype @var{typename}
16185 Print inheritance relationships as well as other information for type
16187 @xref{Symbols, ,Examining the Symbol Table}.
16189 @item info vtbl @var{expression}.
16190 The @code{info vtbl} command can be used to display the virtual
16191 method tables of the object computed by @var{expression}. This shows
16192 one entry per virtual table; there may be multiple virtual tables when
16193 multiple inheritance is in use.
16195 @cindex C@t{++} demangling
16196 @item demangle @var{name}
16197 Demangle @var{name}.
16198 @xref{Symbols}, for a more complete description of the @code{demangle} command.
16200 @cindex C@t{++} symbol display
16201 @item set print demangle
16202 @itemx show print demangle
16203 @itemx set print asm-demangle
16204 @itemx show print asm-demangle
16205 Control whether C@t{++} symbols display in their source form, both when
16206 displaying code as C@t{++} source and when displaying disassemblies.
16207 @xref{Print Settings, ,Print Settings}.
16209 @item set print object
16210 @itemx show print object
16211 Choose whether to print derived (actual) or declared types of objects.
16212 @xref{Print Settings, ,Print Settings}.
16214 @item set print vtbl
16215 @itemx show print vtbl
16216 Control the format for printing virtual function tables.
16217 @xref{Print Settings, ,Print Settings}.
16218 (The @code{vtbl} commands do not work on programs compiled with the HP
16219 ANSI C@t{++} compiler (@code{aCC}).)
16221 @kindex set overload-resolution
16222 @cindex overloaded functions, overload resolution
16223 @item set overload-resolution on
16224 Enable overload resolution for C@t{++} expression evaluation. The default
16225 is on. For overloaded functions, @value{GDBN} evaluates the arguments
16226 and searches for a function whose signature matches the argument types,
16227 using the standard C@t{++} conversion rules (see @ref{C Plus Plus
16228 Expressions, ,C@t{++} Expressions}, for details).
16229 If it cannot find a match, it emits a message.
16231 @item set overload-resolution off
16232 Disable overload resolution for C@t{++} expression evaluation. For
16233 overloaded functions that are not class member functions, @value{GDBN}
16234 chooses the first function of the specified name that it finds in the
16235 symbol table, whether or not its arguments are of the correct type. For
16236 overloaded functions that are class member functions, @value{GDBN}
16237 searches for a function whose signature @emph{exactly} matches the
16240 @kindex show overload-resolution
16241 @item show overload-resolution
16242 Show the current setting of overload resolution.
16244 @item @r{Overloaded symbol names}
16245 You can specify a particular definition of an overloaded symbol, using
16246 the same notation that is used to declare such symbols in C@t{++}: type
16247 @code{@var{symbol}(@var{types})} rather than just @var{symbol}. You can
16248 also use the @value{GDBN} command-line word completion facilities to list the
16249 available choices, or to finish the type list for you.
16250 @xref{Completion,, Command Completion}, for details on how to do this.
16252 @item @r{Breakpoints in functions with ABI tags}
16254 The GNU C@t{++} compiler introduced the notion of ABI ``tags'', which
16255 correspond to changes in the ABI of a type, function, or variable that
16256 would not otherwise be reflected in a mangled name. See
16257 @url{https://developers.redhat.com/blog/2015/02/05/gcc5-and-the-c11-abi/}
16260 The ABI tags are visible in C@t{++} demangled names. For example, a
16261 function that returns a std::string:
16264 std::string function(int);
16268 when compiled for the C++11 ABI is marked with the @code{cxx11} ABI
16269 tag, and @value{GDBN} displays the symbol like this:
16272 function[abi:cxx11](int)
16275 You can set a breakpoint on such functions simply as if they had no
16279 (gdb) b function(int)
16280 Breakpoint 2 at 0x40060d: file main.cc, line 10.
16281 (gdb) info breakpoints
16282 Num Type Disp Enb Address What
16283 1 breakpoint keep y 0x0040060d in function[abi:cxx11](int)
16287 On the rare occasion you need to disambiguate between different ABI
16288 tags, you can do so by simply including the ABI tag in the function
16292 (@value{GDBP}) b ambiguous[abi:other_tag](int)
16296 @node Decimal Floating Point
16297 @subsubsection Decimal Floating Point format
16298 @cindex decimal floating point format
16300 @value{GDBN} can examine, set and perform computations with numbers in
16301 decimal floating point format, which in the C language correspond to the
16302 @code{_Decimal32}, @code{_Decimal64} and @code{_Decimal128} types as
16303 specified by the extension to support decimal floating-point arithmetic.
16305 There are two encodings in use, depending on the architecture: BID (Binary
16306 Integer Decimal) for x86 and x86-64, and DPD (Densely Packed Decimal) for
16307 PowerPC and S/390. @value{GDBN} will use the appropriate encoding for the
16310 Because of a limitation in @file{libdecnumber}, the library used by @value{GDBN}
16311 to manipulate decimal floating point numbers, it is not possible to convert
16312 (using a cast, for example) integers wider than 32-bit to decimal float.
16314 In addition, in order to imitate @value{GDBN}'s behaviour with binary floating
16315 point computations, error checking in decimal float operations ignores
16316 underflow, overflow and divide by zero exceptions.
16318 In the PowerPC architecture, @value{GDBN} provides a set of pseudo-registers
16319 to inspect @code{_Decimal128} values stored in floating point registers.
16320 See @ref{PowerPC,,PowerPC} for more details.
16326 @value{GDBN} can be used to debug programs written in D and compiled with
16327 GDC, LDC or DMD compilers. Currently @value{GDBN} supports only one D
16328 specific feature --- dynamic arrays.
16333 @cindex Go (programming language)
16334 @value{GDBN} can be used to debug programs written in Go and compiled with
16335 @file{gccgo} or @file{6g} compilers.
16337 Here is a summary of the Go-specific features and restrictions:
16340 @cindex current Go package
16341 @item The current Go package
16342 The name of the current package does not need to be specified when
16343 specifying global variables and functions.
16345 For example, given the program:
16349 var myglob = "Shall we?"
16355 When stopped inside @code{main} either of these work:
16359 (gdb) p main.myglob
16362 @cindex builtin Go types
16363 @item Builtin Go types
16364 The @code{string} type is recognized by @value{GDBN} and is printed
16367 @cindex builtin Go functions
16368 @item Builtin Go functions
16369 The @value{GDBN} expression parser recognizes the @code{unsafe.Sizeof}
16370 function and handles it internally.
16372 @cindex restrictions on Go expressions
16373 @item Restrictions on Go expressions
16374 All Go operators are supported except @code{&^}.
16375 The Go @code{_} ``blank identifier'' is not supported.
16376 Automatic dereferencing of pointers is not supported.
16380 @subsection Objective-C
16382 @cindex Objective-C
16383 This section provides information about some commands and command
16384 options that are useful for debugging Objective-C code. See also
16385 @ref{Symbols, info classes}, and @ref{Symbols, info selectors}, for a
16386 few more commands specific to Objective-C support.
16389 * Method Names in Commands::
16390 * The Print Command with Objective-C::
16393 @node Method Names in Commands
16394 @subsubsection Method Names in Commands
16396 The following commands have been extended to accept Objective-C method
16397 names as line specifications:
16399 @kindex clear@r{, and Objective-C}
16400 @kindex break@r{, and Objective-C}
16401 @kindex info line@r{, and Objective-C}
16402 @kindex jump@r{, and Objective-C}
16403 @kindex list@r{, and Objective-C}
16407 @item @code{info line}
16412 A fully qualified Objective-C method name is specified as
16415 -[@var{Class} @var{methodName}]
16418 where the minus sign is used to indicate an instance method and a
16419 plus sign (not shown) is used to indicate a class method. The class
16420 name @var{Class} and method name @var{methodName} are enclosed in
16421 brackets, similar to the way messages are specified in Objective-C
16422 source code. For example, to set a breakpoint at the @code{create}
16423 instance method of class @code{Fruit} in the program currently being
16427 break -[Fruit create]
16430 To list ten program lines around the @code{initialize} class method,
16434 list +[NSText initialize]
16437 In the current version of @value{GDBN}, the plus or minus sign is
16438 required. In future versions of @value{GDBN}, the plus or minus
16439 sign will be optional, but you can use it to narrow the search. It
16440 is also possible to specify just a method name:
16446 You must specify the complete method name, including any colons. If
16447 your program's source files contain more than one @code{create} method,
16448 you'll be presented with a numbered list of classes that implement that
16449 method. Indicate your choice by number, or type @samp{0} to exit if
16452 As another example, to clear a breakpoint established at the
16453 @code{makeKeyAndOrderFront:} method of the @code{NSWindow} class, enter:
16456 clear -[NSWindow makeKeyAndOrderFront:]
16459 @node The Print Command with Objective-C
16460 @subsubsection The Print Command With Objective-C
16461 @cindex Objective-C, print objects
16462 @kindex print-object
16463 @kindex po @r{(@code{print-object})}
16465 The print command has also been extended to accept methods. For example:
16468 print -[@var{object} hash]
16471 @cindex print an Objective-C object description
16472 @cindex @code{_NSPrintForDebugger}, and printing Objective-C objects
16474 will tell @value{GDBN} to send the @code{hash} message to @var{object}
16475 and print the result. Also, an additional command has been added,
16476 @code{print-object} or @code{po} for short, which is meant to print
16477 the description of an object. However, this command may only work
16478 with certain Objective-C libraries that have a particular hook
16479 function, @code{_NSPrintForDebugger}, defined.
16482 @subsection OpenCL C
16485 This section provides information about @value{GDBN}s OpenCL C support.
16488 * OpenCL C Datatypes::
16489 * OpenCL C Expressions::
16490 * OpenCL C Operators::
16493 @node OpenCL C Datatypes
16494 @subsubsection OpenCL C Datatypes
16496 @cindex OpenCL C Datatypes
16497 @value{GDBN} supports the builtin scalar and vector datatypes specified
16498 by OpenCL 1.1. In addition the half- and double-precision floating point
16499 data types of the @code{cl_khr_fp16} and @code{cl_khr_fp64} OpenCL
16500 extensions are also known to @value{GDBN}.
16502 @node OpenCL C Expressions
16503 @subsubsection OpenCL C Expressions
16505 @cindex OpenCL C Expressions
16506 @value{GDBN} supports accesses to vector components including the access as
16507 lvalue where possible. Since OpenCL C is based on C99 most C expressions
16508 supported by @value{GDBN} can be used as well.
16510 @node OpenCL C Operators
16511 @subsubsection OpenCL C Operators
16513 @cindex OpenCL C Operators
16514 @value{GDBN} supports the operators specified by OpenCL 1.1 for scalar and
16518 @subsection Fortran
16519 @cindex Fortran-specific support in @value{GDBN}
16521 @value{GDBN} can be used to debug programs written in Fortran, but it
16522 currently supports only the features of Fortran 77 language.
16524 @cindex trailing underscore, in Fortran symbols
16525 Some Fortran compilers (@sc{gnu} Fortran 77 and Fortran 95 compilers
16526 among them) append an underscore to the names of variables and
16527 functions. When you debug programs compiled by those compilers, you
16528 will need to refer to variables and functions with a trailing
16532 * Fortran Operators:: Fortran operators and expressions
16533 * Fortran Defaults:: Default settings for Fortran
16534 * Special Fortran Commands:: Special @value{GDBN} commands for Fortran
16537 @node Fortran Operators
16538 @subsubsection Fortran Operators and Expressions
16540 @cindex Fortran operators and expressions
16542 Operators must be defined on values of specific types. For instance,
16543 @code{+} is defined on numbers, but not on characters or other non-
16544 arithmetic types. Operators are often defined on groups of types.
16548 The exponentiation operator. It raises the first operand to the power
16552 The range operator. Normally used in the form of array(low:high) to
16553 represent a section of array.
16556 The access component operator. Normally used to access elements in derived
16557 types. Also suitable for unions. As unions aren't part of regular Fortran,
16558 this can only happen when accessing a register that uses a gdbarch-defined
16562 @node Fortran Defaults
16563 @subsubsection Fortran Defaults
16565 @cindex Fortran Defaults
16567 Fortran symbols are usually case-insensitive, so @value{GDBN} by
16568 default uses case-insensitive matches for Fortran symbols. You can
16569 change that with the @samp{set case-insensitive} command, see
16570 @ref{Symbols}, for the details.
16572 @node Special Fortran Commands
16573 @subsubsection Special Fortran Commands
16575 @cindex Special Fortran commands
16577 @value{GDBN} has some commands to support Fortran-specific features,
16578 such as displaying common blocks.
16581 @cindex @code{COMMON} blocks, Fortran
16582 @kindex info common
16583 @item info common @r{[}@var{common-name}@r{]}
16584 This command prints the values contained in the Fortran @code{COMMON}
16585 block whose name is @var{common-name}. With no argument, the names of
16586 all @code{COMMON} blocks visible at the current program location are
16593 @cindex Pascal support in @value{GDBN}, limitations
16594 Debugging Pascal programs which use sets, subranges, file variables, or
16595 nested functions does not currently work. @value{GDBN} does not support
16596 entering expressions, printing values, or similar features using Pascal
16599 The Pascal-specific command @code{set print pascal_static-members}
16600 controls whether static members of Pascal objects are displayed.
16601 @xref{Print Settings, pascal_static-members}.
16606 @value{GDBN} supports the @url{https://www.rust-lang.org/, Rust
16607 Programming Language}. Type- and value-printing, and expression
16608 parsing, are reasonably complete. However, there are a few
16609 peculiarities and holes to be aware of.
16613 Linespecs (@pxref{Specify Location}) are never relative to the current
16614 crate. Instead, they act as if there were a global namespace of
16615 crates, somewhat similar to the way @code{extern crate} behaves.
16617 That is, if @value{GDBN} is stopped at a breakpoint in a function in
16618 crate @samp{A}, module @samp{B}, then @code{break B::f} will attempt
16619 to set a breakpoint in a function named @samp{f} in a crate named
16622 As a consequence of this approach, linespecs also cannot refer to
16623 items using @samp{self::} or @samp{super::}.
16626 Because @value{GDBN} implements Rust name-lookup semantics in
16627 expressions, it will sometimes prepend the current crate to a name.
16628 For example, if @value{GDBN} is stopped at a breakpoint in the crate
16629 @samp{K}, then @code{print ::x::y} will try to find the symbol
16632 However, since it is useful to be able to refer to other crates when
16633 debugging, @value{GDBN} provides the @code{extern} extension to
16634 circumvent this. To use the extension, just put @code{extern} before
16635 a path expression to refer to the otherwise unavailable ``global''
16638 In the above example, if you wanted to refer to the symbol @samp{y} in
16639 the crate @samp{x}, you would use @code{print extern x::y}.
16642 The Rust expression evaluator does not support ``statement-like''
16643 expressions such as @code{if} or @code{match}, or lambda expressions.
16646 Tuple expressions are not implemented.
16649 The Rust expression evaluator does not currently implement the
16650 @code{Drop} trait. Objects that may be created by the evaluator will
16651 never be destroyed.
16654 @value{GDBN} does not implement type inference for generics. In order
16655 to call generic functions or otherwise refer to generic items, you
16656 will have to specify the type parameters manually.
16659 @value{GDBN} currently uses the C@t{++} demangler for Rust. In most
16660 cases this does not cause any problems. However, in an expression
16661 context, completing a generic function name will give syntactically
16662 invalid results. This happens because Rust requires the @samp{::}
16663 operator between the function name and its generic arguments. For
16664 example, @value{GDBN} might provide a completion like
16665 @code{crate::f<u32>}, where the parser would require
16666 @code{crate::f::<u32>}.
16669 As of this writing, the Rust compiler (version 1.8) has a few holes in
16670 the debugging information it generates. These holes prevent certain
16671 features from being implemented by @value{GDBN}:
16675 Method calls cannot be made via traits.
16678 Operator overloading is not implemented.
16681 When debugging in a monomorphized function, you cannot use the generic
16685 The type @code{Self} is not available.
16688 @code{use} statements are not available, so some names may not be
16689 available in the crate.
16694 @subsection Modula-2
16696 @cindex Modula-2, @value{GDBN} support
16698 The extensions made to @value{GDBN} to support Modula-2 only support
16699 output from the @sc{gnu} Modula-2 compiler (which is currently being
16700 developed). Other Modula-2 compilers are not currently supported, and
16701 attempting to debug executables produced by them is most likely
16702 to give an error as @value{GDBN} reads in the executable's symbol
16705 @cindex expressions in Modula-2
16707 * M2 Operators:: Built-in operators
16708 * Built-In Func/Proc:: Built-in functions and procedures
16709 * M2 Constants:: Modula-2 constants
16710 * M2 Types:: Modula-2 types
16711 * M2 Defaults:: Default settings for Modula-2
16712 * Deviations:: Deviations from standard Modula-2
16713 * M2 Checks:: Modula-2 type and range checks
16714 * M2 Scope:: The scope operators @code{::} and @code{.}
16715 * GDB/M2:: @value{GDBN} and Modula-2
16719 @subsubsection Operators
16720 @cindex Modula-2 operators
16722 Operators must be defined on values of specific types. For instance,
16723 @code{+} is defined on numbers, but not on structures. Operators are
16724 often defined on groups of types. For the purposes of Modula-2, the
16725 following definitions hold:
16730 @emph{Integral types} consist of @code{INTEGER}, @code{CARDINAL}, and
16734 @emph{Character types} consist of @code{CHAR} and its subranges.
16737 @emph{Floating-point types} consist of @code{REAL}.
16740 @emph{Pointer types} consist of anything declared as @code{POINTER TO
16744 @emph{Scalar types} consist of all of the above.
16747 @emph{Set types} consist of @code{SET} and @code{BITSET} types.
16750 @emph{Boolean types} consist of @code{BOOLEAN}.
16754 The following operators are supported, and appear in order of
16755 increasing precedence:
16759 Function argument or array index separator.
16762 Assignment. The value of @var{var} @code{:=} @var{value} is
16766 Less than, greater than on integral, floating-point, or enumerated
16770 Less than or equal to, greater than or equal to
16771 on integral, floating-point and enumerated types, or set inclusion on
16772 set types. Same precedence as @code{<}.
16774 @item =@r{, }<>@r{, }#
16775 Equality and two ways of expressing inequality, valid on scalar types.
16776 Same precedence as @code{<}. In @value{GDBN} scripts, only @code{<>} is
16777 available for inequality, since @code{#} conflicts with the script
16781 Set membership. Defined on set types and the types of their members.
16782 Same precedence as @code{<}.
16785 Boolean disjunction. Defined on boolean types.
16788 Boolean conjunction. Defined on boolean types.
16791 The @value{GDBN} ``artificial array'' operator (@pxref{Expressions, ,Expressions}).
16794 Addition and subtraction on integral and floating-point types, or union
16795 and difference on set types.
16798 Multiplication on integral and floating-point types, or set intersection
16802 Division on floating-point types, or symmetric set difference on set
16803 types. Same precedence as @code{*}.
16806 Integer division and remainder. Defined on integral types. Same
16807 precedence as @code{*}.
16810 Negative. Defined on @code{INTEGER} and @code{REAL} data.
16813 Pointer dereferencing. Defined on pointer types.
16816 Boolean negation. Defined on boolean types. Same precedence as
16820 @code{RECORD} field selector. Defined on @code{RECORD} data. Same
16821 precedence as @code{^}.
16824 Array indexing. Defined on @code{ARRAY} data. Same precedence as @code{^}.
16827 Procedure argument list. Defined on @code{PROCEDURE} objects. Same precedence
16831 @value{GDBN} and Modula-2 scope operators.
16835 @emph{Warning:} Set expressions and their operations are not yet supported, so @value{GDBN}
16836 treats the use of the operator @code{IN}, or the use of operators
16837 @code{+}, @code{-}, @code{*}, @code{/}, @code{=}, , @code{<>}, @code{#},
16838 @code{<=}, and @code{>=} on sets as an error.
16842 @node Built-In Func/Proc
16843 @subsubsection Built-in Functions and Procedures
16844 @cindex Modula-2 built-ins
16846 Modula-2 also makes available several built-in procedures and functions.
16847 In describing these, the following metavariables are used:
16852 represents an @code{ARRAY} variable.
16855 represents a @code{CHAR} constant or variable.
16858 represents a variable or constant of integral type.
16861 represents an identifier that belongs to a set. Generally used in the
16862 same function with the metavariable @var{s}. The type of @var{s} should
16863 be @code{SET OF @var{mtype}} (where @var{mtype} is the type of @var{m}).
16866 represents a variable or constant of integral or floating-point type.
16869 represents a variable or constant of floating-point type.
16875 represents a variable.
16878 represents a variable or constant of one of many types. See the
16879 explanation of the function for details.
16882 All Modula-2 built-in procedures also return a result, described below.
16886 Returns the absolute value of @var{n}.
16889 If @var{c} is a lower case letter, it returns its upper case
16890 equivalent, otherwise it returns its argument.
16893 Returns the character whose ordinal value is @var{i}.
16896 Decrements the value in the variable @var{v} by one. Returns the new value.
16898 @item DEC(@var{v},@var{i})
16899 Decrements the value in the variable @var{v} by @var{i}. Returns the
16902 @item EXCL(@var{m},@var{s})
16903 Removes the element @var{m} from the set @var{s}. Returns the new
16906 @item FLOAT(@var{i})
16907 Returns the floating point equivalent of the integer @var{i}.
16909 @item HIGH(@var{a})
16910 Returns the index of the last member of @var{a}.
16913 Increments the value in the variable @var{v} by one. Returns the new value.
16915 @item INC(@var{v},@var{i})
16916 Increments the value in the variable @var{v} by @var{i}. Returns the
16919 @item INCL(@var{m},@var{s})
16920 Adds the element @var{m} to the set @var{s} if it is not already
16921 there. Returns the new set.
16924 Returns the maximum value of the type @var{t}.
16927 Returns the minimum value of the type @var{t}.
16930 Returns boolean TRUE if @var{i} is an odd number.
16933 Returns the ordinal value of its argument. For example, the ordinal
16934 value of a character is its @sc{ascii} value (on machines supporting
16935 the @sc{ascii} character set). The argument @var{x} must be of an
16936 ordered type, which include integral, character and enumerated types.
16938 @item SIZE(@var{x})
16939 Returns the size of its argument. The argument @var{x} can be a
16940 variable or a type.
16942 @item TRUNC(@var{r})
16943 Returns the integral part of @var{r}.
16945 @item TSIZE(@var{x})
16946 Returns the size of its argument. The argument @var{x} can be a
16947 variable or a type.
16949 @item VAL(@var{t},@var{i})
16950 Returns the member of the type @var{t} whose ordinal value is @var{i}.
16954 @emph{Warning:} Sets and their operations are not yet supported, so
16955 @value{GDBN} treats the use of procedures @code{INCL} and @code{EXCL} as
16959 @cindex Modula-2 constants
16961 @subsubsection Constants
16963 @value{GDBN} allows you to express the constants of Modula-2 in the following
16969 Integer constants are simply a sequence of digits. When used in an
16970 expression, a constant is interpreted to be type-compatible with the
16971 rest of the expression. Hexadecimal integers are specified by a
16972 trailing @samp{H}, and octal integers by a trailing @samp{B}.
16975 Floating point constants appear as a sequence of digits, followed by a
16976 decimal point and another sequence of digits. An optional exponent can
16977 then be specified, in the form @samp{E@r{[}+@r{|}-@r{]}@var{nnn}}, where
16978 @samp{@r{[}+@r{|}-@r{]}@var{nnn}} is the desired exponent. All of the
16979 digits of the floating point constant must be valid decimal (base 10)
16983 Character constants consist of a single character enclosed by a pair of
16984 like quotes, either single (@code{'}) or double (@code{"}). They may
16985 also be expressed by their ordinal value (their @sc{ascii} value, usually)
16986 followed by a @samp{C}.
16989 String constants consist of a sequence of characters enclosed by a
16990 pair of like quotes, either single (@code{'}) or double (@code{"}).
16991 Escape sequences in the style of C are also allowed. @xref{C
16992 Constants, ,C and C@t{++} Constants}, for a brief explanation of escape
16996 Enumerated constants consist of an enumerated identifier.
16999 Boolean constants consist of the identifiers @code{TRUE} and
17003 Pointer constants consist of integral values only.
17006 Set constants are not yet supported.
17010 @subsubsection Modula-2 Types
17011 @cindex Modula-2 types
17013 Currently @value{GDBN} can print the following data types in Modula-2
17014 syntax: array types, record types, set types, pointer types, procedure
17015 types, enumerated types, subrange types and base types. You can also
17016 print the contents of variables declared using these type.
17017 This section gives a number of simple source code examples together with
17018 sample @value{GDBN} sessions.
17020 The first example contains the following section of code:
17029 and you can request @value{GDBN} to interrogate the type and value of
17030 @code{r} and @code{s}.
17033 (@value{GDBP}) print s
17035 (@value{GDBP}) ptype s
17037 (@value{GDBP}) print r
17039 (@value{GDBP}) ptype r
17044 Likewise if your source code declares @code{s} as:
17048 s: SET ['A'..'Z'] ;
17052 then you may query the type of @code{s} by:
17055 (@value{GDBP}) ptype s
17056 type = SET ['A'..'Z']
17060 Note that at present you cannot interactively manipulate set
17061 expressions using the debugger.
17063 The following example shows how you might declare an array in Modula-2
17064 and how you can interact with @value{GDBN} to print its type and contents:
17068 s: ARRAY [-10..10] OF CHAR ;
17072 (@value{GDBP}) ptype s
17073 ARRAY [-10..10] OF CHAR
17076 Note that the array handling is not yet complete and although the type
17077 is printed correctly, expression handling still assumes that all
17078 arrays have a lower bound of zero and not @code{-10} as in the example
17081 Here are some more type related Modula-2 examples:
17085 colour = (blue, red, yellow, green) ;
17086 t = [blue..yellow] ;
17094 The @value{GDBN} interaction shows how you can query the data type
17095 and value of a variable.
17098 (@value{GDBP}) print s
17100 (@value{GDBP}) ptype t
17101 type = [blue..yellow]
17105 In this example a Modula-2 array is declared and its contents
17106 displayed. Observe that the contents are written in the same way as
17107 their @code{C} counterparts.
17111 s: ARRAY [1..5] OF CARDINAL ;
17117 (@value{GDBP}) print s
17118 $1 = @{1, 0, 0, 0, 0@}
17119 (@value{GDBP}) ptype s
17120 type = ARRAY [1..5] OF CARDINAL
17123 The Modula-2 language interface to @value{GDBN} also understands
17124 pointer types as shown in this example:
17128 s: POINTER TO ARRAY [1..5] OF CARDINAL ;
17135 and you can request that @value{GDBN} describes the type of @code{s}.
17138 (@value{GDBP}) ptype s
17139 type = POINTER TO ARRAY [1..5] OF CARDINAL
17142 @value{GDBN} handles compound types as we can see in this example.
17143 Here we combine array types, record types, pointer types and subrange
17154 myarray = ARRAY myrange OF CARDINAL ;
17155 myrange = [-2..2] ;
17157 s: POINTER TO ARRAY myrange OF foo ;
17161 and you can ask @value{GDBN} to describe the type of @code{s} as shown
17165 (@value{GDBP}) ptype s
17166 type = POINTER TO ARRAY [-2..2] OF foo = RECORD
17169 f3 : ARRAY [-2..2] OF CARDINAL;
17174 @subsubsection Modula-2 Defaults
17175 @cindex Modula-2 defaults
17177 If type and range checking are set automatically by @value{GDBN}, they
17178 both default to @code{on} whenever the working language changes to
17179 Modula-2. This happens regardless of whether you or @value{GDBN}
17180 selected the working language.
17182 If you allow @value{GDBN} to set the language automatically, then entering
17183 code compiled from a file whose name ends with @file{.mod} sets the
17184 working language to Modula-2. @xref{Automatically, ,Having @value{GDBN}
17185 Infer the Source Language}, for further details.
17188 @subsubsection Deviations from Standard Modula-2
17189 @cindex Modula-2, deviations from
17191 A few changes have been made to make Modula-2 programs easier to debug.
17192 This is done primarily via loosening its type strictness:
17196 Unlike in standard Modula-2, pointer constants can be formed by
17197 integers. This allows you to modify pointer variables during
17198 debugging. (In standard Modula-2, the actual address contained in a
17199 pointer variable is hidden from you; it can only be modified
17200 through direct assignment to another pointer variable or expression that
17201 returned a pointer.)
17204 C escape sequences can be used in strings and characters to represent
17205 non-printable characters. @value{GDBN} prints out strings with these
17206 escape sequences embedded. Single non-printable characters are
17207 printed using the @samp{CHR(@var{nnn})} format.
17210 The assignment operator (@code{:=}) returns the value of its right-hand
17214 All built-in procedures both modify @emph{and} return their argument.
17218 @subsubsection Modula-2 Type and Range Checks
17219 @cindex Modula-2 checks
17222 @emph{Warning:} in this release, @value{GDBN} does not yet perform type or
17225 @c FIXME remove warning when type/range checks added
17227 @value{GDBN} considers two Modula-2 variables type equivalent if:
17231 They are of types that have been declared equivalent via a @code{TYPE
17232 @var{t1} = @var{t2}} statement
17235 They have been declared on the same line. (Note: This is true of the
17236 @sc{gnu} Modula-2 compiler, but it may not be true of other compilers.)
17239 As long as type checking is enabled, any attempt to combine variables
17240 whose types are not equivalent is an error.
17242 Range checking is done on all mathematical operations, assignment, array
17243 index bounds, and all built-in functions and procedures.
17246 @subsubsection The Scope Operators @code{::} and @code{.}
17248 @cindex @code{.}, Modula-2 scope operator
17249 @cindex colon, doubled as scope operator
17251 @vindex colon-colon@r{, in Modula-2}
17252 @c Info cannot handle :: but TeX can.
17255 @vindex ::@r{, in Modula-2}
17258 There are a few subtle differences between the Modula-2 scope operator
17259 (@code{.}) and the @value{GDBN} scope operator (@code{::}). The two have
17264 @var{module} . @var{id}
17265 @var{scope} :: @var{id}
17269 where @var{scope} is the name of a module or a procedure,
17270 @var{module} the name of a module, and @var{id} is any declared
17271 identifier within your program, except another module.
17273 Using the @code{::} operator makes @value{GDBN} search the scope
17274 specified by @var{scope} for the identifier @var{id}. If it is not
17275 found in the specified scope, then @value{GDBN} searches all scopes
17276 enclosing the one specified by @var{scope}.
17278 Using the @code{.} operator makes @value{GDBN} search the current scope for
17279 the identifier specified by @var{id} that was imported from the
17280 definition module specified by @var{module}. With this operator, it is
17281 an error if the identifier @var{id} was not imported from definition
17282 module @var{module}, or if @var{id} is not an identifier in
17286 @subsubsection @value{GDBN} and Modula-2
17288 Some @value{GDBN} commands have little use when debugging Modula-2 programs.
17289 Five subcommands of @code{set print} and @code{show print} apply
17290 specifically to C and C@t{++}: @samp{vtbl}, @samp{demangle},
17291 @samp{asm-demangle}, @samp{object}, and @samp{union}. The first four
17292 apply to C@t{++}, and the last to the C @code{union} type, which has no direct
17293 analogue in Modula-2.
17295 The @code{@@} operator (@pxref{Expressions, ,Expressions}), while available
17296 with any language, is not useful with Modula-2. Its
17297 intent is to aid the debugging of @dfn{dynamic arrays}, which cannot be
17298 created in Modula-2 as they can in C or C@t{++}. However, because an
17299 address can be specified by an integral constant, the construct
17300 @samp{@{@var{type}@}@var{adrexp}} is still useful.
17302 @cindex @code{#} in Modula-2
17303 In @value{GDBN} scripts, the Modula-2 inequality operator @code{#} is
17304 interpreted as the beginning of a comment. Use @code{<>} instead.
17310 The extensions made to @value{GDBN} for Ada only support
17311 output from the @sc{gnu} Ada (GNAT) compiler.
17312 Other Ada compilers are not currently supported, and
17313 attempting to debug executables produced by them is most likely
17317 @cindex expressions in Ada
17319 * Ada Mode Intro:: General remarks on the Ada syntax
17320 and semantics supported by Ada mode
17322 * Omissions from Ada:: Restrictions on the Ada expression syntax.
17323 * Additions to Ada:: Extensions of the Ada expression syntax.
17324 * Overloading support for Ada:: Support for expressions involving overloaded
17326 * Stopping Before Main Program:: Debugging the program during elaboration.
17327 * Ada Exceptions:: Ada Exceptions
17328 * Ada Tasks:: Listing and setting breakpoints in tasks.
17329 * Ada Tasks and Core Files:: Tasking Support when Debugging Core Files
17330 * Ravenscar Profile:: Tasking Support when using the Ravenscar
17332 * Ada Settings:: New settable GDB parameters for Ada.
17333 * Ada Glitches:: Known peculiarities of Ada mode.
17336 @node Ada Mode Intro
17337 @subsubsection Introduction
17338 @cindex Ada mode, general
17340 The Ada mode of @value{GDBN} supports a fairly large subset of Ada expression
17341 syntax, with some extensions.
17342 The philosophy behind the design of this subset is
17346 That @value{GDBN} should provide basic literals and access to operations for
17347 arithmetic, dereferencing, field selection, indexing, and subprogram calls,
17348 leaving more sophisticated computations to subprograms written into the
17349 program (which therefore may be called from @value{GDBN}).
17352 That type safety and strict adherence to Ada language restrictions
17353 are not particularly important to the @value{GDBN} user.
17356 That brevity is important to the @value{GDBN} user.
17359 Thus, for brevity, the debugger acts as if all names declared in
17360 user-written packages are directly visible, even if they are not visible
17361 according to Ada rules, thus making it unnecessary to fully qualify most
17362 names with their packages, regardless of context. Where this causes
17363 ambiguity, @value{GDBN} asks the user's intent.
17365 The debugger will start in Ada mode if it detects an Ada main program.
17366 As for other languages, it will enter Ada mode when stopped in a program that
17367 was translated from an Ada source file.
17369 While in Ada mode, you may use `@t{--}' for comments. This is useful
17370 mostly for documenting command files. The standard @value{GDBN} comment
17371 (@samp{#}) still works at the beginning of a line in Ada mode, but not in the
17372 middle (to allow based literals).
17374 @node Omissions from Ada
17375 @subsubsection Omissions from Ada
17376 @cindex Ada, omissions from
17378 Here are the notable omissions from the subset:
17382 Only a subset of the attributes are supported:
17386 @t{'First}, @t{'Last}, and @t{'Length}
17387 on array objects (not on types and subtypes).
17390 @t{'Min} and @t{'Max}.
17393 @t{'Pos} and @t{'Val}.
17399 @t{'Range} on array objects (not subtypes), but only as the right
17400 operand of the membership (@code{in}) operator.
17403 @t{'Access}, @t{'Unchecked_Access}, and
17404 @t{'Unrestricted_Access} (a GNAT extension).
17412 @code{Characters.Latin_1} are not available and
17413 concatenation is not implemented. Thus, escape characters in strings are
17414 not currently available.
17417 Equality tests (@samp{=} and @samp{/=}) on arrays test for bitwise
17418 equality of representations. They will generally work correctly
17419 for strings and arrays whose elements have integer or enumeration types.
17420 They may not work correctly for arrays whose element
17421 types have user-defined equality, for arrays of real values
17422 (in particular, IEEE-conformant floating point, because of negative
17423 zeroes and NaNs), and for arrays whose elements contain unused bits with
17424 indeterminate values.
17427 The other component-by-component array operations (@code{and}, @code{or},
17428 @code{xor}, @code{not}, and relational tests other than equality)
17429 are not implemented.
17432 @cindex array aggregates (Ada)
17433 @cindex record aggregates (Ada)
17434 @cindex aggregates (Ada)
17435 There is limited support for array and record aggregates. They are
17436 permitted only on the right sides of assignments, as in these examples:
17439 (@value{GDBP}) set An_Array := (1, 2, 3, 4, 5, 6)
17440 (@value{GDBP}) set An_Array := (1, others => 0)
17441 (@value{GDBP}) set An_Array := (0|4 => 1, 1..3 => 2, 5 => 6)
17442 (@value{GDBP}) set A_2D_Array := ((1, 2, 3), (4, 5, 6), (7, 8, 9))
17443 (@value{GDBP}) set A_Record := (1, "Peter", True);
17444 (@value{GDBP}) set A_Record := (Name => "Peter", Id => 1, Alive => True)
17448 discriminant's value by assigning an aggregate has an
17449 undefined effect if that discriminant is used within the record.
17450 However, you can first modify discriminants by directly assigning to
17451 them (which normally would not be allowed in Ada), and then performing an
17452 aggregate assignment. For example, given a variable @code{A_Rec}
17453 declared to have a type such as:
17456 type Rec (Len : Small_Integer := 0) is record
17458 Vals : IntArray (1 .. Len);
17462 you can assign a value with a different size of @code{Vals} with two
17466 (@value{GDBP}) set A_Rec.Len := 4
17467 (@value{GDBP}) set A_Rec := (Id => 42, Vals => (1, 2, 3, 4))
17470 As this example also illustrates, @value{GDBN} is very loose about the usual
17471 rules concerning aggregates. You may leave out some of the
17472 components of an array or record aggregate (such as the @code{Len}
17473 component in the assignment to @code{A_Rec} above); they will retain their
17474 original values upon assignment. You may freely use dynamic values as
17475 indices in component associations. You may even use overlapping or
17476 redundant component associations, although which component values are
17477 assigned in such cases is not defined.
17480 Calls to dispatching subprograms are not implemented.
17483 The overloading algorithm is much more limited (i.e., less selective)
17484 than that of real Ada. It makes only limited use of the context in
17485 which a subexpression appears to resolve its meaning, and it is much
17486 looser in its rules for allowing type matches. As a result, some
17487 function calls will be ambiguous, and the user will be asked to choose
17488 the proper resolution.
17491 The @code{new} operator is not implemented.
17494 Entry calls are not implemented.
17497 Aside from printing, arithmetic operations on the native VAX floating-point
17498 formats are not supported.
17501 It is not possible to slice a packed array.
17504 The names @code{True} and @code{False}, when not part of a qualified name,
17505 are interpreted as if implicitly prefixed by @code{Standard}, regardless of
17507 Should your program
17508 redefine these names in a package or procedure (at best a dubious practice),
17509 you will have to use fully qualified names to access their new definitions.
17512 @node Additions to Ada
17513 @subsubsection Additions to Ada
17514 @cindex Ada, deviations from
17516 As it does for other languages, @value{GDBN} makes certain generic
17517 extensions to Ada (@pxref{Expressions}):
17521 If the expression @var{E} is a variable residing in memory (typically
17522 a local variable or array element) and @var{N} is a positive integer,
17523 then @code{@var{E}@@@var{N}} displays the values of @var{E} and the
17524 @var{N}-1 adjacent variables following it in memory as an array. In
17525 Ada, this operator is generally not necessary, since its prime use is
17526 in displaying parts of an array, and slicing will usually do this in
17527 Ada. However, there are occasional uses when debugging programs in
17528 which certain debugging information has been optimized away.
17531 @code{@var{B}::@var{var}} means ``the variable named @var{var} that
17532 appears in function or file @var{B}.'' When @var{B} is a file name,
17533 you must typically surround it in single quotes.
17536 The expression @code{@{@var{type}@} @var{addr}} means ``the variable of type
17537 @var{type} that appears at address @var{addr}.''
17540 A name starting with @samp{$} is a convenience variable
17541 (@pxref{Convenience Vars}) or a machine register (@pxref{Registers}).
17544 In addition, @value{GDBN} provides a few other shortcuts and outright
17545 additions specific to Ada:
17549 The assignment statement is allowed as an expression, returning
17550 its right-hand operand as its value. Thus, you may enter
17553 (@value{GDBP}) set x := y + 3
17554 (@value{GDBP}) print A(tmp := y + 1)
17558 The semicolon is allowed as an ``operator,'' returning as its value
17559 the value of its right-hand operand.
17560 This allows, for example,
17561 complex conditional breaks:
17564 (@value{GDBP}) break f
17565 (@value{GDBP}) condition 1 (report(i); k += 1; A(k) > 100)
17569 Rather than use catenation and symbolic character names to introduce special
17570 characters into strings, one may instead use a special bracket notation,
17571 which is also used to print strings. A sequence of characters of the form
17572 @samp{["@var{XX}"]} within a string or character literal denotes the
17573 (single) character whose numeric encoding is @var{XX} in hexadecimal. The
17574 sequence of characters @samp{["""]} also denotes a single quotation mark
17575 in strings. For example,
17577 "One line.["0a"]Next line.["0a"]"
17580 contains an ASCII newline character (@code{Ada.Characters.Latin_1.LF})
17584 The subtype used as a prefix for the attributes @t{'Pos}, @t{'Min}, and
17585 @t{'Max} is optional (and is ignored in any case). For example, it is valid
17589 (@value{GDBP}) print 'max(x, y)
17593 When printing arrays, @value{GDBN} uses positional notation when the
17594 array has a lower bound of 1, and uses a modified named notation otherwise.
17595 For example, a one-dimensional array of three integers with a lower bound
17596 of 3 might print as
17603 That is, in contrast to valid Ada, only the first component has a @code{=>}
17607 You may abbreviate attributes in expressions with any unique,
17608 multi-character subsequence of
17609 their names (an exact match gets preference).
17610 For example, you may use @t{a'len}, @t{a'gth}, or @t{a'lh}
17611 in place of @t{a'length}.
17614 @cindex quoting Ada internal identifiers
17615 Since Ada is case-insensitive, the debugger normally maps identifiers you type
17616 to lower case. The GNAT compiler uses upper-case characters for
17617 some of its internal identifiers, which are normally of no interest to users.
17618 For the rare occasions when you actually have to look at them,
17619 enclose them in angle brackets to avoid the lower-case mapping.
17622 (@value{GDBP}) print <JMPBUF_SAVE>[0]
17626 Printing an object of class-wide type or dereferencing an
17627 access-to-class-wide value will display all the components of the object's
17628 specific type (as indicated by its run-time tag). Likewise, component
17629 selection on such a value will operate on the specific type of the
17634 @node Overloading support for Ada
17635 @subsubsection Overloading support for Ada
17636 @cindex overloading, Ada
17638 The debugger supports limited overloading. Given a subprogram call in which
17639 the function symbol has multiple definitions, it will use the number of
17640 actual parameters and some information about their types to attempt to narrow
17641 the set of definitions. It also makes very limited use of context, preferring
17642 procedures to functions in the context of the @code{call} command, and
17643 functions to procedures elsewhere.
17645 If, after narrowing, the set of matching definitions still contains more than
17646 one definition, @value{GDBN} will display a menu to query which one it should
17650 (@value{GDBP}) print f(1)
17651 Multiple matches for f
17653 [1] foo.f (integer) return boolean at foo.adb:23
17654 [2] foo.f (foo.new_integer) return boolean at foo.adb:28
17658 In this case, just select one menu entry either to cancel expression evaluation
17659 (type @kbd{0} and press @key{RET}) or to continue evaluation with a specific
17660 instance (type the corresponding number and press @key{RET}).
17662 Here are a couple of commands to customize @value{GDBN}'s behavior in this
17667 @kindex set ada print-signatures
17668 @item set ada print-signatures
17669 Control whether parameter types and return types are displayed in overloads
17670 selection menus. It is @code{on} by default.
17671 @xref{Overloading support for Ada}.
17673 @kindex show ada print-signatures
17674 @item show ada print-signatures
17675 Show the current setting for displaying parameter types and return types in
17676 overloads selection menu.
17677 @xref{Overloading support for Ada}.
17681 @node Stopping Before Main Program
17682 @subsubsection Stopping at the Very Beginning
17684 @cindex breakpointing Ada elaboration code
17685 It is sometimes necessary to debug the program during elaboration, and
17686 before reaching the main procedure.
17687 As defined in the Ada Reference
17688 Manual, the elaboration code is invoked from a procedure called
17689 @code{adainit}. To run your program up to the beginning of
17690 elaboration, simply use the following two commands:
17691 @code{tbreak adainit} and @code{run}.
17693 @node Ada Exceptions
17694 @subsubsection Ada Exceptions
17696 A command is provided to list all Ada exceptions:
17699 @kindex info exceptions
17700 @item info exceptions
17701 @itemx info exceptions @var{regexp}
17702 The @code{info exceptions} command allows you to list all Ada exceptions
17703 defined within the program being debugged, as well as their addresses.
17704 With a regular expression, @var{regexp}, as argument, only those exceptions
17705 whose names match @var{regexp} are listed.
17708 Below is a small example, showing how the command can be used, first
17709 without argument, and next with a regular expression passed as an
17713 (@value{GDBP}) info exceptions
17714 All defined Ada exceptions:
17715 constraint_error: 0x613da0
17716 program_error: 0x613d20
17717 storage_error: 0x613ce0
17718 tasking_error: 0x613ca0
17719 const.aint_global_e: 0x613b00
17720 (@value{GDBP}) info exceptions const.aint
17721 All Ada exceptions matching regular expression "const.aint":
17722 constraint_error: 0x613da0
17723 const.aint_global_e: 0x613b00
17726 It is also possible to ask @value{GDBN} to stop your program's execution
17727 when an exception is raised. For more details, see @ref{Set Catchpoints}.
17730 @subsubsection Extensions for Ada Tasks
17731 @cindex Ada, tasking
17733 Support for Ada tasks is analogous to that for threads (@pxref{Threads}).
17734 @value{GDBN} provides the following task-related commands:
17739 This command shows a list of current Ada tasks, as in the following example:
17746 (@value{GDBP}) info tasks
17747 ID TID P-ID Pri State Name
17748 1 8088000 0 15 Child Activation Wait main_task
17749 2 80a4000 1 15 Accept Statement b
17750 3 809a800 1 15 Child Activation Wait a
17751 * 4 80ae800 3 15 Runnable c
17756 In this listing, the asterisk before the last task indicates it to be the
17757 task currently being inspected.
17761 Represents @value{GDBN}'s internal task number.
17767 The parent's task ID (@value{GDBN}'s internal task number).
17770 The base priority of the task.
17773 Current state of the task.
17777 The task has been created but has not been activated. It cannot be
17781 The task is not blocked for any reason known to Ada. (It may be waiting
17782 for a mutex, though.) It is conceptually "executing" in normal mode.
17785 The task is terminated, in the sense of ARM 9.3 (5). Any dependents
17786 that were waiting on terminate alternatives have been awakened and have
17787 terminated themselves.
17789 @item Child Activation Wait
17790 The task is waiting for created tasks to complete activation.
17792 @item Accept Statement
17793 The task is waiting on an accept or selective wait statement.
17795 @item Waiting on entry call
17796 The task is waiting on an entry call.
17798 @item Async Select Wait
17799 The task is waiting to start the abortable part of an asynchronous
17803 The task is waiting on a select statement with only a delay
17806 @item Child Termination Wait
17807 The task is sleeping having completed a master within itself, and is
17808 waiting for the tasks dependent on that master to become terminated or
17809 waiting on a terminate Phase.
17811 @item Wait Child in Term Alt
17812 The task is sleeping waiting for tasks on terminate alternatives to
17813 finish terminating.
17815 @item Accepting RV with @var{taskno}
17816 The task is accepting a rendez-vous with the task @var{taskno}.
17820 Name of the task in the program.
17824 @kindex info task @var{taskno}
17825 @item info task @var{taskno}
17826 This command shows detailled informations on the specified task, as in
17827 the following example:
17832 (@value{GDBP}) info tasks
17833 ID TID P-ID Pri State Name
17834 1 8077880 0 15 Child Activation Wait main_task
17835 * 2 807c468 1 15 Runnable task_1
17836 (@value{GDBP}) info task 2
17837 Ada Task: 0x807c468
17841 Parent: 1 ("main_task")
17847 @kindex task@r{ (Ada)}
17848 @cindex current Ada task ID
17849 This command prints the ID and name of the current task.
17855 (@value{GDBP}) info tasks
17856 ID TID P-ID Pri State Name
17857 1 8077870 0 15 Child Activation Wait main_task
17858 * 2 807c458 1 15 Runnable some_task
17859 (@value{GDBP}) task
17860 [Current task is 2 "some_task"]
17863 @item task @var{taskno}
17864 @cindex Ada task switching
17865 This command is like the @code{thread @var{thread-id}}
17866 command (@pxref{Threads}). It switches the context of debugging
17867 from the current task to the given task.
17873 (@value{GDBP}) info tasks
17874 ID TID P-ID Pri State Name
17875 1 8077870 0 15 Child Activation Wait main_task
17876 * 2 807c458 1 15 Runnable some_task
17877 (@value{GDBP}) task 1
17878 [Switching to task 1 "main_task"]
17879 #0 0x8067726 in pthread_cond_wait ()
17881 #0 0x8067726 in pthread_cond_wait ()
17882 #1 0x8056714 in system.os_interface.pthread_cond_wait ()
17883 #2 0x805cb63 in system.task_primitives.operations.sleep ()
17884 #3 0x806153e in system.tasking.stages.activate_tasks ()
17885 #4 0x804aacc in un () at un.adb:5
17888 @item break @var{location} task @var{taskno}
17889 @itemx break @var{location} task @var{taskno} if @dots{}
17890 @cindex breakpoints and tasks, in Ada
17891 @cindex task breakpoints, in Ada
17892 @kindex break @dots{} task @var{taskno}@r{ (Ada)}
17893 These commands are like the @code{break @dots{} thread @dots{}}
17894 command (@pxref{Thread Stops}). The
17895 @var{location} argument specifies source lines, as described
17896 in @ref{Specify Location}.
17898 Use the qualifier @samp{task @var{taskno}} with a breakpoint command
17899 to specify that you only want @value{GDBN} to stop the program when a
17900 particular Ada task reaches this breakpoint. The @var{taskno} is one of the
17901 numeric task identifiers assigned by @value{GDBN}, shown in the first
17902 column of the @samp{info tasks} display.
17904 If you do not specify @samp{task @var{taskno}} when you set a
17905 breakpoint, the breakpoint applies to @emph{all} tasks of your
17908 You can use the @code{task} qualifier on conditional breakpoints as
17909 well; in this case, place @samp{task @var{taskno}} before the
17910 breakpoint condition (before the @code{if}).
17918 (@value{GDBP}) info tasks
17919 ID TID P-ID Pri State Name
17920 1 140022020 0 15 Child Activation Wait main_task
17921 2 140045060 1 15 Accept/Select Wait t2
17922 3 140044840 1 15 Runnable t1
17923 * 4 140056040 1 15 Runnable t3
17924 (@value{GDBP}) b 15 task 2
17925 Breakpoint 5 at 0x120044cb0: file test_task_debug.adb, line 15.
17926 (@value{GDBP}) cont
17931 Breakpoint 5, test_task_debug () at test_task_debug.adb:15
17933 (@value{GDBP}) info tasks
17934 ID TID P-ID Pri State Name
17935 1 140022020 0 15 Child Activation Wait main_task
17936 * 2 140045060 1 15 Runnable t2
17937 3 140044840 1 15 Runnable t1
17938 4 140056040 1 15 Delay Sleep t3
17942 @node Ada Tasks and Core Files
17943 @subsubsection Tasking Support when Debugging Core Files
17944 @cindex Ada tasking and core file debugging
17946 When inspecting a core file, as opposed to debugging a live program,
17947 tasking support may be limited or even unavailable, depending on
17948 the platform being used.
17949 For instance, on x86-linux, the list of tasks is available, but task
17950 switching is not supported.
17952 On certain platforms, the debugger needs to perform some
17953 memory writes in order to provide Ada tasking support. When inspecting
17954 a core file, this means that the core file must be opened with read-write
17955 privileges, using the command @samp{"set write on"} (@pxref{Patching}).
17956 Under these circumstances, you should make a backup copy of the core
17957 file before inspecting it with @value{GDBN}.
17959 @node Ravenscar Profile
17960 @subsubsection Tasking Support when using the Ravenscar Profile
17961 @cindex Ravenscar Profile
17963 The @dfn{Ravenscar Profile} is a subset of the Ada tasking features,
17964 specifically designed for systems with safety-critical real-time
17968 @kindex set ravenscar task-switching on
17969 @cindex task switching with program using Ravenscar Profile
17970 @item set ravenscar task-switching on
17971 Allows task switching when debugging a program that uses the Ravenscar
17972 Profile. This is the default.
17974 @kindex set ravenscar task-switching off
17975 @item set ravenscar task-switching off
17976 Turn off task switching when debugging a program that uses the Ravenscar
17977 Profile. This is mostly intended to disable the code that adds support
17978 for the Ravenscar Profile, in case a bug in either @value{GDBN} or in
17979 the Ravenscar runtime is preventing @value{GDBN} from working properly.
17980 To be effective, this command should be run before the program is started.
17982 @kindex show ravenscar task-switching
17983 @item show ravenscar task-switching
17984 Show whether it is possible to switch from task to task in a program
17985 using the Ravenscar Profile.
17990 @subsubsection Ada Settings
17991 @cindex Ada settings
17994 @kindex set varsize-limit
17995 @item set varsize-limit @var{size}
17996 Prevent @value{GDBN} from attempting to evaluate objects whose size
17997 is above the given limit (@var{size}) when those sizes are computed
17998 from run-time quantities. This is typically the case when the object
17999 has a variable size, such as an array whose bounds are not known at
18000 compile time for example. Setting @var{size} to @code{unlimited}
18001 removes the size limitation. By default, the limit is about 65KB.
18003 The purpose of having such a limit is to prevent @value{GDBN} from
18004 trying to grab enormous chunks of virtual memory when asked to evaluate
18005 a quantity whose bounds have been corrupted or have not yet been fully
18006 initialized. The limit applies to the results of some subexpressions
18007 as well as to complete expressions. For example, an expression denoting
18008 a simple integer component, such as @code{x.y.z}, may fail if the size of
18009 @code{x.y} is variable and exceeds @code{size}. On the other hand,
18010 @value{GDBN} is sometimes clever; the expression @code{A(i)}, where
18011 @code{A} is an array variable with non-constant size, will generally
18012 succeed regardless of the bounds on @code{A}, as long as the component
18013 size is less than @var{size}.
18015 @kindex show varsize-limit
18016 @item show varsize-limit
18017 Show the limit on types whose size is determined by run-time quantities.
18021 @subsubsection Known Peculiarities of Ada Mode
18022 @cindex Ada, problems
18024 Besides the omissions listed previously (@pxref{Omissions from Ada}),
18025 we know of several problems with and limitations of Ada mode in
18027 some of which will be fixed with planned future releases of the debugger
18028 and the GNU Ada compiler.
18032 Static constants that the compiler chooses not to materialize as objects in
18033 storage are invisible to the debugger.
18036 Named parameter associations in function argument lists are ignored (the
18037 argument lists are treated as positional).
18040 Many useful library packages are currently invisible to the debugger.
18043 Fixed-point arithmetic, conversions, input, and output is carried out using
18044 floating-point arithmetic, and may give results that only approximate those on
18048 The GNAT compiler never generates the prefix @code{Standard} for any of
18049 the standard symbols defined by the Ada language. @value{GDBN} knows about
18050 this: it will strip the prefix from names when you use it, and will never
18051 look for a name you have so qualified among local symbols, nor match against
18052 symbols in other packages or subprograms. If you have
18053 defined entities anywhere in your program other than parameters and
18054 local variables whose simple names match names in @code{Standard},
18055 GNAT's lack of qualification here can cause confusion. When this happens,
18056 you can usually resolve the confusion
18057 by qualifying the problematic names with package
18058 @code{Standard} explicitly.
18061 Older versions of the compiler sometimes generate erroneous debugging
18062 information, resulting in the debugger incorrectly printing the value
18063 of affected entities. In some cases, the debugger is able to work
18064 around an issue automatically. In other cases, the debugger is able
18065 to work around the issue, but the work-around has to be specifically
18068 @kindex set ada trust-PAD-over-XVS
18069 @kindex show ada trust-PAD-over-XVS
18072 @item set ada trust-PAD-over-XVS on
18073 Configure GDB to strictly follow the GNAT encoding when computing the
18074 value of Ada entities, particularly when @code{PAD} and @code{PAD___XVS}
18075 types are involved (see @code{ada/exp_dbug.ads} in the GCC sources for
18076 a complete description of the encoding used by the GNAT compiler).
18077 This is the default.
18079 @item set ada trust-PAD-over-XVS off
18080 This is related to the encoding using by the GNAT compiler. If @value{GDBN}
18081 sometimes prints the wrong value for certain entities, changing @code{ada
18082 trust-PAD-over-XVS} to @code{off} activates a work-around which may fix
18083 the issue. It is always safe to set @code{ada trust-PAD-over-XVS} to
18084 @code{off}, but this incurs a slight performance penalty, so it is
18085 recommended to leave this setting to @code{on} unless necessary.
18089 @cindex GNAT descriptive types
18090 @cindex GNAT encoding
18091 Internally, the debugger also relies on the compiler following a number
18092 of conventions known as the @samp{GNAT Encoding}, all documented in
18093 @file{gcc/ada/exp_dbug.ads} in the GCC sources. This encoding describes
18094 how the debugging information should be generated for certain types.
18095 In particular, this convention makes use of @dfn{descriptive types},
18096 which are artificial types generated purely to help the debugger.
18098 These encodings were defined at a time when the debugging information
18099 format used was not powerful enough to describe some of the more complex
18100 types available in Ada. Since DWARF allows us to express nearly all
18101 Ada features, the long-term goal is to slowly replace these descriptive
18102 types by their pure DWARF equivalent. To facilitate that transition,
18103 a new maintenance option is available to force the debugger to ignore
18104 those descriptive types. It allows the user to quickly evaluate how
18105 well @value{GDBN} works without them.
18109 @kindex maint ada set ignore-descriptive-types
18110 @item maintenance ada set ignore-descriptive-types [on|off]
18111 Control whether the debugger should ignore descriptive types.
18112 The default is not to ignore descriptives types (@code{off}).
18114 @kindex maint ada show ignore-descriptive-types
18115 @item maintenance ada show ignore-descriptive-types
18116 Show if descriptive types are ignored by @value{GDBN}.
18120 @node Unsupported Languages
18121 @section Unsupported Languages
18123 @cindex unsupported languages
18124 @cindex minimal language
18125 In addition to the other fully-supported programming languages,
18126 @value{GDBN} also provides a pseudo-language, called @code{minimal}.
18127 It does not represent a real programming language, but provides a set
18128 of capabilities close to what the C or assembly languages provide.
18129 This should allow most simple operations to be performed while debugging
18130 an application that uses a language currently not supported by @value{GDBN}.
18132 If the language is set to @code{auto}, @value{GDBN} will automatically
18133 select this language if the current frame corresponds to an unsupported
18137 @chapter Examining the Symbol Table
18139 The commands described in this chapter allow you to inquire about the
18140 symbols (names of variables, functions and types) defined in your
18141 program. This information is inherent in the text of your program and
18142 does not change as your program executes. @value{GDBN} finds it in your
18143 program's symbol table, in the file indicated when you started @value{GDBN}
18144 (@pxref{File Options, ,Choosing Files}), or by one of the
18145 file-management commands (@pxref{Files, ,Commands to Specify Files}).
18147 @cindex symbol names
18148 @cindex names of symbols
18149 @cindex quoting names
18150 @anchor{quoting names}
18151 Occasionally, you may need to refer to symbols that contain unusual
18152 characters, which @value{GDBN} ordinarily treats as word delimiters. The
18153 most frequent case is in referring to static variables in other
18154 source files (@pxref{Variables,,Program Variables}). File names
18155 are recorded in object files as debugging symbols, but @value{GDBN} would
18156 ordinarily parse a typical file name, like @file{foo.c}, as the three words
18157 @samp{foo} @samp{.} @samp{c}. To allow @value{GDBN} to recognize
18158 @samp{foo.c} as a single symbol, enclose it in single quotes; for example,
18165 looks up the value of @code{x} in the scope of the file @file{foo.c}.
18168 @cindex case-insensitive symbol names
18169 @cindex case sensitivity in symbol names
18170 @kindex set case-sensitive
18171 @item set case-sensitive on
18172 @itemx set case-sensitive off
18173 @itemx set case-sensitive auto
18174 Normally, when @value{GDBN} looks up symbols, it matches their names
18175 with case sensitivity determined by the current source language.
18176 Occasionally, you may wish to control that. The command @code{set
18177 case-sensitive} lets you do that by specifying @code{on} for
18178 case-sensitive matches or @code{off} for case-insensitive ones. If
18179 you specify @code{auto}, case sensitivity is reset to the default
18180 suitable for the source language. The default is case-sensitive
18181 matches for all languages except for Fortran, for which the default is
18182 case-insensitive matches.
18184 @kindex show case-sensitive
18185 @item show case-sensitive
18186 This command shows the current setting of case sensitivity for symbols
18189 @kindex set print type methods
18190 @item set print type methods
18191 @itemx set print type methods on
18192 @itemx set print type methods off
18193 Normally, when @value{GDBN} prints a class, it displays any methods
18194 declared in that class. You can control this behavior either by
18195 passing the appropriate flag to @code{ptype}, or using @command{set
18196 print type methods}. Specifying @code{on} will cause @value{GDBN} to
18197 display the methods; this is the default. Specifying @code{off} will
18198 cause @value{GDBN} to omit the methods.
18200 @kindex show print type methods
18201 @item show print type methods
18202 This command shows the current setting of method display when printing
18205 @kindex set print type nested-type-limit
18206 @item set print type nested-type-limit @var{limit}
18207 @itemx set print type nested-type-limit unlimited
18208 Set the limit of displayed nested types that the type printer will
18209 show. A @var{limit} of @code{unlimited} or @code{-1} will show all
18210 nested definitions. By default, the type printer will not show any nested
18211 types defined in classes.
18213 @kindex show print type nested-type-limit
18214 @item show print type nested-type-limit
18215 This command shows the current display limit of nested types when
18218 @kindex set print type typedefs
18219 @item set print type typedefs
18220 @itemx set print type typedefs on
18221 @itemx set print type typedefs off
18223 Normally, when @value{GDBN} prints a class, it displays any typedefs
18224 defined in that class. You can control this behavior either by
18225 passing the appropriate flag to @code{ptype}, or using @command{set
18226 print type typedefs}. Specifying @code{on} will cause @value{GDBN} to
18227 display the typedef definitions; this is the default. Specifying
18228 @code{off} will cause @value{GDBN} to omit the typedef definitions.
18229 Note that this controls whether the typedef definition itself is
18230 printed, not whether typedef names are substituted when printing other
18233 @kindex show print type typedefs
18234 @item show print type typedefs
18235 This command shows the current setting of typedef display when
18238 @kindex info address
18239 @cindex address of a symbol
18240 @item info address @var{symbol}
18241 Describe where the data for @var{symbol} is stored. For a register
18242 variable, this says which register it is kept in. For a non-register
18243 local variable, this prints the stack-frame offset at which the variable
18246 Note the contrast with @samp{print &@var{symbol}}, which does not work
18247 at all for a register variable, and for a stack local variable prints
18248 the exact address of the current instantiation of the variable.
18250 @kindex info symbol
18251 @cindex symbol from address
18252 @cindex closest symbol and offset for an address
18253 @item info symbol @var{addr}
18254 Print the name of a symbol which is stored at the address @var{addr}.
18255 If no symbol is stored exactly at @var{addr}, @value{GDBN} prints the
18256 nearest symbol and an offset from it:
18259 (@value{GDBP}) info symbol 0x54320
18260 _initialize_vx + 396 in section .text
18264 This is the opposite of the @code{info address} command. You can use
18265 it to find out the name of a variable or a function given its address.
18267 For dynamically linked executables, the name of executable or shared
18268 library containing the symbol is also printed:
18271 (@value{GDBP}) info symbol 0x400225
18272 _start + 5 in section .text of /tmp/a.out
18273 (@value{GDBP}) info symbol 0x2aaaac2811cf
18274 __read_nocancel + 6 in section .text of /usr/lib64/libc.so.6
18279 @item demangle @r{[}-l @var{language}@r{]} @r{[}@var{--}@r{]} @var{name}
18280 Demangle @var{name}.
18281 If @var{language} is provided it is the name of the language to demangle
18282 @var{name} in. Otherwise @var{name} is demangled in the current language.
18284 The @samp{--} option specifies the end of options,
18285 and is useful when @var{name} begins with a dash.
18287 The parameter @code{demangle-style} specifies how to interpret the kind
18288 of mangling used. @xref{Print Settings}.
18291 @item whatis[/@var{flags}] [@var{arg}]
18292 Print the data type of @var{arg}, which can be either an expression
18293 or a name of a data type. With no argument, print the data type of
18294 @code{$}, the last value in the value history.
18296 If @var{arg} is an expression (@pxref{Expressions, ,Expressions}), it
18297 is not actually evaluated, and any side-effecting operations (such as
18298 assignments or function calls) inside it do not take place.
18300 If @var{arg} is a variable or an expression, @code{whatis} prints its
18301 literal type as it is used in the source code. If the type was
18302 defined using a @code{typedef}, @code{whatis} will @emph{not} print
18303 the data type underlying the @code{typedef}. If the type of the
18304 variable or the expression is a compound data type, such as
18305 @code{struct} or @code{class}, @code{whatis} never prints their
18306 fields or methods. It just prints the @code{struct}/@code{class}
18307 name (a.k.a.@: its @dfn{tag}). If you want to see the members of
18308 such a compound data type, use @code{ptype}.
18310 If @var{arg} is a type name that was defined using @code{typedef},
18311 @code{whatis} @dfn{unrolls} only one level of that @code{typedef}.
18312 Unrolling means that @code{whatis} will show the underlying type used
18313 in the @code{typedef} declaration of @var{arg}. However, if that
18314 underlying type is also a @code{typedef}, @code{whatis} will not
18317 For C code, the type names may also have the form @samp{class
18318 @var{class-name}}, @samp{struct @var{struct-tag}}, @samp{union
18319 @var{union-tag}} or @samp{enum @var{enum-tag}}.
18321 @var{flags} can be used to modify how the type is displayed.
18322 Available flags are:
18326 Display in ``raw'' form. Normally, @value{GDBN} substitutes template
18327 parameters and typedefs defined in a class when printing the class'
18328 members. The @code{/r} flag disables this.
18331 Do not print methods defined in the class.
18334 Print methods defined in the class. This is the default, but the flag
18335 exists in case you change the default with @command{set print type methods}.
18338 Do not print typedefs defined in the class. Note that this controls
18339 whether the typedef definition itself is printed, not whether typedef
18340 names are substituted when printing other types.
18343 Print typedefs defined in the class. This is the default, but the flag
18344 exists in case you change the default with @command{set print type typedefs}.
18347 Print the offsets and sizes of fields in a struct, similar to what the
18348 @command{pahole} tool does. This option implies the @code{/tm} flags.
18350 For example, given the following declarations:
18386 Issuing a @kbd{ptype /o struct tuv} command would print:
18389 (@value{GDBP}) ptype /o struct tuv
18390 /* offset | size */ type = struct tuv @{
18391 /* 0 | 4 */ int a1;
18392 /* XXX 4-byte hole */
18393 /* 8 | 8 */ char *a2;
18394 /* 16 | 4 */ int a3;
18396 /* total size (bytes): 24 */
18400 Notice the format of the first column of comments. There, you can
18401 find two parts separated by the @samp{|} character: the @emph{offset},
18402 which indicates where the field is located inside the struct, in
18403 bytes, and the @emph{size} of the field. Another interesting line is
18404 the marker of a @emph{hole} in the struct, indicating that it may be
18405 possible to pack the struct and make it use less space by reorganizing
18408 It is also possible to print offsets inside an union:
18411 (@value{GDBP}) ptype /o union qwe
18412 /* offset | size */ type = union qwe @{
18413 /* 24 */ struct tuv @{
18414 /* 0 | 4 */ int a1;
18415 /* XXX 4-byte hole */
18416 /* 8 | 8 */ char *a2;
18417 /* 16 | 4 */ int a3;
18419 /* total size (bytes): 24 */
18421 /* 40 */ struct xyz @{
18422 /* 0 | 4 */ int f1;
18423 /* 4 | 1 */ char f2;
18424 /* XXX 3-byte hole */
18425 /* 8 | 8 */ void *f3;
18426 /* 16 | 24 */ struct tuv @{
18427 /* 16 | 4 */ int a1;
18428 /* XXX 4-byte hole */
18429 /* 24 | 8 */ char *a2;
18430 /* 32 | 4 */ int a3;
18432 /* total size (bytes): 24 */
18435 /* total size (bytes): 40 */
18438 /* total size (bytes): 40 */
18442 In this case, since @code{struct tuv} and @code{struct xyz} occupy the
18443 same space (because we are dealing with an union), the offset is not
18444 printed for them. However, you can still examine the offset of each
18445 of these structures' fields.
18447 Another useful scenario is printing the offsets of a struct containing
18451 (@value{GDBP}) ptype /o struct tyu
18452 /* offset | size */ type = struct tyu @{
18453 /* 0:31 | 4 */ int a1 : 1;
18454 /* 0:28 | 4 */ int a2 : 3;
18455 /* 0: 5 | 4 */ int a3 : 23;
18456 /* 3: 3 | 1 */ signed char a4 : 2;
18457 /* XXX 3-bit hole */
18458 /* XXX 4-byte hole */
18459 /* 8 | 8 */ int64_t a5;
18460 /* 16: 0 | 4 */ int a6 : 5;
18461 /* 16: 5 | 8 */ int64_t a7 : 3;
18462 "/* XXX 7-byte padding */
18464 /* total size (bytes): 24 */
18468 Note how the offset information is now extended to also include the
18469 first bit of the bitfield.
18473 @item ptype[/@var{flags}] [@var{arg}]
18474 @code{ptype} accepts the same arguments as @code{whatis}, but prints a
18475 detailed description of the type, instead of just the name of the type.
18476 @xref{Expressions, ,Expressions}.
18478 Contrary to @code{whatis}, @code{ptype} always unrolls any
18479 @code{typedef}s in its argument declaration, whether the argument is
18480 a variable, expression, or a data type. This means that @code{ptype}
18481 of a variable or an expression will not print literally its type as
18482 present in the source code---use @code{whatis} for that. @code{typedef}s at
18483 the pointer or reference targets are also unrolled. Only @code{typedef}s of
18484 fields, methods and inner @code{class typedef}s of @code{struct}s,
18485 @code{class}es and @code{union}s are not unrolled even with @code{ptype}.
18487 For example, for this variable declaration:
18490 typedef double real_t;
18491 struct complex @{ real_t real; double imag; @};
18492 typedef struct complex complex_t;
18494 real_t *real_pointer_var;
18498 the two commands give this output:
18502 (@value{GDBP}) whatis var
18504 (@value{GDBP}) ptype var
18505 type = struct complex @{
18509 (@value{GDBP}) whatis complex_t
18510 type = struct complex
18511 (@value{GDBP}) whatis struct complex
18512 type = struct complex
18513 (@value{GDBP}) ptype struct complex
18514 type = struct complex @{
18518 (@value{GDBP}) whatis real_pointer_var
18520 (@value{GDBP}) ptype real_pointer_var
18526 As with @code{whatis}, using @code{ptype} without an argument refers to
18527 the type of @code{$}, the last value in the value history.
18529 @cindex incomplete type
18530 Sometimes, programs use opaque data types or incomplete specifications
18531 of complex data structure. If the debug information included in the
18532 program does not allow @value{GDBN} to display a full declaration of
18533 the data type, it will say @samp{<incomplete type>}. For example,
18534 given these declarations:
18538 struct foo *fooptr;
18542 but no definition for @code{struct foo} itself, @value{GDBN} will say:
18545 (@value{GDBP}) ptype foo
18546 $1 = <incomplete type>
18550 ``Incomplete type'' is C terminology for data types that are not
18551 completely specified.
18553 @cindex unknown type
18554 Othertimes, information about a variable's type is completely absent
18555 from the debug information included in the program. This most often
18556 happens when the program or library where the variable is defined
18557 includes no debug information at all. @value{GDBN} knows the variable
18558 exists from inspecting the linker/loader symbol table (e.g., the ELF
18559 dynamic symbol table), but such symbols do not contain type
18560 information. Inspecting the type of a (global) variable for which
18561 @value{GDBN} has no type information shows:
18564 (@value{GDBP}) ptype var
18565 type = <data variable, no debug info>
18568 @xref{Variables, no debug info variables}, for how to print the values
18572 @item info types [-q] [@var{regexp}]
18573 Print a brief description of all types whose names match the regular
18574 expression @var{regexp} (or all types in your program, if you supply
18575 no argument). Each complete typename is matched as though it were a
18576 complete line; thus, @samp{i type value} gives information on all
18577 types in your program whose names include the string @code{value}, but
18578 @samp{i type ^value$} gives information only on types whose complete
18579 name is @code{value}.
18581 In programs using different languages, @value{GDBN} chooses the syntax
18582 to print the type description according to the
18583 @samp{set language} value: using @samp{set language auto}
18584 (see @ref{Automatically, ,Set Language Automatically}) means to use the
18585 language of the type, other values mean to use
18586 the manually specified language (see @ref{Manually, ,Set Language Manually}).
18588 This command differs from @code{ptype} in two ways: first, like
18589 @code{whatis}, it does not print a detailed description; second, it
18590 lists all source files and line numbers where a type is defined.
18592 The output from @samp{into types} is proceeded with a header line
18593 describing what types are being listed. The optional flag @samp{-q},
18594 which stands for @samp{quiet}, disables printing this header
18597 @kindex info type-printers
18598 @item info type-printers
18599 Versions of @value{GDBN} that ship with Python scripting enabled may
18600 have ``type printers'' available. When using @command{ptype} or
18601 @command{whatis}, these printers are consulted when the name of a type
18602 is needed. @xref{Type Printing API}, for more information on writing
18605 @code{info type-printers} displays all the available type printers.
18607 @kindex enable type-printer
18608 @kindex disable type-printer
18609 @item enable type-printer @var{name}@dots{}
18610 @item disable type-printer @var{name}@dots{}
18611 These commands can be used to enable or disable type printers.
18614 @cindex local variables
18615 @item info scope @var{location}
18616 List all the variables local to a particular scope. This command
18617 accepts a @var{location} argument---a function name, a source line, or
18618 an address preceded by a @samp{*}, and prints all the variables local
18619 to the scope defined by that location. (@xref{Specify Location}, for
18620 details about supported forms of @var{location}.) For example:
18623 (@value{GDBP}) @b{info scope command_line_handler}
18624 Scope for command_line_handler:
18625 Symbol rl is an argument at stack/frame offset 8, length 4.
18626 Symbol linebuffer is in static storage at address 0x150a18, length 4.
18627 Symbol linelength is in static storage at address 0x150a1c, length 4.
18628 Symbol p is a local variable in register $esi, length 4.
18629 Symbol p1 is a local variable in register $ebx, length 4.
18630 Symbol nline is a local variable in register $edx, length 4.
18631 Symbol repeat is a local variable at frame offset -8, length 4.
18635 This command is especially useful for determining what data to collect
18636 during a @dfn{trace experiment}, see @ref{Tracepoint Actions,
18639 @kindex info source
18641 Show information about the current source file---that is, the source file for
18642 the function containing the current point of execution:
18645 the name of the source file, and the directory containing it,
18647 the directory it was compiled in,
18649 its length, in lines,
18651 which programming language it is written in,
18653 if the debug information provides it, the program that compiled the file
18654 (which may include, e.g., the compiler version and command line arguments),
18656 whether the executable includes debugging information for that file, and
18657 if so, what format the information is in (e.g., STABS, Dwarf 2, etc.), and
18659 whether the debugging information includes information about
18660 preprocessor macros.
18664 @kindex info sources
18666 Print the names of all source files in your program for which there is
18667 debugging information, organized into two lists: files whose symbols
18668 have already been read, and files whose symbols will be read when needed.
18670 @item info sources [-dirname | -basename] [--] [@var{regexp}]
18671 Like @samp{info sources}, but only print the names of the files
18672 matching the provided @var{regexp}.
18673 By default, the @var{regexp} is used to match anywhere in the filename.
18674 If @code{-dirname}, only files having a dirname matching @var{regexp} are shown.
18675 If @code{-basename}, only files having a basename matching @var{regexp}
18677 The matching is case-sensitive, except on operating systems that
18678 have case-insensitive filesystem (e.g., MS-Windows).
18680 @kindex info functions
18681 @item info functions [-q] [-n]
18682 Print the names and data types of all defined functions.
18683 Similarly to @samp{info types}, this command groups its output by source
18684 files and annotates each function definition with its source line
18687 In programs using different languages, @value{GDBN} chooses the syntax
18688 to print the function name and type according to the
18689 @samp{set language} value: using @samp{set language auto}
18690 (see @ref{Automatically, ,Set Language Automatically}) means to use the
18691 language of the function, other values mean to use
18692 the manually specified language (see @ref{Manually, ,Set Language Manually}).
18694 The @samp{-n} flag excludes @dfn{non-debugging symbols} from the
18695 results. A non-debugging symbol is a symbol that comes from the
18696 executable's symbol table, not from the debug information (for
18697 example, DWARF) associated with the executable.
18699 The optional flag @samp{-q}, which stands for @samp{quiet}, disables
18700 printing header information and messages explaining why no functions
18703 @item info functions [-q] [-n] [-t @var{type_regexp}] [@var{regexp}]
18704 Like @samp{info functions}, but only print the names and data types
18705 of the functions selected with the provided regexp(s).
18707 If @var{regexp} is provided, print only the functions whose names
18708 match the regular expression @var{regexp}.
18709 Thus, @samp{info fun step} finds all functions whose
18710 names include @code{step}; @samp{info fun ^step} finds those whose names
18711 start with @code{step}. If a function name contains characters that
18712 conflict with the regular expression language (e.g.@:
18713 @samp{operator*()}), they may be quoted with a backslash.
18715 If @var{type_regexp} is provided, print only the functions whose
18716 types, as printed by the @code{whatis} command, match
18717 the regular expression @var{type_regexp}.
18718 If @var{type_regexp} contains space(s), it should be enclosed in
18719 quote characters. If needed, use backslash to escape the meaning
18720 of special characters or quotes.
18721 Thus, @samp{info fun -t '^int ('} finds the functions that return
18722 an integer; @samp{info fun -t '(.*int.*'} finds the functions that
18723 have an argument type containing int; @samp{info fun -t '^int (' ^step}
18724 finds the functions whose names start with @code{step} and that return
18727 If both @var{regexp} and @var{type_regexp} are provided, a function
18728 is printed only if its name matches @var{regexp} and its type matches
18732 @kindex info variables
18733 @item info variables [-q] [-n]
18734 Print the names and data types of all variables that are defined
18735 outside of functions (i.e.@: excluding local variables).
18736 The printed variables are grouped by source files and annotated with
18737 their respective source line numbers.
18739 In programs using different languages, @value{GDBN} chooses the syntax
18740 to print the variable name and type according to the
18741 @samp{set language} value: using @samp{set language auto}
18742 (see @ref{Automatically, ,Set Language Automatically}) means to use the
18743 language of the variable, other values mean to use
18744 the manually specified language (see @ref{Manually, ,Set Language Manually}).
18746 The @samp{-n} flag excludes non-debugging symbols from the results.
18748 The optional flag @samp{-q}, which stands for @samp{quiet}, disables
18749 printing header information and messages explaining why no variables
18752 @item info variables [-q] [-n] [-t @var{type_regexp}] [@var{regexp}]
18753 Like @kbd{info variables}, but only print the variables selected
18754 with the provided regexp(s).
18756 If @var{regexp} is provided, print only the variables whose names
18757 match the regular expression @var{regexp}.
18759 If @var{type_regexp} is provided, print only the variables whose
18760 types, as printed by the @code{whatis} command, match
18761 the regular expression @var{type_regexp}.
18762 If @var{type_regexp} contains space(s), it should be enclosed in
18763 quote characters. If needed, use backslash to escape the meaning
18764 of special characters or quotes.
18766 If both @var{regexp} and @var{type_regexp} are provided, an argument
18767 is printed only if its name matches @var{regexp} and its type matches
18770 @kindex info classes
18771 @cindex Objective-C, classes and selectors
18773 @itemx info classes @var{regexp}
18774 Display all Objective-C classes in your program, or
18775 (with the @var{regexp} argument) all those matching a particular regular
18778 @kindex info selectors
18779 @item info selectors
18780 @itemx info selectors @var{regexp}
18781 Display all Objective-C selectors in your program, or
18782 (with the @var{regexp} argument) all those matching a particular regular
18786 This was never implemented.
18787 @kindex info methods
18789 @itemx info methods @var{regexp}
18790 The @code{info methods} command permits the user to examine all defined
18791 methods within C@t{++} program, or (with the @var{regexp} argument) a
18792 specific set of methods found in the various C@t{++} classes. Many
18793 C@t{++} classes provide a large number of methods. Thus, the output
18794 from the @code{ptype} command can be overwhelming and hard to use. The
18795 @code{info-methods} command filters the methods, printing only those
18796 which match the regular-expression @var{regexp}.
18799 @cindex opaque data types
18800 @kindex set opaque-type-resolution
18801 @item set opaque-type-resolution on
18802 Tell @value{GDBN} to resolve opaque types. An opaque type is a type
18803 declared as a pointer to a @code{struct}, @code{class}, or
18804 @code{union}---for example, @code{struct MyType *}---that is used in one
18805 source file although the full declaration of @code{struct MyType} is in
18806 another source file. The default is on.
18808 A change in the setting of this subcommand will not take effect until
18809 the next time symbols for a file are loaded.
18811 @item set opaque-type-resolution off
18812 Tell @value{GDBN} not to resolve opaque types. In this case, the type
18813 is printed as follows:
18815 @{<no data fields>@}
18818 @kindex show opaque-type-resolution
18819 @item show opaque-type-resolution
18820 Show whether opaque types are resolved or not.
18822 @kindex set print symbol-loading
18823 @cindex print messages when symbols are loaded
18824 @item set print symbol-loading
18825 @itemx set print symbol-loading full
18826 @itemx set print symbol-loading brief
18827 @itemx set print symbol-loading off
18828 The @code{set print symbol-loading} command allows you to control the
18829 printing of messages when @value{GDBN} loads symbol information.
18830 By default a message is printed for the executable and one for each
18831 shared library, and normally this is what you want. However, when
18832 debugging apps with large numbers of shared libraries these messages
18834 When set to @code{brief} a message is printed for each executable,
18835 and when @value{GDBN} loads a collection of shared libraries at once
18836 it will only print one message regardless of the number of shared
18837 libraries. When set to @code{off} no messages are printed.
18839 @kindex show print symbol-loading
18840 @item show print symbol-loading
18841 Show whether messages will be printed when a @value{GDBN} command
18842 entered from the keyboard causes symbol information to be loaded.
18844 @kindex maint print symbols
18845 @cindex symbol dump
18846 @kindex maint print psymbols
18847 @cindex partial symbol dump
18848 @kindex maint print msymbols
18849 @cindex minimal symbol dump
18850 @item maint print symbols @r{[}-pc @var{address}@r{]} @r{[}@var{filename}@r{]}
18851 @itemx maint print symbols @r{[}-objfile @var{objfile}@r{]} @r{[}-source @var{source}@r{]} @r{[}--@r{]} @r{[}@var{filename}@r{]}
18852 @itemx maint print psymbols @r{[}-objfile @var{objfile}@r{]} @r{[}-pc @var{address}@r{]} @r{[}--@r{]} @r{[}@var{filename}@r{]}
18853 @itemx maint print psymbols @r{[}-objfile @var{objfile}@r{]} @r{[}-source @var{source}@r{]} @r{[}--@r{]} @r{[}@var{filename}@r{]}
18854 @itemx maint print msymbols @r{[}-objfile @var{objfile}@r{]} @r{[}--@r{]} @r{[}@var{filename}@r{]}
18855 Write a dump of debugging symbol data into the file @var{filename} or
18856 the terminal if @var{filename} is unspecified.
18857 If @code{-objfile @var{objfile}} is specified, only dump symbols for
18859 If @code{-pc @var{address}} is specified, only dump symbols for the file
18860 with code at that address. Note that @var{address} may be a symbol like
18862 If @code{-source @var{source}} is specified, only dump symbols for that
18865 These commands are used to debug the @value{GDBN} symbol-reading code.
18866 These commands do not modify internal @value{GDBN} state, therefore
18867 @samp{maint print symbols} will only print symbols for already expanded symbol
18869 You can use the command @code{info sources} to find out which files these are.
18870 If you use @samp{maint print psymbols} instead, the dump shows information
18871 about symbols that @value{GDBN} only knows partially---that is, symbols
18872 defined in files that @value{GDBN} has skimmed, but not yet read completely.
18873 Finally, @samp{maint print msymbols} just dumps ``minimal symbols'', e.g.,
18876 @xref{Files, ,Commands to Specify Files}, for a discussion of how
18877 @value{GDBN} reads symbols (in the description of @code{symbol-file}).
18879 @kindex maint info symtabs
18880 @kindex maint info psymtabs
18881 @cindex listing @value{GDBN}'s internal symbol tables
18882 @cindex symbol tables, listing @value{GDBN}'s internal
18883 @cindex full symbol tables, listing @value{GDBN}'s internal
18884 @cindex partial symbol tables, listing @value{GDBN}'s internal
18885 @item maint info symtabs @r{[} @var{regexp} @r{]}
18886 @itemx maint info psymtabs @r{[} @var{regexp} @r{]}
18888 List the @code{struct symtab} or @code{struct partial_symtab}
18889 structures whose names match @var{regexp}. If @var{regexp} is not
18890 given, list them all. The output includes expressions which you can
18891 copy into a @value{GDBN} debugging this one to examine a particular
18892 structure in more detail. For example:
18895 (@value{GDBP}) maint info psymtabs dwarf2read
18896 @{ objfile /home/gnu/build/gdb/gdb
18897 ((struct objfile *) 0x82e69d0)
18898 @{ psymtab /home/gnu/src/gdb/dwarf2read.c
18899 ((struct partial_symtab *) 0x8474b10)
18902 text addresses 0x814d3c8 -- 0x8158074
18903 globals (* (struct partial_symbol **) 0x8507a08 @@ 9)
18904 statics (* (struct partial_symbol **) 0x40e95b78 @@ 2882)
18905 dependencies (none)
18908 (@value{GDBP}) maint info symtabs
18912 We see that there is one partial symbol table whose filename contains
18913 the string @samp{dwarf2read}, belonging to the @samp{gdb} executable;
18914 and we see that @value{GDBN} has not read in any symtabs yet at all.
18915 If we set a breakpoint on a function, that will cause @value{GDBN} to
18916 read the symtab for the compilation unit containing that function:
18919 (@value{GDBP}) break dwarf2_psymtab_to_symtab
18920 Breakpoint 1 at 0x814e5da: file /home/gnu/src/gdb/dwarf2read.c,
18922 (@value{GDBP}) maint info symtabs
18923 @{ objfile /home/gnu/build/gdb/gdb
18924 ((struct objfile *) 0x82e69d0)
18925 @{ symtab /home/gnu/src/gdb/dwarf2read.c
18926 ((struct symtab *) 0x86c1f38)
18929 blockvector ((struct blockvector *) 0x86c1bd0) (primary)
18930 linetable ((struct linetable *) 0x8370fa0)
18931 debugformat DWARF 2
18937 @kindex maint info line-table
18938 @cindex listing @value{GDBN}'s internal line tables
18939 @cindex line tables, listing @value{GDBN}'s internal
18940 @item maint info line-table @r{[} @var{regexp} @r{]}
18942 List the @code{struct linetable} from all @code{struct symtab}
18943 instances whose name matches @var{regexp}. If @var{regexp} is not
18944 given, list the @code{struct linetable} from all @code{struct symtab}.
18946 @kindex maint set symbol-cache-size
18947 @cindex symbol cache size
18948 @item maint set symbol-cache-size @var{size}
18949 Set the size of the symbol cache to @var{size}.
18950 The default size is intended to be good enough for debugging
18951 most applications. This option exists to allow for experimenting
18952 with different sizes.
18954 @kindex maint show symbol-cache-size
18955 @item maint show symbol-cache-size
18956 Show the size of the symbol cache.
18958 @kindex maint print symbol-cache
18959 @cindex symbol cache, printing its contents
18960 @item maint print symbol-cache
18961 Print the contents of the symbol cache.
18962 This is useful when debugging symbol cache issues.
18964 @kindex maint print symbol-cache-statistics
18965 @cindex symbol cache, printing usage statistics
18966 @item maint print symbol-cache-statistics
18967 Print symbol cache usage statistics.
18968 This helps determine how well the cache is being utilized.
18970 @kindex maint flush-symbol-cache
18971 @cindex symbol cache, flushing
18972 @item maint flush-symbol-cache
18973 Flush the contents of the symbol cache, all entries are removed.
18974 This command is useful when debugging the symbol cache.
18975 It is also useful when collecting performance data.
18980 @chapter Altering Execution
18982 Once you think you have found an error in your program, you might want to
18983 find out for certain whether correcting the apparent error would lead to
18984 correct results in the rest of the run. You can find the answer by
18985 experiment, using the @value{GDBN} features for altering execution of the
18988 For example, you can store new values into variables or memory
18989 locations, give your program a signal, restart it at a different
18990 address, or even return prematurely from a function.
18993 * Assignment:: Assignment to variables
18994 * Jumping:: Continuing at a different address
18995 * Signaling:: Giving your program a signal
18996 * Returning:: Returning from a function
18997 * Calling:: Calling your program's functions
18998 * Patching:: Patching your program
18999 * Compiling and Injecting Code:: Compiling and injecting code in @value{GDBN}
19003 @section Assignment to Variables
19006 @cindex setting variables
19007 To alter the value of a variable, evaluate an assignment expression.
19008 @xref{Expressions, ,Expressions}. For example,
19015 stores the value 4 into the variable @code{x}, and then prints the
19016 value of the assignment expression (which is 4).
19017 @xref{Languages, ,Using @value{GDBN} with Different Languages}, for more
19018 information on operators in supported languages.
19020 @kindex set variable
19021 @cindex variables, setting
19022 If you are not interested in seeing the value of the assignment, use the
19023 @code{set} command instead of the @code{print} command. @code{set} is
19024 really the same as @code{print} except that the expression's value is
19025 not printed and is not put in the value history (@pxref{Value History,
19026 ,Value History}). The expression is evaluated only for its effects.
19028 If the beginning of the argument string of the @code{set} command
19029 appears identical to a @code{set} subcommand, use the @code{set
19030 variable} command instead of just @code{set}. This command is identical
19031 to @code{set} except for its lack of subcommands. For example, if your
19032 program has a variable @code{width}, you get an error if you try to set
19033 a new value with just @samp{set width=13}, because @value{GDBN} has the
19034 command @code{set width}:
19037 (@value{GDBP}) whatis width
19039 (@value{GDBP}) p width
19041 (@value{GDBP}) set width=47
19042 Invalid syntax in expression.
19046 The invalid expression, of course, is @samp{=47}. In
19047 order to actually set the program's variable @code{width}, use
19050 (@value{GDBP}) set var width=47
19053 Because the @code{set} command has many subcommands that can conflict
19054 with the names of program variables, it is a good idea to use the
19055 @code{set variable} command instead of just @code{set}. For example, if
19056 your program has a variable @code{g}, you run into problems if you try
19057 to set a new value with just @samp{set g=4}, because @value{GDBN} has
19058 the command @code{set gnutarget}, abbreviated @code{set g}:
19062 (@value{GDBP}) whatis g
19066 (@value{GDBP}) set g=4
19070 The program being debugged has been started already.
19071 Start it from the beginning? (y or n) y
19072 Starting program: /home/smith/cc_progs/a.out
19073 "/home/smith/cc_progs/a.out": can't open to read symbols:
19074 Invalid bfd target.
19075 (@value{GDBP}) show g
19076 The current BFD target is "=4".
19081 The program variable @code{g} did not change, and you silently set the
19082 @code{gnutarget} to an invalid value. In order to set the variable
19086 (@value{GDBP}) set var g=4
19089 @value{GDBN} allows more implicit conversions in assignments than C; you can
19090 freely store an integer value into a pointer variable or vice versa,
19091 and you can convert any structure to any other structure that is the
19092 same length or shorter.
19093 @comment FIXME: how do structs align/pad in these conversions?
19094 @comment /doc@cygnus.com 18dec1990
19096 To store values into arbitrary places in memory, use the @samp{@{@dots{}@}}
19097 construct to generate a value of specified type at a specified address
19098 (@pxref{Expressions, ,Expressions}). For example, @code{@{int@}0x83040} refers
19099 to memory location @code{0x83040} as an integer (which implies a certain size
19100 and representation in memory), and
19103 set @{int@}0x83040 = 4
19107 stores the value 4 into that memory location.
19110 @section Continuing at a Different Address
19112 Ordinarily, when you continue your program, you do so at the place where
19113 it stopped, with the @code{continue} command. You can instead continue at
19114 an address of your own choosing, with the following commands:
19118 @kindex j @r{(@code{jump})}
19119 @item jump @var{location}
19120 @itemx j @var{location}
19121 Resume execution at @var{location}. Execution stops again immediately
19122 if there is a breakpoint there. @xref{Specify Location}, for a description
19123 of the different forms of @var{location}. It is common
19124 practice to use the @code{tbreak} command in conjunction with
19125 @code{jump}. @xref{Set Breaks, ,Setting Breakpoints}.
19127 The @code{jump} command does not change the current stack frame, or
19128 the stack pointer, or the contents of any memory location or any
19129 register other than the program counter. If @var{location} is in
19130 a different function from the one currently executing, the results may
19131 be bizarre if the two functions expect different patterns of arguments or
19132 of local variables. For this reason, the @code{jump} command requests
19133 confirmation if the specified line is not in the function currently
19134 executing. However, even bizarre results are predictable if you are
19135 well acquainted with the machine-language code of your program.
19138 On many systems, you can get much the same effect as the @code{jump}
19139 command by storing a new value into the register @code{$pc}. The
19140 difference is that this does not start your program running; it only
19141 changes the address of where it @emph{will} run when you continue. For
19149 makes the next @code{continue} command or stepping command execute at
19150 address @code{0x485}, rather than at the address where your program stopped.
19151 @xref{Continuing and Stepping, ,Continuing and Stepping}.
19153 The most common occasion to use the @code{jump} command is to back
19154 up---perhaps with more breakpoints set---over a portion of a program
19155 that has already executed, in order to examine its execution in more
19160 @section Giving your Program a Signal
19161 @cindex deliver a signal to a program
19165 @item signal @var{signal}
19166 Resume execution where your program is stopped, but immediately give it the
19167 signal @var{signal}. The @var{signal} can be the name or the number of a
19168 signal. For example, on many systems @code{signal 2} and @code{signal
19169 SIGINT} are both ways of sending an interrupt signal.
19171 Alternatively, if @var{signal} is zero, continue execution without
19172 giving a signal. This is useful when your program stopped on account of
19173 a signal and would ordinarily see the signal when resumed with the
19174 @code{continue} command; @samp{signal 0} causes it to resume without a
19177 @emph{Note:} When resuming a multi-threaded program, @var{signal} is
19178 delivered to the currently selected thread, not the thread that last
19179 reported a stop. This includes the situation where a thread was
19180 stopped due to a signal. So if you want to continue execution
19181 suppressing the signal that stopped a thread, you should select that
19182 same thread before issuing the @samp{signal 0} command. If you issue
19183 the @samp{signal 0} command with another thread as the selected one,
19184 @value{GDBN} detects that and asks for confirmation.
19186 Invoking the @code{signal} command is not the same as invoking the
19187 @code{kill} utility from the shell. Sending a signal with @code{kill}
19188 causes @value{GDBN} to decide what to do with the signal depending on
19189 the signal handling tables (@pxref{Signals}). The @code{signal} command
19190 passes the signal directly to your program.
19192 @code{signal} does not repeat when you press @key{RET} a second time
19193 after executing the command.
19195 @kindex queue-signal
19196 @item queue-signal @var{signal}
19197 Queue @var{signal} to be delivered immediately to the current thread
19198 when execution of the thread resumes. The @var{signal} can be the name or
19199 the number of a signal. For example, on many systems @code{signal 2} and
19200 @code{signal SIGINT} are both ways of sending an interrupt signal.
19201 The handling of the signal must be set to pass the signal to the program,
19202 otherwise @value{GDBN} will report an error.
19203 You can control the handling of signals from @value{GDBN} with the
19204 @code{handle} command (@pxref{Signals}).
19206 Alternatively, if @var{signal} is zero, any currently queued signal
19207 for the current thread is discarded and when execution resumes no signal
19208 will be delivered. This is useful when your program stopped on account
19209 of a signal and would ordinarily see the signal when resumed with the
19210 @code{continue} command.
19212 This command differs from the @code{signal} command in that the signal
19213 is just queued, execution is not resumed. And @code{queue-signal} cannot
19214 be used to pass a signal whose handling state has been set to @code{nopass}
19219 @xref{stepping into signal handlers}, for information on how stepping
19220 commands behave when the thread has a signal queued.
19223 @section Returning from a Function
19226 @cindex returning from a function
19229 @itemx return @var{expression}
19230 You can cancel execution of a function call with the @code{return}
19231 command. If you give an
19232 @var{expression} argument, its value is used as the function's return
19236 When you use @code{return}, @value{GDBN} discards the selected stack frame
19237 (and all frames within it). You can think of this as making the
19238 discarded frame return prematurely. If you wish to specify a value to
19239 be returned, give that value as the argument to @code{return}.
19241 This pops the selected stack frame (@pxref{Selection, ,Selecting a
19242 Frame}), and any other frames inside of it, leaving its caller as the
19243 innermost remaining frame. That frame becomes selected. The
19244 specified value is stored in the registers used for returning values
19247 The @code{return} command does not resume execution; it leaves the
19248 program stopped in the state that would exist if the function had just
19249 returned. In contrast, the @code{finish} command (@pxref{Continuing
19250 and Stepping, ,Continuing and Stepping}) resumes execution until the
19251 selected stack frame returns naturally.
19253 @value{GDBN} needs to know how the @var{expression} argument should be set for
19254 the inferior. The concrete registers assignment depends on the OS ABI and the
19255 type being returned by the selected stack frame. For example it is common for
19256 OS ABI to return floating point values in FPU registers while integer values in
19257 CPU registers. Still some ABIs return even floating point values in CPU
19258 registers. Larger integer widths (such as @code{long long int}) also have
19259 specific placement rules. @value{GDBN} already knows the OS ABI from its
19260 current target so it needs to find out also the type being returned to make the
19261 assignment into the right register(s).
19263 Normally, the selected stack frame has debug info. @value{GDBN} will always
19264 use the debug info instead of the implicit type of @var{expression} when the
19265 debug info is available. For example, if you type @kbd{return -1}, and the
19266 function in the current stack frame is declared to return a @code{long long
19267 int}, @value{GDBN} transparently converts the implicit @code{int} value of -1
19268 into a @code{long long int}:
19271 Breakpoint 1, func () at gdb.base/return-nodebug.c:29
19273 (@value{GDBP}) return -1
19274 Make func return now? (y or n) y
19275 #0 0x004004f6 in main () at gdb.base/return-nodebug.c:43
19276 43 printf ("result=%lld\n", func ());
19280 However, if the selected stack frame does not have a debug info, e.g., if the
19281 function was compiled without debug info, @value{GDBN} has to find out the type
19282 to return from user. Specifying a different type by mistake may set the value
19283 in different inferior registers than the caller code expects. For example,
19284 typing @kbd{return -1} with its implicit type @code{int} would set only a part
19285 of a @code{long long int} result for a debug info less function (on 32-bit
19286 architectures). Therefore the user is required to specify the return type by
19287 an appropriate cast explicitly:
19290 Breakpoint 2, 0x0040050b in func ()
19291 (@value{GDBP}) return -1
19292 Return value type not available for selected stack frame.
19293 Please use an explicit cast of the value to return.
19294 (@value{GDBP}) return (long long int) -1
19295 Make selected stack frame return now? (y or n) y
19296 #0 0x00400526 in main ()
19301 @section Calling Program Functions
19304 @cindex calling functions
19305 @cindex inferior functions, calling
19306 @item print @var{expr}
19307 Evaluate the expression @var{expr} and display the resulting value.
19308 The expression may include calls to functions in the program being
19312 @item call @var{expr}
19313 Evaluate the expression @var{expr} without displaying @code{void}
19316 You can use this variant of the @code{print} command if you want to
19317 execute a function from your program that does not return anything
19318 (a.k.a.@: @dfn{a void function}), but without cluttering the output
19319 with @code{void} returned values that @value{GDBN} will otherwise
19320 print. If the result is not void, it is printed and saved in the
19324 It is possible for the function you call via the @code{print} or
19325 @code{call} command to generate a signal (e.g., if there's a bug in
19326 the function, or if you passed it incorrect arguments). What happens
19327 in that case is controlled by the @code{set unwindonsignal} command.
19329 Similarly, with a C@t{++} program it is possible for the function you
19330 call via the @code{print} or @code{call} command to generate an
19331 exception that is not handled due to the constraints of the dummy
19332 frame. In this case, any exception that is raised in the frame, but has
19333 an out-of-frame exception handler will not be found. GDB builds a
19334 dummy-frame for the inferior function call, and the unwinder cannot
19335 seek for exception handlers outside of this dummy-frame. What happens
19336 in that case is controlled by the
19337 @code{set unwind-on-terminating-exception} command.
19340 @item set unwindonsignal
19341 @kindex set unwindonsignal
19342 @cindex unwind stack in called functions
19343 @cindex call dummy stack unwinding
19344 Set unwinding of the stack if a signal is received while in a function
19345 that @value{GDBN} called in the program being debugged. If set to on,
19346 @value{GDBN} unwinds the stack it created for the call and restores
19347 the context to what it was before the call. If set to off (the
19348 default), @value{GDBN} stops in the frame where the signal was
19351 @item show unwindonsignal
19352 @kindex show unwindonsignal
19353 Show the current setting of stack unwinding in the functions called by
19356 @item set unwind-on-terminating-exception
19357 @kindex set unwind-on-terminating-exception
19358 @cindex unwind stack in called functions with unhandled exceptions
19359 @cindex call dummy stack unwinding on unhandled exception.
19360 Set unwinding of the stack if a C@t{++} exception is raised, but left
19361 unhandled while in a function that @value{GDBN} called in the program being
19362 debugged. If set to on (the default), @value{GDBN} unwinds the stack
19363 it created for the call and restores the context to what it was before
19364 the call. If set to off, @value{GDBN} the exception is delivered to
19365 the default C@t{++} exception handler and the inferior terminated.
19367 @item show unwind-on-terminating-exception
19368 @kindex show unwind-on-terminating-exception
19369 Show the current setting of stack unwinding in the functions called by
19372 @item set may-call-functions
19373 @kindex set may-call-functions
19374 @cindex disabling calling functions in the program
19375 @cindex calling functions in the program, disabling
19376 Set permission to call functions in the program.
19377 This controls whether @value{GDBN} will attempt to call functions in
19378 the program, such as with expressions in the @code{print} command. It
19379 defaults to @code{on}.
19381 To call a function in the program, @value{GDBN} has to temporarily
19382 modify the state of the inferior. This has potentially undesired side
19383 effects. Also, having @value{GDBN} call nested functions is likely to
19384 be erroneous and may even crash the program being debugged. You can
19385 avoid such hazards by forbidding @value{GDBN} from calling functions
19386 in the program being debugged. If calling functions in the program
19387 is forbidden, GDB will throw an error when a command (such as printing
19388 an expression) starts a function call in the program.
19390 @item show may-call-functions
19391 @kindex show may-call-functions
19392 Show permission to call functions in the program.
19396 @subsection Calling functions with no debug info
19398 @cindex no debug info functions
19399 Sometimes, a function you wish to call is missing debug information.
19400 In such case, @value{GDBN} does not know the type of the function,
19401 including the types of the function's parameters. To avoid calling
19402 the inferior function incorrectly, which could result in the called
19403 function functioning erroneously and even crash, @value{GDBN} refuses
19404 to call the function unless you tell it the type of the function.
19406 For prototyped (i.e.@: ANSI/ISO style) functions, there are two ways
19407 to do that. The simplest is to cast the call to the function's
19408 declared return type. For example:
19411 (@value{GDBP}) p getenv ("PATH")
19412 'getenv' has unknown return type; cast the call to its declared return type
19413 (@value{GDBP}) p (char *) getenv ("PATH")
19414 $1 = 0x7fffffffe7ba "/usr/local/bin:/"...
19417 Casting the return type of a no-debug function is equivalent to
19418 casting the function to a pointer to a prototyped function that has a
19419 prototype that matches the types of the passed-in arguments, and
19420 calling that. I.e., the call above is equivalent to:
19423 (@value{GDBP}) p ((char * (*) (const char *)) getenv) ("PATH")
19427 and given this prototyped C or C++ function with float parameters:
19430 float multiply (float v1, float v2) @{ return v1 * v2; @}
19434 these calls are equivalent:
19437 (@value{GDBP}) p (float) multiply (2.0f, 3.0f)
19438 (@value{GDBP}) p ((float (*) (float, float)) multiply) (2.0f, 3.0f)
19441 If the function you wish to call is declared as unprototyped (i.e.@:
19442 old K&R style), you must use the cast-to-function-pointer syntax, so
19443 that @value{GDBN} knows that it needs to apply default argument
19444 promotions (promote float arguments to double). @xref{ABI, float
19445 promotion}. For example, given this unprototyped C function with
19446 float parameters, and no debug info:
19450 multiply_noproto (v1, v2)
19458 you call it like this:
19461 (@value{GDBP}) p ((float (*) ()) multiply_noproto) (2.0f, 3.0f)
19465 @section Patching Programs
19467 @cindex patching binaries
19468 @cindex writing into executables
19469 @cindex writing into corefiles
19471 By default, @value{GDBN} opens the file containing your program's
19472 executable code (or the corefile) read-only. This prevents accidental
19473 alterations to machine code; but it also prevents you from intentionally
19474 patching your program's binary.
19476 If you'd like to be able to patch the binary, you can specify that
19477 explicitly with the @code{set write} command. For example, you might
19478 want to turn on internal debugging flags, or even to make emergency
19484 @itemx set write off
19485 If you specify @samp{set write on}, @value{GDBN} opens executable and
19486 core files for both reading and writing; if you specify @kbd{set write
19487 off} (the default), @value{GDBN} opens them read-only.
19489 If you have already loaded a file, you must load it again (using the
19490 @code{exec-file} or @code{core-file} command) after changing @code{set
19491 write}, for your new setting to take effect.
19495 Display whether executable files and core files are opened for writing
19496 as well as reading.
19499 @node Compiling and Injecting Code
19500 @section Compiling and injecting code in @value{GDBN}
19501 @cindex injecting code
19502 @cindex writing into executables
19503 @cindex compiling code
19505 @value{GDBN} supports on-demand compilation and code injection into
19506 programs running under @value{GDBN}. GCC 5.0 or higher built with
19507 @file{libcc1.so} must be installed for this functionality to be enabled.
19508 This functionality is implemented with the following commands.
19511 @kindex compile code
19512 @item compile code @var{source-code}
19513 @itemx compile code -raw @var{--} @var{source-code}
19514 Compile @var{source-code} with the compiler language found as the current
19515 language in @value{GDBN} (@pxref{Languages}). If compilation and
19516 injection is not supported with the current language specified in
19517 @value{GDBN}, or the compiler does not support this feature, an error
19518 message will be printed. If @var{source-code} compiles and links
19519 successfully, @value{GDBN} will load the object-code emitted,
19520 and execute it within the context of the currently selected inferior.
19521 It is important to note that the compiled code is executed immediately.
19522 After execution, the compiled code is removed from @value{GDBN} and any
19523 new types or variables you have defined will be deleted.
19525 The command allows you to specify @var{source-code} in two ways.
19526 The simplest method is to provide a single line of code to the command.
19530 compile code printf ("hello world\n");
19533 If you specify options on the command line as well as source code, they
19534 may conflict. The @samp{--} delimiter can be used to separate options
19535 from actual source code. E.g.:
19538 compile code -r -- printf ("hello world\n");
19541 Alternatively you can enter source code as multiple lines of text. To
19542 enter this mode, invoke the @samp{compile code} command without any text
19543 following the command. This will start the multiple-line editor and
19544 allow you to type as many lines of source code as required. When you
19545 have completed typing, enter @samp{end} on its own line to exit the
19550 >printf ("hello\n");
19551 >printf ("world\n");
19555 Specifying @samp{-raw}, prohibits @value{GDBN} from wrapping the
19556 provided @var{source-code} in a callable scope. In this case, you must
19557 specify the entry point of the code by defining a function named
19558 @code{_gdb_expr_}. The @samp{-raw} code cannot access variables of the
19559 inferior. Using @samp{-raw} option may be needed for example when
19560 @var{source-code} requires @samp{#include} lines which may conflict with
19561 inferior symbols otherwise.
19563 @kindex compile file
19564 @item compile file @var{filename}
19565 @itemx compile file -raw @var{filename}
19566 Like @code{compile code}, but take the source code from @var{filename}.
19569 compile file /home/user/example.c
19574 @item compile print [[@var{options}] --] @var{expr}
19575 @itemx compile print [[@var{options}] --] /@var{f} @var{expr}
19576 Compile and execute @var{expr} with the compiler language found as the
19577 current language in @value{GDBN} (@pxref{Languages}). By default the
19578 value of @var{expr} is printed in a format appropriate to its data type;
19579 you can choose a different format by specifying @samp{/@var{f}}, where
19580 @var{f} is a letter specifying the format; see @ref{Output Formats,,Output
19581 Formats}. The @code{compile print} command accepts the same options
19582 as the @code{print} command; see @ref{print options}.
19584 @item compile print [[@var{options}] --]
19585 @itemx compile print [[@var{options}] --] /@var{f}
19586 @cindex reprint the last value
19587 Alternatively you can enter the expression (source code producing it) as
19588 multiple lines of text. To enter this mode, invoke the @samp{compile print}
19589 command without any text following the command. This will start the
19590 multiple-line editor.
19594 The process of compiling and injecting the code can be inspected using:
19597 @anchor{set debug compile}
19598 @item set debug compile
19599 @cindex compile command debugging info
19600 Turns on or off display of @value{GDBN} process of compiling and
19601 injecting the code. The default is off.
19603 @item show debug compile
19604 Displays the current state of displaying @value{GDBN} process of
19605 compiling and injecting the code.
19607 @anchor{set debug compile-cplus-types}
19608 @item set debug compile-cplus-types
19609 @cindex compile C@t{++} type conversion
19610 Turns on or off the display of C@t{++} type conversion debugging information.
19611 The default is off.
19613 @item show debug compile-cplus-types
19614 Displays the current state of displaying debugging information for
19615 C@t{++} type conversion.
19618 @subsection Compilation options for the @code{compile} command
19620 @value{GDBN} needs to specify the right compilation options for the code
19621 to be injected, in part to make its ABI compatible with the inferior
19622 and in part to make the injected code compatible with @value{GDBN}'s
19626 The options used, in increasing precedence:
19629 @item target architecture and OS options (@code{gdbarch})
19630 These options depend on target processor type and target operating
19631 system, usually they specify at least 32-bit (@code{-m32}) or 64-bit
19632 (@code{-m64}) compilation option.
19634 @item compilation options recorded in the target
19635 @value{NGCC} (since version 4.7) stores the options used for compilation
19636 into @code{DW_AT_producer} part of DWARF debugging information according
19637 to the @value{NGCC} option @code{-grecord-gcc-switches}. One has to
19638 explicitly specify @code{-g} during inferior compilation otherwise
19639 @value{NGCC} produces no DWARF. This feature is only relevant for
19640 platforms where @code{-g} produces DWARF by default, otherwise one may
19641 try to enforce DWARF by using @code{-gdwarf-4}.
19643 @item compilation options set by @code{set compile-args}
19647 You can override compilation options using the following command:
19650 @item set compile-args
19651 @cindex compile command options override
19652 Set compilation options used for compiling and injecting code with the
19653 @code{compile} commands. These options override any conflicting ones
19654 from the target architecture and/or options stored during inferior
19657 @item show compile-args
19658 Displays the current state of compilation options override.
19659 This does not show all the options actually used during compilation,
19660 use @ref{set debug compile} for that.
19663 @subsection Caveats when using the @code{compile} command
19665 There are a few caveats to keep in mind when using the @code{compile}
19666 command. As the caveats are different per language, the table below
19667 highlights specific issues on a per language basis.
19670 @item C code examples and caveats
19671 When the language in @value{GDBN} is set to @samp{C}, the compiler will
19672 attempt to compile the source code with a @samp{C} compiler. The source
19673 code provided to the @code{compile} command will have much the same
19674 access to variables and types as it normally would if it were part of
19675 the program currently being debugged in @value{GDBN}.
19677 Below is a sample program that forms the basis of the examples that
19678 follow. This program has been compiled and loaded into @value{GDBN},
19679 much like any other normal debugging session.
19682 void function1 (void)
19685 printf ("function 1\n");
19688 void function2 (void)
19703 For the purposes of the examples in this section, the program above has
19704 been compiled, loaded into @value{GDBN}, stopped at the function
19705 @code{main}, and @value{GDBN} is awaiting input from the user.
19707 To access variables and types for any program in @value{GDBN}, the
19708 program must be compiled and packaged with debug information. The
19709 @code{compile} command is not an exception to this rule. Without debug
19710 information, you can still use the @code{compile} command, but you will
19711 be very limited in what variables and types you can access.
19713 So with that in mind, the example above has been compiled with debug
19714 information enabled. The @code{compile} command will have access to
19715 all variables and types (except those that may have been optimized
19716 out). Currently, as @value{GDBN} has stopped the program in the
19717 @code{main} function, the @code{compile} command would have access to
19718 the variable @code{k}. You could invoke the @code{compile} command
19719 and type some source code to set the value of @code{k}. You can also
19720 read it, or do anything with that variable you would normally do in
19721 @code{C}. Be aware that changes to inferior variables in the
19722 @code{compile} command are persistent. In the following example:
19725 compile code k = 3;
19729 the variable @code{k} is now 3. It will retain that value until
19730 something else in the example program changes it, or another
19731 @code{compile} command changes it.
19733 Normal scope and access rules apply to source code compiled and
19734 injected by the @code{compile} command. In the example, the variables
19735 @code{j} and @code{k} are not accessible yet, because the program is
19736 currently stopped in the @code{main} function, where these variables
19737 are not in scope. Therefore, the following command
19740 compile code j = 3;
19744 will result in a compilation error message.
19746 Once the program is continued, execution will bring these variables in
19747 scope, and they will become accessible; then the code you specify via
19748 the @code{compile} command will be able to access them.
19750 You can create variables and types with the @code{compile} command as
19751 part of your source code. Variables and types that are created as part
19752 of the @code{compile} command are not visible to the rest of the program for
19753 the duration of its run. This example is valid:
19756 compile code int ff = 5; printf ("ff is %d\n", ff);
19759 However, if you were to type the following into @value{GDBN} after that
19760 command has completed:
19763 compile code printf ("ff is %d\n'', ff);
19767 a compiler error would be raised as the variable @code{ff} no longer
19768 exists. Object code generated and injected by the @code{compile}
19769 command is removed when its execution ends. Caution is advised
19770 when assigning to program variables values of variables created by the
19771 code submitted to the @code{compile} command. This example is valid:
19774 compile code int ff = 5; k = ff;
19777 The value of the variable @code{ff} is assigned to @code{k}. The variable
19778 @code{k} does not require the existence of @code{ff} to maintain the value
19779 it has been assigned. However, pointers require particular care in
19780 assignment. If the source code compiled with the @code{compile} command
19781 changed the address of a pointer in the example program, perhaps to a
19782 variable created in the @code{compile} command, that pointer would point
19783 to an invalid location when the command exits. The following example
19784 would likely cause issues with your debugged program:
19787 compile code int ff = 5; p = &ff;
19790 In this example, @code{p} would point to @code{ff} when the
19791 @code{compile} command is executing the source code provided to it.
19792 However, as variables in the (example) program persist with their
19793 assigned values, the variable @code{p} would point to an invalid
19794 location when the command exists. A general rule should be followed
19795 in that you should either assign @code{NULL} to any assigned pointers,
19796 or restore a valid location to the pointer before the command exits.
19798 Similar caution must be exercised with any structs, unions, and typedefs
19799 defined in @code{compile} command. Types defined in the @code{compile}
19800 command will no longer be available in the next @code{compile} command.
19801 Therefore, if you cast a variable to a type defined in the
19802 @code{compile} command, care must be taken to ensure that any future
19803 need to resolve the type can be achieved.
19806 (gdb) compile code static struct a @{ int a; @} v = @{ 42 @}; argv = &v;
19807 (gdb) compile code printf ("%d\n", ((struct a *) argv)->a);
19808 gdb command line:1:36: error: dereferencing pointer to incomplete type ‘struct a’
19809 Compilation failed.
19810 (gdb) compile code struct a @{ int a; @}; printf ("%d\n", ((struct a *) argv)->a);
19814 Variables that have been optimized away by the compiler are not
19815 accessible to the code submitted to the @code{compile} command.
19816 Access to those variables will generate a compiler error which @value{GDBN}
19817 will print to the console.
19820 @subsection Compiler search for the @code{compile} command
19822 @value{GDBN} needs to find @value{NGCC} for the inferior being debugged
19823 which may not be obvious for remote targets of different architecture
19824 than where @value{GDBN} is running. Environment variable @code{PATH} on
19825 @value{GDBN} host is searched for @value{NGCC} binary matching the
19826 target architecture and operating system. This search can be overriden
19827 by @code{set compile-gcc} @value{GDBN} command below. @code{PATH} is
19828 taken from shell that executed @value{GDBN}, it is not the value set by
19829 @value{GDBN} command @code{set environment}). @xref{Environment}.
19832 Specifically @code{PATH} is searched for binaries matching regular expression
19833 @code{@var{arch}(-[^-]*)?-@var{os}-gcc} according to the inferior target being
19834 debugged. @var{arch} is processor name --- multiarch is supported, so for
19835 example both @code{i386} and @code{x86_64} targets look for pattern
19836 @code{(x86_64|i.86)} and both @code{s390} and @code{s390x} targets look
19837 for pattern @code{s390x?}. @var{os} is currently supported only for
19838 pattern @code{linux(-gnu)?}.
19840 On Posix hosts the compiler driver @value{GDBN} needs to find also
19841 shared library @file{libcc1.so} from the compiler. It is searched in
19842 default shared library search path (overridable with usual environment
19843 variable @code{LD_LIBRARY_PATH}), unrelated to @code{PATH} or @code{set
19844 compile-gcc} settings. Contrary to it @file{libcc1plugin.so} is found
19845 according to the installation of the found compiler --- as possibly
19846 specified by the @code{set compile-gcc} command.
19849 @item set compile-gcc
19850 @cindex compile command driver filename override
19851 Set compilation command used for compiling and injecting code with the
19852 @code{compile} commands. If this option is not set (it is set to
19853 an empty string), the search described above will occur --- that is the
19856 @item show compile-gcc
19857 Displays the current compile command @value{NGCC} driver filename.
19858 If set, it is the main command @command{gcc}, found usually for example
19859 under name @file{x86_64-linux-gnu-gcc}.
19863 @chapter @value{GDBN} Files
19865 @value{GDBN} needs to know the file name of the program to be debugged,
19866 both in order to read its symbol table and in order to start your
19867 program. To debug a core dump of a previous run, you must also tell
19868 @value{GDBN} the name of the core dump file.
19871 * Files:: Commands to specify files
19872 * File Caching:: Information about @value{GDBN}'s file caching
19873 * Separate Debug Files:: Debugging information in separate files
19874 * MiniDebugInfo:: Debugging information in a special section
19875 * Index Files:: Index files speed up GDB
19876 * Symbol Errors:: Errors reading symbol files
19877 * Data Files:: GDB data files
19881 @section Commands to Specify Files
19883 @cindex symbol table
19884 @cindex core dump file
19886 You may want to specify executable and core dump file names. The usual
19887 way to do this is at start-up time, using the arguments to
19888 @value{GDBN}'s start-up commands (@pxref{Invocation, , Getting In and
19889 Out of @value{GDBN}}).
19891 Occasionally it is necessary to change to a different file during a
19892 @value{GDBN} session. Or you may run @value{GDBN} and forget to
19893 specify a file you want to use. Or you are debugging a remote target
19894 via @code{gdbserver} (@pxref{Server, file, Using the @code{gdbserver}
19895 Program}). In these situations the @value{GDBN} commands to specify
19896 new files are useful.
19899 @cindex executable file
19901 @item file @var{filename}
19902 Use @var{filename} as the program to be debugged. It is read for its
19903 symbols and for the contents of pure memory. It is also the program
19904 executed when you use the @code{run} command. If you do not specify a
19905 directory and the file is not found in the @value{GDBN} working directory,
19906 @value{GDBN} uses the environment variable @code{PATH} as a list of
19907 directories to search, just as the shell does when looking for a program
19908 to run. You can change the value of this variable, for both @value{GDBN}
19909 and your program, using the @code{path} command.
19911 @cindex unlinked object files
19912 @cindex patching object files
19913 You can load unlinked object @file{.o} files into @value{GDBN} using
19914 the @code{file} command. You will not be able to ``run'' an object
19915 file, but you can disassemble functions and inspect variables. Also,
19916 if the underlying BFD functionality supports it, you could use
19917 @kbd{gdb -write} to patch object files using this technique. Note
19918 that @value{GDBN} can neither interpret nor modify relocations in this
19919 case, so branches and some initialized variables will appear to go to
19920 the wrong place. But this feature is still handy from time to time.
19923 @code{file} with no argument makes @value{GDBN} discard any information it
19924 has on both executable file and the symbol table.
19927 @item exec-file @r{[} @var{filename} @r{]}
19928 Specify that the program to be run (but not the symbol table) is found
19929 in @var{filename}. @value{GDBN} searches the environment variable @code{PATH}
19930 if necessary to locate your program. Omitting @var{filename} means to
19931 discard information on the executable file.
19933 @kindex symbol-file
19934 @item symbol-file @r{[} @var{filename} @r{[} -o @var{offset} @r{]]}
19935 Read symbol table information from file @var{filename}. @code{PATH} is
19936 searched when necessary. Use the @code{file} command to get both symbol
19937 table and program to run from the same file.
19939 If an optional @var{offset} is specified, it is added to the start
19940 address of each section in the symbol file. This is useful if the
19941 program is relocated at runtime, such as the Linux kernel with kASLR
19944 @code{symbol-file} with no argument clears out @value{GDBN} information on your
19945 program's symbol table.
19947 The @code{symbol-file} command causes @value{GDBN} to forget the contents of
19948 some breakpoints and auto-display expressions. This is because they may
19949 contain pointers to the internal data recording symbols and data types,
19950 which are part of the old symbol table data being discarded inside
19953 @code{symbol-file} does not repeat if you press @key{RET} again after
19956 When @value{GDBN} is configured for a particular environment, it
19957 understands debugging information in whatever format is the standard
19958 generated for that environment; you may use either a @sc{gnu} compiler, or
19959 other compilers that adhere to the local conventions.
19960 Best results are usually obtained from @sc{gnu} compilers; for example,
19961 using @code{@value{NGCC}} you can generate debugging information for
19964 For most kinds of object files, with the exception of old SVR3 systems
19965 using COFF, the @code{symbol-file} command does not normally read the
19966 symbol table in full right away. Instead, it scans the symbol table
19967 quickly to find which source files and which symbols are present. The
19968 details are read later, one source file at a time, as they are needed.
19970 The purpose of this two-stage reading strategy is to make @value{GDBN}
19971 start up faster. For the most part, it is invisible except for
19972 occasional pauses while the symbol table details for a particular source
19973 file are being read. (The @code{set verbose} command can turn these
19974 pauses into messages if desired. @xref{Messages/Warnings, ,Optional
19975 Warnings and Messages}.)
19977 We have not implemented the two-stage strategy for COFF yet. When the
19978 symbol table is stored in COFF format, @code{symbol-file} reads the
19979 symbol table data in full right away. Note that ``stabs-in-COFF''
19980 still does the two-stage strategy, since the debug info is actually
19984 @cindex reading symbols immediately
19985 @cindex symbols, reading immediately
19986 @item symbol-file @r{[} -readnow @r{]} @var{filename}
19987 @itemx file @r{[} -readnow @r{]} @var{filename}
19988 You can override the @value{GDBN} two-stage strategy for reading symbol
19989 tables by using the @samp{-readnow} option with any of the commands that
19990 load symbol table information, if you want to be sure @value{GDBN} has the
19991 entire symbol table available.
19993 @cindex @code{-readnever}, option for symbol-file command
19994 @cindex never read symbols
19995 @cindex symbols, never read
19996 @item symbol-file @r{[} -readnever @r{]} @var{filename}
19997 @itemx file @r{[} -readnever @r{]} @var{filename}
19998 You can instruct @value{GDBN} to never read the symbolic information
19999 contained in @var{filename} by using the @samp{-readnever} option.
20000 @xref{--readnever}.
20002 @c FIXME: for now no mention of directories, since this seems to be in
20003 @c flux. 13mar1992 status is that in theory GDB would look either in
20004 @c current dir or in same dir as myprog; but issues like competing
20005 @c GDB's, or clutter in system dirs, mean that in practice right now
20006 @c only current dir is used. FFish says maybe a special GDB hierarchy
20007 @c (eg rooted in val of env var GDBSYMS) could exist for mappable symbol
20011 @item core-file @r{[}@var{filename}@r{]}
20013 Specify the whereabouts of a core dump file to be used as the ``contents
20014 of memory''. Traditionally, core files contain only some parts of the
20015 address space of the process that generated them; @value{GDBN} can access the
20016 executable file itself for other parts.
20018 @code{core-file} with no argument specifies that no core file is
20021 Note that the core file is ignored when your program is actually running
20022 under @value{GDBN}. So, if you have been running your program and you
20023 wish to debug a core file instead, you must kill the subprocess in which
20024 the program is running. To do this, use the @code{kill} command
20025 (@pxref{Kill Process, ,Killing the Child Process}).
20027 @kindex add-symbol-file
20028 @cindex dynamic linking
20029 @item add-symbol-file @var{filename} @r{[} -readnow @r{|} -readnever @r{]} @r{[} -o @var{offset} @r{]} @r{[} @var{textaddress} @r{]} @r{[} -s @var{section} @var{address} @dots{} @r{]}
20030 The @code{add-symbol-file} command reads additional symbol table
20031 information from the file @var{filename}. You would use this command
20032 when @var{filename} has been dynamically loaded (by some other means)
20033 into the program that is running. The @var{textaddress} parameter gives
20034 the memory address at which the file's text section has been loaded.
20035 You can additionally specify the base address of other sections using
20036 an arbitrary number of @samp{-s @var{section} @var{address}} pairs.
20037 If a section is omitted, @value{GDBN} will use its default addresses
20038 as found in @var{filename}. Any @var{address} or @var{textaddress}
20039 can be given as an expression.
20041 If an optional @var{offset} is specified, it is added to the start
20042 address of each section, except those for which the address was
20043 specified explicitly.
20045 The symbol table of the file @var{filename} is added to the symbol table
20046 originally read with the @code{symbol-file} command. You can use the
20047 @code{add-symbol-file} command any number of times; the new symbol data
20048 thus read is kept in addition to the old.
20050 Changes can be reverted using the command @code{remove-symbol-file}.
20052 @cindex relocatable object files, reading symbols from
20053 @cindex object files, relocatable, reading symbols from
20054 @cindex reading symbols from relocatable object files
20055 @cindex symbols, reading from relocatable object files
20056 @cindex @file{.o} files, reading symbols from
20057 Although @var{filename} is typically a shared library file, an
20058 executable file, or some other object file which has been fully
20059 relocated for loading into a process, you can also load symbolic
20060 information from relocatable @file{.o} files, as long as:
20064 the file's symbolic information refers only to linker symbols defined in
20065 that file, not to symbols defined by other object files,
20067 every section the file's symbolic information refers to has actually
20068 been loaded into the inferior, as it appears in the file, and
20070 you can determine the address at which every section was loaded, and
20071 provide these to the @code{add-symbol-file} command.
20075 Some embedded operating systems, like Sun Chorus and VxWorks, can load
20076 relocatable files into an already running program; such systems
20077 typically make the requirements above easy to meet. However, it's
20078 important to recognize that many native systems use complex link
20079 procedures (@code{.linkonce} section factoring and C@t{++} constructor table
20080 assembly, for example) that make the requirements difficult to meet. In
20081 general, one cannot assume that using @code{add-symbol-file} to read a
20082 relocatable object file's symbolic information will have the same effect
20083 as linking the relocatable object file into the program in the normal
20086 @code{add-symbol-file} does not repeat if you press @key{RET} after using it.
20088 @kindex remove-symbol-file
20089 @item remove-symbol-file @var{filename}
20090 @item remove-symbol-file -a @var{address}
20091 Remove a symbol file added via the @code{add-symbol-file} command. The
20092 file to remove can be identified by its @var{filename} or by an @var{address}
20093 that lies within the boundaries of this symbol file in memory. Example:
20096 (gdb) add-symbol-file /home/user/gdb/mylib.so 0x7ffff7ff9480
20097 add symbol table from file "/home/user/gdb/mylib.so" at
20098 .text_addr = 0x7ffff7ff9480
20100 Reading symbols from /home/user/gdb/mylib.so...done.
20101 (gdb) remove-symbol-file -a 0x7ffff7ff9480
20102 Remove symbol table from file "/home/user/gdb/mylib.so"? (y or n) y
20107 @code{remove-symbol-file} does not repeat if you press @key{RET} after using it.
20109 @kindex add-symbol-file-from-memory
20110 @cindex @code{syscall DSO}
20111 @cindex load symbols from memory
20112 @item add-symbol-file-from-memory @var{address}
20113 Load symbols from the given @var{address} in a dynamically loaded
20114 object file whose image is mapped directly into the inferior's memory.
20115 For example, the Linux kernel maps a @code{syscall DSO} into each
20116 process's address space; this DSO provides kernel-specific code for
20117 some system calls. The argument can be any expression whose
20118 evaluation yields the address of the file's shared object file header.
20119 For this command to work, you must have used @code{symbol-file} or
20120 @code{exec-file} commands in advance.
20123 @item section @var{section} @var{addr}
20124 The @code{section} command changes the base address of the named
20125 @var{section} of the exec file to @var{addr}. This can be used if the
20126 exec file does not contain section addresses, (such as in the
20127 @code{a.out} format), or when the addresses specified in the file
20128 itself are wrong. Each section must be changed separately. The
20129 @code{info files} command, described below, lists all the sections and
20133 @kindex info target
20136 @code{info files} and @code{info target} are synonymous; both print the
20137 current target (@pxref{Targets, ,Specifying a Debugging Target}),
20138 including the names of the executable and core dump files currently in
20139 use by @value{GDBN}, and the files from which symbols were loaded. The
20140 command @code{help target} lists all possible targets rather than
20143 @kindex maint info sections
20144 @item maint info sections
20145 Another command that can give you extra information about program sections
20146 is @code{maint info sections}. In addition to the section information
20147 displayed by @code{info files}, this command displays the flags and file
20148 offset of each section in the executable and core dump files. In addition,
20149 @code{maint info sections} provides the following command options (which
20150 may be arbitrarily combined):
20154 Display sections for all loaded object files, including shared libraries.
20155 @item @var{sections}
20156 Display info only for named @var{sections}.
20157 @item @var{section-flags}
20158 Display info only for sections for which @var{section-flags} are true.
20159 The section flags that @value{GDBN} currently knows about are:
20162 Section will have space allocated in the process when loaded.
20163 Set for all sections except those containing debug information.
20165 Section will be loaded from the file into the child process memory.
20166 Set for pre-initialized code and data, clear for @code{.bss} sections.
20168 Section needs to be relocated before loading.
20170 Section cannot be modified by the child process.
20172 Section contains executable code only.
20174 Section contains data only (no executable code).
20176 Section will reside in ROM.
20178 Section contains data for constructor/destructor lists.
20180 Section is not empty.
20182 An instruction to the linker to not output the section.
20183 @item COFF_SHARED_LIBRARY
20184 A notification to the linker that the section contains
20185 COFF shared library information.
20187 Section contains common symbols.
20190 @kindex set trust-readonly-sections
20191 @cindex read-only sections
20192 @item set trust-readonly-sections on
20193 Tell @value{GDBN} that readonly sections in your object file
20194 really are read-only (i.e.@: that their contents will not change).
20195 In that case, @value{GDBN} can fetch values from these sections
20196 out of the object file, rather than from the target program.
20197 For some targets (notably embedded ones), this can be a significant
20198 enhancement to debugging performance.
20200 The default is off.
20202 @item set trust-readonly-sections off
20203 Tell @value{GDBN} not to trust readonly sections. This means that
20204 the contents of the section might change while the program is running,
20205 and must therefore be fetched from the target when needed.
20207 @item show trust-readonly-sections
20208 Show the current setting of trusting readonly sections.
20211 All file-specifying commands allow both absolute and relative file names
20212 as arguments. @value{GDBN} always converts the file name to an absolute file
20213 name and remembers it that way.
20215 @cindex shared libraries
20216 @anchor{Shared Libraries}
20217 @value{GDBN} supports @sc{gnu}/Linux, MS-Windows, SunOS,
20218 Darwin/Mach-O, SVr4, IBM RS/6000 AIX, QNX Neutrino, FDPIC (FR-V), and
20219 DSBT (TIC6X) shared libraries.
20221 On MS-Windows @value{GDBN} must be linked with the Expat library to support
20222 shared libraries. @xref{Expat}.
20224 @value{GDBN} automatically loads symbol definitions from shared libraries
20225 when you use the @code{run} command, or when you examine a core file.
20226 (Before you issue the @code{run} command, @value{GDBN} does not understand
20227 references to a function in a shared library, however---unless you are
20228 debugging a core file).
20230 @c FIXME: some @value{GDBN} release may permit some refs to undef
20231 @c FIXME...symbols---eg in a break cmd---assuming they are from a shared
20232 @c FIXME...lib; check this from time to time when updating manual
20234 There are times, however, when you may wish to not automatically load
20235 symbol definitions from shared libraries, such as when they are
20236 particularly large or there are many of them.
20238 To control the automatic loading of shared library symbols, use the
20242 @kindex set auto-solib-add
20243 @item set auto-solib-add @var{mode}
20244 If @var{mode} is @code{on}, symbols from all shared object libraries
20245 will be loaded automatically when the inferior begins execution, you
20246 attach to an independently started inferior, or when the dynamic linker
20247 informs @value{GDBN} that a new library has been loaded. If @var{mode}
20248 is @code{off}, symbols must be loaded manually, using the
20249 @code{sharedlibrary} command. The default value is @code{on}.
20251 @cindex memory used for symbol tables
20252 If your program uses lots of shared libraries with debug info that
20253 takes large amounts of memory, you can decrease the @value{GDBN}
20254 memory footprint by preventing it from automatically loading the
20255 symbols from shared libraries. To that end, type @kbd{set
20256 auto-solib-add off} before running the inferior, then load each
20257 library whose debug symbols you do need with @kbd{sharedlibrary
20258 @var{regexp}}, where @var{regexp} is a regular expression that matches
20259 the libraries whose symbols you want to be loaded.
20261 @kindex show auto-solib-add
20262 @item show auto-solib-add
20263 Display the current autoloading mode.
20266 @cindex load shared library
20267 To explicitly load shared library symbols, use the @code{sharedlibrary}
20271 @kindex info sharedlibrary
20273 @item info share @var{regex}
20274 @itemx info sharedlibrary @var{regex}
20275 Print the names of the shared libraries which are currently loaded
20276 that match @var{regex}. If @var{regex} is omitted then print
20277 all shared libraries that are loaded.
20280 @item info dll @var{regex}
20281 This is an alias of @code{info sharedlibrary}.
20283 @kindex sharedlibrary
20285 @item sharedlibrary @var{regex}
20286 @itemx share @var{regex}
20287 Load shared object library symbols for files matching a
20288 Unix regular expression.
20289 As with files loaded automatically, it only loads shared libraries
20290 required by your program for a core file or after typing @code{run}. If
20291 @var{regex} is omitted all shared libraries required by your program are
20294 @item nosharedlibrary
20295 @kindex nosharedlibrary
20296 @cindex unload symbols from shared libraries
20297 Unload all shared object library symbols. This discards all symbols
20298 that have been loaded from all shared libraries. Symbols from shared
20299 libraries that were loaded by explicit user requests are not
20303 Sometimes you may wish that @value{GDBN} stops and gives you control
20304 when any of shared library events happen. The best way to do this is
20305 to use @code{catch load} and @code{catch unload} (@pxref{Set
20308 @value{GDBN} also supports the the @code{set stop-on-solib-events}
20309 command for this. This command exists for historical reasons. It is
20310 less useful than setting a catchpoint, because it does not allow for
20311 conditions or commands as a catchpoint does.
20314 @item set stop-on-solib-events
20315 @kindex set stop-on-solib-events
20316 This command controls whether @value{GDBN} should give you control
20317 when the dynamic linker notifies it about some shared library event.
20318 The most common event of interest is loading or unloading of a new
20321 @item show stop-on-solib-events
20322 @kindex show stop-on-solib-events
20323 Show whether @value{GDBN} stops and gives you control when shared
20324 library events happen.
20327 Shared libraries are also supported in many cross or remote debugging
20328 configurations. @value{GDBN} needs to have access to the target's libraries;
20329 this can be accomplished either by providing copies of the libraries
20330 on the host system, or by asking @value{GDBN} to automatically retrieve the
20331 libraries from the target. If copies of the target libraries are
20332 provided, they need to be the same as the target libraries, although the
20333 copies on the target can be stripped as long as the copies on the host are
20336 @cindex where to look for shared libraries
20337 For remote debugging, you need to tell @value{GDBN} where the target
20338 libraries are, so that it can load the correct copies---otherwise, it
20339 may try to load the host's libraries. @value{GDBN} has two variables
20340 to specify the search directories for target libraries.
20343 @cindex prefix for executable and shared library file names
20344 @cindex system root, alternate
20345 @kindex set solib-absolute-prefix
20346 @kindex set sysroot
20347 @item set sysroot @var{path}
20348 Use @var{path} as the system root for the program being debugged. Any
20349 absolute shared library paths will be prefixed with @var{path}; many
20350 runtime loaders store the absolute paths to the shared library in the
20351 target program's memory. When starting processes remotely, and when
20352 attaching to already-running processes (local or remote), their
20353 executable filenames will be prefixed with @var{path} if reported to
20354 @value{GDBN} as absolute by the operating system. If you use
20355 @code{set sysroot} to find executables and shared libraries, they need
20356 to be laid out in the same way that they are on the target, with
20357 e.g.@: a @file{/bin}, @file{/lib} and @file{/usr/lib} hierarchy under
20360 If @var{path} starts with the sequence @file{target:} and the target
20361 system is remote then @value{GDBN} will retrieve the target binaries
20362 from the remote system. This is only supported when using a remote
20363 target that supports the @code{remote get} command (@pxref{File
20364 Transfer,,Sending files to a remote system}). The part of @var{path}
20365 following the initial @file{target:} (if present) is used as system
20366 root prefix on the remote file system. If @var{path} starts with the
20367 sequence @file{remote:} this is converted to the sequence
20368 @file{target:} by @code{set sysroot}@footnote{Historically the
20369 functionality to retrieve binaries from the remote system was
20370 provided by prefixing @var{path} with @file{remote:}}. If you want
20371 to specify a local system root using a directory that happens to be
20372 named @file{target:} or @file{remote:}, you need to use some
20373 equivalent variant of the name like @file{./target:}.
20375 For targets with an MS-DOS based filesystem, such as MS-Windows and
20376 SymbianOS, @value{GDBN} tries prefixing a few variants of the target
20377 absolute file name with @var{path}. But first, on Unix hosts,
20378 @value{GDBN} converts all backslash directory separators into forward
20379 slashes, because the backslash is not a directory separator on Unix:
20382 c:\foo\bar.dll @result{} c:/foo/bar.dll
20385 Then, @value{GDBN} attempts prefixing the target file name with
20386 @var{path}, and looks for the resulting file name in the host file
20390 c:/foo/bar.dll @result{} /path/to/sysroot/c:/foo/bar.dll
20393 If that does not find the binary, @value{GDBN} tries removing
20394 the @samp{:} character from the drive spec, both for convenience, and,
20395 for the case of the host file system not supporting file names with
20399 c:/foo/bar.dll @result{} /path/to/sysroot/c/foo/bar.dll
20402 This makes it possible to have a system root that mirrors a target
20403 with more than one drive. E.g., you may want to setup your local
20404 copies of the target system shared libraries like so (note @samp{c} vs
20408 @file{/path/to/sysroot/c/sys/bin/foo.dll}
20409 @file{/path/to/sysroot/c/sys/bin/bar.dll}
20410 @file{/path/to/sysroot/z/sys/bin/bar.dll}
20414 and point the system root at @file{/path/to/sysroot}, so that
20415 @value{GDBN} can find the correct copies of both
20416 @file{c:\sys\bin\foo.dll}, and @file{z:\sys\bin\bar.dll}.
20418 If that still does not find the binary, @value{GDBN} tries
20419 removing the whole drive spec from the target file name:
20422 c:/foo/bar.dll @result{} /path/to/sysroot/foo/bar.dll
20425 This last lookup makes it possible to not care about the drive name,
20426 if you don't want or need to.
20428 The @code{set solib-absolute-prefix} command is an alias for @code{set
20431 @cindex default system root
20432 @cindex @samp{--with-sysroot}
20433 You can set the default system root by using the configure-time
20434 @samp{--with-sysroot} option. If the system root is inside
20435 @value{GDBN}'s configured binary prefix (set with @samp{--prefix} or
20436 @samp{--exec-prefix}), then the default system root will be updated
20437 automatically if the installed @value{GDBN} is moved to a new
20440 @kindex show sysroot
20442 Display the current executable and shared library prefix.
20444 @kindex set solib-search-path
20445 @item set solib-search-path @var{path}
20446 If this variable is set, @var{path} is a colon-separated list of
20447 directories to search for shared libraries. @samp{solib-search-path}
20448 is used after @samp{sysroot} fails to locate the library, or if the
20449 path to the library is relative instead of absolute. If you want to
20450 use @samp{solib-search-path} instead of @samp{sysroot}, be sure to set
20451 @samp{sysroot} to a nonexistent directory to prevent @value{GDBN} from
20452 finding your host's libraries. @samp{sysroot} is preferred; setting
20453 it to a nonexistent directory may interfere with automatic loading
20454 of shared library symbols.
20456 @kindex show solib-search-path
20457 @item show solib-search-path
20458 Display the current shared library search path.
20460 @cindex DOS file-name semantics of file names.
20461 @kindex set target-file-system-kind (unix|dos-based|auto)
20462 @kindex show target-file-system-kind
20463 @item set target-file-system-kind @var{kind}
20464 Set assumed file system kind for target reported file names.
20466 Shared library file names as reported by the target system may not
20467 make sense as is on the system @value{GDBN} is running on. For
20468 example, when remote debugging a target that has MS-DOS based file
20469 system semantics, from a Unix host, the target may be reporting to
20470 @value{GDBN} a list of loaded shared libraries with file names such as
20471 @file{c:\Windows\kernel32.dll}. On Unix hosts, there's no concept of
20472 drive letters, so the @samp{c:\} prefix is not normally understood as
20473 indicating an absolute file name, and neither is the backslash
20474 normally considered a directory separator character. In that case,
20475 the native file system would interpret this whole absolute file name
20476 as a relative file name with no directory components. This would make
20477 it impossible to point @value{GDBN} at a copy of the remote target's
20478 shared libraries on the host using @code{set sysroot}, and impractical
20479 with @code{set solib-search-path}. Setting
20480 @code{target-file-system-kind} to @code{dos-based} tells @value{GDBN}
20481 to interpret such file names similarly to how the target would, and to
20482 map them to file names valid on @value{GDBN}'s native file system
20483 semantics. The value of @var{kind} can be @code{"auto"}, in addition
20484 to one of the supported file system kinds. In that case, @value{GDBN}
20485 tries to determine the appropriate file system variant based on the
20486 current target's operating system (@pxref{ABI, ,Configuring the
20487 Current ABI}). The supported file system settings are:
20491 Instruct @value{GDBN} to assume the target file system is of Unix
20492 kind. Only file names starting the forward slash (@samp{/}) character
20493 are considered absolute, and the directory separator character is also
20497 Instruct @value{GDBN} to assume the target file system is DOS based.
20498 File names starting with either a forward slash, or a drive letter
20499 followed by a colon (e.g., @samp{c:}), are considered absolute, and
20500 both the slash (@samp{/}) and the backslash (@samp{\\}) characters are
20501 considered directory separators.
20504 Instruct @value{GDBN} to use the file system kind associated with the
20505 target operating system (@pxref{ABI, ,Configuring the Current ABI}).
20506 This is the default.
20510 @cindex file name canonicalization
20511 @cindex base name differences
20512 When processing file names provided by the user, @value{GDBN}
20513 frequently needs to compare them to the file names recorded in the
20514 program's debug info. Normally, @value{GDBN} compares just the
20515 @dfn{base names} of the files as strings, which is reasonably fast
20516 even for very large programs. (The base name of a file is the last
20517 portion of its name, after stripping all the leading directories.)
20518 This shortcut in comparison is based upon the assumption that files
20519 cannot have more than one base name. This is usually true, but
20520 references to files that use symlinks or similar filesystem
20521 facilities violate that assumption. If your program records files
20522 using such facilities, or if you provide file names to @value{GDBN}
20523 using symlinks etc., you can set @code{basenames-may-differ} to
20524 @code{true} to instruct @value{GDBN} to completely canonicalize each
20525 pair of file names it needs to compare. This will make file-name
20526 comparisons accurate, but at a price of a significant slowdown.
20529 @item set basenames-may-differ
20530 @kindex set basenames-may-differ
20531 Set whether a source file may have multiple base names.
20533 @item show basenames-may-differ
20534 @kindex show basenames-may-differ
20535 Show whether a source file may have multiple base names.
20539 @section File Caching
20540 @cindex caching of opened files
20541 @cindex caching of bfd objects
20543 To speed up file loading, and reduce memory usage, @value{GDBN} will
20544 reuse the @code{bfd} objects used to track open files. @xref{Top, ,
20545 BFD, bfd, The Binary File Descriptor Library}. The following commands
20546 allow visibility and control of the caching behavior.
20549 @kindex maint info bfds
20550 @item maint info bfds
20551 This prints information about each @code{bfd} object that is known to
20554 @kindex maint set bfd-sharing
20555 @kindex maint show bfd-sharing
20556 @kindex bfd caching
20557 @item maint set bfd-sharing
20558 @item maint show bfd-sharing
20559 Control whether @code{bfd} objects can be shared. When sharing is
20560 enabled @value{GDBN} reuses already open @code{bfd} objects rather
20561 than reopening the same file. Turning sharing off does not cause
20562 already shared @code{bfd} objects to be unshared, but all future files
20563 that are opened will create a new @code{bfd} object. Similarly,
20564 re-enabling sharing does not cause multiple existing @code{bfd}
20565 objects to be collapsed into a single shared @code{bfd} object.
20567 @kindex set debug bfd-cache @var{level}
20568 @kindex bfd caching
20569 @item set debug bfd-cache @var{level}
20570 Turns on debugging of the bfd cache, setting the level to @var{level}.
20572 @kindex show debug bfd-cache
20573 @kindex bfd caching
20574 @item show debug bfd-cache
20575 Show the current debugging level of the bfd cache.
20578 @node Separate Debug Files
20579 @section Debugging Information in Separate Files
20580 @cindex separate debugging information files
20581 @cindex debugging information in separate files
20582 @cindex @file{.debug} subdirectories
20583 @cindex debugging information directory, global
20584 @cindex global debugging information directories
20585 @cindex build ID, and separate debugging files
20586 @cindex @file{.build-id} directory
20588 @value{GDBN} allows you to put a program's debugging information in a
20589 file separate from the executable itself, in a way that allows
20590 @value{GDBN} to find and load the debugging information automatically.
20591 Since debugging information can be very large---sometimes larger
20592 than the executable code itself---some systems distribute debugging
20593 information for their executables in separate files, which users can
20594 install only when they need to debug a problem.
20596 @value{GDBN} supports two ways of specifying the separate debug info
20601 The executable contains a @dfn{debug link} that specifies the name of
20602 the separate debug info file. The separate debug file's name is
20603 usually @file{@var{executable}.debug}, where @var{executable} is the
20604 name of the corresponding executable file without leading directories
20605 (e.g., @file{ls.debug} for @file{/usr/bin/ls}). In addition, the
20606 debug link specifies a 32-bit @dfn{Cyclic Redundancy Check} (CRC)
20607 checksum for the debug file, which @value{GDBN} uses to validate that
20608 the executable and the debug file came from the same build.
20611 The executable contains a @dfn{build ID}, a unique bit string that is
20612 also present in the corresponding debug info file. (This is supported
20613 only on some operating systems, when using the ELF or PE file formats
20614 for binary files and the @sc{gnu} Binutils.) For more details about
20615 this feature, see the description of the @option{--build-id}
20616 command-line option in @ref{Options, , Command Line Options, ld,
20617 The GNU Linker}. The debug info file's name is not specified
20618 explicitly by the build ID, but can be computed from the build ID, see
20622 Depending on the way the debug info file is specified, @value{GDBN}
20623 uses two different methods of looking for the debug file:
20627 For the ``debug link'' method, @value{GDBN} looks up the named file in
20628 the directory of the executable file, then in a subdirectory of that
20629 directory named @file{.debug}, and finally under each one of the
20630 global debug directories, in a subdirectory whose name is identical to
20631 the leading directories of the executable's absolute file name. (On
20632 MS-Windows/MS-DOS, the drive letter of the executable's leading
20633 directories is converted to a one-letter subdirectory, i.e.@:
20634 @file{d:/usr/bin/} is converted to @file{/d/usr/bin/}, because Windows
20635 filesystems disallow colons in file names.)
20638 For the ``build ID'' method, @value{GDBN} looks in the
20639 @file{.build-id} subdirectory of each one of the global debug directories for
20640 a file named @file{@var{nn}/@var{nnnnnnnn}.debug}, where @var{nn} are the
20641 first 2 hex characters of the build ID bit string, and @var{nnnnnnnn}
20642 are the rest of the bit string. (Real build ID strings are 32 or more
20643 hex characters, not 10.)
20646 So, for example, suppose you ask @value{GDBN} to debug
20647 @file{/usr/bin/ls}, which has a debug link that specifies the
20648 file @file{ls.debug}, and a build ID whose value in hex is
20649 @code{abcdef1234}. If the list of the global debug directories includes
20650 @file{/usr/lib/debug}, then @value{GDBN} will look for the following
20651 debug information files, in the indicated order:
20655 @file{/usr/lib/debug/.build-id/ab/cdef1234.debug}
20657 @file{/usr/bin/ls.debug}
20659 @file{/usr/bin/.debug/ls.debug}
20661 @file{/usr/lib/debug/usr/bin/ls.debug}.
20664 @anchor{debug-file-directory}
20665 Global debugging info directories default to what is set by @value{GDBN}
20666 configure option @option{--with-separate-debug-dir}. During @value{GDBN} run
20667 you can also set the global debugging info directories, and view the list
20668 @value{GDBN} is currently using.
20672 @kindex set debug-file-directory
20673 @item set debug-file-directory @var{directories}
20674 Set the directories which @value{GDBN} searches for separate debugging
20675 information files to @var{directory}. Multiple path components can be set
20676 concatenating them by a path separator.
20678 @kindex show debug-file-directory
20679 @item show debug-file-directory
20680 Show the directories @value{GDBN} searches for separate debugging
20685 @cindex @code{.gnu_debuglink} sections
20686 @cindex debug link sections
20687 A debug link is a special section of the executable file named
20688 @code{.gnu_debuglink}. The section must contain:
20692 A filename, with any leading directory components removed, followed by
20695 zero to three bytes of padding, as needed to reach the next four-byte
20696 boundary within the section, and
20698 a four-byte CRC checksum, stored in the same endianness used for the
20699 executable file itself. The checksum is computed on the debugging
20700 information file's full contents by the function given below, passing
20701 zero as the @var{crc} argument.
20704 Any executable file format can carry a debug link, as long as it can
20705 contain a section named @code{.gnu_debuglink} with the contents
20708 @cindex @code{.note.gnu.build-id} sections
20709 @cindex build ID sections
20710 The build ID is a special section in the executable file (and in other
20711 ELF binary files that @value{GDBN} may consider). This section is
20712 often named @code{.note.gnu.build-id}, but that name is not mandatory.
20713 It contains unique identification for the built files---the ID remains
20714 the same across multiple builds of the same build tree. The default
20715 algorithm SHA1 produces 160 bits (40 hexadecimal characters) of the
20716 content for the build ID string. The same section with an identical
20717 value is present in the original built binary with symbols, in its
20718 stripped variant, and in the separate debugging information file.
20720 The debugging information file itself should be an ordinary
20721 executable, containing a full set of linker symbols, sections, and
20722 debugging information. The sections of the debugging information file
20723 should have the same names, addresses, and sizes as the original file,
20724 but they need not contain any data---much like a @code{.bss} section
20725 in an ordinary executable.
20727 The @sc{gnu} binary utilities (Binutils) package includes the
20728 @samp{objcopy} utility that can produce
20729 the separated executable / debugging information file pairs using the
20730 following commands:
20733 @kbd{objcopy --only-keep-debug foo foo.debug}
20738 These commands remove the debugging
20739 information from the executable file @file{foo} and place it in the file
20740 @file{foo.debug}. You can use the first, second or both methods to link the
20745 The debug link method needs the following additional command to also leave
20746 behind a debug link in @file{foo}:
20749 @kbd{objcopy --add-gnu-debuglink=foo.debug foo}
20752 Ulrich Drepper's @file{elfutils} package, starting with version 0.53, contains
20753 a version of the @code{strip} command such that the command @kbd{strip foo -f
20754 foo.debug} has the same functionality as the two @code{objcopy} commands and
20755 the @code{ln -s} command above, together.
20758 Build ID gets embedded into the main executable using @code{ld --build-id} or
20759 the @value{NGCC} counterpart @code{gcc -Wl,--build-id}. Build ID support plus
20760 compatibility fixes for debug files separation are present in @sc{gnu} binary
20761 utilities (Binutils) package since version 2.18.
20766 @cindex CRC algorithm definition
20767 The CRC used in @code{.gnu_debuglink} is the CRC-32 defined in
20768 IEEE 802.3 using the polynomial:
20770 @c TexInfo requires naked braces for multi-digit exponents for Tex
20771 @c output, but this causes HTML output to barf. HTML has to be set using
20772 @c raw commands. So we end up having to specify this equation in 2
20777 <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>
20778 + <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
20784 @math{x^{32} + x^{26} + x^{23} + x^{22} + x^{16} + x^{12} + x^{11}}
20785 @math{+ x^{10} + x^8 + x^7 + x^5 + x^4 + x^2 + x + 1}
20789 The function is computed byte at a time, taking the least
20790 significant bit of each byte first. The initial pattern
20791 @code{0xffffffff} is used, to ensure leading zeros affect the CRC and
20792 the final result is inverted to ensure trailing zeros also affect the
20795 @emph{Note:} This is the same CRC polynomial as used in handling the
20796 @dfn{Remote Serial Protocol} @code{qCRC} packet (@pxref{qCRC packet}).
20797 However in the case of the Remote Serial Protocol, the CRC is computed
20798 @emph{most} significant bit first, and the result is not inverted, so
20799 trailing zeros have no effect on the CRC value.
20801 To complete the description, we show below the code of the function
20802 which produces the CRC used in @code{.gnu_debuglink}. Inverting the
20803 initially supplied @code{crc} argument means that an initial call to
20804 this function passing in zero will start computing the CRC using
20807 @kindex gnu_debuglink_crc32
20810 gnu_debuglink_crc32 (unsigned long crc,
20811 unsigned char *buf, size_t len)
20813 static const unsigned long crc32_table[256] =
20815 0x00000000, 0x77073096, 0xee0e612c, 0x990951ba, 0x076dc419,
20816 0x706af48f, 0xe963a535, 0x9e6495a3, 0x0edb8832, 0x79dcb8a4,
20817 0xe0d5e91e, 0x97d2d988, 0x09b64c2b, 0x7eb17cbd, 0xe7b82d07,
20818 0x90bf1d91, 0x1db71064, 0x6ab020f2, 0xf3b97148, 0x84be41de,
20819 0x1adad47d, 0x6ddde4eb, 0xf4d4b551, 0x83d385c7, 0x136c9856,
20820 0x646ba8c0, 0xfd62f97a, 0x8a65c9ec, 0x14015c4f, 0x63066cd9,
20821 0xfa0f3d63, 0x8d080df5, 0x3b6e20c8, 0x4c69105e, 0xd56041e4,
20822 0xa2677172, 0x3c03e4d1, 0x4b04d447, 0xd20d85fd, 0xa50ab56b,
20823 0x35b5a8fa, 0x42b2986c, 0xdbbbc9d6, 0xacbcf940, 0x32d86ce3,
20824 0x45df5c75, 0xdcd60dcf, 0xabd13d59, 0x26d930ac, 0x51de003a,
20825 0xc8d75180, 0xbfd06116, 0x21b4f4b5, 0x56b3c423, 0xcfba9599,
20826 0xb8bda50f, 0x2802b89e, 0x5f058808, 0xc60cd9b2, 0xb10be924,
20827 0x2f6f7c87, 0x58684c11, 0xc1611dab, 0xb6662d3d, 0x76dc4190,
20828 0x01db7106, 0x98d220bc, 0xefd5102a, 0x71b18589, 0x06b6b51f,
20829 0x9fbfe4a5, 0xe8b8d433, 0x7807c9a2, 0x0f00f934, 0x9609a88e,
20830 0xe10e9818, 0x7f6a0dbb, 0x086d3d2d, 0x91646c97, 0xe6635c01,
20831 0x6b6b51f4, 0x1c6c6162, 0x856530d8, 0xf262004e, 0x6c0695ed,
20832 0x1b01a57b, 0x8208f4c1, 0xf50fc457, 0x65b0d9c6, 0x12b7e950,
20833 0x8bbeb8ea, 0xfcb9887c, 0x62dd1ddf, 0x15da2d49, 0x8cd37cf3,
20834 0xfbd44c65, 0x4db26158, 0x3ab551ce, 0xa3bc0074, 0xd4bb30e2,
20835 0x4adfa541, 0x3dd895d7, 0xa4d1c46d, 0xd3d6f4fb, 0x4369e96a,
20836 0x346ed9fc, 0xad678846, 0xda60b8d0, 0x44042d73, 0x33031de5,
20837 0xaa0a4c5f, 0xdd0d7cc9, 0x5005713c, 0x270241aa, 0xbe0b1010,
20838 0xc90c2086, 0x5768b525, 0x206f85b3, 0xb966d409, 0xce61e49f,
20839 0x5edef90e, 0x29d9c998, 0xb0d09822, 0xc7d7a8b4, 0x59b33d17,
20840 0x2eb40d81, 0xb7bd5c3b, 0xc0ba6cad, 0xedb88320, 0x9abfb3b6,
20841 0x03b6e20c, 0x74b1d29a, 0xead54739, 0x9dd277af, 0x04db2615,
20842 0x73dc1683, 0xe3630b12, 0x94643b84, 0x0d6d6a3e, 0x7a6a5aa8,
20843 0xe40ecf0b, 0x9309ff9d, 0x0a00ae27, 0x7d079eb1, 0xf00f9344,
20844 0x8708a3d2, 0x1e01f268, 0x6906c2fe, 0xf762575d, 0x806567cb,
20845 0x196c3671, 0x6e6b06e7, 0xfed41b76, 0x89d32be0, 0x10da7a5a,
20846 0x67dd4acc, 0xf9b9df6f, 0x8ebeeff9, 0x17b7be43, 0x60b08ed5,
20847 0xd6d6a3e8, 0xa1d1937e, 0x38d8c2c4, 0x4fdff252, 0xd1bb67f1,
20848 0xa6bc5767, 0x3fb506dd, 0x48b2364b, 0xd80d2bda, 0xaf0a1b4c,
20849 0x36034af6, 0x41047a60, 0xdf60efc3, 0xa867df55, 0x316e8eef,
20850 0x4669be79, 0xcb61b38c, 0xbc66831a, 0x256fd2a0, 0x5268e236,
20851 0xcc0c7795, 0xbb0b4703, 0x220216b9, 0x5505262f, 0xc5ba3bbe,
20852 0xb2bd0b28, 0x2bb45a92, 0x5cb36a04, 0xc2d7ffa7, 0xb5d0cf31,
20853 0x2cd99e8b, 0x5bdeae1d, 0x9b64c2b0, 0xec63f226, 0x756aa39c,
20854 0x026d930a, 0x9c0906a9, 0xeb0e363f, 0x72076785, 0x05005713,
20855 0x95bf4a82, 0xe2b87a14, 0x7bb12bae, 0x0cb61b38, 0x92d28e9b,
20856 0xe5d5be0d, 0x7cdcefb7, 0x0bdbdf21, 0x86d3d2d4, 0xf1d4e242,
20857 0x68ddb3f8, 0x1fda836e, 0x81be16cd, 0xf6b9265b, 0x6fb077e1,
20858 0x18b74777, 0x88085ae6, 0xff0f6a70, 0x66063bca, 0x11010b5c,
20859 0x8f659eff, 0xf862ae69, 0x616bffd3, 0x166ccf45, 0xa00ae278,
20860 0xd70dd2ee, 0x4e048354, 0x3903b3c2, 0xa7672661, 0xd06016f7,
20861 0x4969474d, 0x3e6e77db, 0xaed16a4a, 0xd9d65adc, 0x40df0b66,
20862 0x37d83bf0, 0xa9bcae53, 0xdebb9ec5, 0x47b2cf7f, 0x30b5ffe9,
20863 0xbdbdf21c, 0xcabac28a, 0x53b39330, 0x24b4a3a6, 0xbad03605,
20864 0xcdd70693, 0x54de5729, 0x23d967bf, 0xb3667a2e, 0xc4614ab8,
20865 0x5d681b02, 0x2a6f2b94, 0xb40bbe37, 0xc30c8ea1, 0x5a05df1b,
20868 unsigned char *end;
20870 crc = ~crc & 0xffffffff;
20871 for (end = buf + len; buf < end; ++buf)
20872 crc = crc32_table[(crc ^ *buf) & 0xff] ^ (crc >> 8);
20873 return ~crc & 0xffffffff;
20878 This computation does not apply to the ``build ID'' method.
20880 @node MiniDebugInfo
20881 @section Debugging information in a special section
20882 @cindex separate debug sections
20883 @cindex @samp{.gnu_debugdata} section
20885 Some systems ship pre-built executables and libraries that have a
20886 special @samp{.gnu_debugdata} section. This feature is called
20887 @dfn{MiniDebugInfo}. This section holds an LZMA-compressed object and
20888 is used to supply extra symbols for backtraces.
20890 The intent of this section is to provide extra minimal debugging
20891 information for use in simple backtraces. It is not intended to be a
20892 replacement for full separate debugging information (@pxref{Separate
20893 Debug Files}). The example below shows the intended use; however,
20894 @value{GDBN} does not currently put restrictions on what sort of
20895 debugging information might be included in the section.
20897 @value{GDBN} has support for this extension. If the section exists,
20898 then it is used provided that no other source of debugging information
20899 can be found, and that @value{GDBN} was configured with LZMA support.
20901 This section can be easily created using @command{objcopy} and other
20902 standard utilities:
20905 # Extract the dynamic symbols from the main binary, there is no need
20906 # to also have these in the normal symbol table.
20907 nm -D @var{binary} --format=posix --defined-only \
20908 | awk '@{ print $1 @}' | sort > dynsyms
20910 # Extract all the text (i.e. function) symbols from the debuginfo.
20911 # (Note that we actually also accept "D" symbols, for the benefit
20912 # of platforms like PowerPC64 that use function descriptors.)
20913 nm @var{binary} --format=posix --defined-only \
20914 | awk '@{ if ($2 == "T" || $2 == "t" || $2 == "D") print $1 @}' \
20917 # Keep all the function symbols not already in the dynamic symbol
20919 comm -13 dynsyms funcsyms > keep_symbols
20921 # Separate full debug info into debug binary.
20922 objcopy --only-keep-debug @var{binary} debug
20924 # Copy the full debuginfo, keeping only a minimal set of symbols and
20925 # removing some unnecessary sections.
20926 objcopy -S --remove-section .gdb_index --remove-section .comment \
20927 --keep-symbols=keep_symbols debug mini_debuginfo
20929 # Drop the full debug info from the original binary.
20930 strip --strip-all -R .comment @var{binary}
20932 # Inject the compressed data into the .gnu_debugdata section of the
20935 objcopy --add-section .gnu_debugdata=mini_debuginfo.xz @var{binary}
20939 @section Index Files Speed Up @value{GDBN}
20940 @cindex index files
20941 @cindex @samp{.gdb_index} section
20943 When @value{GDBN} finds a symbol file, it scans the symbols in the
20944 file in order to construct an internal symbol table. This lets most
20945 @value{GDBN} operations work quickly---at the cost of a delay early
20946 on. For large programs, this delay can be quite lengthy, so
20947 @value{GDBN} provides a way to build an index, which speeds up
20950 For convenience, @value{GDBN} comes with a program,
20951 @command{gdb-add-index}, which can be used to add the index to a
20952 symbol file. It takes the symbol file as its only argument:
20955 $ gdb-add-index symfile
20958 @xref{gdb-add-index}.
20960 It is also possible to do the work manually. Here is what
20961 @command{gdb-add-index} does behind the curtains.
20963 The index is stored as a section in the symbol file. @value{GDBN} can
20964 write the index to a file, then you can put it into the symbol file
20965 using @command{objcopy}.
20967 To create an index file, use the @code{save gdb-index} command:
20970 @item save gdb-index [-dwarf-5] @var{directory}
20971 @kindex save gdb-index
20972 Create index files for all symbol files currently known by
20973 @value{GDBN}. For each known @var{symbol-file}, this command by
20974 default creates it produces a single file
20975 @file{@var{symbol-file}.gdb-index}. If you invoke this command with
20976 the @option{-dwarf-5} option, it produces 2 files:
20977 @file{@var{symbol-file}.debug_names} and
20978 @file{@var{symbol-file}.debug_str}. The files are created in the
20979 given @var{directory}.
20982 Once you have created an index file you can merge it into your symbol
20983 file, here named @file{symfile}, using @command{objcopy}:
20986 $ objcopy --add-section .gdb_index=symfile.gdb-index \
20987 --set-section-flags .gdb_index=readonly symfile symfile
20990 Or for @code{-dwarf-5}:
20993 $ objcopy --dump-section .debug_str=symfile.debug_str.new symfile
20994 $ cat symfile.debug_str >>symfile.debug_str.new
20995 $ objcopy --add-section .debug_names=symfile.gdb-index \
20996 --set-section-flags .debug_names=readonly \
20997 --update-section .debug_str=symfile.debug_str.new symfile symfile
21000 @value{GDBN} will normally ignore older versions of @file{.gdb_index}
21001 sections that have been deprecated. Usually they are deprecated because
21002 they are missing a new feature or have performance issues.
21003 To tell @value{GDBN} to use a deprecated index section anyway
21004 specify @code{set use-deprecated-index-sections on}.
21005 The default is @code{off}.
21006 This can speed up startup, but may result in some functionality being lost.
21007 @xref{Index Section Format}.
21009 @emph{Warning:} Setting @code{use-deprecated-index-sections} to @code{on}
21010 must be done before gdb reads the file. The following will not work:
21013 $ gdb -ex "set use-deprecated-index-sections on" <program>
21016 Instead you must do, for example,
21019 $ gdb -iex "set use-deprecated-index-sections on" <program>
21022 There are currently some limitation on indices. They only work when
21023 using DWARF debugging information, not stabs. And, only the
21024 @code{-dwarf-5} index works for programs using Ada.
21026 @subsection Automatic symbol index cache
21028 @cindex automatic symbol index cache
21029 It is possible for @value{GDBN} to automatically save a copy of this index in a
21030 cache on disk and retrieve it from there when loading the same binary in the
21031 future. This feature can be turned on with @kbd{set index-cache on}. The
21032 following commands can be used to tweak the behavior of the index cache.
21036 @kindex set index-cache
21037 @item set index-cache on
21038 @itemx set index-cache off
21039 Enable or disable the use of the symbol index cache.
21041 @item set index-cache directory @var{directory}
21042 @kindex show index-cache
21043 @itemx show index-cache directory
21044 Set/show the directory where index files will be saved.
21046 The default value for this directory depends on the host platform. On
21047 most systems, the index is cached in the @file{gdb} subdirectory of
21048 the directory pointed to by the @env{XDG_CACHE_HOME} environment
21049 variable, if it is defined, else in the @file{.cache/gdb} subdirectory
21050 of your home directory. However, on some systems, the default may
21051 differ according to local convention.
21053 There is no limit on the disk space used by index cache. It is perfectly safe
21054 to delete the content of that directory to free up disk space.
21056 @item show index-cache stats
21057 Print the number of cache hits and misses since the launch of @value{GDBN}.
21061 @node Symbol Errors
21062 @section Errors Reading Symbol Files
21064 While reading a symbol file, @value{GDBN} occasionally encounters problems,
21065 such as symbol types it does not recognize, or known bugs in compiler
21066 output. By default, @value{GDBN} does not notify you of such problems, since
21067 they are relatively common and primarily of interest to people
21068 debugging compilers. If you are interested in seeing information
21069 about ill-constructed symbol tables, you can either ask @value{GDBN} to print
21070 only one message about each such type of problem, no matter how many
21071 times the problem occurs; or you can ask @value{GDBN} to print more messages,
21072 to see how many times the problems occur, with the @code{set
21073 complaints} command (@pxref{Messages/Warnings, ,Optional Warnings and
21076 The messages currently printed, and their meanings, include:
21079 @item inner block not inside outer block in @var{symbol}
21081 The symbol information shows where symbol scopes begin and end
21082 (such as at the start of a function or a block of statements). This
21083 error indicates that an inner scope block is not fully contained
21084 in its outer scope blocks.
21086 @value{GDBN} circumvents the problem by treating the inner block as if it had
21087 the same scope as the outer block. In the error message, @var{symbol}
21088 may be shown as ``@code{(don't know)}'' if the outer block is not a
21091 @item block at @var{address} out of order
21093 The symbol information for symbol scope blocks should occur in
21094 order of increasing addresses. This error indicates that it does not
21097 @value{GDBN} does not circumvent this problem, and has trouble
21098 locating symbols in the source file whose symbols it is reading. (You
21099 can often determine what source file is affected by specifying
21100 @code{set verbose on}. @xref{Messages/Warnings, ,Optional Warnings and
21103 @item bad block start address patched
21105 The symbol information for a symbol scope block has a start address
21106 smaller than the address of the preceding source line. This is known
21107 to occur in the SunOS 4.1.1 (and earlier) C compiler.
21109 @value{GDBN} circumvents the problem by treating the symbol scope block as
21110 starting on the previous source line.
21112 @item bad string table offset in symbol @var{n}
21115 Symbol number @var{n} contains a pointer into the string table which is
21116 larger than the size of the string table.
21118 @value{GDBN} circumvents the problem by considering the symbol to have the
21119 name @code{foo}, which may cause other problems if many symbols end up
21122 @item unknown symbol type @code{0x@var{nn}}
21124 The symbol information contains new data types that @value{GDBN} does
21125 not yet know how to read. @code{0x@var{nn}} is the symbol type of the
21126 uncomprehended information, in hexadecimal.
21128 @value{GDBN} circumvents the error by ignoring this symbol information.
21129 This usually allows you to debug your program, though certain symbols
21130 are not accessible. If you encounter such a problem and feel like
21131 debugging it, you can debug @code{@value{GDBP}} with itself, breakpoint
21132 on @code{complain}, then go up to the function @code{read_dbx_symtab}
21133 and examine @code{*bufp} to see the symbol.
21135 @item stub type has NULL name
21137 @value{GDBN} could not find the full definition for a struct or class.
21139 @item const/volatile indicator missing (ok if using g++ v1.x), got@dots{}
21140 The symbol information for a C@t{++} member function is missing some
21141 information that recent versions of the compiler should have output for
21144 @item info mismatch between compiler and debugger
21146 @value{GDBN} could not parse a type specification output by the compiler.
21151 @section GDB Data Files
21153 @cindex prefix for data files
21154 @value{GDBN} will sometimes read an auxiliary data file. These files
21155 are kept in a directory known as the @dfn{data directory}.
21157 You can set the data directory's name, and view the name @value{GDBN}
21158 is currently using.
21161 @kindex set data-directory
21162 @item set data-directory @var{directory}
21163 Set the directory which @value{GDBN} searches for auxiliary data files
21164 to @var{directory}.
21166 @kindex show data-directory
21167 @item show data-directory
21168 Show the directory @value{GDBN} searches for auxiliary data files.
21171 @cindex default data directory
21172 @cindex @samp{--with-gdb-datadir}
21173 You can set the default data directory by using the configure-time
21174 @samp{--with-gdb-datadir} option. If the data directory is inside
21175 @value{GDBN}'s configured binary prefix (set with @samp{--prefix} or
21176 @samp{--exec-prefix}), then the default data directory will be updated
21177 automatically if the installed @value{GDBN} is moved to a new
21180 The data directory may also be specified with the
21181 @code{--data-directory} command line option.
21182 @xref{Mode Options}.
21185 @chapter Specifying a Debugging Target
21187 @cindex debugging target
21188 A @dfn{target} is the execution environment occupied by your program.
21190 Often, @value{GDBN} runs in the same host environment as your program;
21191 in that case, the debugging target is specified as a side effect when
21192 you use the @code{file} or @code{core} commands. When you need more
21193 flexibility---for example, running @value{GDBN} on a physically separate
21194 host, or controlling a standalone system over a serial port or a
21195 realtime system over a TCP/IP connection---you can use the @code{target}
21196 command to specify one of the target types configured for @value{GDBN}
21197 (@pxref{Target Commands, ,Commands for Managing Targets}).
21199 @cindex target architecture
21200 It is possible to build @value{GDBN} for several different @dfn{target
21201 architectures}. When @value{GDBN} is built like that, you can choose
21202 one of the available architectures with the @kbd{set architecture}
21206 @kindex set architecture
21207 @kindex show architecture
21208 @item set architecture @var{arch}
21209 This command sets the current target architecture to @var{arch}. The
21210 value of @var{arch} can be @code{"auto"}, in addition to one of the
21211 supported architectures.
21213 @item show architecture
21214 Show the current target architecture.
21216 @item set processor
21218 @kindex set processor
21219 @kindex show processor
21220 These are alias commands for, respectively, @code{set architecture}
21221 and @code{show architecture}.
21225 * Active Targets:: Active targets
21226 * Target Commands:: Commands for managing targets
21227 * Byte Order:: Choosing target byte order
21230 @node Active Targets
21231 @section Active Targets
21233 @cindex stacking targets
21234 @cindex active targets
21235 @cindex multiple targets
21237 There are multiple classes of targets such as: processes, executable files or
21238 recording sessions. Core files belong to the process class, making core file
21239 and process mutually exclusive. Otherwise, @value{GDBN} can work concurrently
21240 on multiple active targets, one in each class. This allows you to (for
21241 example) start a process and inspect its activity, while still having access to
21242 the executable file after the process finishes. Or if you start process
21243 recording (@pxref{Reverse Execution}) and @code{reverse-step} there, you are
21244 presented a virtual layer of the recording target, while the process target
21245 remains stopped at the chronologically last point of the process execution.
21247 Use the @code{core-file} and @code{exec-file} commands to select a new core
21248 file or executable target (@pxref{Files, ,Commands to Specify Files}). To
21249 specify as a target a process that is already running, use the @code{attach}
21250 command (@pxref{Attach, ,Debugging an Already-running Process}).
21252 @node Target Commands
21253 @section Commands for Managing Targets
21256 @item target @var{type} @var{parameters}
21257 Connects the @value{GDBN} host environment to a target machine or
21258 process. A target is typically a protocol for talking to debugging
21259 facilities. You use the argument @var{type} to specify the type or
21260 protocol of the target machine.
21262 Further @var{parameters} are interpreted by the target protocol, but
21263 typically include things like device names or host names to connect
21264 with, process numbers, and baud rates.
21266 The @code{target} command does not repeat if you press @key{RET} again
21267 after executing the command.
21269 @kindex help target
21271 Displays the names of all targets available. To display targets
21272 currently selected, use either @code{info target} or @code{info files}
21273 (@pxref{Files, ,Commands to Specify Files}).
21275 @item help target @var{name}
21276 Describe a particular target, including any parameters necessary to
21279 @kindex set gnutarget
21280 @item set gnutarget @var{args}
21281 @value{GDBN} uses its own library BFD to read your files. @value{GDBN}
21282 knows whether it is reading an @dfn{executable},
21283 a @dfn{core}, or a @dfn{.o} file; however, you can specify the file format
21284 with the @code{set gnutarget} command. Unlike most @code{target} commands,
21285 with @code{gnutarget} the @code{target} refers to a program, not a machine.
21288 @emph{Warning:} To specify a file format with @code{set gnutarget},
21289 you must know the actual BFD name.
21293 @xref{Files, , Commands to Specify Files}.
21295 @kindex show gnutarget
21296 @item show gnutarget
21297 Use the @code{show gnutarget} command to display what file format
21298 @code{gnutarget} is set to read. If you have not set @code{gnutarget},
21299 @value{GDBN} will determine the file format for each file automatically,
21300 and @code{show gnutarget} displays @samp{The current BFD target is "auto"}.
21303 @cindex common targets
21304 Here are some common targets (available, or not, depending on the GDB
21309 @item target exec @var{program}
21310 @cindex executable file target
21311 An executable file. @samp{target exec @var{program}} is the same as
21312 @samp{exec-file @var{program}}.
21314 @item target core @var{filename}
21315 @cindex core dump file target
21316 A core dump file. @samp{target core @var{filename}} is the same as
21317 @samp{core-file @var{filename}}.
21319 @item target remote @var{medium}
21320 @cindex remote target
21321 A remote system connected to @value{GDBN} via a serial line or network
21322 connection. This command tells @value{GDBN} to use its own remote
21323 protocol over @var{medium} for debugging. @xref{Remote Debugging}.
21325 For example, if you have a board connected to @file{/dev/ttya} on the
21326 machine running @value{GDBN}, you could say:
21329 target remote /dev/ttya
21332 @code{target remote} supports the @code{load} command. This is only
21333 useful if you have some other way of getting the stub to the target
21334 system, and you can put it somewhere in memory where it won't get
21335 clobbered by the download.
21337 @item target sim @r{[}@var{simargs}@r{]} @dots{}
21338 @cindex built-in simulator target
21339 Builtin CPU simulator. @value{GDBN} includes simulators for most architectures.
21347 works; however, you cannot assume that a specific memory map, device
21348 drivers, or even basic I/O is available, although some simulators do
21349 provide these. For info about any processor-specific simulator details,
21350 see the appropriate section in @ref{Embedded Processors, ,Embedded
21353 @item target native
21354 @cindex native target
21355 Setup for local/native process debugging. Useful to make the
21356 @code{run} command spawn native processes (likewise @code{attach},
21357 etc.@:) even when @code{set auto-connect-native-target} is @code{off}
21358 (@pxref{set auto-connect-native-target}).
21362 Different targets are available on different configurations of @value{GDBN};
21363 your configuration may have more or fewer targets.
21365 Many remote targets require you to download the executable's code once
21366 you've successfully established a connection. You may wish to control
21367 various aspects of this process.
21372 @kindex set hash@r{, for remote monitors}
21373 @cindex hash mark while downloading
21374 This command controls whether a hash mark @samp{#} is displayed while
21375 downloading a file to the remote monitor. If on, a hash mark is
21376 displayed after each S-record is successfully downloaded to the
21380 @kindex show hash@r{, for remote monitors}
21381 Show the current status of displaying the hash mark.
21383 @item set debug monitor
21384 @kindex set debug monitor
21385 @cindex display remote monitor communications
21386 Enable or disable display of communications messages between
21387 @value{GDBN} and the remote monitor.
21389 @item show debug monitor
21390 @kindex show debug monitor
21391 Show the current status of displaying communications between
21392 @value{GDBN} and the remote monitor.
21397 @kindex load @var{filename} @var{offset}
21398 @item load @var{filename} @var{offset}
21400 Depending on what remote debugging facilities are configured into
21401 @value{GDBN}, the @code{load} command may be available. Where it exists, it
21402 is meant to make @var{filename} (an executable) available for debugging
21403 on the remote system---by downloading, or dynamic linking, for example.
21404 @code{load} also records the @var{filename} symbol table in @value{GDBN}, like
21405 the @code{add-symbol-file} command.
21407 If your @value{GDBN} does not have a @code{load} command, attempting to
21408 execute it gets the error message ``@code{You can't do that when your
21409 target is @dots{}}''
21411 The file is loaded at whatever address is specified in the executable.
21412 For some object file formats, you can specify the load address when you
21413 link the program; for other formats, like a.out, the object file format
21414 specifies a fixed address.
21415 @c FIXME! This would be a good place for an xref to the GNU linker doc.
21417 It is also possible to tell @value{GDBN} to load the executable file at a
21418 specific offset described by the optional argument @var{offset}. When
21419 @var{offset} is provided, @var{filename} must also be provided.
21421 Depending on the remote side capabilities, @value{GDBN} may be able to
21422 load programs into flash memory.
21424 @code{load} does not repeat if you press @key{RET} again after using it.
21429 @kindex flash-erase
21431 @anchor{flash-erase}
21433 Erases all known flash memory regions on the target.
21438 @section Choosing Target Byte Order
21440 @cindex choosing target byte order
21441 @cindex target byte order
21443 Some types of processors, such as the @acronym{MIPS}, PowerPC, and Renesas SH,
21444 offer the ability to run either big-endian or little-endian byte
21445 orders. Usually the executable or symbol will include a bit to
21446 designate the endian-ness, and you will not need to worry about
21447 which to use. However, you may still find it useful to adjust
21448 @value{GDBN}'s idea of processor endian-ness manually.
21452 @item set endian big
21453 Instruct @value{GDBN} to assume the target is big-endian.
21455 @item set endian little
21456 Instruct @value{GDBN} to assume the target is little-endian.
21458 @item set endian auto
21459 Instruct @value{GDBN} to use the byte order associated with the
21463 Display @value{GDBN}'s current idea of the target byte order.
21467 If the @code{set endian auto} mode is in effect and no executable has
21468 been selected, then the endianness used is the last one chosen either
21469 by one of the @code{set endian big} and @code{set endian little}
21470 commands or by inferring from the last executable used. If no
21471 endianness has been previously chosen, then the default for this mode
21472 is inferred from the target @value{GDBN} has been built for, and is
21473 @code{little} if the name of the target CPU has an @code{el} suffix
21474 and @code{big} otherwise.
21476 Note that these commands merely adjust interpretation of symbolic
21477 data on the host, and that they have absolutely no effect on the
21481 @node Remote Debugging
21482 @chapter Debugging Remote Programs
21483 @cindex remote debugging
21485 If you are trying to debug a program running on a machine that cannot run
21486 @value{GDBN} in the usual way, it is often useful to use remote debugging.
21487 For example, you might use remote debugging on an operating system kernel,
21488 or on a small system which does not have a general purpose operating system
21489 powerful enough to run a full-featured debugger.
21491 Some configurations of @value{GDBN} have special serial or TCP/IP interfaces
21492 to make this work with particular debugging targets. In addition,
21493 @value{GDBN} comes with a generic serial protocol (specific to @value{GDBN},
21494 but not specific to any particular target system) which you can use if you
21495 write the remote stubs---the code that runs on the remote system to
21496 communicate with @value{GDBN}.
21498 Other remote targets may be available in your
21499 configuration of @value{GDBN}; use @code{help target} to list them.
21502 * Connecting:: Connecting to a remote target
21503 * File Transfer:: Sending files to a remote system
21504 * Server:: Using the gdbserver program
21505 * Remote Configuration:: Remote configuration
21506 * Remote Stub:: Implementing a remote stub
21510 @section Connecting to a Remote Target
21511 @cindex remote debugging, connecting
21512 @cindex @code{gdbserver}, connecting
21513 @cindex remote debugging, types of connections
21514 @cindex @code{gdbserver}, types of connections
21515 @cindex @code{gdbserver}, @code{target remote} mode
21516 @cindex @code{gdbserver}, @code{target extended-remote} mode
21518 This section describes how to connect to a remote target, including the
21519 types of connections and their differences, how to set up executable and
21520 symbol files on the host and target, and the commands used for
21521 connecting to and disconnecting from the remote target.
21523 @subsection Types of Remote Connections
21525 @value{GDBN} supports two types of remote connections, @code{target remote}
21526 mode and @code{target extended-remote} mode. Note that many remote targets
21527 support only @code{target remote} mode. There are several major
21528 differences between the two types of connections, enumerated here:
21532 @cindex remote debugging, detach and program exit
21533 @item Result of detach or program exit
21534 @strong{With target remote mode:} When the debugged program exits or you
21535 detach from it, @value{GDBN} disconnects from the target. When using
21536 @code{gdbserver}, @code{gdbserver} will exit.
21538 @strong{With target extended-remote mode:} When the debugged program exits or
21539 you detach from it, @value{GDBN} remains connected to the target, even
21540 though no program is running. You can rerun the program, attach to a
21541 running program, or use @code{monitor} commands specific to the target.
21543 When using @code{gdbserver} in this case, it does not exit unless it was
21544 invoked using the @option{--once} option. If the @option{--once} option
21545 was not used, you can ask @code{gdbserver} to exit using the
21546 @code{monitor exit} command (@pxref{Monitor Commands for gdbserver}).
21548 @item Specifying the program to debug
21549 For both connection types you use the @code{file} command to specify the
21550 program on the host system. If you are using @code{gdbserver} there are
21551 some differences in how to specify the location of the program on the
21554 @strong{With target remote mode:} You must either specify the program to debug
21555 on the @code{gdbserver} command line or use the @option{--attach} option
21556 (@pxref{Attaching to a program,,Attaching to a Running Program}).
21558 @cindex @option{--multi}, @code{gdbserver} option
21559 @strong{With target extended-remote mode:} You may specify the program to debug
21560 on the @code{gdbserver} command line, or you can load the program or attach
21561 to it using @value{GDBN} commands after connecting to @code{gdbserver}.
21563 @anchor{--multi Option in Types of Remote Connnections}
21564 You can start @code{gdbserver} without supplying an initial command to run
21565 or process ID to attach. To do this, use the @option{--multi} command line
21566 option. Then you can connect using @code{target extended-remote} and start
21567 the program you want to debug (see below for details on using the
21568 @code{run} command in this scenario). Note that the conditions under which
21569 @code{gdbserver} terminates depend on how @value{GDBN} connects to it
21570 (@code{target remote} or @code{target extended-remote}). The
21571 @option{--multi} option to @code{gdbserver} has no influence on that.
21573 @item The @code{run} command
21574 @strong{With target remote mode:} The @code{run} command is not
21575 supported. Once a connection has been established, you can use all
21576 the usual @value{GDBN} commands to examine and change data. The
21577 remote program is already running, so you can use commands like
21578 @kbd{step} and @kbd{continue}.
21580 @strong{With target extended-remote mode:} The @code{run} command is
21581 supported. The @code{run} command uses the value set by
21582 @code{set remote exec-file} (@pxref{set remote exec-file}) to select
21583 the program to run. Command line arguments are supported, except for
21584 wildcard expansion and I/O redirection (@pxref{Arguments}).
21586 If you specify the program to debug on the command line, then the
21587 @code{run} command is not required to start execution, and you can
21588 resume using commands like @kbd{step} and @kbd{continue} as with
21589 @code{target remote} mode.
21591 @anchor{Attaching in Types of Remote Connections}
21593 @strong{With target remote mode:} The @value{GDBN} command @code{attach} is
21594 not supported. To attach to a running program using @code{gdbserver}, you
21595 must use the @option{--attach} option (@pxref{Running gdbserver}).
21597 @strong{With target extended-remote mode:} To attach to a running program,
21598 you may use the @code{attach} command after the connection has been
21599 established. If you are using @code{gdbserver}, you may also invoke
21600 @code{gdbserver} using the @option{--attach} option
21601 (@pxref{Running gdbserver}).
21605 @anchor{Host and target files}
21606 @subsection Host and Target Files
21607 @cindex remote debugging, symbol files
21608 @cindex symbol files, remote debugging
21610 @value{GDBN}, running on the host, needs access to symbol and debugging
21611 information for your program running on the target. This requires
21612 access to an unstripped copy of your program, and possibly any associated
21613 symbol files. Note that this section applies equally to both @code{target
21614 remote} mode and @code{target extended-remote} mode.
21616 Some remote targets (@pxref{qXfer executable filename read}, and
21617 @pxref{Host I/O Packets}) allow @value{GDBN} to access program files over
21618 the same connection used to communicate with @value{GDBN}. With such a
21619 target, if the remote program is unstripped, the only command you need is
21620 @code{target remote} (or @code{target extended-remote}).
21622 If the remote program is stripped, or the target does not support remote
21623 program file access, start up @value{GDBN} using the name of the local
21624 unstripped copy of your program as the first argument, or use the
21625 @code{file} command. Use @code{set sysroot} to specify the location (on
21626 the host) of target libraries (unless your @value{GDBN} was compiled with
21627 the correct sysroot using @code{--with-sysroot}). Alternatively, you
21628 may use @code{set solib-search-path} to specify how @value{GDBN} locates
21631 The symbol file and target libraries must exactly match the executable
21632 and libraries on the target, with one exception: the files on the host
21633 system should not be stripped, even if the files on the target system
21634 are. Mismatched or missing files will lead to confusing results
21635 during debugging. On @sc{gnu}/Linux targets, mismatched or missing
21636 files may also prevent @code{gdbserver} from debugging multi-threaded
21639 @subsection Remote Connection Commands
21640 @cindex remote connection commands
21641 @value{GDBN} can communicate with the target over a serial line, a
21642 local Unix domain socket, or
21643 over an @acronym{IP} network using @acronym{TCP} or @acronym{UDP}. In
21644 each case, @value{GDBN} uses the same protocol for debugging your
21645 program; only the medium carrying the debugging packets varies. The
21646 @code{target remote} and @code{target extended-remote} commands
21647 establish a connection to the target. Both commands accept the same
21648 arguments, which indicate the medium to use:
21652 @item target remote @var{serial-device}
21653 @itemx target extended-remote @var{serial-device}
21654 @cindex serial line, @code{target remote}
21655 Use @var{serial-device} to communicate with the target. For example,
21656 to use a serial line connected to the device named @file{/dev/ttyb}:
21659 target remote /dev/ttyb
21662 If you're using a serial line, you may want to give @value{GDBN} the
21663 @samp{--baud} option, or use the @code{set serial baud} command
21664 (@pxref{Remote Configuration, set serial baud}) before the
21665 @code{target} command.
21667 @item target remote @var{local-socket}
21668 @itemx target extended-remote @var{local-socket}
21669 @cindex local socket, @code{target remote}
21670 @cindex Unix domain socket
21671 Use @var{local-socket} to communicate with the target. For example,
21672 to use a local Unix domain socket bound to the file system entry @file{/tmp/gdb-socket0}:
21675 target remote /tmp/gdb-socket0
21678 Note that this command has the same form as the command to connect
21679 to a serial line. @value{GDBN} will automatically determine which
21680 kind of file you have specified and will make the appropriate kind
21682 This feature is not available if the host system does not support
21683 Unix domain sockets.
21685 @item target remote @code{@var{host}:@var{port}}
21686 @itemx target remote @code{@var{[host]}:@var{port}}
21687 @itemx target remote @code{tcp:@var{host}:@var{port}}
21688 @itemx target remote @code{tcp:@var{[host]}:@var{port}}
21689 @itemx target remote @code{tcp4:@var{host}:@var{port}}
21690 @itemx target remote @code{tcp6:@var{host}:@var{port}}
21691 @itemx target remote @code{tcp6:@var{[host]}:@var{port}}
21692 @itemx target extended-remote @code{@var{host}:@var{port}}
21693 @itemx target extended-remote @code{@var{[host]}:@var{port}}
21694 @itemx target extended-remote @code{tcp:@var{host}:@var{port}}
21695 @itemx target extended-remote @code{tcp:@var{[host]}:@var{port}}
21696 @itemx target extended-remote @code{tcp4:@var{host}:@var{port}}
21697 @itemx target extended-remote @code{tcp6:@var{host}:@var{port}}
21698 @itemx target extended-remote @code{tcp6:@var{[host]}:@var{port}}
21699 @cindex @acronym{TCP} port, @code{target remote}
21700 Debug using a @acronym{TCP} connection to @var{port} on @var{host}.
21701 The @var{host} may be either a host name, a numeric @acronym{IPv4}
21702 address, or a numeric @acronym{IPv6} address (with or without the
21703 square brackets to separate the address from the port); @var{port}
21704 must be a decimal number. The @var{host} could be the target machine
21705 itself, if it is directly connected to the net, or it might be a
21706 terminal server which in turn has a serial line to the target.
21708 For example, to connect to port 2828 on a terminal server named
21712 target remote manyfarms:2828
21715 To connect to port 2828 on a terminal server whose address is
21716 @code{2001:0db8:85a3:0000:0000:8a2e:0370:7334}, you can either use the
21717 square bracket syntax:
21720 target remote [2001:0db8:85a3:0000:0000:8a2e:0370:7334]:2828
21724 or explicitly specify the @acronym{IPv6} protocol:
21727 target remote tcp6:2001:0db8:85a3:0000:0000:8a2e:0370:7334:2828
21730 This last example may be confusing to the reader, because there is no
21731 visible separation between the hostname and the port number.
21732 Therefore, we recommend the user to provide @acronym{IPv6} addresses
21733 using square brackets for clarity. However, it is important to
21734 mention that for @value{GDBN} there is no ambiguity: the number after
21735 the last colon is considered to be the port number.
21737 If your remote target is actually running on the same machine as your
21738 debugger session (e.g.@: a simulator for your target running on the
21739 same host), you can omit the hostname. For example, to connect to
21740 port 1234 on your local machine:
21743 target remote :1234
21747 Note that the colon is still required here.
21749 @item target remote @code{udp:@var{host}:@var{port}}
21750 @itemx target remote @code{udp:@var{[host]}:@var{port}}
21751 @itemx target remote @code{udp4:@var{host}:@var{port}}
21752 @itemx target remote @code{udp6:@var{[host]}:@var{port}}
21753 @itemx target extended-remote @code{udp:@var{host}:@var{port}}
21754 @itemx target extended-remote @code{udp:@var{host}:@var{port}}
21755 @itemx target extended-remote @code{udp:@var{[host]}:@var{port}}
21756 @itemx target extended-remote @code{udp4:@var{host}:@var{port}}
21757 @itemx target extended-remote @code{udp6:@var{host}:@var{port}}
21758 @itemx target extended-remote @code{udp6:@var{[host]}:@var{port}}
21759 @cindex @acronym{UDP} port, @code{target remote}
21760 Debug using @acronym{UDP} packets to @var{port} on @var{host}. For example, to
21761 connect to @acronym{UDP} port 2828 on a terminal server named @code{manyfarms}:
21764 target remote udp:manyfarms:2828
21767 When using a @acronym{UDP} connection for remote debugging, you should
21768 keep in mind that the `U' stands for ``Unreliable''. @acronym{UDP}
21769 can silently drop packets on busy or unreliable networks, which will
21770 cause havoc with your debugging session.
21772 @item target remote | @var{command}
21773 @itemx target extended-remote | @var{command}
21774 @cindex pipe, @code{target remote} to
21775 Run @var{command} in the background and communicate with it using a
21776 pipe. The @var{command} is a shell command, to be parsed and expanded
21777 by the system's command shell, @code{/bin/sh}; it should expect remote
21778 protocol packets on its standard input, and send replies on its
21779 standard output. You could use this to run a stand-alone simulator
21780 that speaks the remote debugging protocol, to make net connections
21781 using programs like @code{ssh}, or for other similar tricks.
21783 If @var{command} closes its standard output (perhaps by exiting),
21784 @value{GDBN} will try to send it a @code{SIGTERM} signal. (If the
21785 program has already exited, this will have no effect.)
21789 @cindex interrupting remote programs
21790 @cindex remote programs, interrupting
21791 Whenever @value{GDBN} is waiting for the remote program, if you type the
21792 interrupt character (often @kbd{Ctrl-c}), @value{GDBN} attempts to stop the
21793 program. This may or may not succeed, depending in part on the hardware
21794 and the serial drivers the remote system uses. If you type the
21795 interrupt character once again, @value{GDBN} displays this prompt:
21798 Interrupted while waiting for the program.
21799 Give up (and stop debugging it)? (y or n)
21802 In @code{target remote} mode, if you type @kbd{y}, @value{GDBN} abandons
21803 the remote debugging session. (If you decide you want to try again later,
21804 you can use @kbd{target remote} again to connect once more.) If you type
21805 @kbd{n}, @value{GDBN} goes back to waiting.
21807 In @code{target extended-remote} mode, typing @kbd{n} will leave
21808 @value{GDBN} connected to the target.
21811 @kindex detach (remote)
21813 When you have finished debugging the remote program, you can use the
21814 @code{detach} command to release it from @value{GDBN} control.
21815 Detaching from the target normally resumes its execution, but the results
21816 will depend on your particular remote stub. After the @code{detach}
21817 command in @code{target remote} mode, @value{GDBN} is free to connect to
21818 another target. In @code{target extended-remote} mode, @value{GDBN} is
21819 still connected to the target.
21823 The @code{disconnect} command closes the connection to the target, and
21824 the target is generally not resumed. It will wait for @value{GDBN}
21825 (this instance or another one) to connect and continue debugging. After
21826 the @code{disconnect} command, @value{GDBN} is again free to connect to
21829 @cindex send command to remote monitor
21830 @cindex extend @value{GDBN} for remote targets
21831 @cindex add new commands for external monitor
21833 @item monitor @var{cmd}
21834 This command allows you to send arbitrary commands directly to the
21835 remote monitor. Since @value{GDBN} doesn't care about the commands it
21836 sends like this, this command is the way to extend @value{GDBN}---you
21837 can add new commands that only the external monitor will understand
21841 @node File Transfer
21842 @section Sending files to a remote system
21843 @cindex remote target, file transfer
21844 @cindex file transfer
21845 @cindex sending files to remote systems
21847 Some remote targets offer the ability to transfer files over the same
21848 connection used to communicate with @value{GDBN}. This is convenient
21849 for targets accessible through other means, e.g.@: @sc{gnu}/Linux systems
21850 running @code{gdbserver} over a network interface. For other targets,
21851 e.g.@: embedded devices with only a single serial port, this may be
21852 the only way to upload or download files.
21854 Not all remote targets support these commands.
21858 @item remote put @var{hostfile} @var{targetfile}
21859 Copy file @var{hostfile} from the host system (the machine running
21860 @value{GDBN}) to @var{targetfile} on the target system.
21863 @item remote get @var{targetfile} @var{hostfile}
21864 Copy file @var{targetfile} from the target system to @var{hostfile}
21865 on the host system.
21867 @kindex remote delete
21868 @item remote delete @var{targetfile}
21869 Delete @var{targetfile} from the target system.
21874 @section Using the @code{gdbserver} Program
21877 @cindex remote connection without stubs
21878 @code{gdbserver} is a control program for Unix-like systems, which
21879 allows you to connect your program with a remote @value{GDBN} via
21880 @code{target remote} or @code{target extended-remote}---but without
21881 linking in the usual debugging stub.
21883 @code{gdbserver} is not a complete replacement for the debugging stubs,
21884 because it requires essentially the same operating-system facilities
21885 that @value{GDBN} itself does. In fact, a system that can run
21886 @code{gdbserver} to connect to a remote @value{GDBN} could also run
21887 @value{GDBN} locally! @code{gdbserver} is sometimes useful nevertheless,
21888 because it is a much smaller program than @value{GDBN} itself. It is
21889 also easier to port than all of @value{GDBN}, so you may be able to get
21890 started more quickly on a new system by using @code{gdbserver}.
21891 Finally, if you develop code for real-time systems, you may find that
21892 the tradeoffs involved in real-time operation make it more convenient to
21893 do as much development work as possible on another system, for example
21894 by cross-compiling. You can use @code{gdbserver} to make a similar
21895 choice for debugging.
21897 @value{GDBN} and @code{gdbserver} communicate via either a serial line
21898 or a TCP connection, using the standard @value{GDBN} remote serial
21902 @emph{Warning:} @code{gdbserver} does not have any built-in security.
21903 Do not run @code{gdbserver} connected to any public network; a
21904 @value{GDBN} connection to @code{gdbserver} provides access to the
21905 target system with the same privileges as the user running
21909 @anchor{Running gdbserver}
21910 @subsection Running @code{gdbserver}
21911 @cindex arguments, to @code{gdbserver}
21912 @cindex @code{gdbserver}, command-line arguments
21914 Run @code{gdbserver} on the target system. You need a copy of the
21915 program you want to debug, including any libraries it requires.
21916 @code{gdbserver} does not need your program's symbol table, so you can
21917 strip the program if necessary to save space. @value{GDBN} on the host
21918 system does all the symbol handling.
21920 To use the server, you must tell it how to communicate with @value{GDBN};
21921 the name of your program; and the arguments for your program. The usual
21925 target> gdbserver @var{comm} @var{program} [ @var{args} @dots{} ]
21928 @var{comm} is either a device name (to use a serial line), or a TCP
21929 hostname and portnumber, or @code{-} or @code{stdio} to use
21930 stdin/stdout of @code{gdbserver}.
21931 For example, to debug Emacs with the argument
21932 @samp{foo.txt} and communicate with @value{GDBN} over the serial port
21936 target> gdbserver /dev/com1 emacs foo.txt
21939 @code{gdbserver} waits passively for the host @value{GDBN} to communicate
21942 To use a TCP connection instead of a serial line:
21945 target> gdbserver host:2345 emacs foo.txt
21948 The only difference from the previous example is the first argument,
21949 specifying that you are communicating with the host @value{GDBN} via
21950 TCP. The @samp{host:2345} argument means that @code{gdbserver} is to
21951 expect a TCP connection from machine @samp{host} to local TCP port 2345.
21952 (Currently, the @samp{host} part is ignored.) You can choose any number
21953 you want for the port number as long as it does not conflict with any
21954 TCP ports already in use on the target system (for example, @code{23} is
21955 reserved for @code{telnet}).@footnote{If you choose a port number that
21956 conflicts with another service, @code{gdbserver} prints an error message
21957 and exits.} You must use the same port number with the host @value{GDBN}
21958 @code{target remote} command.
21960 The @code{stdio} connection is useful when starting @code{gdbserver}
21964 (gdb) target remote | ssh -T hostname gdbserver - hello
21967 The @samp{-T} option to ssh is provided because we don't need a remote pty,
21968 and we don't want escape-character handling. Ssh does this by default when
21969 a command is provided, the flag is provided to make it explicit.
21970 You could elide it if you want to.
21972 Programs started with stdio-connected gdbserver have @file{/dev/null} for
21973 @code{stdin}, and @code{stdout},@code{stderr} are sent back to gdb for
21974 display through a pipe connected to gdbserver.
21975 Both @code{stdout} and @code{stderr} use the same pipe.
21977 @anchor{Attaching to a program}
21978 @subsubsection Attaching to a Running Program
21979 @cindex attach to a program, @code{gdbserver}
21980 @cindex @option{--attach}, @code{gdbserver} option
21982 On some targets, @code{gdbserver} can also attach to running programs.
21983 This is accomplished via the @code{--attach} argument. The syntax is:
21986 target> gdbserver --attach @var{comm} @var{pid}
21989 @var{pid} is the process ID of a currently running process. It isn't
21990 necessary to point @code{gdbserver} at a binary for the running process.
21992 In @code{target extended-remote} mode, you can also attach using the
21993 @value{GDBN} attach command
21994 (@pxref{Attaching in Types of Remote Connections}).
21997 You can debug processes by name instead of process ID if your target has the
21998 @code{pidof} utility:
22001 target> gdbserver --attach @var{comm} `pidof @var{program}`
22004 In case more than one copy of @var{program} is running, or @var{program}
22005 has multiple threads, most versions of @code{pidof} support the
22006 @code{-s} option to only return the first process ID.
22008 @subsubsection TCP port allocation lifecycle of @code{gdbserver}
22010 This section applies only when @code{gdbserver} is run to listen on a TCP
22013 @code{gdbserver} normally terminates after all of its debugged processes have
22014 terminated in @kbd{target remote} mode. On the other hand, for @kbd{target
22015 extended-remote}, @code{gdbserver} stays running even with no processes left.
22016 @value{GDBN} normally terminates the spawned debugged process on its exit,
22017 which normally also terminates @code{gdbserver} in the @kbd{target remote}
22018 mode. Therefore, when the connection drops unexpectedly, and @value{GDBN}
22019 cannot ask @code{gdbserver} to kill its debugged processes, @code{gdbserver}
22020 stays running even in the @kbd{target remote} mode.
22022 When @code{gdbserver} stays running, @value{GDBN} can connect to it again later.
22023 Such reconnecting is useful for features like @ref{disconnected tracing}. For
22024 completeness, at most one @value{GDBN} can be connected at a time.
22026 @cindex @option{--once}, @code{gdbserver} option
22027 By default, @code{gdbserver} keeps the listening TCP port open, so that
22028 subsequent connections are possible. However, if you start @code{gdbserver}
22029 with the @option{--once} option, it will stop listening for any further
22030 connection attempts after connecting to the first @value{GDBN} session. This
22031 means no further connections to @code{gdbserver} will be possible after the
22032 first one. It also means @code{gdbserver} will terminate after the first
22033 connection with remote @value{GDBN} has closed, even for unexpectedly closed
22034 connections and even in the @kbd{target extended-remote} mode. The
22035 @option{--once} option allows reusing the same port number for connecting to
22036 multiple instances of @code{gdbserver} running on the same host, since each
22037 instance closes its port after the first connection.
22039 @anchor{Other Command-Line Arguments for gdbserver}
22040 @subsubsection Other Command-Line Arguments for @code{gdbserver}
22042 You can use the @option{--multi} option to start @code{gdbserver} without
22043 specifying a program to debug or a process to attach to. Then you can
22044 attach in @code{target extended-remote} mode and run or attach to a
22045 program. For more information,
22046 @pxref{--multi Option in Types of Remote Connnections}.
22048 @cindex @option{--debug}, @code{gdbserver} option
22049 The @option{--debug} option tells @code{gdbserver} to display extra
22050 status information about the debugging process.
22051 @cindex @option{--remote-debug}, @code{gdbserver} option
22052 The @option{--remote-debug} option tells @code{gdbserver} to display
22053 remote protocol debug output.
22054 @cindex @option{--debug-file}, @code{gdbserver} option
22055 @cindex @code{gdbserver}, send all debug output to a single file
22056 The @option{--debug-file=@var{filename}} option tells @code{gdbserver} to
22057 write any debug output to the given @var{filename}. These options are intended
22058 for @code{gdbserver} development and for bug reports to the developers.
22060 @cindex @option{--debug-format}, @code{gdbserver} option
22061 The @option{--debug-format=option1[,option2,...]} option tells
22062 @code{gdbserver} to include additional information in each output.
22063 Possible options are:
22067 Turn off all extra information in debugging output.
22069 Turn on all extra information in debugging output.
22071 Include a timestamp in each line of debugging output.
22074 Options are processed in order. Thus, for example, if @option{none}
22075 appears last then no additional information is added to debugging output.
22077 @cindex @option{--wrapper}, @code{gdbserver} option
22078 The @option{--wrapper} option specifies a wrapper to launch programs
22079 for debugging. The option should be followed by the name of the
22080 wrapper, then any command-line arguments to pass to the wrapper, then
22081 @kbd{--} indicating the end of the wrapper arguments.
22083 @code{gdbserver} runs the specified wrapper program with a combined
22084 command line including the wrapper arguments, then the name of the
22085 program to debug, then any arguments to the program. The wrapper
22086 runs until it executes your program, and then @value{GDBN} gains control.
22088 You can use any program that eventually calls @code{execve} with
22089 its arguments as a wrapper. Several standard Unix utilities do
22090 this, e.g.@: @code{env} and @code{nohup}. Any Unix shell script ending
22091 with @code{exec "$@@"} will also work.
22093 For example, you can use @code{env} to pass an environment variable to
22094 the debugged program, without setting the variable in @code{gdbserver}'s
22098 $ gdbserver --wrapper env LD_PRELOAD=libtest.so -- :2222 ./testprog
22101 @cindex @option{--selftest}
22102 The @option{--selftest} option runs the self tests in @code{gdbserver}:
22105 $ gdbserver --selftest
22106 Ran 2 unit tests, 0 failed
22109 These tests are disabled in release.
22110 @subsection Connecting to @code{gdbserver}
22112 The basic procedure for connecting to the remote target is:
22116 Run @value{GDBN} on the host system.
22119 Make sure you have the necessary symbol files
22120 (@pxref{Host and target files}).
22121 Load symbols for your application using the @code{file} command before you
22122 connect. Use @code{set sysroot} to locate target libraries (unless your
22123 @value{GDBN} was compiled with the correct sysroot using
22124 @code{--with-sysroot}).
22127 Connect to your target (@pxref{Connecting,,Connecting to a Remote Target}).
22128 For TCP connections, you must start up @code{gdbserver} prior to using
22129 the @code{target} command. Otherwise you may get an error whose
22130 text depends on the host system, but which usually looks something like
22131 @samp{Connection refused}. Don't use the @code{load}
22132 command in @value{GDBN} when using @code{target remote} mode, since the
22133 program is already on the target.
22137 @anchor{Monitor Commands for gdbserver}
22138 @subsection Monitor Commands for @code{gdbserver}
22139 @cindex monitor commands, for @code{gdbserver}
22141 During a @value{GDBN} session using @code{gdbserver}, you can use the
22142 @code{monitor} command to send special requests to @code{gdbserver}.
22143 Here are the available commands.
22147 List the available monitor commands.
22149 @item monitor set debug 0
22150 @itemx monitor set debug 1
22151 Disable or enable general debugging messages.
22153 @item monitor set remote-debug 0
22154 @itemx monitor set remote-debug 1
22155 Disable or enable specific debugging messages associated with the remote
22156 protocol (@pxref{Remote Protocol}).
22158 @item monitor set debug-file filename
22159 @itemx monitor set debug-file
22160 Send any debug output to the given file, or to stderr.
22162 @item monitor set debug-format option1@r{[},option2,...@r{]}
22163 Specify additional text to add to debugging messages.
22164 Possible options are:
22168 Turn off all extra information in debugging output.
22170 Turn on all extra information in debugging output.
22172 Include a timestamp in each line of debugging output.
22175 Options are processed in order. Thus, for example, if @option{none}
22176 appears last then no additional information is added to debugging output.
22178 @item monitor set libthread-db-search-path [PATH]
22179 @cindex gdbserver, search path for @code{libthread_db}
22180 When this command is issued, @var{path} is a colon-separated list of
22181 directories to search for @code{libthread_db} (@pxref{Threads,,set
22182 libthread-db-search-path}). If you omit @var{path},
22183 @samp{libthread-db-search-path} will be reset to its default value.
22185 The special entry @samp{$pdir} for @samp{libthread-db-search-path} is
22186 not supported in @code{gdbserver}.
22189 Tell gdbserver to exit immediately. This command should be followed by
22190 @code{disconnect} to close the debugging session. @code{gdbserver} will
22191 detach from any attached processes and kill any processes it created.
22192 Use @code{monitor exit} to terminate @code{gdbserver} at the end
22193 of a multi-process mode debug session.
22197 @subsection Tracepoints support in @code{gdbserver}
22198 @cindex tracepoints support in @code{gdbserver}
22200 On some targets, @code{gdbserver} supports tracepoints, fast
22201 tracepoints and static tracepoints.
22203 For fast or static tracepoints to work, a special library called the
22204 @dfn{in-process agent} (IPA), must be loaded in the inferior process.
22205 This library is built and distributed as an integral part of
22206 @code{gdbserver}. In addition, support for static tracepoints
22207 requires building the in-process agent library with static tracepoints
22208 support. At present, the UST (LTTng Userspace Tracer,
22209 @url{http://lttng.org/ust}) tracing engine is supported. This support
22210 is automatically available if UST development headers are found in the
22211 standard include path when @code{gdbserver} is built, or if
22212 @code{gdbserver} was explicitly configured using @option{--with-ust}
22213 to point at such headers. You can explicitly disable the support
22214 using @option{--with-ust=no}.
22216 There are several ways to load the in-process agent in your program:
22219 @item Specifying it as dependency at link time
22221 You can link your program dynamically with the in-process agent
22222 library. On most systems, this is accomplished by adding
22223 @code{-linproctrace} to the link command.
22225 @item Using the system's preloading mechanisms
22227 You can force loading the in-process agent at startup time by using
22228 your system's support for preloading shared libraries. Many Unixes
22229 support the concept of preloading user defined libraries. In most
22230 cases, you do that by specifying @code{LD_PRELOAD=libinproctrace.so}
22231 in the environment. See also the description of @code{gdbserver}'s
22232 @option{--wrapper} command line option.
22234 @item Using @value{GDBN} to force loading the agent at run time
22236 On some systems, you can force the inferior to load a shared library,
22237 by calling a dynamic loader function in the inferior that takes care
22238 of dynamically looking up and loading a shared library. On most Unix
22239 systems, the function is @code{dlopen}. You'll use the @code{call}
22240 command for that. For example:
22243 (@value{GDBP}) call dlopen ("libinproctrace.so", ...)
22246 Note that on most Unix systems, for the @code{dlopen} function to be
22247 available, the program needs to be linked with @code{-ldl}.
22250 On systems that have a userspace dynamic loader, like most Unix
22251 systems, when you connect to @code{gdbserver} using @code{target
22252 remote}, you'll find that the program is stopped at the dynamic
22253 loader's entry point, and no shared library has been loaded in the
22254 program's address space yet, including the in-process agent. In that
22255 case, before being able to use any of the fast or static tracepoints
22256 features, you need to let the loader run and load the shared
22257 libraries. The simplest way to do that is to run the program to the
22258 main procedure. E.g., if debugging a C or C@t{++} program, start
22259 @code{gdbserver} like so:
22262 $ gdbserver :9999 myprogram
22265 Start GDB and connect to @code{gdbserver} like so, and run to main:
22269 (@value{GDBP}) target remote myhost:9999
22270 0x00007f215893ba60 in ?? () from /lib64/ld-linux-x86-64.so.2
22271 (@value{GDBP}) b main
22272 (@value{GDBP}) continue
22275 The in-process tracing agent library should now be loaded into the
22276 process; you can confirm it with the @code{info sharedlibrary}
22277 command, which will list @file{libinproctrace.so} as loaded in the
22278 process. You are now ready to install fast tracepoints, list static
22279 tracepoint markers, probe static tracepoints markers, and start
22282 @node Remote Configuration
22283 @section Remote Configuration
22286 @kindex show remote
22287 This section documents the configuration options available when
22288 debugging remote programs. For the options related to the File I/O
22289 extensions of the remote protocol, see @ref{system,
22290 system-call-allowed}.
22293 @item set remoteaddresssize @var{bits}
22294 @cindex address size for remote targets
22295 @cindex bits in remote address
22296 Set the maximum size of address in a memory packet to the specified
22297 number of bits. @value{GDBN} will mask off the address bits above
22298 that number, when it passes addresses to the remote target. The
22299 default value is the number of bits in the target's address.
22301 @item show remoteaddresssize
22302 Show the current value of remote address size in bits.
22304 @item set serial baud @var{n}
22305 @cindex baud rate for remote targets
22306 Set the baud rate for the remote serial I/O to @var{n} baud. The
22307 value is used to set the speed of the serial port used for debugging
22310 @item show serial baud
22311 Show the current speed of the remote connection.
22313 @item set serial parity @var{parity}
22314 Set the parity for the remote serial I/O. Supported values of @var{parity} are:
22315 @code{even}, @code{none}, and @code{odd}. The default is @code{none}.
22317 @item show serial parity
22318 Show the current parity of the serial port.
22320 @item set remotebreak
22321 @cindex interrupt remote programs
22322 @cindex BREAK signal instead of Ctrl-C
22323 @anchor{set remotebreak}
22324 If set to on, @value{GDBN} sends a @code{BREAK} signal to the remote
22325 when you type @kbd{Ctrl-c} to interrupt the program running
22326 on the remote. If set to off, @value{GDBN} sends the @samp{Ctrl-C}
22327 character instead. The default is off, since most remote systems
22328 expect to see @samp{Ctrl-C} as the interrupt signal.
22330 @item show remotebreak
22331 Show whether @value{GDBN} sends @code{BREAK} or @samp{Ctrl-C} to
22332 interrupt the remote program.
22334 @item set remoteflow on
22335 @itemx set remoteflow off
22336 @kindex set remoteflow
22337 Enable or disable hardware flow control (@code{RTS}/@code{CTS})
22338 on the serial port used to communicate to the remote target.
22340 @item show remoteflow
22341 @kindex show remoteflow
22342 Show the current setting of hardware flow control.
22344 @item set remotelogbase @var{base}
22345 Set the base (a.k.a.@: radix) of logging serial protocol
22346 communications to @var{base}. Supported values of @var{base} are:
22347 @code{ascii}, @code{octal}, and @code{hex}. The default is
22350 @item show remotelogbase
22351 Show the current setting of the radix for logging remote serial
22354 @item set remotelogfile @var{file}
22355 @cindex record serial communications on file
22356 Record remote serial communications on the named @var{file}. The
22357 default is not to record at all.
22359 @item show remotelogfile
22360 Show the current setting of the file name on which to record the
22361 serial communications.
22363 @item set remotetimeout @var{num}
22364 @cindex timeout for serial communications
22365 @cindex remote timeout
22366 Set the timeout limit to wait for the remote target to respond to
22367 @var{num} seconds. The default is 2 seconds.
22369 @item show remotetimeout
22370 Show the current number of seconds to wait for the remote target
22373 @cindex limit hardware breakpoints and watchpoints
22374 @cindex remote target, limit break- and watchpoints
22375 @anchor{set remote hardware-watchpoint-limit}
22376 @anchor{set remote hardware-breakpoint-limit}
22377 @item set remote hardware-watchpoint-limit @var{limit}
22378 @itemx set remote hardware-breakpoint-limit @var{limit}
22379 Restrict @value{GDBN} to using @var{limit} remote hardware watchpoints
22380 or breakpoints. The @var{limit} can be set to 0 to disable hardware
22381 watchpoints or breakpoints, and @code{unlimited} for unlimited
22382 watchpoints or breakpoints.
22384 @item show remote hardware-watchpoint-limit
22385 @itemx show remote hardware-breakpoint-limit
22386 Show the current limit for the number of hardware watchpoints or
22387 breakpoints that @value{GDBN} can use.
22389 @cindex limit hardware watchpoints length
22390 @cindex remote target, limit watchpoints length
22391 @anchor{set remote hardware-watchpoint-length-limit}
22392 @item set remote hardware-watchpoint-length-limit @var{limit}
22393 Restrict @value{GDBN} to using @var{limit} bytes for the maximum
22394 length of a remote hardware watchpoint. A @var{limit} of 0 disables
22395 hardware watchpoints and @code{unlimited} allows watchpoints of any
22398 @item show remote hardware-watchpoint-length-limit
22399 Show the current limit (in bytes) of the maximum length of
22400 a remote hardware watchpoint.
22402 @item set remote exec-file @var{filename}
22403 @itemx show remote exec-file
22404 @anchor{set remote exec-file}
22405 @cindex executable file, for remote target
22406 Select the file used for @code{run} with @code{target
22407 extended-remote}. This should be set to a filename valid on the
22408 target system. If it is not set, the target will use a default
22409 filename (e.g.@: the last program run).
22411 @item set remote interrupt-sequence
22412 @cindex interrupt remote programs
22413 @cindex select Ctrl-C, BREAK or BREAK-g
22414 Allow the user to select one of @samp{Ctrl-C}, a @code{BREAK} or
22415 @samp{BREAK-g} as the
22416 sequence to the remote target in order to interrupt the execution.
22417 @samp{Ctrl-C} is a default. Some system prefers @code{BREAK} which
22418 is high level of serial line for some certain time.
22419 Linux kernel prefers @samp{BREAK-g}, a.k.a Magic SysRq g.
22420 It is @code{BREAK} signal followed by character @code{g}.
22422 @item show interrupt-sequence
22423 Show which of @samp{Ctrl-C}, @code{BREAK} or @code{BREAK-g}
22424 is sent by @value{GDBN} to interrupt the remote program.
22425 @code{BREAK-g} is BREAK signal followed by @code{g} and
22426 also known as Magic SysRq g.
22428 @item set remote interrupt-on-connect
22429 @cindex send interrupt-sequence on start
22430 Specify whether interrupt-sequence is sent to remote target when
22431 @value{GDBN} connects to it. This is mostly needed when you debug
22432 Linux kernel. Linux kernel expects @code{BREAK} followed by @code{g}
22433 which is known as Magic SysRq g in order to connect @value{GDBN}.
22435 @item show interrupt-on-connect
22436 Show whether interrupt-sequence is sent
22437 to remote target when @value{GDBN} connects to it.
22441 @item set tcp auto-retry on
22442 @cindex auto-retry, for remote TCP target
22443 Enable auto-retry for remote TCP connections. This is useful if the remote
22444 debugging agent is launched in parallel with @value{GDBN}; there is a race
22445 condition because the agent may not become ready to accept the connection
22446 before @value{GDBN} attempts to connect. When auto-retry is
22447 enabled, if the initial attempt to connect fails, @value{GDBN} reattempts
22448 to establish the connection using the timeout specified by
22449 @code{set tcp connect-timeout}.
22451 @item set tcp auto-retry off
22452 Do not auto-retry failed TCP connections.
22454 @item show tcp auto-retry
22455 Show the current auto-retry setting.
22457 @item set tcp connect-timeout @var{seconds}
22458 @itemx set tcp connect-timeout unlimited
22459 @cindex connection timeout, for remote TCP target
22460 @cindex timeout, for remote target connection
22461 Set the timeout for establishing a TCP connection to the remote target to
22462 @var{seconds}. The timeout affects both polling to retry failed connections
22463 (enabled by @code{set tcp auto-retry on}) and waiting for connections
22464 that are merely slow to complete, and represents an approximate cumulative
22465 value. If @var{seconds} is @code{unlimited}, there is no timeout and
22466 @value{GDBN} will keep attempting to establish a connection forever,
22467 unless interrupted with @kbd{Ctrl-c}. The default is 15 seconds.
22469 @item show tcp connect-timeout
22470 Show the current connection timeout setting.
22473 @cindex remote packets, enabling and disabling
22474 The @value{GDBN} remote protocol autodetects the packets supported by
22475 your debugging stub. If you need to override the autodetection, you
22476 can use these commands to enable or disable individual packets. Each
22477 packet can be set to @samp{on} (the remote target supports this
22478 packet), @samp{off} (the remote target does not support this packet),
22479 or @samp{auto} (detect remote target support for this packet). They
22480 all default to @samp{auto}. For more information about each packet,
22481 see @ref{Remote Protocol}.
22483 During normal use, you should not have to use any of these commands.
22484 If you do, that may be a bug in your remote debugging stub, or a bug
22485 in @value{GDBN}. You may want to report the problem to the
22486 @value{GDBN} developers.
22488 For each packet @var{name}, the command to enable or disable the
22489 packet is @code{set remote @var{name}-packet}. The available settings
22492 @multitable @columnfractions 0.28 0.32 0.25
22495 @tab Related Features
22497 @item @code{fetch-register}
22499 @tab @code{info registers}
22501 @item @code{set-register}
22505 @item @code{binary-download}
22507 @tab @code{load}, @code{set}
22509 @item @code{read-aux-vector}
22510 @tab @code{qXfer:auxv:read}
22511 @tab @code{info auxv}
22513 @item @code{symbol-lookup}
22514 @tab @code{qSymbol}
22515 @tab Detecting multiple threads
22517 @item @code{attach}
22518 @tab @code{vAttach}
22521 @item @code{verbose-resume}
22523 @tab Stepping or resuming multiple threads
22529 @item @code{software-breakpoint}
22533 @item @code{hardware-breakpoint}
22537 @item @code{write-watchpoint}
22541 @item @code{read-watchpoint}
22545 @item @code{access-watchpoint}
22549 @item @code{pid-to-exec-file}
22550 @tab @code{qXfer:exec-file:read}
22551 @tab @code{attach}, @code{run}
22553 @item @code{target-features}
22554 @tab @code{qXfer:features:read}
22555 @tab @code{set architecture}
22557 @item @code{library-info}
22558 @tab @code{qXfer:libraries:read}
22559 @tab @code{info sharedlibrary}
22561 @item @code{memory-map}
22562 @tab @code{qXfer:memory-map:read}
22563 @tab @code{info mem}
22565 @item @code{read-sdata-object}
22566 @tab @code{qXfer:sdata:read}
22567 @tab @code{print $_sdata}
22569 @item @code{read-siginfo-object}
22570 @tab @code{qXfer:siginfo:read}
22571 @tab @code{print $_siginfo}
22573 @item @code{write-siginfo-object}
22574 @tab @code{qXfer:siginfo:write}
22575 @tab @code{set $_siginfo}
22577 @item @code{threads}
22578 @tab @code{qXfer:threads:read}
22579 @tab @code{info threads}
22581 @item @code{get-thread-local-@*storage-address}
22582 @tab @code{qGetTLSAddr}
22583 @tab Displaying @code{__thread} variables
22585 @item @code{get-thread-information-block-address}
22586 @tab @code{qGetTIBAddr}
22587 @tab Display MS-Windows Thread Information Block.
22589 @item @code{search-memory}
22590 @tab @code{qSearch:memory}
22593 @item @code{supported-packets}
22594 @tab @code{qSupported}
22595 @tab Remote communications parameters
22597 @item @code{catch-syscalls}
22598 @tab @code{QCatchSyscalls}
22599 @tab @code{catch syscall}
22601 @item @code{pass-signals}
22602 @tab @code{QPassSignals}
22603 @tab @code{handle @var{signal}}
22605 @item @code{program-signals}
22606 @tab @code{QProgramSignals}
22607 @tab @code{handle @var{signal}}
22609 @item @code{hostio-close-packet}
22610 @tab @code{vFile:close}
22611 @tab @code{remote get}, @code{remote put}
22613 @item @code{hostio-open-packet}
22614 @tab @code{vFile:open}
22615 @tab @code{remote get}, @code{remote put}
22617 @item @code{hostio-pread-packet}
22618 @tab @code{vFile:pread}
22619 @tab @code{remote get}, @code{remote put}
22621 @item @code{hostio-pwrite-packet}
22622 @tab @code{vFile:pwrite}
22623 @tab @code{remote get}, @code{remote put}
22625 @item @code{hostio-unlink-packet}
22626 @tab @code{vFile:unlink}
22627 @tab @code{remote delete}
22629 @item @code{hostio-readlink-packet}
22630 @tab @code{vFile:readlink}
22633 @item @code{hostio-fstat-packet}
22634 @tab @code{vFile:fstat}
22637 @item @code{hostio-setfs-packet}
22638 @tab @code{vFile:setfs}
22641 @item @code{noack-packet}
22642 @tab @code{QStartNoAckMode}
22643 @tab Packet acknowledgment
22645 @item @code{osdata}
22646 @tab @code{qXfer:osdata:read}
22647 @tab @code{info os}
22649 @item @code{query-attached}
22650 @tab @code{qAttached}
22651 @tab Querying remote process attach state.
22653 @item @code{trace-buffer-size}
22654 @tab @code{QTBuffer:size}
22655 @tab @code{set trace-buffer-size}
22657 @item @code{trace-status}
22658 @tab @code{qTStatus}
22659 @tab @code{tstatus}
22661 @item @code{traceframe-info}
22662 @tab @code{qXfer:traceframe-info:read}
22663 @tab Traceframe info
22665 @item @code{install-in-trace}
22666 @tab @code{InstallInTrace}
22667 @tab Install tracepoint in tracing
22669 @item @code{disable-randomization}
22670 @tab @code{QDisableRandomization}
22671 @tab @code{set disable-randomization}
22673 @item @code{startup-with-shell}
22674 @tab @code{QStartupWithShell}
22675 @tab @code{set startup-with-shell}
22677 @item @code{environment-hex-encoded}
22678 @tab @code{QEnvironmentHexEncoded}
22679 @tab @code{set environment}
22681 @item @code{environment-unset}
22682 @tab @code{QEnvironmentUnset}
22683 @tab @code{unset environment}
22685 @item @code{environment-reset}
22686 @tab @code{QEnvironmentReset}
22687 @tab @code{Reset the inferior environment (i.e., unset user-set variables)}
22689 @item @code{set-working-dir}
22690 @tab @code{QSetWorkingDir}
22691 @tab @code{set cwd}
22693 @item @code{conditional-breakpoints-packet}
22694 @tab @code{Z0 and Z1}
22695 @tab @code{Support for target-side breakpoint condition evaluation}
22697 @item @code{multiprocess-extensions}
22698 @tab @code{multiprocess extensions}
22699 @tab Debug multiple processes and remote process PID awareness
22701 @item @code{swbreak-feature}
22702 @tab @code{swbreak stop reason}
22705 @item @code{hwbreak-feature}
22706 @tab @code{hwbreak stop reason}
22709 @item @code{fork-event-feature}
22710 @tab @code{fork stop reason}
22713 @item @code{vfork-event-feature}
22714 @tab @code{vfork stop reason}
22717 @item @code{exec-event-feature}
22718 @tab @code{exec stop reason}
22721 @item @code{thread-events}
22722 @tab @code{QThreadEvents}
22723 @tab Tracking thread lifetime.
22725 @item @code{no-resumed-stop-reply}
22726 @tab @code{no resumed thread left stop reply}
22727 @tab Tracking thread lifetime.
22732 @section Implementing a Remote Stub
22734 @cindex debugging stub, example
22735 @cindex remote stub, example
22736 @cindex stub example, remote debugging
22737 The stub files provided with @value{GDBN} implement the target side of the
22738 communication protocol, and the @value{GDBN} side is implemented in the
22739 @value{GDBN} source file @file{remote.c}. Normally, you can simply allow
22740 these subroutines to communicate, and ignore the details. (If you're
22741 implementing your own stub file, you can still ignore the details: start
22742 with one of the existing stub files. @file{sparc-stub.c} is the best
22743 organized, and therefore the easiest to read.)
22745 @cindex remote serial debugging, overview
22746 To debug a program running on another machine (the debugging
22747 @dfn{target} machine), you must first arrange for all the usual
22748 prerequisites for the program to run by itself. For example, for a C
22753 A startup routine to set up the C runtime environment; these usually
22754 have a name like @file{crt0}. The startup routine may be supplied by
22755 your hardware supplier, or you may have to write your own.
22758 A C subroutine library to support your program's
22759 subroutine calls, notably managing input and output.
22762 A way of getting your program to the other machine---for example, a
22763 download program. These are often supplied by the hardware
22764 manufacturer, but you may have to write your own from hardware
22768 The next step is to arrange for your program to use a serial port to
22769 communicate with the machine where @value{GDBN} is running (the @dfn{host}
22770 machine). In general terms, the scheme looks like this:
22774 @value{GDBN} already understands how to use this protocol; when everything
22775 else is set up, you can simply use the @samp{target remote} command
22776 (@pxref{Targets,,Specifying a Debugging Target}).
22778 @item On the target,
22779 you must link with your program a few special-purpose subroutines that
22780 implement the @value{GDBN} remote serial protocol. The file containing these
22781 subroutines is called a @dfn{debugging stub}.
22783 On certain remote targets, you can use an auxiliary program
22784 @code{gdbserver} instead of linking a stub into your program.
22785 @xref{Server,,Using the @code{gdbserver} Program}, for details.
22788 The debugging stub is specific to the architecture of the remote
22789 machine; for example, use @file{sparc-stub.c} to debug programs on
22792 @cindex remote serial stub list
22793 These working remote stubs are distributed with @value{GDBN}:
22798 @cindex @file{i386-stub.c}
22801 For Intel 386 and compatible architectures.
22804 @cindex @file{m68k-stub.c}
22805 @cindex Motorola 680x0
22807 For Motorola 680x0 architectures.
22810 @cindex @file{sh-stub.c}
22813 For Renesas SH architectures.
22816 @cindex @file{sparc-stub.c}
22818 For @sc{sparc} architectures.
22820 @item sparcl-stub.c
22821 @cindex @file{sparcl-stub.c}
22824 For Fujitsu @sc{sparclite} architectures.
22828 The @file{README} file in the @value{GDBN} distribution may list other
22829 recently added stubs.
22832 * Stub Contents:: What the stub can do for you
22833 * Bootstrapping:: What you must do for the stub
22834 * Debug Session:: Putting it all together
22837 @node Stub Contents
22838 @subsection What the Stub Can Do for You
22840 @cindex remote serial stub
22841 The debugging stub for your architecture supplies these three
22845 @item set_debug_traps
22846 @findex set_debug_traps
22847 @cindex remote serial stub, initialization
22848 This routine arranges for @code{handle_exception} to run when your
22849 program stops. You must call this subroutine explicitly in your
22850 program's startup code.
22852 @item handle_exception
22853 @findex handle_exception
22854 @cindex remote serial stub, main routine
22855 This is the central workhorse, but your program never calls it
22856 explicitly---the setup code arranges for @code{handle_exception} to
22857 run when a trap is triggered.
22859 @code{handle_exception} takes control when your program stops during
22860 execution (for example, on a breakpoint), and mediates communications
22861 with @value{GDBN} on the host machine. This is where the communications
22862 protocol is implemented; @code{handle_exception} acts as the @value{GDBN}
22863 representative on the target machine. It begins by sending summary
22864 information on the state of your program, then continues to execute,
22865 retrieving and transmitting any information @value{GDBN} needs, until you
22866 execute a @value{GDBN} command that makes your program resume; at that point,
22867 @code{handle_exception} returns control to your own code on the target
22871 @cindex @code{breakpoint} subroutine, remote
22872 Use this auxiliary subroutine to make your program contain a
22873 breakpoint. Depending on the particular situation, this may be the only
22874 way for @value{GDBN} to get control. For instance, if your target
22875 machine has some sort of interrupt button, you won't need to call this;
22876 pressing the interrupt button transfers control to
22877 @code{handle_exception}---in effect, to @value{GDBN}. On some machines,
22878 simply receiving characters on the serial port may also trigger a trap;
22879 again, in that situation, you don't need to call @code{breakpoint} from
22880 your own program---simply running @samp{target remote} from the host
22881 @value{GDBN} session gets control.
22883 Call @code{breakpoint} if none of these is true, or if you simply want
22884 to make certain your program stops at a predetermined point for the
22885 start of your debugging session.
22888 @node Bootstrapping
22889 @subsection What You Must Do for the Stub
22891 @cindex remote stub, support routines
22892 The debugging stubs that come with @value{GDBN} are set up for a particular
22893 chip architecture, but they have no information about the rest of your
22894 debugging target machine.
22896 First of all you need to tell the stub how to communicate with the
22900 @item int getDebugChar()
22901 @findex getDebugChar
22902 Write this subroutine to read a single character from the serial port.
22903 It may be identical to @code{getchar} for your target system; a
22904 different name is used to allow you to distinguish the two if you wish.
22906 @item void putDebugChar(int)
22907 @findex putDebugChar
22908 Write this subroutine to write a single character to the serial port.
22909 It may be identical to @code{putchar} for your target system; a
22910 different name is used to allow you to distinguish the two if you wish.
22913 @cindex control C, and remote debugging
22914 @cindex interrupting remote targets
22915 If you want @value{GDBN} to be able to stop your program while it is
22916 running, you need to use an interrupt-driven serial driver, and arrange
22917 for it to stop when it receives a @code{^C} (@samp{\003}, the control-C
22918 character). That is the character which @value{GDBN} uses to tell the
22919 remote system to stop.
22921 Getting the debugging target to return the proper status to @value{GDBN}
22922 probably requires changes to the standard stub; one quick and dirty way
22923 is to just execute a breakpoint instruction (the ``dirty'' part is that
22924 @value{GDBN} reports a @code{SIGTRAP} instead of a @code{SIGINT}).
22926 Other routines you need to supply are:
22929 @item void exceptionHandler (int @var{exception_number}, void *@var{exception_address})
22930 @findex exceptionHandler
22931 Write this function to install @var{exception_address} in the exception
22932 handling tables. You need to do this because the stub does not have any
22933 way of knowing what the exception handling tables on your target system
22934 are like (for example, the processor's table might be in @sc{rom},
22935 containing entries which point to a table in @sc{ram}).
22936 The @var{exception_number} specifies the exception which should be changed;
22937 its meaning is architecture-dependent (for example, different numbers
22938 might represent divide by zero, misaligned access, etc). When this
22939 exception occurs, control should be transferred directly to
22940 @var{exception_address}, and the processor state (stack, registers,
22941 and so on) should be just as it is when a processor exception occurs. So if
22942 you want to use a jump instruction to reach @var{exception_address}, it
22943 should be a simple jump, not a jump to subroutine.
22945 For the 386, @var{exception_address} should be installed as an interrupt
22946 gate so that interrupts are masked while the handler runs. The gate
22947 should be at privilege level 0 (the most privileged level). The
22948 @sc{sparc} and 68k stubs are able to mask interrupts themselves without
22949 help from @code{exceptionHandler}.
22951 @item void flush_i_cache()
22952 @findex flush_i_cache
22953 On @sc{sparc} and @sc{sparclite} only, write this subroutine to flush the
22954 instruction cache, if any, on your target machine. If there is no
22955 instruction cache, this subroutine may be a no-op.
22957 On target machines that have instruction caches, @value{GDBN} requires this
22958 function to make certain that the state of your program is stable.
22962 You must also make sure this library routine is available:
22965 @item void *memset(void *, int, int)
22967 This is the standard library function @code{memset} that sets an area of
22968 memory to a known value. If you have one of the free versions of
22969 @code{libc.a}, @code{memset} can be found there; otherwise, you must
22970 either obtain it from your hardware manufacturer, or write your own.
22973 If you do not use the GNU C compiler, you may need other standard
22974 library subroutines as well; this varies from one stub to another,
22975 but in general the stubs are likely to use any of the common library
22976 subroutines which @code{@value{NGCC}} generates as inline code.
22979 @node Debug Session
22980 @subsection Putting it All Together
22982 @cindex remote serial debugging summary
22983 In summary, when your program is ready to debug, you must follow these
22988 Make sure you have defined the supporting low-level routines
22989 (@pxref{Bootstrapping,,What You Must Do for the Stub}):
22991 @code{getDebugChar}, @code{putDebugChar},
22992 @code{flush_i_cache}, @code{memset}, @code{exceptionHandler}.
22996 Insert these lines in your program's startup code, before the main
22997 procedure is called:
23004 On some machines, when a breakpoint trap is raised, the hardware
23005 automatically makes the PC point to the instruction after the
23006 breakpoint. If your machine doesn't do that, you may need to adjust
23007 @code{handle_exception} to arrange for it to return to the instruction
23008 after the breakpoint on this first invocation, so that your program
23009 doesn't keep hitting the initial breakpoint instead of making
23013 For the 680x0 stub only, you need to provide a variable called
23014 @code{exceptionHook}. Normally you just use:
23017 void (*exceptionHook)() = 0;
23021 but if before calling @code{set_debug_traps}, you set it to point to a
23022 function in your program, that function is called when
23023 @code{@value{GDBN}} continues after stopping on a trap (for example, bus
23024 error). The function indicated by @code{exceptionHook} is called with
23025 one parameter: an @code{int} which is the exception number.
23028 Compile and link together: your program, the @value{GDBN} debugging stub for
23029 your target architecture, and the supporting subroutines.
23032 Make sure you have a serial connection between your target machine and
23033 the @value{GDBN} host, and identify the serial port on the host.
23036 @c The "remote" target now provides a `load' command, so we should
23037 @c document that. FIXME.
23038 Download your program to your target machine (or get it there by
23039 whatever means the manufacturer provides), and start it.
23042 Start @value{GDBN} on the host, and connect to the target
23043 (@pxref{Connecting,,Connecting to a Remote Target}).
23047 @node Configurations
23048 @chapter Configuration-Specific Information
23050 While nearly all @value{GDBN} commands are available for all native and
23051 cross versions of the debugger, there are some exceptions. This chapter
23052 describes things that are only available in certain configurations.
23054 There are three major categories of configurations: native
23055 configurations, where the host and target are the same, embedded
23056 operating system configurations, which are usually the same for several
23057 different processor architectures, and bare embedded processors, which
23058 are quite different from each other.
23063 * Embedded Processors::
23070 This section describes details specific to particular native
23074 * BSD libkvm Interface:: Debugging BSD kernel memory images
23075 * Process Information:: Process information
23076 * DJGPP Native:: Features specific to the DJGPP port
23077 * Cygwin Native:: Features specific to the Cygwin port
23078 * Hurd Native:: Features specific to @sc{gnu} Hurd
23079 * Darwin:: Features specific to Darwin
23080 * FreeBSD:: Features specific to FreeBSD
23083 @node BSD libkvm Interface
23084 @subsection BSD libkvm Interface
23087 @cindex kernel memory image
23088 @cindex kernel crash dump
23090 BSD-derived systems (FreeBSD/NetBSD/OpenBSD) have a kernel memory
23091 interface that provides a uniform interface for accessing kernel virtual
23092 memory images, including live systems and crash dumps. @value{GDBN}
23093 uses this interface to allow you to debug live kernels and kernel crash
23094 dumps on many native BSD configurations. This is implemented as a
23095 special @code{kvm} debugging target. For debugging a live system, load
23096 the currently running kernel into @value{GDBN} and connect to the
23100 (@value{GDBP}) @b{target kvm}
23103 For debugging crash dumps, provide the file name of the crash dump as an
23107 (@value{GDBP}) @b{target kvm /var/crash/bsd.0}
23110 Once connected to the @code{kvm} target, the following commands are
23116 Set current context from the @dfn{Process Control Block} (PCB) address.
23119 Set current context from proc address. This command isn't available on
23120 modern FreeBSD systems.
23123 @node Process Information
23124 @subsection Process Information
23126 @cindex examine process image
23127 @cindex process info via @file{/proc}
23129 Some operating systems provide interfaces to fetch additional
23130 information about running processes beyond memory and per-thread
23131 register state. If @value{GDBN} is configured for an operating system
23132 with a supported interface, the command @code{info proc} is available
23133 to report information about the process running your program, or about
23134 any process running on your system.
23136 One supported interface is a facility called @samp{/proc} that can be
23137 used to examine the image of a running process using file-system
23138 subroutines. This facility is supported on @sc{gnu}/Linux and Solaris
23141 On FreeBSD systems, system control nodes are used to query process
23144 In addition, some systems may provide additional process information
23145 in core files. Note that a core file may include a subset of the
23146 information available from a live process. Process information is
23147 currently avaiable from cores created on @sc{gnu}/Linux and FreeBSD
23154 @itemx info proc @var{process-id}
23155 Summarize available information about a process. If a
23156 process ID is specified by @var{process-id}, display information about
23157 that process; otherwise display information about the program being
23158 debugged. The summary includes the debugged process ID, the command
23159 line used to invoke it, its current working directory, and its
23160 executable file's absolute file name.
23162 On some systems, @var{process-id} can be of the form
23163 @samp{[@var{pid}]/@var{tid}} which specifies a certain thread ID
23164 within a process. If the optional @var{pid} part is missing, it means
23165 a thread from the process being debugged (the leading @samp{/} still
23166 needs to be present, or else @value{GDBN} will interpret the number as
23167 a process ID rather than a thread ID).
23169 @item info proc cmdline
23170 @cindex info proc cmdline
23171 Show the original command line of the process. This command is
23172 supported on @sc{gnu}/Linux and FreeBSD.
23174 @item info proc cwd
23175 @cindex info proc cwd
23176 Show the current working directory of the process. This command is
23177 supported on @sc{gnu}/Linux and FreeBSD.
23179 @item info proc exe
23180 @cindex info proc exe
23181 Show the name of executable of the process. This command is supported
23182 on @sc{gnu}/Linux and FreeBSD.
23184 @item info proc files
23185 @cindex info proc files
23186 Show the file descriptors open by the process. For each open file
23187 descriptor, @value{GDBN} shows its number, type (file, directory,
23188 character device, socket), file pointer offset, and the name of the
23189 resource open on the descriptor. The resource name can be a file name
23190 (for files, directories, and devices) or a protocol followed by socket
23191 address (for network connections). This command is supported on
23194 This example shows the open file descriptors for a process using a
23195 tty for standard input and output as well as two network sockets:
23198 (gdb) info proc files 22136
23202 FD Type Offset Flags Name
23203 text file - r-------- /usr/bin/ssh
23204 ctty chr - rw------- /dev/pts/20
23205 cwd dir - r-------- /usr/home/john
23206 root dir - r-------- /
23207 0 chr 0x32933a4 rw------- /dev/pts/20
23208 1 chr 0x32933a4 rw------- /dev/pts/20
23209 2 chr 0x32933a4 rw------- /dev/pts/20
23210 3 socket 0x0 rw----n-- tcp4 10.0.1.2:53014 -> 10.0.1.10:22
23211 4 socket 0x0 rw------- unix stream:/tmp/ssh-FIt89oAzOn5f/agent.2456
23214 @item info proc mappings
23215 @cindex memory address space mappings
23216 Report the memory address space ranges accessible in a process. On
23217 Solaris and FreeBSD systems, each memory range includes information on
23218 whether the process has read, write, or execute access rights to each
23219 range. On @sc{gnu}/Linux and FreeBSD systems, each memory range
23220 includes the object file which is mapped to that range.
23222 @item info proc stat
23223 @itemx info proc status
23224 @cindex process detailed status information
23225 Show additional process-related information, including the user ID and
23226 group ID; virtual memory usage; the signals that are pending, blocked,
23227 and ignored; its TTY; its consumption of system and user time; its
23228 stack size; its @samp{nice} value; etc. These commands are supported
23229 on @sc{gnu}/Linux and FreeBSD.
23231 For @sc{gnu}/Linux systems, see the @samp{proc} man page for more
23232 information (type @kbd{man 5 proc} from your shell prompt).
23234 For FreeBSD systems, @code{info proc stat} is an alias for @code{info
23237 @item info proc all
23238 Show all the information about the process described under all of the
23239 above @code{info proc} subcommands.
23242 @comment These sub-options of 'info proc' were not included when
23243 @comment procfs.c was re-written. Keep their descriptions around
23244 @comment against the day when someone finds the time to put them back in.
23245 @kindex info proc times
23246 @item info proc times
23247 Starting time, user CPU time, and system CPU time for your program and
23250 @kindex info proc id
23252 Report on the process IDs related to your program: its own process ID,
23253 the ID of its parent, the process group ID, and the session ID.
23256 @item set procfs-trace
23257 @kindex set procfs-trace
23258 @cindex @code{procfs} API calls
23259 This command enables and disables tracing of @code{procfs} API calls.
23261 @item show procfs-trace
23262 @kindex show procfs-trace
23263 Show the current state of @code{procfs} API call tracing.
23265 @item set procfs-file @var{file}
23266 @kindex set procfs-file
23267 Tell @value{GDBN} to write @code{procfs} API trace to the named
23268 @var{file}. @value{GDBN} appends the trace info to the previous
23269 contents of the file. The default is to display the trace on the
23272 @item show procfs-file
23273 @kindex show procfs-file
23274 Show the file to which @code{procfs} API trace is written.
23276 @item proc-trace-entry
23277 @itemx proc-trace-exit
23278 @itemx proc-untrace-entry
23279 @itemx proc-untrace-exit
23280 @kindex proc-trace-entry
23281 @kindex proc-trace-exit
23282 @kindex proc-untrace-entry
23283 @kindex proc-untrace-exit
23284 These commands enable and disable tracing of entries into and exits
23285 from the @code{syscall} interface.
23288 @kindex info pidlist
23289 @cindex process list, QNX Neutrino
23290 For QNX Neutrino only, this command displays the list of all the
23291 processes and all the threads within each process.
23294 @kindex info meminfo
23295 @cindex mapinfo list, QNX Neutrino
23296 For QNX Neutrino only, this command displays the list of all mapinfos.
23300 @subsection Features for Debugging @sc{djgpp} Programs
23301 @cindex @sc{djgpp} debugging
23302 @cindex native @sc{djgpp} debugging
23303 @cindex MS-DOS-specific commands
23306 @sc{djgpp} is a port of the @sc{gnu} development tools to MS-DOS and
23307 MS-Windows. @sc{djgpp} programs are 32-bit protected-mode programs
23308 that use the @dfn{DPMI} (DOS Protected-Mode Interface) API to run on
23309 top of real-mode DOS systems and their emulations.
23311 @value{GDBN} supports native debugging of @sc{djgpp} programs, and
23312 defines a few commands specific to the @sc{djgpp} port. This
23313 subsection describes those commands.
23318 This is a prefix of @sc{djgpp}-specific commands which print
23319 information about the target system and important OS structures.
23322 @cindex MS-DOS system info
23323 @cindex free memory information (MS-DOS)
23324 @item info dos sysinfo
23325 This command displays assorted information about the underlying
23326 platform: the CPU type and features, the OS version and flavor, the
23327 DPMI version, and the available conventional and DPMI memory.
23332 @cindex segment descriptor tables
23333 @cindex descriptor tables display
23335 @itemx info dos ldt
23336 @itemx info dos idt
23337 These 3 commands display entries from, respectively, Global, Local,
23338 and Interrupt Descriptor Tables (GDT, LDT, and IDT). The descriptor
23339 tables are data structures which store a descriptor for each segment
23340 that is currently in use. The segment's selector is an index into a
23341 descriptor table; the table entry for that index holds the
23342 descriptor's base address and limit, and its attributes and access
23345 A typical @sc{djgpp} program uses 3 segments: a code segment, a data
23346 segment (used for both data and the stack), and a DOS segment (which
23347 allows access to DOS/BIOS data structures and absolute addresses in
23348 conventional memory). However, the DPMI host will usually define
23349 additional segments in order to support the DPMI environment.
23351 @cindex garbled pointers
23352 These commands allow to display entries from the descriptor tables.
23353 Without an argument, all entries from the specified table are
23354 displayed. An argument, which should be an integer expression, means
23355 display a single entry whose index is given by the argument. For
23356 example, here's a convenient way to display information about the
23357 debugged program's data segment:
23360 @exdent @code{(@value{GDBP}) info dos ldt $ds}
23361 @exdent @code{0x13f: base=0x11970000 limit=0x0009ffff 32-Bit Data (Read/Write, Exp-up)}
23365 This comes in handy when you want to see whether a pointer is outside
23366 the data segment's limit (i.e.@: @dfn{garbled}).
23368 @cindex page tables display (MS-DOS)
23370 @itemx info dos pte
23371 These two commands display entries from, respectively, the Page
23372 Directory and the Page Tables. Page Directories and Page Tables are
23373 data structures which control how virtual memory addresses are mapped
23374 into physical addresses. A Page Table includes an entry for every
23375 page of memory that is mapped into the program's address space; there
23376 may be several Page Tables, each one holding up to 4096 entries. A
23377 Page Directory has up to 4096 entries, one each for every Page Table
23378 that is currently in use.
23380 Without an argument, @kbd{info dos pde} displays the entire Page
23381 Directory, and @kbd{info dos pte} displays all the entries in all of
23382 the Page Tables. An argument, an integer expression, given to the
23383 @kbd{info dos pde} command means display only that entry from the Page
23384 Directory table. An argument given to the @kbd{info dos pte} command
23385 means display entries from a single Page Table, the one pointed to by
23386 the specified entry in the Page Directory.
23388 @cindex direct memory access (DMA) on MS-DOS
23389 These commands are useful when your program uses @dfn{DMA} (Direct
23390 Memory Access), which needs physical addresses to program the DMA
23393 These commands are supported only with some DPMI servers.
23395 @cindex physical address from linear address
23396 @item info dos address-pte @var{addr}
23397 This command displays the Page Table entry for a specified linear
23398 address. The argument @var{addr} is a linear address which should
23399 already have the appropriate segment's base address added to it,
23400 because this command accepts addresses which may belong to @emph{any}
23401 segment. For example, here's how to display the Page Table entry for
23402 the page where a variable @code{i} is stored:
23405 @exdent @code{(@value{GDBP}) info dos address-pte __djgpp_base_address + (char *)&i}
23406 @exdent @code{Page Table entry for address 0x11a00d30:}
23407 @exdent @code{Base=0x02698000 Dirty Acc. Not-Cached Write-Back Usr Read-Write +0xd30}
23411 This says that @code{i} is stored at offset @code{0xd30} from the page
23412 whose physical base address is @code{0x02698000}, and shows all the
23413 attributes of that page.
23415 Note that you must cast the addresses of variables to a @code{char *},
23416 since otherwise the value of @code{__djgpp_base_address}, the base
23417 address of all variables and functions in a @sc{djgpp} program, will
23418 be added using the rules of C pointer arithmetics: if @code{i} is
23419 declared an @code{int}, @value{GDBN} will add 4 times the value of
23420 @code{__djgpp_base_address} to the address of @code{i}.
23422 Here's another example, it displays the Page Table entry for the
23426 @exdent @code{(@value{GDBP}) info dos address-pte *((unsigned *)&_go32_info_block + 3)}
23427 @exdent @code{Page Table entry for address 0x29110:}
23428 @exdent @code{Base=0x00029000 Dirty Acc. Not-Cached Write-Back Usr Read-Write +0x110}
23432 (The @code{+ 3} offset is because the transfer buffer's address is the
23433 3rd member of the @code{_go32_info_block} structure.) The output
23434 clearly shows that this DPMI server maps the addresses in conventional
23435 memory 1:1, i.e.@: the physical (@code{0x00029000} + @code{0x110}) and
23436 linear (@code{0x29110}) addresses are identical.
23438 This command is supported only with some DPMI servers.
23441 @cindex DOS serial data link, remote debugging
23442 In addition to native debugging, the DJGPP port supports remote
23443 debugging via a serial data link. The following commands are specific
23444 to remote serial debugging in the DJGPP port of @value{GDBN}.
23447 @kindex set com1base
23448 @kindex set com1irq
23449 @kindex set com2base
23450 @kindex set com2irq
23451 @kindex set com3base
23452 @kindex set com3irq
23453 @kindex set com4base
23454 @kindex set com4irq
23455 @item set com1base @var{addr}
23456 This command sets the base I/O port address of the @file{COM1} serial
23459 @item set com1irq @var{irq}
23460 This command sets the @dfn{Interrupt Request} (@code{IRQ}) line to use
23461 for the @file{COM1} serial port.
23463 There are similar commands @samp{set com2base}, @samp{set com3irq},
23464 etc.@: for setting the port address and the @code{IRQ} lines for the
23467 @kindex show com1base
23468 @kindex show com1irq
23469 @kindex show com2base
23470 @kindex show com2irq
23471 @kindex show com3base
23472 @kindex show com3irq
23473 @kindex show com4base
23474 @kindex show com4irq
23475 The related commands @samp{show com1base}, @samp{show com1irq} etc.@:
23476 display the current settings of the base address and the @code{IRQ}
23477 lines used by the COM ports.
23480 @kindex info serial
23481 @cindex DOS serial port status
23482 This command prints the status of the 4 DOS serial ports. For each
23483 port, it prints whether it's active or not, its I/O base address and
23484 IRQ number, whether it uses a 16550-style FIFO, its baudrate, and the
23485 counts of various errors encountered so far.
23489 @node Cygwin Native
23490 @subsection Features for Debugging MS Windows PE Executables
23491 @cindex MS Windows debugging
23492 @cindex native Cygwin debugging
23493 @cindex Cygwin-specific commands
23495 @value{GDBN} supports native debugging of MS Windows programs, including
23496 DLLs with and without symbolic debugging information.
23498 @cindex Ctrl-BREAK, MS-Windows
23499 @cindex interrupt debuggee on MS-Windows
23500 MS-Windows programs that call @code{SetConsoleMode} to switch off the
23501 special meaning of the @samp{Ctrl-C} keystroke cannot be interrupted
23502 by typing @kbd{C-c}. For this reason, @value{GDBN} on MS-Windows
23503 supports @kbd{C-@key{BREAK}} as an alternative interrupt key
23504 sequence, which can be used to interrupt the debuggee even if it
23507 There are various additional Cygwin-specific commands, described in
23508 this section. Working with DLLs that have no debugging symbols is
23509 described in @ref{Non-debug DLL Symbols}.
23514 This is a prefix of MS Windows-specific commands which print
23515 information about the target system and important OS structures.
23517 @item info w32 selector
23518 This command displays information returned by
23519 the Win32 API @code{GetThreadSelectorEntry} function.
23520 It takes an optional argument that is evaluated to
23521 a long value to give the information about this given selector.
23522 Without argument, this command displays information
23523 about the six segment registers.
23525 @item info w32 thread-information-block
23526 This command displays thread specific information stored in the
23527 Thread Information Block (readable on the X86 CPU family using @code{$fs}
23528 selector for 32-bit programs and @code{$gs} for 64-bit programs).
23530 @kindex signal-event
23531 @item signal-event @var{id}
23532 This command signals an event with user-provided @var{id}. Used to resume
23533 crashing process when attached to it using MS-Windows JIT debugging (AeDebug).
23535 To use it, create or edit the following keys in
23536 @code{HKLM\SOFTWARE\Microsoft\Windows NT\CurrentVersion\AeDebug} and/or
23537 @code{HKLM\SOFTWARE\Wow6432Node\Microsoft\Windows NT\CurrentVersion\AeDebug}
23538 (for x86_64 versions):
23542 @code{Debugger} (REG_SZ) --- a command to launch the debugger.
23543 Suggested command is: @code{@var{fully-qualified-path-to-gdb.exe} -ex
23544 "attach %ld" -ex "signal-event %ld" -ex "continue"}.
23546 The first @code{%ld} will be replaced by the process ID of the
23547 crashing process, the second @code{%ld} will be replaced by the ID of
23548 the event that blocks the crashing process, waiting for @value{GDBN}
23552 @code{Auto} (REG_SZ) --- either @code{1} or @code{0}. @code{1} will
23553 make the system run debugger specified by the Debugger key
23554 automatically, @code{0} will cause a dialog box with ``OK'' and
23555 ``Cancel'' buttons to appear, which allows the user to either
23556 terminate the crashing process (OK) or debug it (Cancel).
23559 @kindex set cygwin-exceptions
23560 @cindex debugging the Cygwin DLL
23561 @cindex Cygwin DLL, debugging
23562 @item set cygwin-exceptions @var{mode}
23563 If @var{mode} is @code{on}, @value{GDBN} will break on exceptions that
23564 happen inside the Cygwin DLL. If @var{mode} is @code{off},
23565 @value{GDBN} will delay recognition of exceptions, and may ignore some
23566 exceptions which seem to be caused by internal Cygwin DLL
23567 ``bookkeeping''. This option is meant primarily for debugging the
23568 Cygwin DLL itself; the default value is @code{off} to avoid annoying
23569 @value{GDBN} users with false @code{SIGSEGV} signals.
23571 @kindex show cygwin-exceptions
23572 @item show cygwin-exceptions
23573 Displays whether @value{GDBN} will break on exceptions that happen
23574 inside the Cygwin DLL itself.
23576 @kindex set new-console
23577 @item set new-console @var{mode}
23578 If @var{mode} is @code{on} the debuggee will
23579 be started in a new console on next start.
23580 If @var{mode} is @code{off}, the debuggee will
23581 be started in the same console as the debugger.
23583 @kindex show new-console
23584 @item show new-console
23585 Displays whether a new console is used
23586 when the debuggee is started.
23588 @kindex set new-group
23589 @item set new-group @var{mode}
23590 This boolean value controls whether the debuggee should
23591 start a new group or stay in the same group as the debugger.
23592 This affects the way the Windows OS handles
23595 @kindex show new-group
23596 @item show new-group
23597 Displays current value of new-group boolean.
23599 @kindex set debugevents
23600 @item set debugevents
23601 This boolean value adds debug output concerning kernel events related
23602 to the debuggee seen by the debugger. This includes events that
23603 signal thread and process creation and exit, DLL loading and
23604 unloading, console interrupts, and debugging messages produced by the
23605 Windows @code{OutputDebugString} API call.
23607 @kindex set debugexec
23608 @item set debugexec
23609 This boolean value adds debug output concerning execute events
23610 (such as resume thread) seen by the debugger.
23612 @kindex set debugexceptions
23613 @item set debugexceptions
23614 This boolean value adds debug output concerning exceptions in the
23615 debuggee seen by the debugger.
23617 @kindex set debugmemory
23618 @item set debugmemory
23619 This boolean value adds debug output concerning debuggee memory reads
23620 and writes by the debugger.
23624 This boolean values specifies whether the debuggee is called
23625 via a shell or directly (default value is on).
23629 Displays if the debuggee will be started with a shell.
23634 * Non-debug DLL Symbols:: Support for DLLs without debugging symbols
23637 @node Non-debug DLL Symbols
23638 @subsubsection Support for DLLs without Debugging Symbols
23639 @cindex DLLs with no debugging symbols
23640 @cindex Minimal symbols and DLLs
23642 Very often on windows, some of the DLLs that your program relies on do
23643 not include symbolic debugging information (for example,
23644 @file{kernel32.dll}). When @value{GDBN} doesn't recognize any debugging
23645 symbols in a DLL, it relies on the minimal amount of symbolic
23646 information contained in the DLL's export table. This section
23647 describes working with such symbols, known internally to @value{GDBN} as
23648 ``minimal symbols''.
23650 Note that before the debugged program has started execution, no DLLs
23651 will have been loaded. The easiest way around this problem is simply to
23652 start the program --- either by setting a breakpoint or letting the
23653 program run once to completion.
23655 @subsubsection DLL Name Prefixes
23657 In keeping with the naming conventions used by the Microsoft debugging
23658 tools, DLL export symbols are made available with a prefix based on the
23659 DLL name, for instance @code{KERNEL32!CreateFileA}. The plain name is
23660 also entered into the symbol table, so @code{CreateFileA} is often
23661 sufficient. In some cases there will be name clashes within a program
23662 (particularly if the executable itself includes full debugging symbols)
23663 necessitating the use of the fully qualified name when referring to the
23664 contents of the DLL. Use single-quotes around the name to avoid the
23665 exclamation mark (``!'') being interpreted as a language operator.
23667 Note that the internal name of the DLL may be all upper-case, even
23668 though the file name of the DLL is lower-case, or vice-versa. Since
23669 symbols within @value{GDBN} are @emph{case-sensitive} this may cause
23670 some confusion. If in doubt, try the @code{info functions} and
23671 @code{info variables} commands or even @code{maint print msymbols}
23672 (@pxref{Symbols}). Here's an example:
23675 (@value{GDBP}) info function CreateFileA
23676 All functions matching regular expression "CreateFileA":
23678 Non-debugging symbols:
23679 0x77e885f4 CreateFileA
23680 0x77e885f4 KERNEL32!CreateFileA
23684 (@value{GDBP}) info function !
23685 All functions matching regular expression "!":
23687 Non-debugging symbols:
23688 0x6100114c cygwin1!__assert
23689 0x61004034 cygwin1!_dll_crt0@@0
23690 0x61004240 cygwin1!dll_crt0(per_process *)
23694 @subsubsection Working with Minimal Symbols
23696 Symbols extracted from a DLL's export table do not contain very much
23697 type information. All that @value{GDBN} can do is guess whether a symbol
23698 refers to a function or variable depending on the linker section that
23699 contains the symbol. Also note that the actual contents of the memory
23700 contained in a DLL are not available unless the program is running. This
23701 means that you cannot examine the contents of a variable or disassemble
23702 a function within a DLL without a running program.
23704 Variables are generally treated as pointers and dereferenced
23705 automatically. For this reason, it is often necessary to prefix a
23706 variable name with the address-of operator (``&'') and provide explicit
23707 type information in the command. Here's an example of the type of
23711 (@value{GDBP}) print 'cygwin1!__argv'
23712 'cygwin1!__argv' has unknown type; cast it to its declared type
23716 (@value{GDBP}) x 'cygwin1!__argv'
23717 'cygwin1!__argv' has unknown type; cast it to its declared type
23720 And two possible solutions:
23723 (@value{GDBP}) print ((char **)'cygwin1!__argv')[0]
23724 $2 = 0x22fd98 "/cygdrive/c/mydirectory/myprogram"
23728 (@value{GDBP}) x/2x &'cygwin1!__argv'
23729 0x610c0aa8 <cygwin1!__argv>: 0x10021608 0x00000000
23730 (@value{GDBP}) x/x 0x10021608
23731 0x10021608: 0x0022fd98
23732 (@value{GDBP}) x/s 0x0022fd98
23733 0x22fd98: "/cygdrive/c/mydirectory/myprogram"
23736 Setting a break point within a DLL is possible even before the program
23737 starts execution. However, under these circumstances, @value{GDBN} can't
23738 examine the initial instructions of the function in order to skip the
23739 function's frame set-up code. You can work around this by using ``*&''
23740 to set the breakpoint at a raw memory address:
23743 (@value{GDBP}) break *&'python22!PyOS_Readline'
23744 Breakpoint 1 at 0x1e04eff0
23747 The author of these extensions is not entirely convinced that setting a
23748 break point within a shared DLL like @file{kernel32.dll} is completely
23752 @subsection Commands Specific to @sc{gnu} Hurd Systems
23753 @cindex @sc{gnu} Hurd debugging
23755 This subsection describes @value{GDBN} commands specific to the
23756 @sc{gnu} Hurd native debugging.
23761 @kindex set signals@r{, Hurd command}
23762 @kindex set sigs@r{, Hurd command}
23763 This command toggles the state of inferior signal interception by
23764 @value{GDBN}. Mach exceptions, such as breakpoint traps, are not
23765 affected by this command. @code{sigs} is a shorthand alias for
23770 @kindex show signals@r{, Hurd command}
23771 @kindex show sigs@r{, Hurd command}
23772 Show the current state of intercepting inferior's signals.
23774 @item set signal-thread
23775 @itemx set sigthread
23776 @kindex set signal-thread
23777 @kindex set sigthread
23778 This command tells @value{GDBN} which thread is the @code{libc} signal
23779 thread. That thread is run when a signal is delivered to a running
23780 process. @code{set sigthread} is the shorthand alias of @code{set
23783 @item show signal-thread
23784 @itemx show sigthread
23785 @kindex show signal-thread
23786 @kindex show sigthread
23787 These two commands show which thread will run when the inferior is
23788 delivered a signal.
23791 @kindex set stopped@r{, Hurd command}
23792 This commands tells @value{GDBN} that the inferior process is stopped,
23793 as with the @code{SIGSTOP} signal. The stopped process can be
23794 continued by delivering a signal to it.
23797 @kindex show stopped@r{, Hurd command}
23798 This command shows whether @value{GDBN} thinks the debuggee is
23801 @item set exceptions
23802 @kindex set exceptions@r{, Hurd command}
23803 Use this command to turn off trapping of exceptions in the inferior.
23804 When exception trapping is off, neither breakpoints nor
23805 single-stepping will work. To restore the default, set exception
23808 @item show exceptions
23809 @kindex show exceptions@r{, Hurd command}
23810 Show the current state of trapping exceptions in the inferior.
23812 @item set task pause
23813 @kindex set task@r{, Hurd commands}
23814 @cindex task attributes (@sc{gnu} Hurd)
23815 @cindex pause current task (@sc{gnu} Hurd)
23816 This command toggles task suspension when @value{GDBN} has control.
23817 Setting it to on takes effect immediately, and the task is suspended
23818 whenever @value{GDBN} gets control. Setting it to off will take
23819 effect the next time the inferior is continued. If this option is set
23820 to off, you can use @code{set thread default pause on} or @code{set
23821 thread pause on} (see below) to pause individual threads.
23823 @item show task pause
23824 @kindex show task@r{, Hurd commands}
23825 Show the current state of task suspension.
23827 @item set task detach-suspend-count
23828 @cindex task suspend count
23829 @cindex detach from task, @sc{gnu} Hurd
23830 This command sets the suspend count the task will be left with when
23831 @value{GDBN} detaches from it.
23833 @item show task detach-suspend-count
23834 Show the suspend count the task will be left with when detaching.
23836 @item set task exception-port
23837 @itemx set task excp
23838 @cindex task exception port, @sc{gnu} Hurd
23839 This command sets the task exception port to which @value{GDBN} will
23840 forward exceptions. The argument should be the value of the @dfn{send
23841 rights} of the task. @code{set task excp} is a shorthand alias.
23843 @item set noninvasive
23844 @cindex noninvasive task options
23845 This command switches @value{GDBN} to a mode that is the least
23846 invasive as far as interfering with the inferior is concerned. This
23847 is the same as using @code{set task pause}, @code{set exceptions}, and
23848 @code{set signals} to values opposite to the defaults.
23850 @item info send-rights
23851 @itemx info receive-rights
23852 @itemx info port-rights
23853 @itemx info port-sets
23854 @itemx info dead-names
23857 @cindex send rights, @sc{gnu} Hurd
23858 @cindex receive rights, @sc{gnu} Hurd
23859 @cindex port rights, @sc{gnu} Hurd
23860 @cindex port sets, @sc{gnu} Hurd
23861 @cindex dead names, @sc{gnu} Hurd
23862 These commands display information about, respectively, send rights,
23863 receive rights, port rights, port sets, and dead names of a task.
23864 There are also shorthand aliases: @code{info ports} for @code{info
23865 port-rights} and @code{info psets} for @code{info port-sets}.
23867 @item set thread pause
23868 @kindex set thread@r{, Hurd command}
23869 @cindex thread properties, @sc{gnu} Hurd
23870 @cindex pause current thread (@sc{gnu} Hurd)
23871 This command toggles current thread suspension when @value{GDBN} has
23872 control. Setting it to on takes effect immediately, and the current
23873 thread is suspended whenever @value{GDBN} gets control. Setting it to
23874 off will take effect the next time the inferior is continued.
23875 Normally, this command has no effect, since when @value{GDBN} has
23876 control, the whole task is suspended. However, if you used @code{set
23877 task pause off} (see above), this command comes in handy to suspend
23878 only the current thread.
23880 @item show thread pause
23881 @kindex show thread@r{, Hurd command}
23882 This command shows the state of current thread suspension.
23884 @item set thread run
23885 This command sets whether the current thread is allowed to run.
23887 @item show thread run
23888 Show whether the current thread is allowed to run.
23890 @item set thread detach-suspend-count
23891 @cindex thread suspend count, @sc{gnu} Hurd
23892 @cindex detach from thread, @sc{gnu} Hurd
23893 This command sets the suspend count @value{GDBN} will leave on a
23894 thread when detaching. This number is relative to the suspend count
23895 found by @value{GDBN} when it notices the thread; use @code{set thread
23896 takeover-suspend-count} to force it to an absolute value.
23898 @item show thread detach-suspend-count
23899 Show the suspend count @value{GDBN} will leave on the thread when
23902 @item set thread exception-port
23903 @itemx set thread excp
23904 Set the thread exception port to which to forward exceptions. This
23905 overrides the port set by @code{set task exception-port} (see above).
23906 @code{set thread excp} is the shorthand alias.
23908 @item set thread takeover-suspend-count
23909 Normally, @value{GDBN}'s thread suspend counts are relative to the
23910 value @value{GDBN} finds when it notices each thread. This command
23911 changes the suspend counts to be absolute instead.
23913 @item set thread default
23914 @itemx show thread default
23915 @cindex thread default settings, @sc{gnu} Hurd
23916 Each of the above @code{set thread} commands has a @code{set thread
23917 default} counterpart (e.g., @code{set thread default pause}, @code{set
23918 thread default exception-port}, etc.). The @code{thread default}
23919 variety of commands sets the default thread properties for all
23920 threads; you can then change the properties of individual threads with
23921 the non-default commands.
23928 @value{GDBN} provides the following commands specific to the Darwin target:
23931 @item set debug darwin @var{num}
23932 @kindex set debug darwin
23933 When set to a non zero value, enables debugging messages specific to
23934 the Darwin support. Higher values produce more verbose output.
23936 @item show debug darwin
23937 @kindex show debug darwin
23938 Show the current state of Darwin messages.
23940 @item set debug mach-o @var{num}
23941 @kindex set debug mach-o
23942 When set to a non zero value, enables debugging messages while
23943 @value{GDBN} is reading Darwin object files. (@dfn{Mach-O} is the
23944 file format used on Darwin for object and executable files.) Higher
23945 values produce more verbose output. This is a command to diagnose
23946 problems internal to @value{GDBN} and should not be needed in normal
23949 @item show debug mach-o
23950 @kindex show debug mach-o
23951 Show the current state of Mach-O file messages.
23953 @item set mach-exceptions on
23954 @itemx set mach-exceptions off
23955 @kindex set mach-exceptions
23956 On Darwin, faults are first reported as a Mach exception and are then
23957 mapped to a Posix signal. Use this command to turn on trapping of
23958 Mach exceptions in the inferior. This might be sometimes useful to
23959 better understand the cause of a fault. The default is off.
23961 @item show mach-exceptions
23962 @kindex show mach-exceptions
23963 Show the current state of exceptions trapping.
23967 @subsection FreeBSD
23970 When the ABI of a system call is changed in the FreeBSD kernel, this
23971 is implemented by leaving a compatibility system call using the old
23972 ABI at the existing number and allocating a new system call number for
23973 the version using the new ABI. As a convenience, when a system call
23974 is caught by name (@pxref{catch syscall}), compatibility system calls
23977 For example, FreeBSD 12 introduced a new variant of the @code{kevent}
23978 system call and catching the @code{kevent} system call by name catches
23982 (@value{GDBP}) catch syscall kevent
23983 Catchpoint 1 (syscalls 'freebsd11_kevent' [363] 'kevent' [560])
23989 @section Embedded Operating Systems
23991 This section describes configurations involving the debugging of
23992 embedded operating systems that are available for several different
23995 @value{GDBN} includes the ability to debug programs running on
23996 various real-time operating systems.
23998 @node Embedded Processors
23999 @section Embedded Processors
24001 This section goes into details specific to particular embedded
24004 @cindex send command to simulator
24005 Whenever a specific embedded processor has a simulator, @value{GDBN}
24006 allows to send an arbitrary command to the simulator.
24009 @item sim @var{command}
24010 @kindex sim@r{, a command}
24011 Send an arbitrary @var{command} string to the simulator. Consult the
24012 documentation for the specific simulator in use for information about
24013 acceptable commands.
24018 * ARC:: Synopsys ARC
24020 * M68K:: Motorola M68K
24021 * MicroBlaze:: Xilinx MicroBlaze
24022 * MIPS Embedded:: MIPS Embedded
24023 * OpenRISC 1000:: OpenRISC 1000 (or1k)
24024 * PowerPC Embedded:: PowerPC Embedded
24027 * Super-H:: Renesas Super-H
24031 @subsection Synopsys ARC
24032 @cindex Synopsys ARC
24033 @cindex ARC specific commands
24039 @value{GDBN} provides the following ARC-specific commands:
24042 @item set debug arc
24043 @kindex set debug arc
24044 Control the level of ARC specific debug messages. Use 0 for no messages (the
24045 default), 1 for debug messages, and 2 for even more debug messages.
24047 @item show debug arc
24048 @kindex show debug arc
24049 Show the level of ARC specific debugging in operation.
24051 @item maint print arc arc-instruction @var{address}
24052 @kindex maint print arc arc-instruction
24053 Print internal disassembler information about instruction at a given address.
24060 @value{GDBN} provides the following ARM-specific commands:
24063 @item set arm disassembler
24065 This commands selects from a list of disassembly styles. The
24066 @code{"std"} style is the standard style.
24068 @item show arm disassembler
24070 Show the current disassembly style.
24072 @item set arm apcs32
24073 @cindex ARM 32-bit mode
24074 This command toggles ARM operation mode between 32-bit and 26-bit.
24076 @item show arm apcs32
24077 Display the current usage of the ARM 32-bit mode.
24079 @item set arm fpu @var{fputype}
24080 This command sets the ARM floating-point unit (FPU) type. The
24081 argument @var{fputype} can be one of these:
24085 Determine the FPU type by querying the OS ABI.
24087 Software FPU, with mixed-endian doubles on little-endian ARM
24090 GCC-compiled FPA co-processor.
24092 Software FPU with pure-endian doubles.
24098 Show the current type of the FPU.
24101 This command forces @value{GDBN} to use the specified ABI.
24104 Show the currently used ABI.
24106 @item set arm fallback-mode (arm|thumb|auto)
24107 @value{GDBN} uses the symbol table, when available, to determine
24108 whether instructions are ARM or Thumb. This command controls
24109 @value{GDBN}'s default behavior when the symbol table is not
24110 available. The default is @samp{auto}, which causes @value{GDBN} to
24111 use the current execution mode (from the @code{T} bit in the @code{CPSR}
24114 @item show arm fallback-mode
24115 Show the current fallback instruction mode.
24117 @item set arm force-mode (arm|thumb|auto)
24118 This command overrides use of the symbol table to determine whether
24119 instructions are ARM or Thumb. The default is @samp{auto}, which
24120 causes @value{GDBN} to use the symbol table and then the setting
24121 of @samp{set arm fallback-mode}.
24123 @item show arm force-mode
24124 Show the current forced instruction mode.
24126 @item set debug arm
24127 Toggle whether to display ARM-specific debugging messages from the ARM
24128 target support subsystem.
24130 @item show debug arm
24131 Show whether ARM-specific debugging messages are enabled.
24135 @item target sim @r{[}@var{simargs}@r{]} @dots{}
24136 The @value{GDBN} ARM simulator accepts the following optional arguments.
24139 @item --swi-support=@var{type}
24140 Tell the simulator which SWI interfaces to support. The argument
24141 @var{type} may be a comma separated list of the following values.
24142 The default value is @code{all}.
24157 The Motorola m68k configuration includes ColdFire support.
24160 @subsection MicroBlaze
24161 @cindex Xilinx MicroBlaze
24162 @cindex XMD, Xilinx Microprocessor Debugger
24164 The MicroBlaze is a soft-core processor supported on various Xilinx
24165 FPGAs, such as Spartan or Virtex series. Boards with these processors
24166 usually have JTAG ports which connect to a host system running the Xilinx
24167 Embedded Development Kit (EDK) or Software Development Kit (SDK).
24168 This host system is used to download the configuration bitstream to
24169 the target FPGA. The Xilinx Microprocessor Debugger (XMD) program
24170 communicates with the target board using the JTAG interface and
24171 presents a @code{gdbserver} interface to the board. By default
24172 @code{xmd} uses port @code{1234}. (While it is possible to change
24173 this default port, it requires the use of undocumented @code{xmd}
24174 commands. Contact Xilinx support if you need to do this.)
24176 Use these GDB commands to connect to the MicroBlaze target processor.
24179 @item target remote :1234
24180 Use this command to connect to the target if you are running @value{GDBN}
24181 on the same system as @code{xmd}.
24183 @item target remote @var{xmd-host}:1234
24184 Use this command to connect to the target if it is connected to @code{xmd}
24185 running on a different system named @var{xmd-host}.
24188 Use this command to download a program to the MicroBlaze target.
24190 @item set debug microblaze @var{n}
24191 Enable MicroBlaze-specific debugging messages if non-zero.
24193 @item show debug microblaze @var{n}
24194 Show MicroBlaze-specific debugging level.
24197 @node MIPS Embedded
24198 @subsection @acronym{MIPS} Embedded
24201 @value{GDBN} supports these special commands for @acronym{MIPS} targets:
24204 @item set mipsfpu double
24205 @itemx set mipsfpu single
24206 @itemx set mipsfpu none
24207 @itemx set mipsfpu auto
24208 @itemx show mipsfpu
24209 @kindex set mipsfpu
24210 @kindex show mipsfpu
24211 @cindex @acronym{MIPS} remote floating point
24212 @cindex floating point, @acronym{MIPS} remote
24213 If your target board does not support the @acronym{MIPS} floating point
24214 coprocessor, you should use the command @samp{set mipsfpu none} (if you
24215 need this, you may wish to put the command in your @value{GDBN} init
24216 file). This tells @value{GDBN} how to find the return value of
24217 functions which return floating point values. It also allows
24218 @value{GDBN} to avoid saving the floating point registers when calling
24219 functions on the board. If you are using a floating point coprocessor
24220 with only single precision floating point support, as on the @sc{r4650}
24221 processor, use the command @samp{set mipsfpu single}. The default
24222 double precision floating point coprocessor may be selected using
24223 @samp{set mipsfpu double}.
24225 In previous versions the only choices were double precision or no
24226 floating point, so @samp{set mipsfpu on} will select double precision
24227 and @samp{set mipsfpu off} will select no floating point.
24229 As usual, you can inquire about the @code{mipsfpu} variable with
24230 @samp{show mipsfpu}.
24233 @node OpenRISC 1000
24234 @subsection OpenRISC 1000
24235 @cindex OpenRISC 1000
24238 The OpenRISC 1000 provides a free RISC instruction set architecture. It is
24239 mainly provided as a soft-core which can run on Xilinx, Altera and other
24242 @value{GDBN} for OpenRISC supports the below commands when connecting to
24250 Runs the builtin CPU simulator which can run very basic
24251 programs but does not support most hardware functions like MMU.
24252 For more complex use cases the user is advised to run an external
24253 target, and connect using @samp{target remote}.
24255 Example: @code{target sim}
24257 @item set debug or1k
24258 Toggle whether to display OpenRISC-specific debugging messages from the
24259 OpenRISC target support subsystem.
24261 @item show debug or1k
24262 Show whether OpenRISC-specific debugging messages are enabled.
24265 @node PowerPC Embedded
24266 @subsection PowerPC Embedded
24268 @cindex DVC register
24269 @value{GDBN} supports using the DVC (Data Value Compare) register to
24270 implement in hardware simple hardware watchpoint conditions of the form:
24273 (@value{GDBP}) watch @var{ADDRESS|VARIABLE} \
24274 if @var{ADDRESS|VARIABLE} == @var{CONSTANT EXPRESSION}
24277 The DVC register will be automatically used when @value{GDBN} detects
24278 such pattern in a condition expression, and the created watchpoint uses one
24279 debug register (either the @code{exact-watchpoints} option is on and the
24280 variable is scalar, or the variable has a length of one byte). This feature
24281 is available in native @value{GDBN} running on a Linux kernel version 2.6.34
24284 When running on PowerPC embedded processors, @value{GDBN} automatically uses
24285 ranged hardware watchpoints, unless the @code{exact-watchpoints} option is on,
24286 in which case watchpoints using only one debug register are created when
24287 watching variables of scalar types.
24289 You can create an artificial array to watch an arbitrary memory
24290 region using one of the following commands (@pxref{Expressions}):
24293 (@value{GDBP}) watch *((char *) @var{address})@@@var{length}
24294 (@value{GDBP}) watch @{char[@var{length}]@} @var{address}
24297 PowerPC embedded processors support masked watchpoints. See the discussion
24298 about the @code{mask} argument in @ref{Set Watchpoints}.
24300 @cindex ranged breakpoint
24301 PowerPC embedded processors support hardware accelerated
24302 @dfn{ranged breakpoints}. A ranged breakpoint stops execution of
24303 the inferior whenever it executes an instruction at any address within
24304 the range it specifies. To set a ranged breakpoint in @value{GDBN},
24305 use the @code{break-range} command.
24307 @value{GDBN} provides the following PowerPC-specific commands:
24310 @kindex break-range
24311 @item break-range @var{start-location}, @var{end-location}
24312 Set a breakpoint for an address range given by
24313 @var{start-location} and @var{end-location}, which can specify a function name,
24314 a line number, an offset of lines from the current line or from the start
24315 location, or an address of an instruction (see @ref{Specify Location},
24316 for a list of all the possible ways to specify a @var{location}.)
24317 The breakpoint will stop execution of the inferior whenever it
24318 executes an instruction at any address within the specified range,
24319 (including @var{start-location} and @var{end-location}.)
24321 @kindex set powerpc
24322 @item set powerpc soft-float
24323 @itemx show powerpc soft-float
24324 Force @value{GDBN} to use (or not use) a software floating point calling
24325 convention. By default, @value{GDBN} selects the calling convention based
24326 on the selected architecture and the provided executable file.
24328 @item set powerpc vector-abi
24329 @itemx show powerpc vector-abi
24330 Force @value{GDBN} to use the specified calling convention for vector
24331 arguments and return values. The valid options are @samp{auto};
24332 @samp{generic}, to avoid vector registers even if they are present;
24333 @samp{altivec}, to use AltiVec registers; and @samp{spe} to use SPE
24334 registers. By default, @value{GDBN} selects the calling convention
24335 based on the selected architecture and the provided executable file.
24337 @item set powerpc exact-watchpoints
24338 @itemx show powerpc exact-watchpoints
24339 Allow @value{GDBN} to use only one debug register when watching a variable
24340 of scalar type, thus assuming that the variable is accessed through the
24341 address of its first byte.
24346 @subsection Atmel AVR
24349 When configured for debugging the Atmel AVR, @value{GDBN} supports the
24350 following AVR-specific commands:
24353 @item info io_registers
24354 @kindex info io_registers@r{, AVR}
24355 @cindex I/O registers (Atmel AVR)
24356 This command displays information about the AVR I/O registers. For
24357 each register, @value{GDBN} prints its number and value.
24364 When configured for debugging CRIS, @value{GDBN} provides the
24365 following CRIS-specific commands:
24368 @item set cris-version @var{ver}
24369 @cindex CRIS version
24370 Set the current CRIS version to @var{ver}, either @samp{10} or @samp{32}.
24371 The CRIS version affects register names and sizes. This command is useful in
24372 case autodetection of the CRIS version fails.
24374 @item show cris-version
24375 Show the current CRIS version.
24377 @item set cris-dwarf2-cfi
24378 @cindex DWARF-2 CFI and CRIS
24379 Set the usage of DWARF-2 CFI for CRIS debugging. The default is @samp{on}.
24380 Change to @samp{off} when using @code{gcc-cris} whose version is below
24383 @item show cris-dwarf2-cfi
24384 Show the current state of using DWARF-2 CFI.
24386 @item set cris-mode @var{mode}
24388 Set the current CRIS mode to @var{mode}. It should only be changed when
24389 debugging in guru mode, in which case it should be set to
24390 @samp{guru} (the default is @samp{normal}).
24392 @item show cris-mode
24393 Show the current CRIS mode.
24397 @subsection Renesas Super-H
24400 For the Renesas Super-H processor, @value{GDBN} provides these
24404 @item set sh calling-convention @var{convention}
24405 @kindex set sh calling-convention
24406 Set the calling-convention used when calling functions from @value{GDBN}.
24407 Allowed values are @samp{gcc}, which is the default setting, and @samp{renesas}.
24408 With the @samp{gcc} setting, functions are called using the @value{NGCC} calling
24409 convention. If the DWARF-2 information of the called function specifies
24410 that the function follows the Renesas calling convention, the function
24411 is called using the Renesas calling convention. If the calling convention
24412 is set to @samp{renesas}, the Renesas calling convention is always used,
24413 regardless of the DWARF-2 information. This can be used to override the
24414 default of @samp{gcc} if debug information is missing, or the compiler
24415 does not emit the DWARF-2 calling convention entry for a function.
24417 @item show sh calling-convention
24418 @kindex show sh calling-convention
24419 Show the current calling convention setting.
24424 @node Architectures
24425 @section Architectures
24427 This section describes characteristics of architectures that affect
24428 all uses of @value{GDBN} with the architecture, both native and cross.
24435 * HPPA:: HP PA architecture
24443 @subsection AArch64
24444 @cindex AArch64 support
24446 When @value{GDBN} is debugging the AArch64 architecture, it provides the
24447 following special commands:
24450 @item set debug aarch64
24451 @kindex set debug aarch64
24452 This command determines whether AArch64 architecture-specific debugging
24453 messages are to be displayed.
24455 @item show debug aarch64
24456 Show whether AArch64 debugging messages are displayed.
24460 @subsubsection AArch64 SVE.
24461 @cindex AArch64 SVE.
24463 When @value{GDBN} is debugging the AArch64 architecture, if the Scalable Vector
24464 Extension (SVE) is present, then @value{GDBN} will provide the vector registers
24465 @code{$z0} through @code{$z31}, vector predicate registers @code{$p0} through
24466 @code{$p15}, and the @code{$ffr} register. In addition, the pseudo register
24467 @code{$vg} will be provided. This is the vector granule for the current thread
24468 and represents the number of 64-bit chunks in an SVE @code{z} register.
24470 If the vector length changes, then the @code{$vg} register will be updated,
24471 but the lengths of the @code{z} and @code{p} registers will not change. This
24472 is a known limitation of @value{GDBN} and does not affect the execution of the
24475 @subsubsection AArch64 Pointer Authentication.
24476 @cindex AArch64 Pointer Authentication.
24478 When @value{GDBN} is debugging the AArch64 architecture, and the program is
24479 using the v8.3-A feature Pointer Authentication (PAC), then whenever the link
24480 register @code{$lr} is pointing to an PAC function its value will be masked.
24481 When GDB prints a backtrace, any addresses that required unmasking will be
24482 postfixed with the marker [PAC]. When using the MI, this is printed as part
24483 of the @code{addr_flags} field.
24486 @subsection x86 Architecture-specific Issues
24489 @item set struct-convention @var{mode}
24490 @kindex set struct-convention
24491 @cindex struct return convention
24492 @cindex struct/union returned in registers
24493 Set the convention used by the inferior to return @code{struct}s and
24494 @code{union}s from functions to @var{mode}. Possible values of
24495 @var{mode} are @code{"pcc"}, @code{"reg"}, and @code{"default"} (the
24496 default). @code{"default"} or @code{"pcc"} means that @code{struct}s
24497 are returned on the stack, while @code{"reg"} means that a
24498 @code{struct} or a @code{union} whose size is 1, 2, 4, or 8 bytes will
24499 be returned in a register.
24501 @item show struct-convention
24502 @kindex show struct-convention
24503 Show the current setting of the convention to return @code{struct}s
24508 @subsubsection Intel @dfn{Memory Protection Extensions} (MPX).
24509 @cindex Intel Memory Protection Extensions (MPX).
24511 Memory Protection Extension (MPX) adds the bound registers @samp{BND0}
24512 @footnote{The register named with capital letters represent the architecture
24513 registers.} through @samp{BND3}. Bound registers store a pair of 64-bit values
24514 which are the lower bound and upper bound. Bounds are effective addresses or
24515 memory locations. The upper bounds are architecturally represented in 1's
24516 complement form. A bound having lower bound = 0, and upper bound = 0
24517 (1's complement of all bits set) will allow access to the entire address space.
24519 @samp{BND0} through @samp{BND3} are represented in @value{GDBN} as @samp{bnd0raw}
24520 through @samp{bnd3raw}. Pseudo registers @samp{bnd0} through @samp{bnd3}
24521 display the upper bound performing the complement of one operation on the
24522 upper bound value, i.e.@ when upper bound in @samp{bnd0raw} is 0 in the
24523 @value{GDBN} @samp{bnd0} it will be @code{0xfff@dots{}}. In this sense it
24524 can also be noted that the upper bounds are inclusive.
24526 As an example, assume that the register BND0 holds bounds for a pointer having
24527 access allowed for the range between 0x32 and 0x71. The values present on
24528 bnd0raw and bnd registers are presented as follows:
24531 bnd0raw = @{0x32, 0xffffffff8e@}
24532 bnd0 = @{lbound = 0x32, ubound = 0x71@} : size 64
24535 This way the raw value can be accessed via bnd0raw@dots{}bnd3raw. Any
24536 change on bnd0@dots{}bnd3 or bnd0raw@dots{}bnd3raw is reflect on its
24537 counterpart. When the bnd0@dots{}bnd3 registers are displayed via
24538 Python, the display includes the memory size, in bits, accessible to
24541 Bounds can also be stored in bounds tables, which are stored in
24542 application memory. These tables store bounds for pointers by specifying
24543 the bounds pointer's value along with its bounds. Evaluating and changing
24544 bounds located in bound tables is therefore interesting while investigating
24545 bugs on MPX context. @value{GDBN} provides commands for this purpose:
24548 @item show mpx bound @var{pointer}
24549 @kindex show mpx bound
24550 Display bounds of the given @var{pointer}.
24552 @item set mpx bound @var{pointer}, @var{lbound}, @var{ubound}
24553 @kindex set mpx bound
24554 Set the bounds of a pointer in the bound table.
24555 This command takes three parameters: @var{pointer} is the pointers
24556 whose bounds are to be changed, @var{lbound} and @var{ubound} are new values
24557 for lower and upper bounds respectively.
24560 When you call an inferior function on an Intel MPX enabled program,
24561 GDB sets the inferior's bound registers to the init (disabled) state
24562 before calling the function. As a consequence, bounds checks for the
24563 pointer arguments passed to the function will always pass.
24565 This is necessary because when you call an inferior function, the
24566 program is usually in the middle of the execution of other function.
24567 Since at that point bound registers are in an arbitrary state, not
24568 clearing them would lead to random bound violations in the called
24571 You can still examine the influence of the bound registers on the
24572 execution of the called function by stopping the execution of the
24573 called function at its prologue, setting bound registers, and
24574 continuing the execution. For example:
24578 Breakpoint 2 at 0x4009de: file i386-mpx-call.c, line 47.
24579 $ print upper (a, b, c, d, 1)
24580 Breakpoint 2, upper (a=0x0, b=0x6e0000005b, c=0x0, d=0x0, len=48)....
24582 @{lbound = 0x0, ubound = ffffffff@} : size -1
24585 At this last step the value of bnd0 can be changed for investigation of bound
24586 violations caused along the execution of the call. In order to know how to
24587 set the bound registers or bound table for the call consult the ABI.
24592 See the following section.
24595 @subsection @acronym{MIPS}
24597 @cindex stack on Alpha
24598 @cindex stack on @acronym{MIPS}
24599 @cindex Alpha stack
24600 @cindex @acronym{MIPS} stack
24601 Alpha- and @acronym{MIPS}-based computers use an unusual stack frame, which
24602 sometimes requires @value{GDBN} to search backward in the object code to
24603 find the beginning of a function.
24605 @cindex response time, @acronym{MIPS} debugging
24606 To improve response time (especially for embedded applications, where
24607 @value{GDBN} may be restricted to a slow serial line for this search)
24608 you may want to limit the size of this search, using one of these
24612 @cindex @code{heuristic-fence-post} (Alpha, @acronym{MIPS})
24613 @item set heuristic-fence-post @var{limit}
24614 Restrict @value{GDBN} to examining at most @var{limit} bytes in its
24615 search for the beginning of a function. A value of @var{0} (the
24616 default) means there is no limit. However, except for @var{0}, the
24617 larger the limit the more bytes @code{heuristic-fence-post} must search
24618 and therefore the longer it takes to run. You should only need to use
24619 this command when debugging a stripped executable.
24621 @item show heuristic-fence-post
24622 Display the current limit.
24626 These commands are available @emph{only} when @value{GDBN} is configured
24627 for debugging programs on Alpha or @acronym{MIPS} processors.
24629 Several @acronym{MIPS}-specific commands are available when debugging @acronym{MIPS}
24633 @item set mips abi @var{arg}
24634 @kindex set mips abi
24635 @cindex set ABI for @acronym{MIPS}
24636 Tell @value{GDBN} which @acronym{MIPS} ABI is used by the inferior. Possible
24637 values of @var{arg} are:
24641 The default ABI associated with the current binary (this is the
24651 @item show mips abi
24652 @kindex show mips abi
24653 Show the @acronym{MIPS} ABI used by @value{GDBN} to debug the inferior.
24655 @item set mips compression @var{arg}
24656 @kindex set mips compression
24657 @cindex code compression, @acronym{MIPS}
24658 Tell @value{GDBN} which @acronym{MIPS} compressed
24659 @acronym{ISA, Instruction Set Architecture} encoding is used by the
24660 inferior. @value{GDBN} uses this for code disassembly and other
24661 internal interpretation purposes. This setting is only referred to
24662 when no executable has been associated with the debugging session or
24663 the executable does not provide information about the encoding it uses.
24664 Otherwise this setting is automatically updated from information
24665 provided by the executable.
24667 Possible values of @var{arg} are @samp{mips16} and @samp{micromips}.
24668 The default compressed @acronym{ISA} encoding is @samp{mips16}, as
24669 executables containing @acronym{MIPS16} code frequently are not
24670 identified as such.
24672 This setting is ``sticky''; that is, it retains its value across
24673 debugging sessions until reset either explicitly with this command or
24674 implicitly from an executable.
24676 The compiler and/or assembler typically add symbol table annotations to
24677 identify functions compiled for the @acronym{MIPS16} or
24678 @acronym{microMIPS} @acronym{ISA}s. If these function-scope annotations
24679 are present, @value{GDBN} uses them in preference to the global
24680 compressed @acronym{ISA} encoding setting.
24682 @item show mips compression
24683 @kindex show mips compression
24684 Show the @acronym{MIPS} compressed @acronym{ISA} encoding used by
24685 @value{GDBN} to debug the inferior.
24688 @itemx show mipsfpu
24689 @xref{MIPS Embedded, set mipsfpu}.
24691 @item set mips mask-address @var{arg}
24692 @kindex set mips mask-address
24693 @cindex @acronym{MIPS} addresses, masking
24694 This command determines whether the most-significant 32 bits of 64-bit
24695 @acronym{MIPS} addresses are masked off. The argument @var{arg} can be
24696 @samp{on}, @samp{off}, or @samp{auto}. The latter is the default
24697 setting, which lets @value{GDBN} determine the correct value.
24699 @item show mips mask-address
24700 @kindex show mips mask-address
24701 Show whether the upper 32 bits of @acronym{MIPS} addresses are masked off or
24704 @item set remote-mips64-transfers-32bit-regs
24705 @kindex set remote-mips64-transfers-32bit-regs
24706 This command controls compatibility with 64-bit @acronym{MIPS} targets that
24707 transfer data in 32-bit quantities. If you have an old @acronym{MIPS} 64 target
24708 that transfers 32 bits for some registers, like @sc{sr} and @sc{fsr},
24709 and 64 bits for other registers, set this option to @samp{on}.
24711 @item show remote-mips64-transfers-32bit-regs
24712 @kindex show remote-mips64-transfers-32bit-regs
24713 Show the current setting of compatibility with older @acronym{MIPS} 64 targets.
24715 @item set debug mips
24716 @kindex set debug mips
24717 This command turns on and off debugging messages for the @acronym{MIPS}-specific
24718 target code in @value{GDBN}.
24720 @item show debug mips
24721 @kindex show debug mips
24722 Show the current setting of @acronym{MIPS} debugging messages.
24728 @cindex HPPA support
24730 When @value{GDBN} is debugging the HP PA architecture, it provides the
24731 following special commands:
24734 @item set debug hppa
24735 @kindex set debug hppa
24736 This command determines whether HPPA architecture-specific debugging
24737 messages are to be displayed.
24739 @item show debug hppa
24740 Show whether HPPA debugging messages are displayed.
24742 @item maint print unwind @var{address}
24743 @kindex maint print unwind@r{, HPPA}
24744 This command displays the contents of the unwind table entry at the
24745 given @var{address}.
24751 @subsection PowerPC
24752 @cindex PowerPC architecture
24754 When @value{GDBN} is debugging the PowerPC architecture, it provides a set of
24755 pseudo-registers to enable inspection of 128-bit wide Decimal Floating Point
24756 numbers stored in the floating point registers. These values must be stored
24757 in two consecutive registers, always starting at an even register like
24758 @code{f0} or @code{f2}.
24760 The pseudo-registers go from @code{$dl0} through @code{$dl15}, and are formed
24761 by joining the even/odd register pairs @code{f0} and @code{f1} for @code{$dl0},
24762 @code{f2} and @code{f3} for @code{$dl1} and so on.
24764 For POWER7 processors, @value{GDBN} provides a set of pseudo-registers, the 64-bit
24765 wide Extended Floating Point Registers (@samp{f32} through @samp{f63}).
24768 @subsection Nios II
24769 @cindex Nios II architecture
24771 When @value{GDBN} is debugging the Nios II architecture,
24772 it provides the following special commands:
24776 @item set debug nios2
24777 @kindex set debug nios2
24778 This command turns on and off debugging messages for the Nios II
24779 target code in @value{GDBN}.
24781 @item show debug nios2
24782 @kindex show debug nios2
24783 Show the current setting of Nios II debugging messages.
24787 @subsection Sparc64
24788 @cindex Sparc64 support
24789 @cindex Application Data Integrity
24790 @subsubsection ADI Support
24792 The M7 processor supports an Application Data Integrity (ADI) feature that
24793 detects invalid data accesses. When software allocates memory and enables
24794 ADI on the allocated memory, it chooses a 4-bit version number, sets the
24795 version in the upper 4 bits of the 64-bit pointer to that data, and stores
24796 the 4-bit version in every cacheline of that data. Hardware saves the latter
24797 in spare bits in the cache and memory hierarchy. On each load and store,
24798 the processor compares the upper 4 VA (virtual address) bits to the
24799 cacheline's version. If there is a mismatch, the processor generates a
24800 version mismatch trap which can be either precise or disrupting. The trap
24801 is an error condition which the kernel delivers to the process as a SIGSEGV
24804 Note that only 64-bit applications can use ADI and need to be built with
24807 Values of the ADI version tags, which are in granularity of a
24808 cacheline (64 bytes), can be viewed or modified.
24812 @kindex adi examine
24813 @item adi (examine | x) [ / @var{n} ] @var{addr}
24815 The @code{adi examine} command displays the value of one ADI version tag per
24818 @var{n} is a decimal integer specifying the number in bytes; the default
24819 is 1. It specifies how much ADI version information, at the ratio of 1:ADI
24820 block size, to display.
24822 @var{addr} is the address in user address space where you want @value{GDBN}
24823 to begin displaying the ADI version tags.
24825 Below is an example of displaying ADI versions of variable "shmaddr".
24828 (@value{GDBP}) adi x/100 shmaddr
24829 0xfff800010002c000: 0 0
24833 @item adi (assign | a) [ / @var{n} ] @var{addr} = @var{tag}
24835 The @code{adi assign} command is used to assign new ADI version tag
24838 @var{n} is a decimal integer specifying the number in bytes;
24839 the default is 1. It specifies how much ADI version information, at the
24840 ratio of 1:ADI block size, to modify.
24842 @var{addr} is the address in user address space where you want @value{GDBN}
24843 to begin modifying the ADI version tags.
24845 @var{tag} is the new ADI version tag.
24847 For example, do the following to modify then verify ADI versions of
24848 variable "shmaddr":
24851 (@value{GDBP}) adi a/100 shmaddr = 7
24852 (@value{GDBP}) adi x/100 shmaddr
24853 0xfff800010002c000: 7 7
24860 @cindex S12Z support
24862 When @value{GDBN} is debugging the S12Z architecture,
24863 it provides the following special command:
24866 @item maint info bdccsr
24867 @kindex maint info bdccsr@r{, S12Z}
24868 This command displays the current value of the microprocessor's
24873 @node Controlling GDB
24874 @chapter Controlling @value{GDBN}
24876 You can alter the way @value{GDBN} interacts with you by using the
24877 @code{set} command. For commands controlling how @value{GDBN} displays
24878 data, see @ref{Print Settings, ,Print Settings}. Other settings are
24883 * Editing:: Command editing
24884 * Command History:: Command history
24885 * Screen Size:: Screen size
24886 * Output Styling:: Output styling
24887 * Numbers:: Numbers
24888 * ABI:: Configuring the current ABI
24889 * Auto-loading:: Automatically loading associated files
24890 * Messages/Warnings:: Optional warnings and messages
24891 * Debugging Output:: Optional messages about internal happenings
24892 * Other Misc Settings:: Other Miscellaneous Settings
24900 @value{GDBN} indicates its readiness to read a command by printing a string
24901 called the @dfn{prompt}. This string is normally @samp{(@value{GDBP})}. You
24902 can change the prompt string with the @code{set prompt} command. For
24903 instance, when debugging @value{GDBN} with @value{GDBN}, it is useful to change
24904 the prompt in one of the @value{GDBN} sessions so that you can always tell
24905 which one you are talking to.
24907 @emph{Note:} @code{set prompt} does not add a space for you after the
24908 prompt you set. This allows you to set a prompt which ends in a space
24909 or a prompt that does not.
24913 @item set prompt @var{newprompt}
24914 Directs @value{GDBN} to use @var{newprompt} as its prompt string henceforth.
24916 @kindex show prompt
24918 Prints a line of the form: @samp{Gdb's prompt is: @var{your-prompt}}
24921 Versions of @value{GDBN} that ship with Python scripting enabled have
24922 prompt extensions. The commands for interacting with these extensions
24926 @kindex set extended-prompt
24927 @item set extended-prompt @var{prompt}
24928 Set an extended prompt that allows for substitutions.
24929 @xref{gdb.prompt}, for a list of escape sequences that can be used for
24930 substitution. Any escape sequences specified as part of the prompt
24931 string are replaced with the corresponding strings each time the prompt
24937 set extended-prompt Current working directory: \w (gdb)
24940 Note that when an extended-prompt is set, it takes control of the
24941 @var{prompt_hook} hook. @xref{prompt_hook}, for further information.
24943 @kindex show extended-prompt
24944 @item show extended-prompt
24945 Prints the extended prompt. Any escape sequences specified as part of
24946 the prompt string with @code{set extended-prompt}, are replaced with the
24947 corresponding strings each time the prompt is displayed.
24951 @section Command Editing
24953 @cindex command line editing
24955 @value{GDBN} reads its input commands via the @dfn{Readline} interface. This
24956 @sc{gnu} library provides consistent behavior for programs which provide a
24957 command line interface to the user. Advantages are @sc{gnu} Emacs-style
24958 or @dfn{vi}-style inline editing of commands, @code{csh}-like history
24959 substitution, and a storage and recall of command history across
24960 debugging sessions.
24962 You may control the behavior of command line editing in @value{GDBN} with the
24963 command @code{set}.
24966 @kindex set editing
24969 @itemx set editing on
24970 Enable command line editing (enabled by default).
24972 @item set editing off
24973 Disable command line editing.
24975 @kindex show editing
24977 Show whether command line editing is enabled.
24980 @ifset SYSTEM_READLINE
24981 @xref{Command Line Editing, , , rluserman, GNU Readline Library},
24983 @ifclear SYSTEM_READLINE
24984 @xref{Command Line Editing},
24986 for more details about the Readline
24987 interface. Users unfamiliar with @sc{gnu} Emacs or @code{vi} are
24988 encouraged to read that chapter.
24990 @cindex Readline application name
24991 @value{GDBN} sets the Readline application name to @samp{gdb}. This
24992 is useful for conditions in @file{.inputrc}.
24994 @node Command History
24995 @section Command History
24996 @cindex command history
24998 @value{GDBN} can keep track of the commands you type during your
24999 debugging sessions, so that you can be certain of precisely what
25000 happened. Use these commands to manage the @value{GDBN} command
25003 @value{GDBN} uses the @sc{gnu} History library, a part of the Readline
25004 package, to provide the history facility.
25005 @ifset SYSTEM_READLINE
25006 @xref{Using History Interactively, , , history, GNU History Library},
25008 @ifclear SYSTEM_READLINE
25009 @xref{Using History Interactively},
25011 for the detailed description of the History library.
25013 To issue a command to @value{GDBN} without affecting certain aspects of
25014 the state which is seen by users, prefix it with @samp{server }
25015 (@pxref{Server Prefix}). This
25016 means that this command will not affect the command history, nor will it
25017 affect @value{GDBN}'s notion of which command to repeat if @key{RET} is
25018 pressed on a line by itself.
25020 @cindex @code{server}, command prefix
25021 The server prefix does not affect the recording of values into the value
25022 history; to print a value without recording it into the value history,
25023 use the @code{output} command instead of the @code{print} command.
25025 Here is the description of @value{GDBN} commands related to command
25029 @cindex history substitution
25030 @cindex history file
25031 @kindex set history filename
25032 @cindex @env{GDBHISTFILE}, environment variable
25033 @item set history filename @var{fname}
25034 Set the name of the @value{GDBN} command history file to @var{fname}.
25035 This is the file where @value{GDBN} reads an initial command history
25036 list, and where it writes the command history from this session when it
25037 exits. You can access this list through history expansion or through
25038 the history command editing characters listed below. This file defaults
25039 to the value of the environment variable @code{GDBHISTFILE}, or to
25040 @file{./.gdb_history} (@file{./_gdb_history} on MS-DOS) if this variable
25043 @cindex save command history
25044 @kindex set history save
25045 @item set history save
25046 @itemx set history save on
25047 Record command history in a file, whose name may be specified with the
25048 @code{set history filename} command. By default, this option is disabled.
25050 @item set history save off
25051 Stop recording command history in a file.
25053 @cindex history size
25054 @kindex set history size
25055 @cindex @env{GDBHISTSIZE}, environment variable
25056 @item set history size @var{size}
25057 @itemx set history size unlimited
25058 Set the number of commands which @value{GDBN} keeps in its history list.
25059 This defaults to the value of the environment variable @env{GDBHISTSIZE}, or
25060 to 256 if this variable is not set. Non-numeric values of @env{GDBHISTSIZE}
25061 are ignored. If @var{size} is @code{unlimited} or if @env{GDBHISTSIZE} is
25062 either a negative number or the empty string, then the number of commands
25063 @value{GDBN} keeps in the history list is unlimited.
25065 @cindex remove duplicate history
25066 @kindex set history remove-duplicates
25067 @item set history remove-duplicates @var{count}
25068 @itemx set history remove-duplicates unlimited
25069 Control the removal of duplicate history entries in the command history list.
25070 If @var{count} is non-zero, @value{GDBN} will look back at the last @var{count}
25071 history entries and remove the first entry that is a duplicate of the current
25072 entry being added to the command history list. If @var{count} is
25073 @code{unlimited} then this lookbehind is unbounded. If @var{count} is 0, then
25074 removal of duplicate history entries is disabled.
25076 Only history entries added during the current session are considered for
25077 removal. This option is set to 0 by default.
25081 History expansion assigns special meaning to the character @kbd{!}.
25082 @ifset SYSTEM_READLINE
25083 @xref{Event Designators, , , history, GNU History Library},
25085 @ifclear SYSTEM_READLINE
25086 @xref{Event Designators},
25090 @cindex history expansion, turn on/off
25091 Since @kbd{!} is also the logical not operator in C, history expansion
25092 is off by default. If you decide to enable history expansion with the
25093 @code{set history expansion on} command, you may sometimes need to
25094 follow @kbd{!} (when it is used as logical not, in an expression) with
25095 a space or a tab to prevent it from being expanded. The readline
25096 history facilities do not attempt substitution on the strings
25097 @kbd{!=} and @kbd{!(}, even when history expansion is enabled.
25099 The commands to control history expansion are:
25102 @item set history expansion on
25103 @itemx set history expansion
25104 @kindex set history expansion
25105 Enable history expansion. History expansion is off by default.
25107 @item set history expansion off
25108 Disable history expansion.
25111 @kindex show history
25113 @itemx show history filename
25114 @itemx show history save
25115 @itemx show history size
25116 @itemx show history expansion
25117 These commands display the state of the @value{GDBN} history parameters.
25118 @code{show history} by itself displays all four states.
25123 @kindex show commands
25124 @cindex show last commands
25125 @cindex display command history
25126 @item show commands
25127 Display the last ten commands in the command history.
25129 @item show commands @var{n}
25130 Print ten commands centered on command number @var{n}.
25132 @item show commands +
25133 Print ten commands just after the commands last printed.
25137 @section Screen Size
25138 @cindex size of screen
25139 @cindex screen size
25142 @cindex pauses in output
25144 Certain commands to @value{GDBN} may produce large amounts of
25145 information output to the screen. To help you read all of it,
25146 @value{GDBN} pauses and asks you for input at the end of each page of
25147 output. Type @key{RET} when you want to see one more page of output,
25148 @kbd{q} to discard the remaining output, or @kbd{c} to continue
25149 without paging for the rest of the current command. Also, the screen
25150 width setting determines when to wrap lines of output. Depending on
25151 what is being printed, @value{GDBN} tries to break the line at a
25152 readable place, rather than simply letting it overflow onto the
25155 Normally @value{GDBN} knows the size of the screen from the terminal
25156 driver software. For example, on Unix @value{GDBN} uses the termcap data base
25157 together with the value of the @code{TERM} environment variable and the
25158 @code{stty rows} and @code{stty cols} settings. If this is not correct,
25159 you can override it with the @code{set height} and @code{set
25166 @kindex show height
25167 @item set height @var{lpp}
25168 @itemx set height unlimited
25170 @itemx set width @var{cpl}
25171 @itemx set width unlimited
25173 These @code{set} commands specify a screen height of @var{lpp} lines and
25174 a screen width of @var{cpl} characters. The associated @code{show}
25175 commands display the current settings.
25177 If you specify a height of either @code{unlimited} or zero lines,
25178 @value{GDBN} does not pause during output no matter how long the
25179 output is. This is useful if output is to a file or to an editor
25182 Likewise, you can specify @samp{set width unlimited} or @samp{set
25183 width 0} to prevent @value{GDBN} from wrapping its output.
25185 @item set pagination on
25186 @itemx set pagination off
25187 @kindex set pagination
25188 Turn the output pagination on or off; the default is on. Turning
25189 pagination off is the alternative to @code{set height unlimited}. Note that
25190 running @value{GDBN} with the @option{--batch} option (@pxref{Mode
25191 Options, -batch}) also automatically disables pagination.
25193 @item show pagination
25194 @kindex show pagination
25195 Show the current pagination mode.
25198 @node Output Styling
25199 @section Output Styling
25205 @value{GDBN} can style its output on a capable terminal. This is
25206 enabled by default on most systems, but disabled by default when in
25207 batch mode (@pxref{Mode Options}). Various style settings are available;
25208 and styles can also be disabled entirely.
25211 @item set style enabled @samp{on|off}
25212 Enable or disable all styling. The default is host-dependent, with
25213 most hosts defaulting to @samp{on}.
25215 @item show style enabled
25216 Show the current state of styling.
25218 @item set style sources @samp{on|off}
25219 Enable or disable source code styling. This affects whether source
25220 code, such as the output of the @code{list} command, is styled. Note
25221 that source styling only works if styling in general is enabled, and
25222 if @value{GDBN} was linked with the GNU Source Highlight library. The
25223 default is @samp{on}.
25225 @item show style sources
25226 Show the current state of source code styling.
25229 Subcommands of @code{set style} control specific forms of styling.
25230 These subcommands all follow the same pattern: each style-able object
25231 can be styled with a foreground color, a background color, and an
25234 For example, the style of file names can be controlled using the
25235 @code{set style filename} group of commands:
25238 @item set style filename background @var{color}
25239 Set the background to @var{color}. Valid colors are @samp{none}
25240 (meaning the terminal's default color), @samp{black}, @samp{red},
25241 @samp{green}, @samp{yellow}, @samp{blue}, @samp{magenta}, @samp{cyan},
25244 @item set style filename foreground @var{color}
25245 Set the foreground to @var{color}. Valid colors are @samp{none}
25246 (meaning the terminal's default color), @samp{black}, @samp{red},
25247 @samp{green}, @samp{yellow}, @samp{blue}, @samp{magenta}, @samp{cyan},
25250 @item set style filename intensity @var{value}
25251 Set the intensity to @var{value}. Valid intensities are @samp{normal}
25252 (the default), @samp{bold}, and @samp{dim}.
25255 The @code{show style} command and its subcommands are styling
25256 a style name in their output using its own style.
25257 So, use @command{show style} to see the complete list of styles,
25258 their characteristics and the visual aspect of each style.
25260 The style-able objects are:
25263 Control the styling of file names. By default, this style's
25264 foreground color is green.
25267 Control the styling of function names. These are managed with the
25268 @code{set style function} family of commands. By default, this
25269 style's foreground color is yellow.
25272 Control the styling of variable names. These are managed with the
25273 @code{set style variable} family of commands. By default, this style's
25274 foreground color is cyan.
25277 Control the styling of addresses. These are managed with the
25278 @code{set style address} family of commands. By default, this style's
25279 foreground color is blue.
25282 Control the styling of titles. These are managed with the
25283 @code{set style title} family of commands. By default, this style's
25284 intensity is bold. Commands are using the title style to improve
25285 the readibility of large output. For example, the commands
25286 @command{apropos} and @command{help} are using the title style
25287 for the command names.
25290 Control the styling of highlightings. These are managed with the
25291 @code{set style highlight} family of commands. By default, this style's
25292 foreground color is red. Commands are using the highlight style to draw
25293 the user attention to some specific parts of their output. For example,
25294 the command @command{apropos -v REGEXP} uses the highlight style to
25295 mark the documentation parts matching @var{regexp}.
25301 @cindex number representation
25302 @cindex entering numbers
25304 You can always enter numbers in octal, decimal, or hexadecimal in
25305 @value{GDBN} by the usual conventions: octal numbers begin with
25306 @samp{0}, decimal numbers end with @samp{.}, and hexadecimal numbers
25307 begin with @samp{0x}. Numbers that neither begin with @samp{0} or
25308 @samp{0x}, nor end with a @samp{.} are, by default, entered in base
25309 10; likewise, the default display for numbers---when no particular
25310 format is specified---is base 10. You can change the default base for
25311 both input and output with the commands described below.
25314 @kindex set input-radix
25315 @item set input-radix @var{base}
25316 Set the default base for numeric input. Supported choices
25317 for @var{base} are decimal 8, 10, or 16. The base must itself be
25318 specified either unambiguously or using the current input radix; for
25322 set input-radix 012
25323 set input-radix 10.
25324 set input-radix 0xa
25328 sets the input base to decimal. On the other hand, @samp{set input-radix 10}
25329 leaves the input radix unchanged, no matter what it was, since
25330 @samp{10}, being without any leading or trailing signs of its base, is
25331 interpreted in the current radix. Thus, if the current radix is 16,
25332 @samp{10} is interpreted in hex, i.e.@: as 16 decimal, which doesn't
25335 @kindex set output-radix
25336 @item set output-radix @var{base}
25337 Set the default base for numeric display. Supported choices
25338 for @var{base} are decimal 8, 10, or 16. The base must itself be
25339 specified either unambiguously or using the current input radix.
25341 @kindex show input-radix
25342 @item show input-radix
25343 Display the current default base for numeric input.
25345 @kindex show output-radix
25346 @item show output-radix
25347 Display the current default base for numeric display.
25349 @item set radix @r{[}@var{base}@r{]}
25353 These commands set and show the default base for both input and output
25354 of numbers. @code{set radix} sets the radix of input and output to
25355 the same base; without an argument, it resets the radix back to its
25356 default value of 10.
25361 @section Configuring the Current ABI
25363 @value{GDBN} can determine the @dfn{ABI} (Application Binary Interface) of your
25364 application automatically. However, sometimes you need to override its
25365 conclusions. Use these commands to manage @value{GDBN}'s view of the
25371 @cindex Newlib OS ABI and its influence on the longjmp handling
25373 One @value{GDBN} configuration can debug binaries for multiple operating
25374 system targets, either via remote debugging or native emulation.
25375 @value{GDBN} will autodetect the @dfn{OS ABI} (Operating System ABI) in use,
25376 but you can override its conclusion using the @code{set osabi} command.
25377 One example where this is useful is in debugging of binaries which use
25378 an alternate C library (e.g.@: @sc{uClibc} for @sc{gnu}/Linux) which does
25379 not have the same identifying marks that the standard C library for your
25382 When @value{GDBN} is debugging the AArch64 architecture, it provides a
25383 ``Newlib'' OS ABI. This is useful for handling @code{setjmp} and
25384 @code{longjmp} when debugging binaries that use the @sc{newlib} C library.
25385 The ``Newlib'' OS ABI can be selected by @code{set osabi Newlib}.
25389 Show the OS ABI currently in use.
25392 With no argument, show the list of registered available OS ABI's.
25394 @item set osabi @var{abi}
25395 Set the current OS ABI to @var{abi}.
25398 @cindex float promotion
25400 Generally, the way that an argument of type @code{float} is passed to a
25401 function depends on whether the function is prototyped. For a prototyped
25402 (i.e.@: ANSI/ISO style) function, @code{float} arguments are passed unchanged,
25403 according to the architecture's convention for @code{float}. For unprototyped
25404 (i.e.@: K&R style) functions, @code{float} arguments are first promoted to type
25405 @code{double} and then passed.
25407 Unfortunately, some forms of debug information do not reliably indicate whether
25408 a function is prototyped. If @value{GDBN} calls a function that is not marked
25409 as prototyped, it consults @kbd{set coerce-float-to-double}.
25412 @kindex set coerce-float-to-double
25413 @item set coerce-float-to-double
25414 @itemx set coerce-float-to-double on
25415 Arguments of type @code{float} will be promoted to @code{double} when passed
25416 to an unprototyped function. This is the default setting.
25418 @item set coerce-float-to-double off
25419 Arguments of type @code{float} will be passed directly to unprototyped
25422 @kindex show coerce-float-to-double
25423 @item show coerce-float-to-double
25424 Show the current setting of promoting @code{float} to @code{double}.
25428 @kindex show cp-abi
25429 @value{GDBN} needs to know the ABI used for your program's C@t{++}
25430 objects. The correct C@t{++} ABI depends on which C@t{++} compiler was
25431 used to build your application. @value{GDBN} only fully supports
25432 programs with a single C@t{++} ABI; if your program contains code using
25433 multiple C@t{++} ABI's or if @value{GDBN} can not identify your
25434 program's ABI correctly, you can tell @value{GDBN} which ABI to use.
25435 Currently supported ABI's include ``gnu-v2'', for @code{g++} versions
25436 before 3.0, ``gnu-v3'', for @code{g++} versions 3.0 and later, and
25437 ``hpaCC'' for the HP ANSI C@t{++} compiler. Other C@t{++} compilers may
25438 use the ``gnu-v2'' or ``gnu-v3'' ABI's as well. The default setting is
25443 Show the C@t{++} ABI currently in use.
25446 With no argument, show the list of supported C@t{++} ABI's.
25448 @item set cp-abi @var{abi}
25449 @itemx set cp-abi auto
25450 Set the current C@t{++} ABI to @var{abi}, or return to automatic detection.
25454 @section Automatically loading associated files
25455 @cindex auto-loading
25457 @value{GDBN} sometimes reads files with commands and settings automatically,
25458 without being explicitly told so by the user. We call this feature
25459 @dfn{auto-loading}. While auto-loading is useful for automatically adapting
25460 @value{GDBN} to the needs of your project, it can sometimes produce unexpected
25461 results or introduce security risks (e.g., if the file comes from untrusted
25465 * Init File in the Current Directory:: @samp{set/show/info auto-load local-gdbinit}
25466 * libthread_db.so.1 file:: @samp{set/show/info auto-load libthread-db}
25468 * Auto-loading safe path:: @samp{set/show/info auto-load safe-path}
25469 * Auto-loading verbose mode:: @samp{set/show debug auto-load}
25472 There are various kinds of files @value{GDBN} can automatically load.
25473 In addition to these files, @value{GDBN} supports auto-loading code written
25474 in various extension languages. @xref{Auto-loading extensions}.
25476 Note that loading of these associated files (including the local @file{.gdbinit}
25477 file) requires accordingly configured @code{auto-load safe-path}
25478 (@pxref{Auto-loading safe path}).
25480 For these reasons, @value{GDBN} includes commands and options to let you
25481 control when to auto-load files and which files should be auto-loaded.
25484 @anchor{set auto-load off}
25485 @kindex set auto-load off
25486 @item set auto-load off
25487 Globally disable loading of all auto-loaded files.
25488 You may want to use this command with the @samp{-iex} option
25489 (@pxref{Option -init-eval-command}) such as:
25491 $ @kbd{gdb -iex "set auto-load off" untrusted-executable corefile}
25494 Be aware that system init file (@pxref{System-wide configuration})
25495 and init files from your home directory (@pxref{Home Directory Init File})
25496 still get read (as they come from generally trusted directories).
25497 To prevent @value{GDBN} from auto-loading even those init files, use the
25498 @option{-nx} option (@pxref{Mode Options}), in addition to
25499 @code{set auto-load no}.
25501 @anchor{show auto-load}
25502 @kindex show auto-load
25503 @item show auto-load
25504 Show whether auto-loading of each specific @samp{auto-load} file(s) is enabled
25508 (gdb) show auto-load
25509 gdb-scripts: Auto-loading of canned sequences of commands scripts is on.
25510 libthread-db: Auto-loading of inferior specific libthread_db is on.
25511 local-gdbinit: Auto-loading of .gdbinit script from current directory
25513 python-scripts: Auto-loading of Python scripts is on.
25514 safe-path: List of directories from which it is safe to auto-load files
25515 is $debugdir:$datadir/auto-load.
25516 scripts-directory: List of directories from which to load auto-loaded scripts
25517 is $debugdir:$datadir/auto-load.
25520 @anchor{info auto-load}
25521 @kindex info auto-load
25522 @item info auto-load
25523 Print whether each specific @samp{auto-load} file(s) have been auto-loaded or
25527 (gdb) info auto-load
25530 Yes /home/user/gdb/gdb-gdb.gdb
25531 libthread-db: No auto-loaded libthread-db.
25532 local-gdbinit: Local .gdbinit file "/home/user/gdb/.gdbinit" has been
25536 Yes /home/user/gdb/gdb-gdb.py
25540 These are @value{GDBN} control commands for the auto-loading:
25542 @multitable @columnfractions .5 .5
25543 @item @xref{set auto-load off}.
25544 @tab Disable auto-loading globally.
25545 @item @xref{show auto-load}.
25546 @tab Show setting of all kinds of files.
25547 @item @xref{info auto-load}.
25548 @tab Show state of all kinds of files.
25549 @item @xref{set auto-load gdb-scripts}.
25550 @tab Control for @value{GDBN} command scripts.
25551 @item @xref{show auto-load gdb-scripts}.
25552 @tab Show setting of @value{GDBN} command scripts.
25553 @item @xref{info auto-load gdb-scripts}.
25554 @tab Show state of @value{GDBN} command scripts.
25555 @item @xref{set auto-load python-scripts}.
25556 @tab Control for @value{GDBN} Python scripts.
25557 @item @xref{show auto-load python-scripts}.
25558 @tab Show setting of @value{GDBN} Python scripts.
25559 @item @xref{info auto-load python-scripts}.
25560 @tab Show state of @value{GDBN} Python scripts.
25561 @item @xref{set auto-load guile-scripts}.
25562 @tab Control for @value{GDBN} Guile scripts.
25563 @item @xref{show auto-load guile-scripts}.
25564 @tab Show setting of @value{GDBN} Guile scripts.
25565 @item @xref{info auto-load guile-scripts}.
25566 @tab Show state of @value{GDBN} Guile scripts.
25567 @item @xref{set auto-load scripts-directory}.
25568 @tab Control for @value{GDBN} auto-loaded scripts location.
25569 @item @xref{show auto-load scripts-directory}.
25570 @tab Show @value{GDBN} auto-loaded scripts location.
25571 @item @xref{add-auto-load-scripts-directory}.
25572 @tab Add directory for auto-loaded scripts location list.
25573 @item @xref{set auto-load local-gdbinit}.
25574 @tab Control for init file in the current directory.
25575 @item @xref{show auto-load local-gdbinit}.
25576 @tab Show setting of init file in the current directory.
25577 @item @xref{info auto-load local-gdbinit}.
25578 @tab Show state of init file in the current directory.
25579 @item @xref{set auto-load libthread-db}.
25580 @tab Control for thread debugging library.
25581 @item @xref{show auto-load libthread-db}.
25582 @tab Show setting of thread debugging library.
25583 @item @xref{info auto-load libthread-db}.
25584 @tab Show state of thread debugging library.
25585 @item @xref{set auto-load safe-path}.
25586 @tab Control directories trusted for automatic loading.
25587 @item @xref{show auto-load safe-path}.
25588 @tab Show directories trusted for automatic loading.
25589 @item @xref{add-auto-load-safe-path}.
25590 @tab Add directory trusted for automatic loading.
25593 @node Init File in the Current Directory
25594 @subsection Automatically loading init file in the current directory
25595 @cindex auto-loading init file in the current directory
25597 By default, @value{GDBN} reads and executes the canned sequences of commands
25598 from init file (if any) in the current working directory,
25599 see @ref{Init File in the Current Directory during Startup}.
25601 Note that loading of this local @file{.gdbinit} file also requires accordingly
25602 configured @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
25605 @anchor{set auto-load local-gdbinit}
25606 @kindex set auto-load local-gdbinit
25607 @item set auto-load local-gdbinit [on|off]
25608 Enable or disable the auto-loading of canned sequences of commands
25609 (@pxref{Sequences}) found in init file in the current directory.
25611 @anchor{show auto-load local-gdbinit}
25612 @kindex show auto-load local-gdbinit
25613 @item show auto-load local-gdbinit
25614 Show whether auto-loading of canned sequences of commands from init file in the
25615 current directory is enabled or disabled.
25617 @anchor{info auto-load local-gdbinit}
25618 @kindex info auto-load local-gdbinit
25619 @item info auto-load local-gdbinit
25620 Print whether canned sequences of commands from init file in the
25621 current directory have been auto-loaded.
25624 @node libthread_db.so.1 file
25625 @subsection Automatically loading thread debugging library
25626 @cindex auto-loading libthread_db.so.1
25628 This feature is currently present only on @sc{gnu}/Linux native hosts.
25630 @value{GDBN} reads in some cases thread debugging library from places specific
25631 to the inferior (@pxref{set libthread-db-search-path}).
25633 The special @samp{libthread-db-search-path} entry @samp{$sdir} is processed
25634 without checking this @samp{set auto-load libthread-db} switch as system
25635 libraries have to be trusted in general. In all other cases of
25636 @samp{libthread-db-search-path} entries @value{GDBN} checks first if @samp{set
25637 auto-load libthread-db} is enabled before trying to open such thread debugging
25640 Note that loading of this debugging library also requires accordingly configured
25641 @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
25644 @anchor{set auto-load libthread-db}
25645 @kindex set auto-load libthread-db
25646 @item set auto-load libthread-db [on|off]
25647 Enable or disable the auto-loading of inferior specific thread debugging library.
25649 @anchor{show auto-load libthread-db}
25650 @kindex show auto-load libthread-db
25651 @item show auto-load libthread-db
25652 Show whether auto-loading of inferior specific thread debugging library is
25653 enabled or disabled.
25655 @anchor{info auto-load libthread-db}
25656 @kindex info auto-load libthread-db
25657 @item info auto-load libthread-db
25658 Print the list of all loaded inferior specific thread debugging libraries and
25659 for each such library print list of inferior @var{pid}s using it.
25662 @node Auto-loading safe path
25663 @subsection Security restriction for auto-loading
25664 @cindex auto-loading safe-path
25666 As the files of inferior can come from untrusted source (such as submitted by
25667 an application user) @value{GDBN} does not always load any files automatically.
25668 @value{GDBN} provides the @samp{set auto-load safe-path} setting to list
25669 directories trusted for loading files not explicitly requested by user.
25670 Each directory can also be a shell wildcard pattern.
25672 If the path is not set properly you will see a warning and the file will not
25677 Reading symbols from /home/user/gdb/gdb...done.
25678 warning: File "/home/user/gdb/gdb-gdb.gdb" auto-loading has been
25679 declined by your `auto-load safe-path' set
25680 to "$debugdir:$datadir/auto-load".
25681 warning: File "/home/user/gdb/gdb-gdb.py" auto-loading has been
25682 declined by your `auto-load safe-path' set
25683 to "$debugdir:$datadir/auto-load".
25687 To instruct @value{GDBN} to go ahead and use the init files anyway,
25688 invoke @value{GDBN} like this:
25691 $ gdb -q -iex "set auto-load safe-path /home/user/gdb" ./gdb
25694 The list of trusted directories is controlled by the following commands:
25697 @anchor{set auto-load safe-path}
25698 @kindex set auto-load safe-path
25699 @item set auto-load safe-path @r{[}@var{directories}@r{]}
25700 Set the list of directories (and their subdirectories) trusted for automatic
25701 loading and execution of scripts. You can also enter a specific trusted file.
25702 Each directory can also be a shell wildcard pattern; wildcards do not match
25703 directory separator - see @code{FNM_PATHNAME} for system function @code{fnmatch}
25704 (@pxref{Wildcard Matching, fnmatch, , libc, GNU C Library Reference Manual}).
25705 If you omit @var{directories}, @samp{auto-load safe-path} will be reset to
25706 its default value as specified during @value{GDBN} compilation.
25708 The list of directories uses path separator (@samp{:} on GNU and Unix
25709 systems, @samp{;} on MS-Windows and MS-DOS) to separate directories, similarly
25710 to the @env{PATH} environment variable.
25712 @anchor{show auto-load safe-path}
25713 @kindex show auto-load safe-path
25714 @item show auto-load safe-path
25715 Show the list of directories trusted for automatic loading and execution of
25718 @anchor{add-auto-load-safe-path}
25719 @kindex add-auto-load-safe-path
25720 @item add-auto-load-safe-path
25721 Add an entry (or list of entries) to the list of directories trusted for
25722 automatic loading and execution of scripts. Multiple entries may be delimited
25723 by the host platform path separator in use.
25726 This variable defaults to what @code{--with-auto-load-dir} has been configured
25727 to (@pxref{with-auto-load-dir}). @file{$debugdir} and @file{$datadir}
25728 substitution applies the same as for @ref{set auto-load scripts-directory}.
25729 The default @code{set auto-load safe-path} value can be also overriden by
25730 @value{GDBN} configuration option @option{--with-auto-load-safe-path}.
25732 Setting this variable to @file{/} disables this security protection,
25733 corresponding @value{GDBN} configuration option is
25734 @option{--without-auto-load-safe-path}.
25735 This variable is supposed to be set to the system directories writable by the
25736 system superuser only. Users can add their source directories in init files in
25737 their home directories (@pxref{Home Directory Init File}). See also deprecated
25738 init file in the current directory
25739 (@pxref{Init File in the Current Directory during Startup}).
25741 To force @value{GDBN} to load the files it declined to load in the previous
25742 example, you could use one of the following ways:
25745 @item @file{~/.gdbinit}: @samp{add-auto-load-safe-path ~/src/gdb}
25746 Specify this trusted directory (or a file) as additional component of the list.
25747 You have to specify also any existing directories displayed by
25748 by @samp{show auto-load safe-path} (such as @samp{/usr:/bin} in this example).
25750 @item @kbd{gdb -iex "set auto-load safe-path /usr:/bin:~/src/gdb" @dots{}}
25751 Specify this directory as in the previous case but just for a single
25752 @value{GDBN} session.
25754 @item @kbd{gdb -iex "set auto-load safe-path /" @dots{}}
25755 Disable auto-loading safety for a single @value{GDBN} session.
25756 This assumes all the files you debug during this @value{GDBN} session will come
25757 from trusted sources.
25759 @item @kbd{./configure --without-auto-load-safe-path}
25760 During compilation of @value{GDBN} you may disable any auto-loading safety.
25761 This assumes all the files you will ever debug with this @value{GDBN} come from
25765 On the other hand you can also explicitly forbid automatic files loading which
25766 also suppresses any such warning messages:
25769 @item @kbd{gdb -iex "set auto-load no" @dots{}}
25770 You can use @value{GDBN} command-line option for a single @value{GDBN} session.
25772 @item @file{~/.gdbinit}: @samp{set auto-load no}
25773 Disable auto-loading globally for the user
25774 (@pxref{Home Directory Init File}). While it is improbable, you could also
25775 use system init file instead (@pxref{System-wide configuration}).
25778 This setting applies to the file names as entered by user. If no entry matches
25779 @value{GDBN} tries as a last resort to also resolve all the file names into
25780 their canonical form (typically resolving symbolic links) and compare the
25781 entries again. @value{GDBN} already canonicalizes most of the filenames on its
25782 own before starting the comparison so a canonical form of directories is
25783 recommended to be entered.
25785 @node Auto-loading verbose mode
25786 @subsection Displaying files tried for auto-load
25787 @cindex auto-loading verbose mode
25789 For better visibility of all the file locations where you can place scripts to
25790 be auto-loaded with inferior --- or to protect yourself against accidental
25791 execution of untrusted scripts --- @value{GDBN} provides a feature for printing
25792 all the files attempted to be loaded. Both existing and non-existing files may
25795 For example the list of directories from which it is safe to auto-load files
25796 (@pxref{Auto-loading safe path}) applies also to canonicalized filenames which
25797 may not be too obvious while setting it up.
25800 (gdb) set debug auto-load on
25801 (gdb) file ~/src/t/true
25802 auto-load: Loading canned sequences of commands script "/tmp/true-gdb.gdb"
25803 for objfile "/tmp/true".
25804 auto-load: Updating directories of "/usr:/opt".
25805 auto-load: Using directory "/usr".
25806 auto-load: Using directory "/opt".
25807 warning: File "/tmp/true-gdb.gdb" auto-loading has been declined
25808 by your `auto-load safe-path' set to "/usr:/opt".
25812 @anchor{set debug auto-load}
25813 @kindex set debug auto-load
25814 @item set debug auto-load [on|off]
25815 Set whether to print the filenames attempted to be auto-loaded.
25817 @anchor{show debug auto-load}
25818 @kindex show debug auto-load
25819 @item show debug auto-load
25820 Show whether printing of the filenames attempted to be auto-loaded is turned
25824 @node Messages/Warnings
25825 @section Optional Warnings and Messages
25827 @cindex verbose operation
25828 @cindex optional warnings
25829 By default, @value{GDBN} is silent about its inner workings. If you are
25830 running on a slow machine, you may want to use the @code{set verbose}
25831 command. This makes @value{GDBN} tell you when it does a lengthy
25832 internal operation, so you will not think it has crashed.
25834 Currently, the messages controlled by @code{set verbose} are those
25835 which announce that the symbol table for a source file is being read;
25836 see @code{symbol-file} in @ref{Files, ,Commands to Specify Files}.
25839 @kindex set verbose
25840 @item set verbose on
25841 Enables @value{GDBN} output of certain informational messages.
25843 @item set verbose off
25844 Disables @value{GDBN} output of certain informational messages.
25846 @kindex show verbose
25848 Displays whether @code{set verbose} is on or off.
25851 By default, if @value{GDBN} encounters bugs in the symbol table of an
25852 object file, it is silent; but if you are debugging a compiler, you may
25853 find this information useful (@pxref{Symbol Errors, ,Errors Reading
25858 @kindex set complaints
25859 @item set complaints @var{limit}
25860 Permits @value{GDBN} to output @var{limit} complaints about each type of
25861 unusual symbols before becoming silent about the problem. Set
25862 @var{limit} to zero to suppress all complaints; set it to a large number
25863 to prevent complaints from being suppressed.
25865 @kindex show complaints
25866 @item show complaints
25867 Displays how many symbol complaints @value{GDBN} is permitted to produce.
25871 @anchor{confirmation requests}
25872 By default, @value{GDBN} is cautious, and asks what sometimes seems to be a
25873 lot of stupid questions to confirm certain commands. For example, if
25874 you try to run a program which is already running:
25878 The program being debugged has been started already.
25879 Start it from the beginning? (y or n)
25882 If you are willing to unflinchingly face the consequences of your own
25883 commands, you can disable this ``feature'':
25887 @kindex set confirm
25889 @cindex confirmation
25890 @cindex stupid questions
25891 @item set confirm off
25892 Disables confirmation requests. Note that running @value{GDBN} with
25893 the @option{--batch} option (@pxref{Mode Options, -batch}) also
25894 automatically disables confirmation requests.
25896 @item set confirm on
25897 Enables confirmation requests (the default).
25899 @kindex show confirm
25901 Displays state of confirmation requests.
25905 @cindex command tracing
25906 If you need to debug user-defined commands or sourced files you may find it
25907 useful to enable @dfn{command tracing}. In this mode each command will be
25908 printed as it is executed, prefixed with one or more @samp{+} symbols, the
25909 quantity denoting the call depth of each command.
25912 @kindex set trace-commands
25913 @cindex command scripts, debugging
25914 @item set trace-commands on
25915 Enable command tracing.
25916 @item set trace-commands off
25917 Disable command tracing.
25918 @item show trace-commands
25919 Display the current state of command tracing.
25922 @node Debugging Output
25923 @section Optional Messages about Internal Happenings
25924 @cindex optional debugging messages
25926 @value{GDBN} has commands that enable optional debugging messages from
25927 various @value{GDBN} subsystems; normally these commands are of
25928 interest to @value{GDBN} maintainers, or when reporting a bug. This
25929 section documents those commands.
25932 @kindex set exec-done-display
25933 @item set exec-done-display
25934 Turns on or off the notification of asynchronous commands'
25935 completion. When on, @value{GDBN} will print a message when an
25936 asynchronous command finishes its execution. The default is off.
25937 @kindex show exec-done-display
25938 @item show exec-done-display
25939 Displays the current setting of asynchronous command completion
25942 @cindex ARM AArch64
25943 @item set debug aarch64
25944 Turns on or off display of debugging messages related to ARM AArch64.
25945 The default is off.
25947 @item show debug aarch64
25948 Displays the current state of displaying debugging messages related to
25950 @cindex gdbarch debugging info
25951 @cindex architecture debugging info
25952 @item set debug arch
25953 Turns on or off display of gdbarch debugging info. The default is off
25954 @item show debug arch
25955 Displays the current state of displaying gdbarch debugging info.
25956 @item set debug aix-solib
25957 @cindex AIX shared library debugging
25958 Control display of debugging messages from the AIX shared library
25959 support module. The default is off.
25960 @item show debug aix-thread
25961 Show the current state of displaying AIX shared library debugging messages.
25962 @item set debug aix-thread
25963 @cindex AIX threads
25964 Display debugging messages about inner workings of the AIX thread
25966 @item show debug aix-thread
25967 Show the current state of AIX thread debugging info display.
25968 @item set debug check-physname
25970 Check the results of the ``physname'' computation. When reading DWARF
25971 debugging information for C@t{++}, @value{GDBN} attempts to compute
25972 each entity's name. @value{GDBN} can do this computation in two
25973 different ways, depending on exactly what information is present.
25974 When enabled, this setting causes @value{GDBN} to compute the names
25975 both ways and display any discrepancies.
25976 @item show debug check-physname
25977 Show the current state of ``physname'' checking.
25978 @item set debug coff-pe-read
25979 @cindex COFF/PE exported symbols
25980 Control display of debugging messages related to reading of COFF/PE
25981 exported symbols. The default is off.
25982 @item show debug coff-pe-read
25983 Displays the current state of displaying debugging messages related to
25984 reading of COFF/PE exported symbols.
25985 @item set debug dwarf-die
25987 Dump DWARF DIEs after they are read in.
25988 The value is the number of nesting levels to print.
25989 A value of zero turns off the display.
25990 @item show debug dwarf-die
25991 Show the current state of DWARF DIE debugging.
25992 @item set debug dwarf-line
25993 @cindex DWARF Line Tables
25994 Turns on or off display of debugging messages related to reading
25995 DWARF line tables. The default is 0 (off).
25996 A value of 1 provides basic information.
25997 A value greater than 1 provides more verbose information.
25998 @item show debug dwarf-line
25999 Show the current state of DWARF line table debugging.
26000 @item set debug dwarf-read
26001 @cindex DWARF Reading
26002 Turns on or off display of debugging messages related to reading
26003 DWARF debug info. The default is 0 (off).
26004 A value of 1 provides basic information.
26005 A value greater than 1 provides more verbose information.
26006 @item show debug dwarf-read
26007 Show the current state of DWARF reader debugging.
26008 @item set debug displaced
26009 @cindex displaced stepping debugging info
26010 Turns on or off display of @value{GDBN} debugging info for the
26011 displaced stepping support. The default is off.
26012 @item show debug displaced
26013 Displays the current state of displaying @value{GDBN} debugging info
26014 related to displaced stepping.
26015 @item set debug event
26016 @cindex event debugging info
26017 Turns on or off display of @value{GDBN} event debugging info. The
26019 @item show debug event
26020 Displays the current state of displaying @value{GDBN} event debugging
26022 @item set debug expression
26023 @cindex expression debugging info
26024 Turns on or off display of debugging info about @value{GDBN}
26025 expression parsing. The default is off.
26026 @item show debug expression
26027 Displays the current state of displaying debugging info about
26028 @value{GDBN} expression parsing.
26029 @item set debug fbsd-lwp
26030 @cindex FreeBSD LWP debug messages
26031 Turns on or off debugging messages from the FreeBSD LWP debug support.
26032 @item show debug fbsd-lwp
26033 Show the current state of FreeBSD LWP debugging messages.
26034 @item set debug fbsd-nat
26035 @cindex FreeBSD native target debug messages
26036 Turns on or off debugging messages from the FreeBSD native target.
26037 @item show debug fbsd-nat
26038 Show the current state of FreeBSD native target debugging messages.
26039 @item set debug frame
26040 @cindex frame debugging info
26041 Turns on or off display of @value{GDBN} frame debugging info. The
26043 @item show debug frame
26044 Displays the current state of displaying @value{GDBN} frame debugging
26046 @item set debug gnu-nat
26047 @cindex @sc{gnu}/Hurd debug messages
26048 Turn on or off debugging messages from the @sc{gnu}/Hurd debug support.
26049 @item show debug gnu-nat
26050 Show the current state of @sc{gnu}/Hurd debugging messages.
26051 @item set debug infrun
26052 @cindex inferior debugging info
26053 Turns on or off display of @value{GDBN} debugging info for running the inferior.
26054 The default is off. @file{infrun.c} contains GDB's runtime state machine used
26055 for implementing operations such as single-stepping the inferior.
26056 @item show debug infrun
26057 Displays the current state of @value{GDBN} inferior debugging.
26058 @item set debug jit
26059 @cindex just-in-time compilation, debugging messages
26060 Turn on or off debugging messages from JIT debug support.
26061 @item show debug jit
26062 Displays the current state of @value{GDBN} JIT debugging.
26063 @item set debug lin-lwp
26064 @cindex @sc{gnu}/Linux LWP debug messages
26065 @cindex Linux lightweight processes
26066 Turn on or off debugging messages from the Linux LWP debug support.
26067 @item show debug lin-lwp
26068 Show the current state of Linux LWP debugging messages.
26069 @item set debug linux-namespaces
26070 @cindex @sc{gnu}/Linux namespaces debug messages
26071 Turn on or off debugging messages from the Linux namespaces debug support.
26072 @item show debug linux-namespaces
26073 Show the current state of Linux namespaces debugging messages.
26074 @item set debug mach-o
26075 @cindex Mach-O symbols processing
26076 Control display of debugging messages related to Mach-O symbols
26077 processing. The default is off.
26078 @item show debug mach-o
26079 Displays the current state of displaying debugging messages related to
26080 reading of COFF/PE exported symbols.
26081 @item set debug notification
26082 @cindex remote async notification debugging info
26083 Turn on or off debugging messages about remote async notification.
26084 The default is off.
26085 @item show debug notification
26086 Displays the current state of remote async notification debugging messages.
26087 @item set debug observer
26088 @cindex observer debugging info
26089 Turns on or off display of @value{GDBN} observer debugging. This
26090 includes info such as the notification of observable events.
26091 @item show debug observer
26092 Displays the current state of observer debugging.
26093 @item set debug overload
26094 @cindex C@t{++} overload debugging info
26095 Turns on or off display of @value{GDBN} C@t{++} overload debugging
26096 info. This includes info such as ranking of functions, etc. The default
26098 @item show debug overload
26099 Displays the current state of displaying @value{GDBN} C@t{++} overload
26101 @cindex expression parser, debugging info
26102 @cindex debug expression parser
26103 @item set debug parser
26104 Turns on or off the display of expression parser debugging output.
26105 Internally, this sets the @code{yydebug} variable in the expression
26106 parser. @xref{Tracing, , Tracing Your Parser, bison, Bison}, for
26107 details. The default is off.
26108 @item show debug parser
26109 Show the current state of expression parser debugging.
26110 @cindex packets, reporting on stdout
26111 @cindex serial connections, debugging
26112 @cindex debug remote protocol
26113 @cindex remote protocol debugging
26114 @cindex display remote packets
26115 @item set debug remote
26116 Turns on or off display of reports on all packets sent back and forth across
26117 the serial line to the remote machine. The info is printed on the
26118 @value{GDBN} standard output stream. The default is off.
26119 @item show debug remote
26120 Displays the state of display of remote packets.
26122 @item set debug separate-debug-file
26123 Turns on or off display of debug output about separate debug file search.
26124 @item show debug separate-debug-file
26125 Displays the state of separate debug file search debug output.
26127 @item set debug serial
26128 Turns on or off display of @value{GDBN} serial debugging info. The
26130 @item show debug serial
26131 Displays the current state of displaying @value{GDBN} serial debugging
26133 @item set debug solib-frv
26134 @cindex FR-V shared-library debugging
26135 Turn on or off debugging messages for FR-V shared-library code.
26136 @item show debug solib-frv
26137 Display the current state of FR-V shared-library code debugging
26139 @item set debug symbol-lookup
26140 @cindex symbol lookup
26141 Turns on or off display of debugging messages related to symbol lookup.
26142 The default is 0 (off).
26143 A value of 1 provides basic information.
26144 A value greater than 1 provides more verbose information.
26145 @item show debug symbol-lookup
26146 Show the current state of symbol lookup debugging messages.
26147 @item set debug symfile
26148 @cindex symbol file functions
26149 Turns on or off display of debugging messages related to symbol file functions.
26150 The default is off. @xref{Files}.
26151 @item show debug symfile
26152 Show the current state of symbol file debugging messages.
26153 @item set debug symtab-create
26154 @cindex symbol table creation
26155 Turns on or off display of debugging messages related to symbol table creation.
26156 The default is 0 (off).
26157 A value of 1 provides basic information.
26158 A value greater than 1 provides more verbose information.
26159 @item show debug symtab-create
26160 Show the current state of symbol table creation debugging.
26161 @item set debug target
26162 @cindex target debugging info
26163 Turns on or off display of @value{GDBN} target debugging info. This info
26164 includes what is going on at the target level of GDB, as it happens. The
26165 default is 0. Set it to 1 to track events, and to 2 to also track the
26166 value of large memory transfers.
26167 @item show debug target
26168 Displays the current state of displaying @value{GDBN} target debugging
26170 @item set debug timestamp
26171 @cindex timestampping debugging info
26172 Turns on or off display of timestamps with @value{GDBN} debugging info.
26173 When enabled, seconds and microseconds are displayed before each debugging
26175 @item show debug timestamp
26176 Displays the current state of displaying timestamps with @value{GDBN}
26178 @item set debug varobj
26179 @cindex variable object debugging info
26180 Turns on or off display of @value{GDBN} variable object debugging
26181 info. The default is off.
26182 @item show debug varobj
26183 Displays the current state of displaying @value{GDBN} variable object
26185 @item set debug xml
26186 @cindex XML parser debugging
26187 Turn on or off debugging messages for built-in XML parsers.
26188 @item show debug xml
26189 Displays the current state of XML debugging messages.
26192 @node Other Misc Settings
26193 @section Other Miscellaneous Settings
26194 @cindex miscellaneous settings
26197 @kindex set interactive-mode
26198 @item set interactive-mode
26199 If @code{on}, forces @value{GDBN} to assume that GDB was started
26200 in a terminal. In practice, this means that @value{GDBN} should wait
26201 for the user to answer queries generated by commands entered at
26202 the command prompt. If @code{off}, forces @value{GDBN} to operate
26203 in the opposite mode, and it uses the default answers to all queries.
26204 If @code{auto} (the default), @value{GDBN} tries to determine whether
26205 its standard input is a terminal, and works in interactive-mode if it
26206 is, non-interactively otherwise.
26208 In the vast majority of cases, the debugger should be able to guess
26209 correctly which mode should be used. But this setting can be useful
26210 in certain specific cases, such as running a MinGW @value{GDBN}
26211 inside a cygwin window.
26213 @kindex show interactive-mode
26214 @item show interactive-mode
26215 Displays whether the debugger is operating in interactive mode or not.
26218 @node Extending GDB
26219 @chapter Extending @value{GDBN}
26220 @cindex extending GDB
26222 @value{GDBN} provides several mechanisms for extension.
26223 @value{GDBN} also provides the ability to automatically load
26224 extensions when it reads a file for debugging. This allows the
26225 user to automatically customize @value{GDBN} for the program
26229 * Sequences:: Canned Sequences of @value{GDBN} Commands
26230 * Python:: Extending @value{GDBN} using Python
26231 * Guile:: Extending @value{GDBN} using Guile
26232 * Auto-loading extensions:: Automatically loading extensions
26233 * Multiple Extension Languages:: Working with multiple extension languages
26234 * Aliases:: Creating new spellings of existing commands
26237 To facilitate the use of extension languages, @value{GDBN} is capable
26238 of evaluating the contents of a file. When doing so, @value{GDBN}
26239 can recognize which extension language is being used by looking at
26240 the filename extension. Files with an unrecognized filename extension
26241 are always treated as a @value{GDBN} Command Files.
26242 @xref{Command Files,, Command files}.
26244 You can control how @value{GDBN} evaluates these files with the following
26248 @kindex set script-extension
26249 @kindex show script-extension
26250 @item set script-extension off
26251 All scripts are always evaluated as @value{GDBN} Command Files.
26253 @item set script-extension soft
26254 The debugger determines the scripting language based on filename
26255 extension. If this scripting language is supported, @value{GDBN}
26256 evaluates the script using that language. Otherwise, it evaluates
26257 the file as a @value{GDBN} Command File.
26259 @item set script-extension strict
26260 The debugger determines the scripting language based on filename
26261 extension, and evaluates the script using that language. If the
26262 language is not supported, then the evaluation fails.
26264 @item show script-extension
26265 Display the current value of the @code{script-extension} option.
26270 @section Canned Sequences of Commands
26272 Aside from breakpoint commands (@pxref{Break Commands, ,Breakpoint
26273 Command Lists}), @value{GDBN} provides two ways to store sequences of
26274 commands for execution as a unit: user-defined commands and command
26278 * Define:: How to define your own commands
26279 * Hooks:: Hooks for user-defined commands
26280 * Command Files:: How to write scripts of commands to be stored in a file
26281 * Output:: Commands for controlled output
26282 * Auto-loading sequences:: Controlling auto-loaded command files
26286 @subsection User-defined Commands
26288 @cindex user-defined command
26289 @cindex arguments, to user-defined commands
26290 A @dfn{user-defined command} is a sequence of @value{GDBN} commands to
26291 which you assign a new name as a command. This is done with the
26292 @code{define} command. User commands may accept an unlimited number of arguments
26293 separated by whitespace. Arguments are accessed within the user command
26294 via @code{$arg0@dots{}$argN}. A trivial example:
26298 print $arg0 + $arg1 + $arg2
26303 To execute the command use:
26310 This defines the command @code{adder}, which prints the sum of
26311 its three arguments. Note the arguments are text substitutions, so they may
26312 reference variables, use complex expressions, or even perform inferior
26315 @cindex argument count in user-defined commands
26316 @cindex how many arguments (user-defined commands)
26317 In addition, @code{$argc} may be used to find out how many arguments have
26323 print $arg0 + $arg1
26326 print $arg0 + $arg1 + $arg2
26331 Combining with the @code{eval} command (@pxref{eval}) makes it easier
26332 to process a variable number of arguments:
26339 eval "set $sum = $sum + $arg%d", $i
26349 @item define @var{commandname}
26350 Define a command named @var{commandname}. If there is already a command
26351 by that name, you are asked to confirm that you want to redefine it.
26352 The argument @var{commandname} may be a bare command name consisting of letters,
26353 numbers, dashes, and underscores. It may also start with any predefined
26354 prefix command. For example, @samp{define target my-target} creates
26355 a user-defined @samp{target my-target} command.
26357 The definition of the command is made up of other @value{GDBN} command lines,
26358 which are given following the @code{define} command. The end of these
26359 commands is marked by a line containing @code{end}.
26362 @kindex end@r{ (user-defined commands)}
26363 @item document @var{commandname}
26364 Document the user-defined command @var{commandname}, so that it can be
26365 accessed by @code{help}. The command @var{commandname} must already be
26366 defined. This command reads lines of documentation just as @code{define}
26367 reads the lines of the command definition, ending with @code{end}.
26368 After the @code{document} command is finished, @code{help} on command
26369 @var{commandname} displays the documentation you have written.
26371 You may use the @code{document} command again to change the
26372 documentation of a command. Redefining the command with @code{define}
26373 does not change the documentation.
26375 @kindex dont-repeat
26376 @cindex don't repeat command
26378 Used inside a user-defined command, this tells @value{GDBN} that this
26379 command should not be repeated when the user hits @key{RET}
26380 (@pxref{Command Syntax, repeat last command}).
26382 @kindex help user-defined
26383 @item help user-defined
26384 List all user-defined commands and all python commands defined in class
26385 COMAND_USER. The first line of the documentation or docstring is
26390 @itemx show user @var{commandname}
26391 Display the @value{GDBN} commands used to define @var{commandname} (but
26392 not its documentation). If no @var{commandname} is given, display the
26393 definitions for all user-defined commands.
26394 This does not work for user-defined python commands.
26396 @cindex infinite recursion in user-defined commands
26397 @kindex show max-user-call-depth
26398 @kindex set max-user-call-depth
26399 @item show max-user-call-depth
26400 @itemx set max-user-call-depth
26401 The value of @code{max-user-call-depth} controls how many recursion
26402 levels are allowed in user-defined commands before @value{GDBN} suspects an
26403 infinite recursion and aborts the command.
26404 This does not apply to user-defined python commands.
26407 In addition to the above commands, user-defined commands frequently
26408 use control flow commands, described in @ref{Command Files}.
26410 When user-defined commands are executed, the
26411 commands of the definition are not printed. An error in any command
26412 stops execution of the user-defined command.
26414 If used interactively, commands that would ask for confirmation proceed
26415 without asking when used inside a user-defined command. Many @value{GDBN}
26416 commands that normally print messages to say what they are doing omit the
26417 messages when used in a user-defined command.
26420 @subsection User-defined Command Hooks
26421 @cindex command hooks
26422 @cindex hooks, for commands
26423 @cindex hooks, pre-command
26426 You may define @dfn{hooks}, which are a special kind of user-defined
26427 command. Whenever you run the command @samp{foo}, if the user-defined
26428 command @samp{hook-foo} exists, it is executed (with no arguments)
26429 before that command.
26431 @cindex hooks, post-command
26433 A hook may also be defined which is run after the command you executed.
26434 Whenever you run the command @samp{foo}, if the user-defined command
26435 @samp{hookpost-foo} exists, it is executed (with no arguments) after
26436 that command. Post-execution hooks may exist simultaneously with
26437 pre-execution hooks, for the same command.
26439 It is valid for a hook to call the command which it hooks. If this
26440 occurs, the hook is not re-executed, thereby avoiding infinite recursion.
26442 @c It would be nice if hookpost could be passed a parameter indicating
26443 @c if the command it hooks executed properly or not. FIXME!
26445 @kindex stop@r{, a pseudo-command}
26446 In addition, a pseudo-command, @samp{stop} exists. Defining
26447 (@samp{hook-stop}) makes the associated commands execute every time
26448 execution stops in your program: before breakpoint commands are run,
26449 displays are printed, or the stack frame is printed.
26451 For example, to ignore @code{SIGALRM} signals while
26452 single-stepping, but treat them normally during normal execution,
26457 handle SIGALRM nopass
26461 handle SIGALRM pass
26464 define hook-continue
26465 handle SIGALRM pass
26469 As a further example, to hook at the beginning and end of the @code{echo}
26470 command, and to add extra text to the beginning and end of the message,
26478 define hookpost-echo
26482 (@value{GDBP}) echo Hello World
26483 <<<---Hello World--->>>
26488 You can define a hook for any single-word command in @value{GDBN}, but
26489 not for command aliases; you should define a hook for the basic command
26490 name, e.g.@: @code{backtrace} rather than @code{bt}.
26491 @c FIXME! So how does Joe User discover whether a command is an alias
26493 You can hook a multi-word command by adding @code{hook-} or
26494 @code{hookpost-} to the last word of the command, e.g.@:
26495 @samp{define target hook-remote} to add a hook to @samp{target remote}.
26497 If an error occurs during the execution of your hook, execution of
26498 @value{GDBN} commands stops and @value{GDBN} issues a prompt
26499 (before the command that you actually typed had a chance to run).
26501 If you try to define a hook which does not match any known command, you
26502 get a warning from the @code{define} command.
26504 @node Command Files
26505 @subsection Command Files
26507 @cindex command files
26508 @cindex scripting commands
26509 A command file for @value{GDBN} is a text file made of lines that are
26510 @value{GDBN} commands. Comments (lines starting with @kbd{#}) may
26511 also be included. An empty line in a command file does nothing; it
26512 does not mean to repeat the last command, as it would from the
26515 You can request the execution of a command file with the @code{source}
26516 command. Note that the @code{source} command is also used to evaluate
26517 scripts that are not Command Files. The exact behavior can be configured
26518 using the @code{script-extension} setting.
26519 @xref{Extending GDB,, Extending GDB}.
26523 @cindex execute commands from a file
26524 @item source [-s] [-v] @var{filename}
26525 Execute the command file @var{filename}.
26528 The lines in a command file are generally executed sequentially,
26529 unless the order of execution is changed by one of the
26530 @emph{flow-control commands} described below. The commands are not
26531 printed as they are executed. An error in any command terminates
26532 execution of the command file and control is returned to the console.
26534 @value{GDBN} first searches for @var{filename} in the current directory.
26535 If the file is not found there, and @var{filename} does not specify a
26536 directory, then @value{GDBN} also looks for the file on the source search path
26537 (specified with the @samp{directory} command);
26538 except that @file{$cdir} is not searched because the compilation directory
26539 is not relevant to scripts.
26541 If @code{-s} is specified, then @value{GDBN} searches for @var{filename}
26542 on the search path even if @var{filename} specifies a directory.
26543 The search is done by appending @var{filename} to each element of the
26544 search path. So, for example, if @var{filename} is @file{mylib/myscript}
26545 and the search path contains @file{/home/user} then @value{GDBN} will
26546 look for the script @file{/home/user/mylib/myscript}.
26547 The search is also done if @var{filename} is an absolute path.
26548 For example, if @var{filename} is @file{/tmp/myscript} and
26549 the search path contains @file{/home/user} then @value{GDBN} will
26550 look for the script @file{/home/user/tmp/myscript}.
26551 For DOS-like systems, if @var{filename} contains a drive specification,
26552 it is stripped before concatenation. For example, if @var{filename} is
26553 @file{d:myscript} and the search path contains @file{c:/tmp} then @value{GDBN}
26554 will look for the script @file{c:/tmp/myscript}.
26556 If @code{-v}, for verbose mode, is given then @value{GDBN} displays
26557 each command as it is executed. The option must be given before
26558 @var{filename}, and is interpreted as part of the filename anywhere else.
26560 Commands that would ask for confirmation if used interactively proceed
26561 without asking when used in a command file. Many @value{GDBN} commands that
26562 normally print messages to say what they are doing omit the messages
26563 when called from command files.
26565 @value{GDBN} also accepts command input from standard input. In this
26566 mode, normal output goes to standard output and error output goes to
26567 standard error. Errors in a command file supplied on standard input do
26568 not terminate execution of the command file---execution continues with
26572 gdb < cmds > log 2>&1
26575 (The syntax above will vary depending on the shell used.) This example
26576 will execute commands from the file @file{cmds}. All output and errors
26577 would be directed to @file{log}.
26579 Since commands stored on command files tend to be more general than
26580 commands typed interactively, they frequently need to deal with
26581 complicated situations, such as different or unexpected values of
26582 variables and symbols, changes in how the program being debugged is
26583 built, etc. @value{GDBN} provides a set of flow-control commands to
26584 deal with these complexities. Using these commands, you can write
26585 complex scripts that loop over data structures, execute commands
26586 conditionally, etc.
26593 This command allows to include in your script conditionally executed
26594 commands. The @code{if} command takes a single argument, which is an
26595 expression to evaluate. It is followed by a series of commands that
26596 are executed only if the expression is true (its value is nonzero).
26597 There can then optionally be an @code{else} line, followed by a series
26598 of commands that are only executed if the expression was false. The
26599 end of the list is marked by a line containing @code{end}.
26603 This command allows to write loops. Its syntax is similar to
26604 @code{if}: the command takes a single argument, which is an expression
26605 to evaluate, and must be followed by the commands to execute, one per
26606 line, terminated by an @code{end}. These commands are called the
26607 @dfn{body} of the loop. The commands in the body of @code{while} are
26608 executed repeatedly as long as the expression evaluates to true.
26612 This command exits the @code{while} loop in whose body it is included.
26613 Execution of the script continues after that @code{while}s @code{end}
26616 @kindex loop_continue
26617 @item loop_continue
26618 This command skips the execution of the rest of the body of commands
26619 in the @code{while} loop in whose body it is included. Execution
26620 branches to the beginning of the @code{while} loop, where it evaluates
26621 the controlling expression.
26623 @kindex end@r{ (if/else/while commands)}
26625 Terminate the block of commands that are the body of @code{if},
26626 @code{else}, or @code{while} flow-control commands.
26631 @subsection Commands for Controlled Output
26633 During the execution of a command file or a user-defined command, normal
26634 @value{GDBN} output is suppressed; the only output that appears is what is
26635 explicitly printed by the commands in the definition. This section
26636 describes three commands useful for generating exactly the output you
26641 @item echo @var{text}
26642 @c I do not consider backslash-space a standard C escape sequence
26643 @c because it is not in ANSI.
26644 Print @var{text}. Nonprinting characters can be included in
26645 @var{text} using C escape sequences, such as @samp{\n} to print a
26646 newline. @strong{No newline is printed unless you specify one.}
26647 In addition to the standard C escape sequences, a backslash followed
26648 by a space stands for a space. This is useful for displaying a
26649 string with spaces at the beginning or the end, since leading and
26650 trailing spaces are otherwise trimmed from all arguments.
26651 To print @samp{@w{ }and foo =@w{ }}, use the command
26652 @samp{echo \@w{ }and foo = \@w{ }}.
26654 A backslash at the end of @var{text} can be used, as in C, to continue
26655 the command onto subsequent lines. For example,
26658 echo This is some text\n\
26659 which is continued\n\
26660 onto several lines.\n
26663 produces the same output as
26666 echo This is some text\n
26667 echo which is continued\n
26668 echo onto several lines.\n
26672 @item output @var{expression}
26673 Print the value of @var{expression} and nothing but that value: no
26674 newlines, no @samp{$@var{nn} = }. The value is not entered in the
26675 value history either. @xref{Expressions, ,Expressions}, for more information
26678 @item output/@var{fmt} @var{expression}
26679 Print the value of @var{expression} in format @var{fmt}. You can use
26680 the same formats as for @code{print}. @xref{Output Formats,,Output
26681 Formats}, for more information.
26684 @item printf @var{template}, @var{expressions}@dots{}
26685 Print the values of one or more @var{expressions} under the control of
26686 the string @var{template}. To print several values, make
26687 @var{expressions} be a comma-separated list of individual expressions,
26688 which may be either numbers or pointers. Their values are printed as
26689 specified by @var{template}, exactly as a C program would do by
26690 executing the code below:
26693 printf (@var{template}, @var{expressions}@dots{});
26696 As in @code{C} @code{printf}, ordinary characters in @var{template}
26697 are printed verbatim, while @dfn{conversion specification} introduced
26698 by the @samp{%} character cause subsequent @var{expressions} to be
26699 evaluated, their values converted and formatted according to type and
26700 style information encoded in the conversion specifications, and then
26703 For example, you can print two values in hex like this:
26706 printf "foo, bar-foo = 0x%x, 0x%x\n", foo, bar-foo
26709 @code{printf} supports all the standard @code{C} conversion
26710 specifications, including the flags and modifiers between the @samp{%}
26711 character and the conversion letter, with the following exceptions:
26715 The argument-ordering modifiers, such as @samp{2$}, are not supported.
26718 The modifier @samp{*} is not supported for specifying precision or
26722 The @samp{'} flag (for separation of digits into groups according to
26723 @code{LC_NUMERIC'}) is not supported.
26726 The type modifiers @samp{hh}, @samp{j}, @samp{t}, and @samp{z} are not
26730 The conversion letter @samp{n} (as in @samp{%n}) is not supported.
26733 The conversion letters @samp{a} and @samp{A} are not supported.
26737 Note that the @samp{ll} type modifier is supported only if the
26738 underlying @code{C} implementation used to build @value{GDBN} supports
26739 the @code{long long int} type, and the @samp{L} type modifier is
26740 supported only if @code{long double} type is available.
26742 As in @code{C}, @code{printf} supports simple backslash-escape
26743 sequences, such as @code{\n}, @samp{\t}, @samp{\\}, @samp{\"},
26744 @samp{\a}, and @samp{\f}, that consist of backslash followed by a
26745 single character. Octal and hexadecimal escape sequences are not
26748 Additionally, @code{printf} supports conversion specifications for DFP
26749 (@dfn{Decimal Floating Point}) types using the following length modifiers
26750 together with a floating point specifier.
26755 @samp{H} for printing @code{Decimal32} types.
26758 @samp{D} for printing @code{Decimal64} types.
26761 @samp{DD} for printing @code{Decimal128} types.
26764 If the underlying @code{C} implementation used to build @value{GDBN} has
26765 support for the three length modifiers for DFP types, other modifiers
26766 such as width and precision will also be available for @value{GDBN} to use.
26768 In case there is no such @code{C} support, no additional modifiers will be
26769 available and the value will be printed in the standard way.
26771 Here's an example of printing DFP types using the above conversion letters:
26773 printf "D32: %Hf - D64: %Df - D128: %DDf\n",1.2345df,1.2E10dd,1.2E1dl
26778 @item eval @var{template}, @var{expressions}@dots{}
26779 Convert the values of one or more @var{expressions} under the control of
26780 the string @var{template} to a command line, and call it.
26784 @node Auto-loading sequences
26785 @subsection Controlling auto-loading native @value{GDBN} scripts
26786 @cindex native script auto-loading
26788 When a new object file is read (for example, due to the @code{file}
26789 command, or because the inferior has loaded a shared library),
26790 @value{GDBN} will look for the command file @file{@var{objfile}-gdb.gdb}.
26791 @xref{Auto-loading extensions}.
26793 Auto-loading can be enabled or disabled,
26794 and the list of auto-loaded scripts can be printed.
26797 @anchor{set auto-load gdb-scripts}
26798 @kindex set auto-load gdb-scripts
26799 @item set auto-load gdb-scripts [on|off]
26800 Enable or disable the auto-loading of canned sequences of commands scripts.
26802 @anchor{show auto-load gdb-scripts}
26803 @kindex show auto-load gdb-scripts
26804 @item show auto-load gdb-scripts
26805 Show whether auto-loading of canned sequences of commands scripts is enabled or
26808 @anchor{info auto-load gdb-scripts}
26809 @kindex info auto-load gdb-scripts
26810 @cindex print list of auto-loaded canned sequences of commands scripts
26811 @item info auto-load gdb-scripts [@var{regexp}]
26812 Print the list of all canned sequences of commands scripts that @value{GDBN}
26816 If @var{regexp} is supplied only canned sequences of commands scripts with
26817 matching names are printed.
26819 @c Python docs live in a separate file.
26820 @include python.texi
26822 @c Guile docs live in a separate file.
26823 @include guile.texi
26825 @node Auto-loading extensions
26826 @section Auto-loading extensions
26827 @cindex auto-loading extensions
26829 @value{GDBN} provides two mechanisms for automatically loading extensions
26830 when a new object file is read (for example, due to the @code{file}
26831 command, or because the inferior has loaded a shared library):
26832 @file{@var{objfile}-gdb.@var{ext}} and the @code{.debug_gdb_scripts}
26833 section of modern file formats like ELF.
26836 * objfile-gdb.ext file: objfile-gdbdotext file. The @file{@var{objfile}-gdb.@var{ext}} file
26837 * .debug_gdb_scripts section: dotdebug_gdb_scripts section. The @code{.debug_gdb_scripts} section
26838 * Which flavor to choose?::
26841 The auto-loading feature is useful for supplying application-specific
26842 debugging commands and features.
26844 Auto-loading can be enabled or disabled,
26845 and the list of auto-loaded scripts can be printed.
26846 See the @samp{auto-loading} section of each extension language
26847 for more information.
26848 For @value{GDBN} command files see @ref{Auto-loading sequences}.
26849 For Python files see @ref{Python Auto-loading}.
26851 Note that loading of this script file also requires accordingly configured
26852 @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
26854 @node objfile-gdbdotext file
26855 @subsection The @file{@var{objfile}-gdb.@var{ext}} file
26856 @cindex @file{@var{objfile}-gdb.gdb}
26857 @cindex @file{@var{objfile}-gdb.py}
26858 @cindex @file{@var{objfile}-gdb.scm}
26860 When a new object file is read, @value{GDBN} looks for a file named
26861 @file{@var{objfile}-gdb.@var{ext}} (we call it @var{script-name} below),
26862 where @var{objfile} is the object file's name and
26863 where @var{ext} is the file extension for the extension language:
26866 @item @file{@var{objfile}-gdb.gdb}
26867 GDB's own command language
26868 @item @file{@var{objfile}-gdb.py}
26870 @item @file{@var{objfile}-gdb.scm}
26874 @var{script-name} is formed by ensuring that the file name of @var{objfile}
26875 is absolute, following all symlinks, and resolving @code{.} and @code{..}
26876 components, and appending the @file{-gdb.@var{ext}} suffix.
26877 If this file exists and is readable, @value{GDBN} will evaluate it as a
26878 script in the specified extension language.
26880 If this file does not exist, then @value{GDBN} will look for
26881 @var{script-name} file in all of the directories as specified below.
26883 Note that loading of these files requires an accordingly configured
26884 @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
26886 For object files using @file{.exe} suffix @value{GDBN} tries to load first the
26887 scripts normally according to its @file{.exe} filename. But if no scripts are
26888 found @value{GDBN} also tries script filenames matching the object file without
26889 its @file{.exe} suffix. This @file{.exe} stripping is case insensitive and it
26890 is attempted on any platform. This makes the script filenames compatible
26891 between Unix and MS-Windows hosts.
26894 @anchor{set auto-load scripts-directory}
26895 @kindex set auto-load scripts-directory
26896 @item set auto-load scripts-directory @r{[}@var{directories}@r{]}
26897 Control @value{GDBN} auto-loaded scripts location. Multiple directory entries
26898 may be delimited by the host platform path separator in use
26899 (@samp{:} on Unix, @samp{;} on MS-Windows and MS-DOS).
26901 Each entry here needs to be covered also by the security setting
26902 @code{set auto-load safe-path} (@pxref{set auto-load safe-path}).
26904 @anchor{with-auto-load-dir}
26905 This variable defaults to @file{$debugdir:$datadir/auto-load}. The default
26906 @code{set auto-load safe-path} value can be also overriden by @value{GDBN}
26907 configuration option @option{--with-auto-load-dir}.
26909 Any reference to @file{$debugdir} will get replaced by
26910 @var{debug-file-directory} value (@pxref{Separate Debug Files}) and any
26911 reference to @file{$datadir} will get replaced by @var{data-directory} which is
26912 determined at @value{GDBN} startup (@pxref{Data Files}). @file{$debugdir} and
26913 @file{$datadir} must be placed as a directory component --- either alone or
26914 delimited by @file{/} or @file{\} directory separators, depending on the host
26917 The list of directories uses path separator (@samp{:} on GNU and Unix
26918 systems, @samp{;} on MS-Windows and MS-DOS) to separate directories, similarly
26919 to the @env{PATH} environment variable.
26921 @anchor{show auto-load scripts-directory}
26922 @kindex show auto-load scripts-directory
26923 @item show auto-load scripts-directory
26924 Show @value{GDBN} auto-loaded scripts location.
26926 @anchor{add-auto-load-scripts-directory}
26927 @kindex add-auto-load-scripts-directory
26928 @item add-auto-load-scripts-directory @r{[}@var{directories}@dots{}@r{]}
26929 Add an entry (or list of entries) to the list of auto-loaded scripts locations.
26930 Multiple entries may be delimited by the host platform path separator in use.
26933 @value{GDBN} does not track which files it has already auto-loaded this way.
26934 @value{GDBN} will load the associated script every time the corresponding
26935 @var{objfile} is opened.
26936 So your @file{-gdb.@var{ext}} file should be careful to avoid errors if it
26937 is evaluated more than once.
26939 @node dotdebug_gdb_scripts section
26940 @subsection The @code{.debug_gdb_scripts} section
26941 @cindex @code{.debug_gdb_scripts} section
26943 For systems using file formats like ELF and COFF,
26944 when @value{GDBN} loads a new object file
26945 it will look for a special section named @code{.debug_gdb_scripts}.
26946 If this section exists, its contents is a list of null-terminated entries
26947 specifying scripts to load. Each entry begins with a non-null prefix byte that
26948 specifies the kind of entry, typically the extension language and whether the
26949 script is in a file or inlined in @code{.debug_gdb_scripts}.
26951 The following entries are supported:
26954 @item SECTION_SCRIPT_ID_PYTHON_FILE = 1
26955 @item SECTION_SCRIPT_ID_SCHEME_FILE = 3
26956 @item SECTION_SCRIPT_ID_PYTHON_TEXT = 4
26957 @item SECTION_SCRIPT_ID_SCHEME_TEXT = 6
26960 @subsubsection Script File Entries
26962 If the entry specifies a file, @value{GDBN} will look for the file first
26963 in the current directory and then along the source search path
26964 (@pxref{Source Path, ,Specifying Source Directories}),
26965 except that @file{$cdir} is not searched, since the compilation
26966 directory is not relevant to scripts.
26968 File entries can be placed in section @code{.debug_gdb_scripts} with,
26969 for example, this GCC macro for Python scripts.
26972 /* Note: The "MS" section flags are to remove duplicates. */
26973 #define DEFINE_GDB_PY_SCRIPT(script_name) \
26975 .pushsection \".debug_gdb_scripts\", \"MS\",@@progbits,1\n\
26976 .byte 1 /* Python */\n\
26977 .asciz \"" script_name "\"\n\
26983 For Guile scripts, replace @code{.byte 1} with @code{.byte 3}.
26984 Then one can reference the macro in a header or source file like this:
26987 DEFINE_GDB_PY_SCRIPT ("my-app-scripts.py")
26990 The script name may include directories if desired.
26992 Note that loading of this script file also requires accordingly configured
26993 @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
26995 If the macro invocation is put in a header, any application or library
26996 using this header will get a reference to the specified script,
26997 and with the use of @code{"MS"} attributes on the section, the linker
26998 will remove duplicates.
27000 @subsubsection Script Text Entries
27002 Script text entries allow to put the executable script in the entry
27003 itself instead of loading it from a file.
27004 The first line of the entry, everything after the prefix byte and up to
27005 the first newline (@code{0xa}) character, is the script name, and must not
27006 contain any kind of space character, e.g., spaces or tabs.
27007 The rest of the entry, up to the trailing null byte, is the script to
27008 execute in the specified language. The name needs to be unique among
27009 all script names, as @value{GDBN} executes each script only once based
27012 Here is an example from file @file{py-section-script.c} in the @value{GDBN}
27016 #include "symcat.h"
27017 #include "gdb/section-scripts.h"
27019 ".pushsection \".debug_gdb_scripts\", \"MS\",@@progbits,1\n"
27020 ".byte " XSTRING (SECTION_SCRIPT_ID_PYTHON_TEXT) "\n"
27021 ".ascii \"gdb.inlined-script\\n\"\n"
27022 ".ascii \"class test_cmd (gdb.Command):\\n\"\n"
27023 ".ascii \" def __init__ (self):\\n\"\n"
27024 ".ascii \" super (test_cmd, self).__init__ ("
27025 "\\\"test-cmd\\\", gdb.COMMAND_OBSCURE)\\n\"\n"
27026 ".ascii \" def invoke (self, arg, from_tty):\\n\"\n"
27027 ".ascii \" print (\\\"test-cmd output, arg = %s\\\" % arg)\\n\"\n"
27028 ".ascii \"test_cmd ()\\n\"\n"
27034 Loading of inlined scripts requires a properly configured
27035 @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
27036 The path to specify in @code{auto-load safe-path} is the path of the file
27037 containing the @code{.debug_gdb_scripts} section.
27039 @node Which flavor to choose?
27040 @subsection Which flavor to choose?
27042 Given the multiple ways of auto-loading extensions, it might not always
27043 be clear which one to choose. This section provides some guidance.
27046 Benefits of the @file{-gdb.@var{ext}} way:
27050 Can be used with file formats that don't support multiple sections.
27053 Ease of finding scripts for public libraries.
27055 Scripts specified in the @code{.debug_gdb_scripts} section are searched for
27056 in the source search path.
27057 For publicly installed libraries, e.g., @file{libstdc++}, there typically
27058 isn't a source directory in which to find the script.
27061 Doesn't require source code additions.
27065 Benefits of the @code{.debug_gdb_scripts} way:
27069 Works with static linking.
27071 Scripts for libraries done the @file{-gdb.@var{ext}} way require an objfile to
27072 trigger their loading. When an application is statically linked the only
27073 objfile available is the executable, and it is cumbersome to attach all the
27074 scripts from all the input libraries to the executable's
27075 @file{-gdb.@var{ext}} script.
27078 Works with classes that are entirely inlined.
27080 Some classes can be entirely inlined, and thus there may not be an associated
27081 shared library to attach a @file{-gdb.@var{ext}} script to.
27084 Scripts needn't be copied out of the source tree.
27086 In some circumstances, apps can be built out of large collections of internal
27087 libraries, and the build infrastructure necessary to install the
27088 @file{-gdb.@var{ext}} scripts in a place where @value{GDBN} can find them is
27089 cumbersome. It may be easier to specify the scripts in the
27090 @code{.debug_gdb_scripts} section as relative paths, and add a path to the
27091 top of the source tree to the source search path.
27094 @node Multiple Extension Languages
27095 @section Multiple Extension Languages
27097 The Guile and Python extension languages do not share any state,
27098 and generally do not interfere with each other.
27099 There are some things to be aware of, however.
27101 @subsection Python comes first
27103 Python was @value{GDBN}'s first extension language, and to avoid breaking
27104 existing behaviour Python comes first. This is generally solved by the
27105 ``first one wins'' principle. @value{GDBN} maintains a list of enabled
27106 extension languages, and when it makes a call to an extension language,
27107 (say to pretty-print a value), it tries each in turn until an extension
27108 language indicates it has performed the request (e.g., has returned the
27109 pretty-printed form of a value).
27110 This extends to errors while performing such requests: If an error happens
27111 while, for example, trying to pretty-print an object then the error is
27112 reported and any following extension languages are not tried.
27115 @section Creating new spellings of existing commands
27116 @cindex aliases for commands
27118 It is often useful to define alternate spellings of existing commands.
27119 For example, if a new @value{GDBN} command defined in Python has
27120 a long name to type, it is handy to have an abbreviated version of it
27121 that involves less typing.
27123 @value{GDBN} itself uses aliases. For example @samp{s} is an alias
27124 of the @samp{step} command even though it is otherwise an ambiguous
27125 abbreviation of other commands like @samp{set} and @samp{show}.
27127 Aliases are also used to provide shortened or more common versions
27128 of multi-word commands. For example, @value{GDBN} provides the
27129 @samp{tty} alias of the @samp{set inferior-tty} command.
27131 You can define a new alias with the @samp{alias} command.
27136 @item alias [-a] [--] @var{ALIAS} = @var{COMMAND}
27140 @var{ALIAS} specifies the name of the new alias.
27141 Each word of @var{ALIAS} must consist of letters, numbers, dashes and
27144 @var{COMMAND} specifies the name of an existing command
27145 that is being aliased.
27147 The @samp{-a} option specifies that the new alias is an abbreviation
27148 of the command. Abbreviations are not shown in command
27149 lists displayed by the @samp{help} command.
27151 The @samp{--} option specifies the end of options,
27152 and is useful when @var{ALIAS} begins with a dash.
27154 Here is a simple example showing how to make an abbreviation
27155 of a command so that there is less to type.
27156 Suppose you were tired of typing @samp{disas}, the current
27157 shortest unambiguous abbreviation of the @samp{disassemble} command
27158 and you wanted an even shorter version named @samp{di}.
27159 The following will accomplish this.
27162 (gdb) alias -a di = disas
27165 Note that aliases are different from user-defined commands.
27166 With a user-defined command, you also need to write documentation
27167 for it with the @samp{document} command.
27168 An alias automatically picks up the documentation of the existing command.
27170 Here is an example where we make @samp{elms} an abbreviation of
27171 @samp{elements} in the @samp{set print elements} command.
27172 This is to show that you can make an abbreviation of any part
27176 (gdb) alias -a set print elms = set print elements
27177 (gdb) alias -a show print elms = show print elements
27178 (gdb) set p elms 20
27180 Limit on string chars or array elements to print is 200.
27183 Note that if you are defining an alias of a @samp{set} command,
27184 and you want to have an alias for the corresponding @samp{show}
27185 command, then you need to define the latter separately.
27187 Unambiguously abbreviated commands are allowed in @var{COMMAND} and
27188 @var{ALIAS}, just as they are normally.
27191 (gdb) alias -a set pr elms = set p ele
27194 Finally, here is an example showing the creation of a one word
27195 alias for a more complex command.
27196 This creates alias @samp{spe} of the command @samp{set print elements}.
27199 (gdb) alias spe = set print elements
27204 @chapter Command Interpreters
27205 @cindex command interpreters
27207 @value{GDBN} supports multiple command interpreters, and some command
27208 infrastructure to allow users or user interface writers to switch
27209 between interpreters or run commands in other interpreters.
27211 @value{GDBN} currently supports two command interpreters, the console
27212 interpreter (sometimes called the command-line interpreter or @sc{cli})
27213 and the machine interface interpreter (or @sc{gdb/mi}). This manual
27214 describes both of these interfaces in great detail.
27216 By default, @value{GDBN} will start with the console interpreter.
27217 However, the user may choose to start @value{GDBN} with another
27218 interpreter by specifying the @option{-i} or @option{--interpreter}
27219 startup options. Defined interpreters include:
27223 @cindex console interpreter
27224 The traditional console or command-line interpreter. This is the most often
27225 used interpreter with @value{GDBN}. With no interpreter specified at runtime,
27226 @value{GDBN} will use this interpreter.
27229 @cindex mi interpreter
27230 The newest @sc{gdb/mi} interface (currently @code{mi3}). Used primarily
27231 by programs wishing to use @value{GDBN} as a backend for a debugger GUI
27232 or an IDE. For more information, see @ref{GDB/MI, ,The @sc{gdb/mi}
27236 @cindex mi3 interpreter
27237 The @sc{gdb/mi} interface introduced in @value{GDBN} 9.1.
27240 @cindex mi2 interpreter
27241 The @sc{gdb/mi} interface introduced in @value{GDBN} 6.0.
27244 @cindex mi1 interpreter
27245 The @sc{gdb/mi} interface introduced in @value{GDBN} 5.1.
27249 @cindex invoke another interpreter
27251 @kindex interpreter-exec
27252 You may execute commands in any interpreter from the current
27253 interpreter using the appropriate command. If you are running the
27254 console interpreter, simply use the @code{interpreter-exec} command:
27257 interpreter-exec mi "-data-list-register-names"
27260 @sc{gdb/mi} has a similar command, although it is only available in versions of
27261 @value{GDBN} which support @sc{gdb/mi} version 2 (or greater).
27263 Note that @code{interpreter-exec} only changes the interpreter for the
27264 duration of the specified command. It does not change the interpreter
27267 @cindex start a new independent interpreter
27269 Although you may only choose a single interpreter at startup, it is
27270 possible to run an independent interpreter on a specified input/output
27271 device (usually a tty).
27273 For example, consider a debugger GUI or IDE that wants to provide a
27274 @value{GDBN} console view. It may do so by embedding a terminal
27275 emulator widget in its GUI, starting @value{GDBN} in the traditional
27276 command-line mode with stdin/stdout/stderr redirected to that
27277 terminal, and then creating an MI interpreter running on a specified
27278 input/output device. The console interpreter created by @value{GDBN}
27279 at startup handles commands the user types in the terminal widget,
27280 while the GUI controls and synchronizes state with @value{GDBN} using
27281 the separate MI interpreter.
27283 To start a new secondary @dfn{user interface} running MI, use the
27284 @code{new-ui} command:
27287 @cindex new user interface
27289 new-ui @var{interpreter} @var{tty}
27292 The @var{interpreter} parameter specifies the interpreter to run.
27293 This accepts the same values as the @code{interpreter-exec} command.
27294 For example, @samp{console}, @samp{mi}, @samp{mi2}, etc. The
27295 @var{tty} parameter specifies the name of the bidirectional file the
27296 interpreter uses for input/output, usually the name of a
27297 pseudoterminal slave on Unix systems. For example:
27300 (@value{GDBP}) new-ui mi /dev/pts/9
27304 runs an MI interpreter on @file{/dev/pts/9}.
27307 @chapter @value{GDBN} Text User Interface
27309 @cindex Text User Interface
27312 * TUI Overview:: TUI overview
27313 * TUI Keys:: TUI key bindings
27314 * TUI Single Key Mode:: TUI single key mode
27315 * TUI Commands:: TUI-specific commands
27316 * TUI Configuration:: TUI configuration variables
27319 The @value{GDBN} Text User Interface (TUI) is a terminal
27320 interface which uses the @code{curses} library to show the source
27321 file, the assembly output, the program registers and @value{GDBN}
27322 commands in separate text windows. The TUI mode is supported only
27323 on platforms where a suitable version of the @code{curses} library
27326 The TUI mode is enabled by default when you invoke @value{GDBN} as
27327 @samp{@value{GDBP} -tui}.
27328 You can also switch in and out of TUI mode while @value{GDBN} runs by
27329 using various TUI commands and key bindings, such as @command{tui
27330 enable} or @kbd{C-x C-a}. @xref{TUI Commands, ,TUI Commands}, and
27331 @ref{TUI Keys, ,TUI Key Bindings}.
27334 @section TUI Overview
27336 In TUI mode, @value{GDBN} can display several text windows:
27340 This window is the @value{GDBN} command window with the @value{GDBN}
27341 prompt and the @value{GDBN} output. The @value{GDBN} input is still
27342 managed using readline.
27345 The source window shows the source file of the program. The current
27346 line and active breakpoints are displayed in this window.
27349 The assembly window shows the disassembly output of the program.
27352 This window shows the processor registers. Registers are highlighted
27353 when their values change.
27356 The source and assembly windows show the current program position
27357 by highlighting the current line and marking it with a @samp{>} marker.
27358 Breakpoints are indicated with two markers. The first marker
27359 indicates the breakpoint type:
27363 Breakpoint which was hit at least once.
27366 Breakpoint which was never hit.
27369 Hardware breakpoint which was hit at least once.
27372 Hardware breakpoint which was never hit.
27375 The second marker indicates whether the breakpoint is enabled or not:
27379 Breakpoint is enabled.
27382 Breakpoint is disabled.
27385 The source, assembly and register windows are updated when the current
27386 thread changes, when the frame changes, or when the program counter
27389 These windows are not all visible at the same time. The command
27390 window is always visible. The others can be arranged in several
27401 source and assembly,
27404 source and registers, or
27407 assembly and registers.
27410 A status line above the command window shows the following information:
27414 Indicates the current @value{GDBN} target.
27415 (@pxref{Targets, ,Specifying a Debugging Target}).
27418 Gives the current process or thread number.
27419 When no process is being debugged, this field is set to @code{No process}.
27422 Gives the current function name for the selected frame.
27423 The name is demangled if demangling is turned on (@pxref{Print Settings}).
27424 When there is no symbol corresponding to the current program counter,
27425 the string @code{??} is displayed.
27428 Indicates the current line number for the selected frame.
27429 When the current line number is not known, the string @code{??} is displayed.
27432 Indicates the current program counter address.
27436 @section TUI Key Bindings
27437 @cindex TUI key bindings
27439 The TUI installs several key bindings in the readline keymaps
27440 @ifset SYSTEM_READLINE
27441 (@pxref{Command Line Editing, , , rluserman, GNU Readline Library}).
27443 @ifclear SYSTEM_READLINE
27444 (@pxref{Command Line Editing}).
27446 The following key bindings are installed for both TUI mode and the
27447 @value{GDBN} standard mode.
27456 Enter or leave the TUI mode. When leaving the TUI mode,
27457 the curses window management stops and @value{GDBN} operates using
27458 its standard mode, writing on the terminal directly. When reentering
27459 the TUI mode, control is given back to the curses windows.
27460 The screen is then refreshed.
27464 Use a TUI layout with only one window. The layout will
27465 either be @samp{source} or @samp{assembly}. When the TUI mode
27466 is not active, it will switch to the TUI mode.
27468 Think of this key binding as the Emacs @kbd{C-x 1} binding.
27472 Use a TUI layout with at least two windows. When the current
27473 layout already has two windows, the next layout with two windows is used.
27474 When a new layout is chosen, one window will always be common to the
27475 previous layout and the new one.
27477 Think of it as the Emacs @kbd{C-x 2} binding.
27481 Change the active window. The TUI associates several key bindings
27482 (like scrolling and arrow keys) with the active window. This command
27483 gives the focus to the next TUI window.
27485 Think of it as the Emacs @kbd{C-x o} binding.
27489 Switch in and out of the TUI SingleKey mode that binds single
27490 keys to @value{GDBN} commands (@pxref{TUI Single Key Mode}).
27493 The following key bindings only work in the TUI mode:
27498 Scroll the active window one page up.
27502 Scroll the active window one page down.
27506 Scroll the active window one line up.
27510 Scroll the active window one line down.
27514 Scroll the active window one column left.
27518 Scroll the active window one column right.
27522 Refresh the screen.
27525 Because the arrow keys scroll the active window in the TUI mode, they
27526 are not available for their normal use by readline unless the command
27527 window has the focus. When another window is active, you must use
27528 other readline key bindings such as @kbd{C-p}, @kbd{C-n}, @kbd{C-b}
27529 and @kbd{C-f} to control the command window.
27531 @node TUI Single Key Mode
27532 @section TUI Single Key Mode
27533 @cindex TUI single key mode
27535 The TUI also provides a @dfn{SingleKey} mode, which binds several
27536 frequently used @value{GDBN} commands to single keys. Type @kbd{C-x s} to
27537 switch into this mode, where the following key bindings are used:
27540 @kindex c @r{(SingleKey TUI key)}
27544 @kindex d @r{(SingleKey TUI key)}
27548 @kindex f @r{(SingleKey TUI key)}
27552 @kindex n @r{(SingleKey TUI key)}
27556 @kindex o @r{(SingleKey TUI key)}
27558 nexti. The shortcut letter @samp{o} stands for ``step Over''.
27560 @kindex q @r{(SingleKey TUI key)}
27562 exit the SingleKey mode.
27564 @kindex r @r{(SingleKey TUI key)}
27568 @kindex s @r{(SingleKey TUI key)}
27572 @kindex i @r{(SingleKey TUI key)}
27574 stepi. The shortcut letter @samp{i} stands for ``step Into''.
27576 @kindex u @r{(SingleKey TUI key)}
27580 @kindex v @r{(SingleKey TUI key)}
27584 @kindex w @r{(SingleKey TUI key)}
27589 Other keys temporarily switch to the @value{GDBN} command prompt.
27590 The key that was pressed is inserted in the editing buffer so that
27591 it is possible to type most @value{GDBN} commands without interaction
27592 with the TUI SingleKey mode. Once the command is entered the TUI
27593 SingleKey mode is restored. The only way to permanently leave
27594 this mode is by typing @kbd{q} or @kbd{C-x s}.
27596 @cindex SingleKey keymap name
27597 If @value{GDBN} was built with Readline 8.0 or later, the TUI
27598 SingleKey keymap will be named @samp{SingleKey}. This can be used in
27599 @file{.inputrc} to add additional bindings to this keymap.
27602 @section TUI-specific Commands
27603 @cindex TUI commands
27605 The TUI has specific commands to control the text windows.
27606 These commands are always available, even when @value{GDBN} is not in
27607 the TUI mode. When @value{GDBN} is in the standard mode, most
27608 of these commands will automatically switch to the TUI mode.
27610 Note that if @value{GDBN}'s @code{stdout} is not connected to a
27611 terminal, or @value{GDBN} has been started with the machine interface
27612 interpreter (@pxref{GDB/MI, ,The @sc{gdb/mi} Interface}), most of
27613 these commands will fail with an error, because it would not be
27614 possible or desirable to enable curses window management.
27619 Activate TUI mode. The last active TUI window layout will be used if
27620 TUI mode has prevsiouly been used in the current debugging session,
27621 otherwise a default layout is used.
27624 @kindex tui disable
27625 Disable TUI mode, returning to the console interpreter.
27629 List and give the size of all displayed windows.
27631 @item layout @var{name}
27633 Changes which TUI windows are displayed. In each layout the command
27634 window is always displayed, the @var{name} parameter controls which
27635 additional windows are displayed, and can be any of the following:
27639 Display the next layout.
27642 Display the previous layout.
27645 Display the source and command windows.
27648 Display the assembly and command windows.
27651 Display the source, assembly, and command windows.
27654 When in @code{src} layout display the register, source, and command
27655 windows. When in @code{asm} or @code{split} layout display the
27656 register, assembler, and command windows.
27659 @item focus @var{name}
27661 Changes which TUI window is currently active for scrolling. The
27662 @var{name} parameter can be any of the following:
27666 Make the next window active for scrolling.
27669 Make the previous window active for scrolling.
27672 Make the source window active for scrolling.
27675 Make the assembly window active for scrolling.
27678 Make the register window active for scrolling.
27681 Make the command window active for scrolling.
27686 Refresh the screen. This is similar to typing @kbd{C-L}.
27688 @item tui reg @var{group}
27690 Changes the register group displayed in the tui register window to
27691 @var{group}. If the register window is not currently displayed this
27692 command will cause the register window to be displayed. The list of
27693 register groups, as well as their order is target specific. The
27694 following groups are available on most targets:
27697 Repeatedly selecting this group will cause the display to cycle
27698 through all of the available register groups.
27701 Repeatedly selecting this group will cause the display to cycle
27702 through all of the available register groups in the reverse order to
27706 Display the general registers.
27708 Display the floating point registers.
27710 Display the system registers.
27712 Display the vector registers.
27714 Display all registers.
27719 Update the source window and the current execution point.
27721 @item winheight @var{name} +@var{count}
27722 @itemx winheight @var{name} -@var{count}
27724 Change the height of the window @var{name} by @var{count}
27725 lines. Positive counts increase the height, while negative counts
27726 decrease it. The @var{name} parameter can be one of @code{src} (the
27727 source window), @code{cmd} (the command window), @code{asm} (the
27728 disassembly window), or @code{regs} (the register display window).
27731 @node TUI Configuration
27732 @section TUI Configuration Variables
27733 @cindex TUI configuration variables
27735 Several configuration variables control the appearance of TUI windows.
27738 @item set tui border-kind @var{kind}
27739 @kindex set tui border-kind
27740 Select the border appearance for the source, assembly and register windows.
27741 The possible values are the following:
27744 Use a space character to draw the border.
27747 Use @sc{ascii} characters @samp{+}, @samp{-} and @samp{|} to draw the border.
27750 Use the Alternate Character Set to draw the border. The border is
27751 drawn using character line graphics if the terminal supports them.
27754 @item set tui border-mode @var{mode}
27755 @kindex set tui border-mode
27756 @itemx set tui active-border-mode @var{mode}
27757 @kindex set tui active-border-mode
27758 Select the display attributes for the borders of the inactive windows
27759 or the active window. The @var{mode} can be one of the following:
27762 Use normal attributes to display the border.
27768 Use reverse video mode.
27771 Use half bright mode.
27773 @item half-standout
27774 Use half bright and standout mode.
27777 Use extra bright or bold mode.
27779 @item bold-standout
27780 Use extra bright or bold and standout mode.
27783 @item set tui tab-width @var{nchars}
27784 @kindex set tui tab-width
27786 Set the width of tab stops to be @var{nchars} characters. This
27787 setting affects the display of TAB characters in the source and
27792 @chapter Using @value{GDBN} under @sc{gnu} Emacs
27795 @cindex @sc{gnu} Emacs
27796 A special interface allows you to use @sc{gnu} Emacs to view (and
27797 edit) the source files for the program you are debugging with
27800 To use this interface, use the command @kbd{M-x gdb} in Emacs. Give the
27801 executable file you want to debug as an argument. This command starts
27802 @value{GDBN} as a subprocess of Emacs, with input and output through a newly
27803 created Emacs buffer.
27804 @c (Do not use the @code{-tui} option to run @value{GDBN} from Emacs.)
27806 Running @value{GDBN} under Emacs can be just like running @value{GDBN} normally except for two
27811 All ``terminal'' input and output goes through an Emacs buffer, called
27814 This applies both to @value{GDBN} commands and their output, and to the input
27815 and output done by the program you are debugging.
27817 This is useful because it means that you can copy the text of previous
27818 commands and input them again; you can even use parts of the output
27821 All the facilities of Emacs' Shell mode are available for interacting
27822 with your program. In particular, you can send signals the usual
27823 way---for example, @kbd{C-c C-c} for an interrupt, @kbd{C-c C-z} for a
27827 @value{GDBN} displays source code through Emacs.
27829 Each time @value{GDBN} displays a stack frame, Emacs automatically finds the
27830 source file for that frame and puts an arrow (@samp{=>}) at the
27831 left margin of the current line. Emacs uses a separate buffer for
27832 source display, and splits the screen to show both your @value{GDBN} session
27835 Explicit @value{GDBN} @code{list} or search commands still produce output as
27836 usual, but you probably have no reason to use them from Emacs.
27839 We call this @dfn{text command mode}. Emacs 22.1, and later, also uses
27840 a graphical mode, enabled by default, which provides further buffers
27841 that can control the execution and describe the state of your program.
27842 @xref{GDB Graphical Interface,,, Emacs, The @sc{gnu} Emacs Manual}.
27844 If you specify an absolute file name when prompted for the @kbd{M-x
27845 gdb} argument, then Emacs sets your current working directory to where
27846 your program resides. If you only specify the file name, then Emacs
27847 sets your current working directory to the directory associated
27848 with the previous buffer. In this case, @value{GDBN} may find your
27849 program by searching your environment's @code{PATH} variable, but on
27850 some operating systems it might not find the source. So, although the
27851 @value{GDBN} input and output session proceeds normally, the auxiliary
27852 buffer does not display the current source and line of execution.
27854 The initial working directory of @value{GDBN} is printed on the top
27855 line of the GUD buffer and this serves as a default for the commands
27856 that specify files for @value{GDBN} to operate on. @xref{Files,
27857 ,Commands to Specify Files}.
27859 By default, @kbd{M-x gdb} calls the program called @file{gdb}. If you
27860 need to call @value{GDBN} by a different name (for example, if you
27861 keep several configurations around, with different names) you can
27862 customize the Emacs variable @code{gud-gdb-command-name} to run the
27865 In the GUD buffer, you can use these special Emacs commands in
27866 addition to the standard Shell mode commands:
27870 Describe the features of Emacs' GUD Mode.
27873 Execute to another source line, like the @value{GDBN} @code{step} command; also
27874 update the display window to show the current file and location.
27877 Execute to next source line in this function, skipping all function
27878 calls, like the @value{GDBN} @code{next} command. Then update the display window
27879 to show the current file and location.
27882 Execute one instruction, like the @value{GDBN} @code{stepi} command; update
27883 display window accordingly.
27886 Execute until exit from the selected stack frame, like the @value{GDBN}
27887 @code{finish} command.
27890 Continue execution of your program, like the @value{GDBN} @code{continue}
27894 Go up the number of frames indicated by the numeric argument
27895 (@pxref{Arguments, , Numeric Arguments, Emacs, The @sc{gnu} Emacs Manual}),
27896 like the @value{GDBN} @code{up} command.
27899 Go down the number of frames indicated by the numeric argument, like the
27900 @value{GDBN} @code{down} command.
27903 In any source file, the Emacs command @kbd{C-x @key{SPC}} (@code{gud-break})
27904 tells @value{GDBN} to set a breakpoint on the source line point is on.
27906 In text command mode, if you type @kbd{M-x speedbar}, Emacs displays a
27907 separate frame which shows a backtrace when the GUD buffer is current.
27908 Move point to any frame in the stack and type @key{RET} to make it
27909 become the current frame and display the associated source in the
27910 source buffer. Alternatively, click @kbd{Mouse-2} to make the
27911 selected frame become the current one. In graphical mode, the
27912 speedbar displays watch expressions.
27914 If you accidentally delete the source-display buffer, an easy way to get
27915 it back is to type the command @code{f} in the @value{GDBN} buffer, to
27916 request a frame display; when you run under Emacs, this recreates
27917 the source buffer if necessary to show you the context of the current
27920 The source files displayed in Emacs are in ordinary Emacs buffers
27921 which are visiting the source files in the usual way. You can edit
27922 the files with these buffers if you wish; but keep in mind that @value{GDBN}
27923 communicates with Emacs in terms of line numbers. If you add or
27924 delete lines from the text, the line numbers that @value{GDBN} knows cease
27925 to correspond properly with the code.
27927 A more detailed description of Emacs' interaction with @value{GDBN} is
27928 given in the Emacs manual (@pxref{Debuggers,,, Emacs, The @sc{gnu}
27932 @chapter The @sc{gdb/mi} Interface
27934 @unnumberedsec Function and Purpose
27936 @cindex @sc{gdb/mi}, its purpose
27937 @sc{gdb/mi} is a line based machine oriented text interface to
27938 @value{GDBN} and is activated by specifying using the
27939 @option{--interpreter} command line option (@pxref{Mode Options}). It
27940 is specifically intended to support the development of systems which
27941 use the debugger as just one small component of a larger system.
27943 This chapter is a specification of the @sc{gdb/mi} interface. It is written
27944 in the form of a reference manual.
27946 Note that @sc{gdb/mi} is still under construction, so some of the
27947 features described below are incomplete and subject to change
27948 (@pxref{GDB/MI Development and Front Ends, , @sc{gdb/mi} Development and Front Ends}).
27950 @unnumberedsec Notation and Terminology
27952 @cindex notational conventions, for @sc{gdb/mi}
27953 This chapter uses the following notation:
27957 @code{|} separates two alternatives.
27960 @code{[ @var{something} ]} indicates that @var{something} is optional:
27961 it may or may not be given.
27964 @code{( @var{group} )*} means that @var{group} inside the parentheses
27965 may repeat zero or more times.
27968 @code{( @var{group} )+} means that @var{group} inside the parentheses
27969 may repeat one or more times.
27972 @code{"@var{string}"} means a literal @var{string}.
27976 @heading Dependencies
27980 * GDB/MI General Design::
27981 * GDB/MI Command Syntax::
27982 * GDB/MI Compatibility with CLI::
27983 * GDB/MI Development and Front Ends::
27984 * GDB/MI Output Records::
27985 * GDB/MI Simple Examples::
27986 * GDB/MI Command Description Format::
27987 * GDB/MI Breakpoint Commands::
27988 * GDB/MI Catchpoint Commands::
27989 * GDB/MI Program Context::
27990 * GDB/MI Thread Commands::
27991 * GDB/MI Ada Tasking Commands::
27992 * GDB/MI Program Execution::
27993 * GDB/MI Stack Manipulation::
27994 * GDB/MI Variable Objects::
27995 * GDB/MI Data Manipulation::
27996 * GDB/MI Tracepoint Commands::
27997 * GDB/MI Symbol Query::
27998 * GDB/MI File Commands::
28000 * GDB/MI Kod Commands::
28001 * GDB/MI Memory Overlay Commands::
28002 * GDB/MI Signal Handling Commands::
28004 * GDB/MI Target Manipulation::
28005 * GDB/MI File Transfer Commands::
28006 * GDB/MI Ada Exceptions Commands::
28007 * GDB/MI Support Commands::
28008 * GDB/MI Miscellaneous Commands::
28011 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
28012 @node GDB/MI General Design
28013 @section @sc{gdb/mi} General Design
28014 @cindex GDB/MI General Design
28016 Interaction of a @sc{GDB/MI} frontend with @value{GDBN} involves three
28017 parts---commands sent to @value{GDBN}, responses to those commands
28018 and notifications. Each command results in exactly one response,
28019 indicating either successful completion of the command, or an error.
28020 For the commands that do not resume the target, the response contains the
28021 requested information. For the commands that resume the target, the
28022 response only indicates whether the target was successfully resumed.
28023 Notifications is the mechanism for reporting changes in the state of the
28024 target, or in @value{GDBN} state, that cannot conveniently be associated with
28025 a command and reported as part of that command response.
28027 The important examples of notifications are:
28031 Exec notifications. These are used to report changes in
28032 target state---when a target is resumed, or stopped. It would not
28033 be feasible to include this information in response of resuming
28034 commands, because one resume commands can result in multiple events in
28035 different threads. Also, quite some time may pass before any event
28036 happens in the target, while a frontend needs to know whether the resuming
28037 command itself was successfully executed.
28040 Console output, and status notifications. Console output
28041 notifications are used to report output of CLI commands, as well as
28042 diagnostics for other commands. Status notifications are used to
28043 report the progress of a long-running operation. Naturally, including
28044 this information in command response would mean no output is produced
28045 until the command is finished, which is undesirable.
28048 General notifications. Commands may have various side effects on
28049 the @value{GDBN} or target state beyond their official purpose. For example,
28050 a command may change the selected thread. Although such changes can
28051 be included in command response, using notification allows for more
28052 orthogonal frontend design.
28056 There's no guarantee that whenever an MI command reports an error,
28057 @value{GDBN} or the target are in any specific state, and especially,
28058 the state is not reverted to the state before the MI command was
28059 processed. Therefore, whenever an MI command results in an error,
28060 we recommend that the frontend refreshes all the information shown in
28061 the user interface.
28065 * Context management::
28066 * Asynchronous and non-stop modes::
28070 @node Context management
28071 @subsection Context management
28073 @subsubsection Threads and Frames
28075 In most cases when @value{GDBN} accesses the target, this access is
28076 done in context of a specific thread and frame (@pxref{Frames}).
28077 Often, even when accessing global data, the target requires that a thread
28078 be specified. The CLI interface maintains the selected thread and frame,
28079 and supplies them to target on each command. This is convenient,
28080 because a command line user would not want to specify that information
28081 explicitly on each command, and because user interacts with
28082 @value{GDBN} via a single terminal, so no confusion is possible as
28083 to what thread and frame are the current ones.
28085 In the case of MI, the concept of selected thread and frame is less
28086 useful. First, a frontend can easily remember this information
28087 itself. Second, a graphical frontend can have more than one window,
28088 each one used for debugging a different thread, and the frontend might
28089 want to access additional threads for internal purposes. This
28090 increases the risk that by relying on implicitly selected thread, the
28091 frontend may be operating on a wrong one. Therefore, each MI command
28092 should explicitly specify which thread and frame to operate on. To
28093 make it possible, each MI command accepts the @samp{--thread} and
28094 @samp{--frame} options, the value to each is @value{GDBN} global
28095 identifier for thread and frame to operate on.
28097 Usually, each top-level window in a frontend allows the user to select
28098 a thread and a frame, and remembers the user selection for further
28099 operations. However, in some cases @value{GDBN} may suggest that the
28100 current thread or frame be changed. For example, when stopping on a
28101 breakpoint it is reasonable to switch to the thread where breakpoint is
28102 hit. For another example, if the user issues the CLI @samp{thread} or
28103 @samp{frame} commands via the frontend, it is desirable to change the
28104 frontend's selection to the one specified by user. @value{GDBN}
28105 communicates the suggestion to change current thread and frame using the
28106 @samp{=thread-selected} notification.
28108 Note that historically, MI shares the selected thread with CLI, so
28109 frontends used the @code{-thread-select} to execute commands in the
28110 right context. However, getting this to work right is cumbersome. The
28111 simplest way is for frontend to emit @code{-thread-select} command
28112 before every command. This doubles the number of commands that need
28113 to be sent. The alternative approach is to suppress @code{-thread-select}
28114 if the selected thread in @value{GDBN} is supposed to be identical to the
28115 thread the frontend wants to operate on. However, getting this
28116 optimization right can be tricky. In particular, if the frontend
28117 sends several commands to @value{GDBN}, and one of the commands changes the
28118 selected thread, then the behaviour of subsequent commands will
28119 change. So, a frontend should either wait for response from such
28120 problematic commands, or explicitly add @code{-thread-select} for
28121 all subsequent commands. No frontend is known to do this exactly
28122 right, so it is suggested to just always pass the @samp{--thread} and
28123 @samp{--frame} options.
28125 @subsubsection Language
28127 The execution of several commands depends on which language is selected.
28128 By default, the current language (@pxref{show language}) is used.
28129 But for commands known to be language-sensitive, it is recommended
28130 to use the @samp{--language} option. This option takes one argument,
28131 which is the name of the language to use while executing the command.
28135 -data-evaluate-expression --language c "sizeof (void*)"
28140 The valid language names are the same names accepted by the
28141 @samp{set language} command (@pxref{Manually}), excluding @samp{auto},
28142 @samp{local} or @samp{unknown}.
28144 @node Asynchronous and non-stop modes
28145 @subsection Asynchronous command execution and non-stop mode
28147 On some targets, @value{GDBN} is capable of processing MI commands
28148 even while the target is running. This is called @dfn{asynchronous
28149 command execution} (@pxref{Background Execution}). The frontend may
28150 specify a preferrence for asynchronous execution using the
28151 @code{-gdb-set mi-async 1} command, which should be emitted before
28152 either running the executable or attaching to the target. After the
28153 frontend has started the executable or attached to the target, it can
28154 find if asynchronous execution is enabled using the
28155 @code{-list-target-features} command.
28158 @item -gdb-set mi-async on
28159 @item -gdb-set mi-async off
28160 Set whether MI is in asynchronous mode.
28162 When @code{off}, which is the default, MI execution commands (e.g.,
28163 @code{-exec-continue}) are foreground commands, and @value{GDBN} waits
28164 for the program to stop before processing further commands.
28166 When @code{on}, MI execution commands are background execution
28167 commands (e.g., @code{-exec-continue} becomes the equivalent of the
28168 @code{c&} CLI command), and so @value{GDBN} is capable of processing
28169 MI commands even while the target is running.
28171 @item -gdb-show mi-async
28172 Show whether MI asynchronous mode is enabled.
28175 Note: In @value{GDBN} version 7.7 and earlier, this option was called
28176 @code{target-async} instead of @code{mi-async}, and it had the effect
28177 of both putting MI in asynchronous mode and making CLI background
28178 commands possible. CLI background commands are now always possible
28179 ``out of the box'' if the target supports them. The old spelling is
28180 kept as a deprecated alias for backwards compatibility.
28182 Even if @value{GDBN} can accept a command while target is running,
28183 many commands that access the target do not work when the target is
28184 running. Therefore, asynchronous command execution is most useful
28185 when combined with non-stop mode (@pxref{Non-Stop Mode}). Then,
28186 it is possible to examine the state of one thread, while other threads
28189 When a given thread is running, MI commands that try to access the
28190 target in the context of that thread may not work, or may work only on
28191 some targets. In particular, commands that try to operate on thread's
28192 stack will not work, on any target. Commands that read memory, or
28193 modify breakpoints, may work or not work, depending on the target. Note
28194 that even commands that operate on global state, such as @code{print},
28195 @code{set}, and breakpoint commands, still access the target in the
28196 context of a specific thread, so frontend should try to find a
28197 stopped thread and perform the operation on that thread (using the
28198 @samp{--thread} option).
28200 Which commands will work in the context of a running thread is
28201 highly target dependent. However, the two commands
28202 @code{-exec-interrupt}, to stop a thread, and @code{-thread-info},
28203 to find the state of a thread, will always work.
28205 @node Thread groups
28206 @subsection Thread groups
28207 @value{GDBN} may be used to debug several processes at the same time.
28208 On some platfroms, @value{GDBN} may support debugging of several
28209 hardware systems, each one having several cores with several different
28210 processes running on each core. This section describes the MI
28211 mechanism to support such debugging scenarios.
28213 The key observation is that regardless of the structure of the
28214 target, MI can have a global list of threads, because most commands that
28215 accept the @samp{--thread} option do not need to know what process that
28216 thread belongs to. Therefore, it is not necessary to introduce
28217 neither additional @samp{--process} option, nor an notion of the
28218 current process in the MI interface. The only strictly new feature
28219 that is required is the ability to find how the threads are grouped
28222 To allow the user to discover such grouping, and to support arbitrary
28223 hierarchy of machines/cores/processes, MI introduces the concept of a
28224 @dfn{thread group}. Thread group is a collection of threads and other
28225 thread groups. A thread group always has a string identifier, a type,
28226 and may have additional attributes specific to the type. A new
28227 command, @code{-list-thread-groups}, returns the list of top-level
28228 thread groups, which correspond to processes that @value{GDBN} is
28229 debugging at the moment. By passing an identifier of a thread group
28230 to the @code{-list-thread-groups} command, it is possible to obtain
28231 the members of specific thread group.
28233 To allow the user to easily discover processes, and other objects, he
28234 wishes to debug, a concept of @dfn{available thread group} is
28235 introduced. Available thread group is an thread group that
28236 @value{GDBN} is not debugging, but that can be attached to, using the
28237 @code{-target-attach} command. The list of available top-level thread
28238 groups can be obtained using @samp{-list-thread-groups --available}.
28239 In general, the content of a thread group may be only retrieved only
28240 after attaching to that thread group.
28242 Thread groups are related to inferiors (@pxref{Inferiors and
28243 Programs}). Each inferior corresponds to a thread group of a special
28244 type @samp{process}, and some additional operations are permitted on
28245 such thread groups.
28247 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
28248 @node GDB/MI Command Syntax
28249 @section @sc{gdb/mi} Command Syntax
28252 * GDB/MI Input Syntax::
28253 * GDB/MI Output Syntax::
28256 @node GDB/MI Input Syntax
28257 @subsection @sc{gdb/mi} Input Syntax
28259 @cindex input syntax for @sc{gdb/mi}
28260 @cindex @sc{gdb/mi}, input syntax
28262 @item @var{command} @expansion{}
28263 @code{@var{cli-command} | @var{mi-command}}
28265 @item @var{cli-command} @expansion{}
28266 @code{[ @var{token} ] @var{cli-command} @var{nl}}, where
28267 @var{cli-command} is any existing @value{GDBN} CLI command.
28269 @item @var{mi-command} @expansion{}
28270 @code{[ @var{token} ] "-" @var{operation} ( " " @var{option} )*
28271 @code{[} " --" @code{]} ( " " @var{parameter} )* @var{nl}}
28273 @item @var{token} @expansion{}
28274 "any sequence of digits"
28276 @item @var{option} @expansion{}
28277 @code{"-" @var{parameter} [ " " @var{parameter} ]}
28279 @item @var{parameter} @expansion{}
28280 @code{@var{non-blank-sequence} | @var{c-string}}
28282 @item @var{operation} @expansion{}
28283 @emph{any of the operations described in this chapter}
28285 @item @var{non-blank-sequence} @expansion{}
28286 @emph{anything, provided it doesn't contain special characters such as
28287 "-", @var{nl}, """ and of course " "}
28289 @item @var{c-string} @expansion{}
28290 @code{""" @var{seven-bit-iso-c-string-content} """}
28292 @item @var{nl} @expansion{}
28301 The CLI commands are still handled by the @sc{mi} interpreter; their
28302 output is described below.
28305 The @code{@var{token}}, when present, is passed back when the command
28309 Some @sc{mi} commands accept optional arguments as part of the parameter
28310 list. Each option is identified by a leading @samp{-} (dash) and may be
28311 followed by an optional argument parameter. Options occur first in the
28312 parameter list and can be delimited from normal parameters using
28313 @samp{--} (this is useful when some parameters begin with a dash).
28320 We want easy access to the existing CLI syntax (for debugging).
28323 We want it to be easy to spot a @sc{mi} operation.
28326 @node GDB/MI Output Syntax
28327 @subsection @sc{gdb/mi} Output Syntax
28329 @cindex output syntax of @sc{gdb/mi}
28330 @cindex @sc{gdb/mi}, output syntax
28331 The output from @sc{gdb/mi} consists of zero or more out-of-band records
28332 followed, optionally, by a single result record. This result record
28333 is for the most recent command. The sequence of output records is
28334 terminated by @samp{(gdb)}.
28336 If an input command was prefixed with a @code{@var{token}} then the
28337 corresponding output for that command will also be prefixed by that same
28341 @item @var{output} @expansion{}
28342 @code{( @var{out-of-band-record} )* [ @var{result-record} ] "(gdb)" @var{nl}}
28344 @item @var{result-record} @expansion{}
28345 @code{ [ @var{token} ] "^" @var{result-class} ( "," @var{result} )* @var{nl}}
28347 @item @var{out-of-band-record} @expansion{}
28348 @code{@var{async-record} | @var{stream-record}}
28350 @item @var{async-record} @expansion{}
28351 @code{@var{exec-async-output} | @var{status-async-output} | @var{notify-async-output}}
28353 @item @var{exec-async-output} @expansion{}
28354 @code{[ @var{token} ] "*" @var{async-output nl}}
28356 @item @var{status-async-output} @expansion{}
28357 @code{[ @var{token} ] "+" @var{async-output nl}}
28359 @item @var{notify-async-output} @expansion{}
28360 @code{[ @var{token} ] "=" @var{async-output nl}}
28362 @item @var{async-output} @expansion{}
28363 @code{@var{async-class} ( "," @var{result} )*}
28365 @item @var{result-class} @expansion{}
28366 @code{"done" | "running" | "connected" | "error" | "exit"}
28368 @item @var{async-class} @expansion{}
28369 @code{"stopped" | @var{others}} (where @var{others} will be added
28370 depending on the needs---this is still in development).
28372 @item @var{result} @expansion{}
28373 @code{ @var{variable} "=" @var{value}}
28375 @item @var{variable} @expansion{}
28376 @code{ @var{string} }
28378 @item @var{value} @expansion{}
28379 @code{ @var{const} | @var{tuple} | @var{list} }
28381 @item @var{const} @expansion{}
28382 @code{@var{c-string}}
28384 @item @var{tuple} @expansion{}
28385 @code{ "@{@}" | "@{" @var{result} ( "," @var{result} )* "@}" }
28387 @item @var{list} @expansion{}
28388 @code{ "[]" | "[" @var{value} ( "," @var{value} )* "]" | "["
28389 @var{result} ( "," @var{result} )* "]" }
28391 @item @var{stream-record} @expansion{}
28392 @code{@var{console-stream-output} | @var{target-stream-output} | @var{log-stream-output}}
28394 @item @var{console-stream-output} @expansion{}
28395 @code{"~" @var{c-string nl}}
28397 @item @var{target-stream-output} @expansion{}
28398 @code{"@@" @var{c-string nl}}
28400 @item @var{log-stream-output} @expansion{}
28401 @code{"&" @var{c-string nl}}
28403 @item @var{nl} @expansion{}
28406 @item @var{token} @expansion{}
28407 @emph{any sequence of digits}.
28415 All output sequences end in a single line containing a period.
28418 The @code{@var{token}} is from the corresponding request. Note that
28419 for all async output, while the token is allowed by the grammar and
28420 may be output by future versions of @value{GDBN} for select async
28421 output messages, it is generally omitted. Frontends should treat
28422 all async output as reporting general changes in the state of the
28423 target and there should be no need to associate async output to any
28427 @cindex status output in @sc{gdb/mi}
28428 @var{status-async-output} contains on-going status information about the
28429 progress of a slow operation. It can be discarded. All status output is
28430 prefixed by @samp{+}.
28433 @cindex async output in @sc{gdb/mi}
28434 @var{exec-async-output} contains asynchronous state change on the target
28435 (stopped, started, disappeared). All async output is prefixed by
28439 @cindex notify output in @sc{gdb/mi}
28440 @var{notify-async-output} contains supplementary information that the
28441 client should handle (e.g., a new breakpoint information). All notify
28442 output is prefixed by @samp{=}.
28445 @cindex console output in @sc{gdb/mi}
28446 @var{console-stream-output} is output that should be displayed as is in the
28447 console. It is the textual response to a CLI command. All the console
28448 output is prefixed by @samp{~}.
28451 @cindex target output in @sc{gdb/mi}
28452 @var{target-stream-output} is the output produced by the target program.
28453 All the target output is prefixed by @samp{@@}.
28456 @cindex log output in @sc{gdb/mi}
28457 @var{log-stream-output} is output text coming from @value{GDBN}'s internals, for
28458 instance messages that should be displayed as part of an error log. All
28459 the log output is prefixed by @samp{&}.
28462 @cindex list output in @sc{gdb/mi}
28463 New @sc{gdb/mi} commands should only output @var{lists} containing
28469 @xref{GDB/MI Stream Records, , @sc{gdb/mi} Stream Records}, for more
28470 details about the various output records.
28472 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
28473 @node GDB/MI Compatibility with CLI
28474 @section @sc{gdb/mi} Compatibility with CLI
28476 @cindex compatibility, @sc{gdb/mi} and CLI
28477 @cindex @sc{gdb/mi}, compatibility with CLI
28479 For the developers convenience CLI commands can be entered directly,
28480 but there may be some unexpected behaviour. For example, commands
28481 that query the user will behave as if the user replied yes, breakpoint
28482 command lists are not executed and some CLI commands, such as
28483 @code{if}, @code{when} and @code{define}, prompt for further input with
28484 @samp{>}, which is not valid MI output.
28486 This feature may be removed at some stage in the future and it is
28487 recommended that front ends use the @code{-interpreter-exec} command
28488 (@pxref{-interpreter-exec}).
28490 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
28491 @node GDB/MI Development and Front Ends
28492 @section @sc{gdb/mi} Development and Front Ends
28493 @cindex @sc{gdb/mi} development
28495 The application which takes the MI output and presents the state of the
28496 program being debugged to the user is called a @dfn{front end}.
28498 Since @sc{gdb/mi} is used by a variety of front ends to @value{GDBN}, changes
28499 to the MI interface may break existing usage. This section describes how the
28500 protocol changes and how to request previous version of the protocol when it
28503 Some changes in MI need not break a carefully designed front end, and
28504 for these the MI version will remain unchanged. The following is a
28505 list of changes that may occur within one level, so front ends should
28506 parse MI output in a way that can handle them:
28510 New MI commands may be added.
28513 New fields may be added to the output of any MI command.
28516 The range of values for fields with specified values, e.g.,
28517 @code{in_scope} (@pxref{-var-update}) may be extended.
28519 @c The format of field's content e.g type prefix, may change so parse it
28520 @c at your own risk. Yes, in general?
28522 @c The order of fields may change? Shouldn't really matter but it might
28523 @c resolve inconsistencies.
28526 If the changes are likely to break front ends, the MI version level
28527 will be increased by one. The new versions of the MI protocol are not compatible
28528 with the old versions. Old versions of MI remain available, allowing front ends
28529 to keep using them until they are modified to use the latest MI version.
28531 Since @code{--interpreter=mi} always points to the latest MI version, it is
28532 recommended that front ends request a specific version of MI when launching
28533 @value{GDBN} (e.g. @code{--interpreter=mi2}) to make sure they get an
28534 interpreter with the MI version they expect.
28536 The following table gives a summary of the the released versions of the MI
28537 interface: the version number, the version of GDB in which it first appeared
28538 and the breaking changes compared to the previous version.
28540 @multitable @columnfractions .05 .05 .9
28541 @headitem MI version @tab GDB version @tab Breaking changes
28558 The @code{-environment-pwd}, @code{-environment-directory} and
28559 @code{-environment-path} commands now returns values using the MI output
28560 syntax, rather than CLI output syntax.
28563 @code{-var-list-children}'s @code{children} result field is now a list, rather
28567 @code{-var-update}'s @code{changelist} result field is now a list, rather than
28579 The output of information about multi-location breakpoints has changed in the
28580 responses to the @code{-break-insert} and @code{-break-info} commands, as well
28581 as in the @code{=breakpoint-created} and @code{=breakpoint-modified} events.
28582 The multiple locations are now placed in a @code{locations} field, whose value
28588 If your front end cannot yet migrate to a more recent version of the
28589 MI protocol, you can nevertheless selectively enable specific features
28590 available in those recent MI versions, using the following commands:
28594 @item -fix-multi-location-breakpoint-output
28595 Use the output for multi-location breakpoints which was introduced by
28596 MI 3, even when using MI versions 2 or 1. This command has no
28597 effect when using MI version 3 or later.
28601 The best way to avoid unexpected changes in MI that might break your front
28602 end is to make your project known to @value{GDBN} developers and
28603 follow development on @email{gdb@@sourceware.org} and
28604 @email{gdb-patches@@sourceware.org}.
28605 @cindex mailing lists
28607 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
28608 @node GDB/MI Output Records
28609 @section @sc{gdb/mi} Output Records
28612 * GDB/MI Result Records::
28613 * GDB/MI Stream Records::
28614 * GDB/MI Async Records::
28615 * GDB/MI Breakpoint Information::
28616 * GDB/MI Frame Information::
28617 * GDB/MI Thread Information::
28618 * GDB/MI Ada Exception Information::
28621 @node GDB/MI Result Records
28622 @subsection @sc{gdb/mi} Result Records
28624 @cindex result records in @sc{gdb/mi}
28625 @cindex @sc{gdb/mi}, result records
28626 In addition to a number of out-of-band notifications, the response to a
28627 @sc{gdb/mi} command includes one of the following result indications:
28631 @item "^done" [ "," @var{results} ]
28632 The synchronous operation was successful, @code{@var{results}} are the return
28637 This result record is equivalent to @samp{^done}. Historically, it
28638 was output instead of @samp{^done} if the command has resumed the
28639 target. This behaviour is maintained for backward compatibility, but
28640 all frontends should treat @samp{^done} and @samp{^running}
28641 identically and rely on the @samp{*running} output record to determine
28642 which threads are resumed.
28646 @value{GDBN} has connected to a remote target.
28648 @item "^error" "," "msg=" @var{c-string} [ "," "code=" @var{c-string} ]
28650 The operation failed. The @code{msg=@var{c-string}} variable contains
28651 the corresponding error message.
28653 If present, the @code{code=@var{c-string}} variable provides an error
28654 code on which consumers can rely on to detect the corresponding
28655 error condition. At present, only one error code is defined:
28658 @item "undefined-command"
28659 Indicates that the command causing the error does not exist.
28664 @value{GDBN} has terminated.
28668 @node GDB/MI Stream Records
28669 @subsection @sc{gdb/mi} Stream Records
28671 @cindex @sc{gdb/mi}, stream records
28672 @cindex stream records in @sc{gdb/mi}
28673 @value{GDBN} internally maintains a number of output streams: the console, the
28674 target, and the log. The output intended for each of these streams is
28675 funneled through the @sc{gdb/mi} interface using @dfn{stream records}.
28677 Each stream record begins with a unique @dfn{prefix character} which
28678 identifies its stream (@pxref{GDB/MI Output Syntax, , @sc{gdb/mi} Output
28679 Syntax}). In addition to the prefix, each stream record contains a
28680 @code{@var{string-output}}. This is either raw text (with an implicit new
28681 line) or a quoted C string (which does not contain an implicit newline).
28684 @item "~" @var{string-output}
28685 The console output stream contains text that should be displayed in the
28686 CLI console window. It contains the textual responses to CLI commands.
28688 @item "@@" @var{string-output}
28689 The target output stream contains any textual output from the running
28690 target. This is only present when GDB's event loop is truly
28691 asynchronous, which is currently only the case for remote targets.
28693 @item "&" @var{string-output}
28694 The log stream contains debugging messages being produced by @value{GDBN}'s
28698 @node GDB/MI Async Records
28699 @subsection @sc{gdb/mi} Async Records
28701 @cindex async records in @sc{gdb/mi}
28702 @cindex @sc{gdb/mi}, async records
28703 @dfn{Async} records are used to notify the @sc{gdb/mi} client of
28704 additional changes that have occurred. Those changes can either be a
28705 consequence of @sc{gdb/mi} commands (e.g., a breakpoint modified) or a result of
28706 target activity (e.g., target stopped).
28708 The following is the list of possible async records:
28712 @item *running,thread-id="@var{thread}"
28713 The target is now running. The @var{thread} field can be the global
28714 thread ID of the the thread that is now running, and it can be
28715 @samp{all} if all threads are running. The frontend should assume
28716 that no interaction with a running thread is possible after this
28717 notification is produced. The frontend should not assume that this
28718 notification is output only once for any command. @value{GDBN} may
28719 emit this notification several times, either for different threads,
28720 because it cannot resume all threads together, or even for a single
28721 thread, if the thread must be stepped though some code before letting
28724 @item *stopped,reason="@var{reason}",thread-id="@var{id}",stopped-threads="@var{stopped}",core="@var{core}"
28725 The target has stopped. The @var{reason} field can have one of the
28729 @item breakpoint-hit
28730 A breakpoint was reached.
28731 @item watchpoint-trigger
28732 A watchpoint was triggered.
28733 @item read-watchpoint-trigger
28734 A read watchpoint was triggered.
28735 @item access-watchpoint-trigger
28736 An access watchpoint was triggered.
28737 @item function-finished
28738 An -exec-finish or similar CLI command was accomplished.
28739 @item location-reached
28740 An -exec-until or similar CLI command was accomplished.
28741 @item watchpoint-scope
28742 A watchpoint has gone out of scope.
28743 @item end-stepping-range
28744 An -exec-next, -exec-next-instruction, -exec-step, -exec-step-instruction or
28745 similar CLI command was accomplished.
28746 @item exited-signalled
28747 The inferior exited because of a signal.
28749 The inferior exited.
28750 @item exited-normally
28751 The inferior exited normally.
28752 @item signal-received
28753 A signal was received by the inferior.
28755 The inferior has stopped due to a library being loaded or unloaded.
28756 This can happen when @code{stop-on-solib-events} (@pxref{Files}) is
28757 set or when a @code{catch load} or @code{catch unload} catchpoint is
28758 in use (@pxref{Set Catchpoints}).
28760 The inferior has forked. This is reported when @code{catch fork}
28761 (@pxref{Set Catchpoints}) has been used.
28763 The inferior has vforked. This is reported in when @code{catch vfork}
28764 (@pxref{Set Catchpoints}) has been used.
28765 @item syscall-entry
28766 The inferior entered a system call. This is reported when @code{catch
28767 syscall} (@pxref{Set Catchpoints}) has been used.
28768 @item syscall-return
28769 The inferior returned from a system call. This is reported when
28770 @code{catch syscall} (@pxref{Set Catchpoints}) has been used.
28772 The inferior called @code{exec}. This is reported when @code{catch exec}
28773 (@pxref{Set Catchpoints}) has been used.
28776 The @var{id} field identifies the global thread ID of the thread
28777 that directly caused the stop -- for example by hitting a breakpoint.
28778 Depending on whether all-stop
28779 mode is in effect (@pxref{All-Stop Mode}), @value{GDBN} may either
28780 stop all threads, or only the thread that directly triggered the stop.
28781 If all threads are stopped, the @var{stopped} field will have the
28782 value of @code{"all"}. Otherwise, the value of the @var{stopped}
28783 field will be a list of thread identifiers. Presently, this list will
28784 always include a single thread, but frontend should be prepared to see
28785 several threads in the list. The @var{core} field reports the
28786 processor core on which the stop event has happened. This field may be absent
28787 if such information is not available.
28789 @item =thread-group-added,id="@var{id}"
28790 @itemx =thread-group-removed,id="@var{id}"
28791 A thread group was either added or removed. The @var{id} field
28792 contains the @value{GDBN} identifier of the thread group. When a thread
28793 group is added, it generally might not be associated with a running
28794 process. When a thread group is removed, its id becomes invalid and
28795 cannot be used in any way.
28797 @item =thread-group-started,id="@var{id}",pid="@var{pid}"
28798 A thread group became associated with a running program,
28799 either because the program was just started or the thread group
28800 was attached to a program. The @var{id} field contains the
28801 @value{GDBN} identifier of the thread group. The @var{pid} field
28802 contains process identifier, specific to the operating system.
28804 @item =thread-group-exited,id="@var{id}"[,exit-code="@var{code}"]
28805 A thread group is no longer associated with a running program,
28806 either because the program has exited, or because it was detached
28807 from. The @var{id} field contains the @value{GDBN} identifier of the
28808 thread group. The @var{code} field is the exit code of the inferior; it exists
28809 only when the inferior exited with some code.
28811 @item =thread-created,id="@var{id}",group-id="@var{gid}"
28812 @itemx =thread-exited,id="@var{id}",group-id="@var{gid}"
28813 A thread either was created, or has exited. The @var{id} field
28814 contains the global @value{GDBN} identifier of the thread. The @var{gid}
28815 field identifies the thread group this thread belongs to.
28817 @item =thread-selected,id="@var{id}"[,frame="@var{frame}"]
28818 Informs that the selected thread or frame were changed. This notification
28819 is not emitted as result of the @code{-thread-select} or
28820 @code{-stack-select-frame} commands, but is emitted whenever an MI command
28821 that is not documented to change the selected thread and frame actually
28822 changes them. In particular, invoking, directly or indirectly
28823 (via user-defined command), the CLI @code{thread} or @code{frame} commands,
28824 will generate this notification. Changing the thread or frame from another
28825 user interface (see @ref{Interpreters}) will also generate this notification.
28827 The @var{frame} field is only present if the newly selected thread is
28828 stopped. See @ref{GDB/MI Frame Information} for the format of its value.
28830 We suggest that in response to this notification, front ends
28831 highlight the selected thread and cause subsequent commands to apply to
28834 @item =library-loaded,...
28835 Reports that a new library file was loaded by the program. This
28836 notification has 5 fields---@var{id}, @var{target-name},
28837 @var{host-name}, @var{symbols-loaded} and @var{ranges}. The @var{id} field is an
28838 opaque identifier of the library. For remote debugging case,
28839 @var{target-name} and @var{host-name} fields give the name of the
28840 library file on the target, and on the host respectively. For native
28841 debugging, both those fields have the same value. The
28842 @var{symbols-loaded} field is emitted only for backward compatibility
28843 and should not be relied on to convey any useful information. The
28844 @var{thread-group} field, if present, specifies the id of the thread
28845 group in whose context the library was loaded. If the field is
28846 absent, it means the library was loaded in the context of all present
28847 thread groups. The @var{ranges} field specifies the ranges of addresses belonging
28850 @item =library-unloaded,...
28851 Reports that a library was unloaded by the program. This notification
28852 has 3 fields---@var{id}, @var{target-name} and @var{host-name} with
28853 the same meaning as for the @code{=library-loaded} notification.
28854 The @var{thread-group} field, if present, specifies the id of the
28855 thread group in whose context the library was unloaded. If the field is
28856 absent, it means the library was unloaded in the context of all present
28859 @item =traceframe-changed,num=@var{tfnum},tracepoint=@var{tpnum}
28860 @itemx =traceframe-changed,end
28861 Reports that the trace frame was changed and its new number is
28862 @var{tfnum}. The number of the tracepoint associated with this trace
28863 frame is @var{tpnum}.
28865 @item =tsv-created,name=@var{name},initial=@var{initial}
28866 Reports that the new trace state variable @var{name} is created with
28867 initial value @var{initial}.
28869 @item =tsv-deleted,name=@var{name}
28870 @itemx =tsv-deleted
28871 Reports that the trace state variable @var{name} is deleted or all
28872 trace state variables are deleted.
28874 @item =tsv-modified,name=@var{name},initial=@var{initial}[,current=@var{current}]
28875 Reports that the trace state variable @var{name} is modified with
28876 the initial value @var{initial}. The current value @var{current} of
28877 trace state variable is optional and is reported if the current
28878 value of trace state variable is known.
28880 @item =breakpoint-created,bkpt=@{...@}
28881 @itemx =breakpoint-modified,bkpt=@{...@}
28882 @itemx =breakpoint-deleted,id=@var{number}
28883 Reports that a breakpoint was created, modified, or deleted,
28884 respectively. Only user-visible breakpoints are reported to the MI
28887 The @var{bkpt} argument is of the same form as returned by the various
28888 breakpoint commands; @xref{GDB/MI Breakpoint Commands}. The
28889 @var{number} is the ordinal number of the breakpoint.
28891 Note that if a breakpoint is emitted in the result record of a
28892 command, then it will not also be emitted in an async record.
28894 @item =record-started,thread-group="@var{id}",method="@var{method}"[,format="@var{format}"]
28895 @itemx =record-stopped,thread-group="@var{id}"
28896 Execution log recording was either started or stopped on an
28897 inferior. The @var{id} is the @value{GDBN} identifier of the thread
28898 group corresponding to the affected inferior.
28900 The @var{method} field indicates the method used to record execution. If the
28901 method in use supports multiple recording formats, @var{format} will be present
28902 and contain the currently used format. @xref{Process Record and Replay},
28903 for existing method and format values.
28905 @item =cmd-param-changed,param=@var{param},value=@var{value}
28906 Reports that a parameter of the command @code{set @var{param}} is
28907 changed to @var{value}. In the multi-word @code{set} command,
28908 the @var{param} is the whole parameter list to @code{set} command.
28909 For example, In command @code{set check type on}, @var{param}
28910 is @code{check type} and @var{value} is @code{on}.
28912 @item =memory-changed,thread-group=@var{id},addr=@var{addr},len=@var{len}[,type="code"]
28913 Reports that bytes from @var{addr} to @var{data} + @var{len} were
28914 written in an inferior. The @var{id} is the identifier of the
28915 thread group corresponding to the affected inferior. The optional
28916 @code{type="code"} part is reported if the memory written to holds
28920 @node GDB/MI Breakpoint Information
28921 @subsection @sc{gdb/mi} Breakpoint Information
28923 When @value{GDBN} reports information about a breakpoint, a
28924 tracepoint, a watchpoint, or a catchpoint, it uses a tuple with the
28929 The breakpoint number.
28932 The type of the breakpoint. For ordinary breakpoints this will be
28933 @samp{breakpoint}, but many values are possible.
28936 If the type of the breakpoint is @samp{catchpoint}, then this
28937 indicates the exact type of catchpoint.
28940 This is the breakpoint disposition---either @samp{del}, meaning that
28941 the breakpoint will be deleted at the next stop, or @samp{keep},
28942 meaning that the breakpoint will not be deleted.
28945 This indicates whether the breakpoint is enabled, in which case the
28946 value is @samp{y}, or disabled, in which case the value is @samp{n}.
28947 Note that this is not the same as the field @code{enable}.
28950 The address of the breakpoint. This may be a hexidecimal number,
28951 giving the address; or the string @samp{<PENDING>}, for a pending
28952 breakpoint; or the string @samp{<MULTIPLE>}, for a breakpoint with
28953 multiple locations. This field will not be present if no address can
28954 be determined. For example, a watchpoint does not have an address.
28957 Optional field containing any flags related to the address. These flags are
28958 architecture-dependent; see @ref{Architectures} for their meaning for a
28962 If known, the function in which the breakpoint appears.
28963 If not known, this field is not present.
28966 The name of the source file which contains this function, if known.
28967 If not known, this field is not present.
28970 The full file name of the source file which contains this function, if
28971 known. If not known, this field is not present.
28974 The line number at which this breakpoint appears, if known.
28975 If not known, this field is not present.
28978 If the source file is not known, this field may be provided. If
28979 provided, this holds the address of the breakpoint, possibly followed
28983 If this breakpoint is pending, this field is present and holds the
28984 text used to set the breakpoint, as entered by the user.
28987 Where this breakpoint's condition is evaluated, either @samp{host} or
28991 If this is a thread-specific breakpoint, then this identifies the
28992 thread in which the breakpoint can trigger.
28995 If this breakpoint is restricted to a particular Ada task, then this
28996 field will hold the task identifier.
28999 If the breakpoint is conditional, this is the condition expression.
29002 The ignore count of the breakpoint.
29005 The enable count of the breakpoint.
29007 @item traceframe-usage
29010 @item static-tracepoint-marker-string-id
29011 For a static tracepoint, the name of the static tracepoint marker.
29014 For a masked watchpoint, this is the mask.
29017 A tracepoint's pass count.
29019 @item original-location
29020 The location of the breakpoint as originally specified by the user.
29021 This field is optional.
29024 The number of times the breakpoint has been hit.
29027 This field is only given for tracepoints. This is either @samp{y},
29028 meaning that the tracepoint is installed, or @samp{n}, meaning that it
29032 Some extra data, the exact contents of which are type-dependent.
29035 This field is present if the breakpoint has multiple locations. It is also
29036 exceptionally present if the breakpoint is enabled and has a single, disabled
29039 The value is a list of locations. The format of a location is decribed below.
29043 A location in a multi-location breakpoint is represented as a tuple with the
29049 The location number as a dotted pair, like @samp{1.2}. The first digit is the
29050 number of the parent breakpoint. The second digit is the number of the
29051 location within that breakpoint.
29054 This indicates whether the location is enabled, in which case the
29055 value is @samp{y}, or disabled, in which case the value is @samp{n}.
29056 Note that this is not the same as the field @code{enable}.
29059 The address of this location as an hexidecimal number.
29062 Optional field containing any flags related to the address. These flags are
29063 architecture-dependent; see @ref{Architectures} for their meaning for a
29067 If known, the function in which the location appears.
29068 If not known, this field is not present.
29071 The name of the source file which contains this location, if known.
29072 If not known, this field is not present.
29075 The full file name of the source file which contains this location, if
29076 known. If not known, this field is not present.
29079 The line number at which this location appears, if known.
29080 If not known, this field is not present.
29082 @item thread-groups
29083 The thread groups this location is in.
29087 For example, here is what the output of @code{-break-insert}
29088 (@pxref{GDB/MI Breakpoint Commands}) might be:
29091 -> -break-insert main
29092 <- ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
29093 enabled="y",addr="0x08048564",func="main",file="myprog.c",
29094 fullname="/home/nickrob/myprog.c",line="68",thread-groups=["i1"],
29099 @node GDB/MI Frame Information
29100 @subsection @sc{gdb/mi} Frame Information
29102 Response from many MI commands includes an information about stack
29103 frame. This information is a tuple that may have the following
29108 The level of the stack frame. The innermost frame has the level of
29109 zero. This field is always present.
29112 The name of the function corresponding to the frame. This field may
29113 be absent if @value{GDBN} is unable to determine the function name.
29116 The code address for the frame. This field is always present.
29119 Optional field containing any flags related to the address. These flags are
29120 architecture-dependent; see @ref{Architectures} for their meaning for a
29124 The name of the source files that correspond to the frame's code
29125 address. This field may be absent.
29128 The source line corresponding to the frames' code address. This field
29132 The name of the binary file (either executable or shared library) the
29133 corresponds to the frame's code address. This field may be absent.
29137 @node GDB/MI Thread Information
29138 @subsection @sc{gdb/mi} Thread Information
29140 Whenever @value{GDBN} has to report an information about a thread, it
29141 uses a tuple with the following fields. The fields are always present unless
29146 The global numeric id assigned to the thread by @value{GDBN}.
29149 The target-specific string identifying the thread.
29152 Additional information about the thread provided by the target.
29153 It is supposed to be human-readable and not interpreted by the
29154 frontend. This field is optional.
29157 The name of the thread. If the user specified a name using the
29158 @code{thread name} command, then this name is given. Otherwise, if
29159 @value{GDBN} can extract the thread name from the target, then that
29160 name is given. If @value{GDBN} cannot find the thread name, then this
29164 The execution state of the thread, either @samp{stopped} or @samp{running},
29165 depending on whether the thread is presently running.
29168 The stack frame currently executing in the thread. This field is only present
29169 if the thread is stopped. Its format is documented in
29170 @ref{GDB/MI Frame Information}.
29173 The value of this field is an integer number of the processor core the
29174 thread was last seen on. This field is optional.
29177 @node GDB/MI Ada Exception Information
29178 @subsection @sc{gdb/mi} Ada Exception Information
29180 Whenever a @code{*stopped} record is emitted because the program
29181 stopped after hitting an exception catchpoint (@pxref{Set Catchpoints}),
29182 @value{GDBN} provides the name of the exception that was raised via
29183 the @code{exception-name} field. Also, for exceptions that were raised
29184 with an exception message, @value{GDBN} provides that message via
29185 the @code{exception-message} field.
29187 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
29188 @node GDB/MI Simple Examples
29189 @section Simple Examples of @sc{gdb/mi} Interaction
29190 @cindex @sc{gdb/mi}, simple examples
29192 This subsection presents several simple examples of interaction using
29193 the @sc{gdb/mi} interface. In these examples, @samp{->} means that the
29194 following line is passed to @sc{gdb/mi} as input, while @samp{<-} means
29195 the output received from @sc{gdb/mi}.
29197 Note the line breaks shown in the examples are here only for
29198 readability, they don't appear in the real output.
29200 @subheading Setting a Breakpoint
29202 Setting a breakpoint generates synchronous output which contains detailed
29203 information of the breakpoint.
29206 -> -break-insert main
29207 <- ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
29208 enabled="y",addr="0x08048564",func="main",file="myprog.c",
29209 fullname="/home/nickrob/myprog.c",line="68",thread-groups=["i1"],
29214 @subheading Program Execution
29216 Program execution generates asynchronous records and MI gives the
29217 reason that execution stopped.
29223 <- *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",thread-id="0",
29224 frame=@{addr="0x08048564",func="main",
29225 args=[@{name="argc",value="1"@},@{name="argv",value="0xbfc4d4d4"@}],
29226 file="myprog.c",fullname="/home/nickrob/myprog.c",line="68",
29227 arch="i386:x86_64"@}
29232 <- *stopped,reason="exited-normally"
29236 @subheading Quitting @value{GDBN}
29238 Quitting @value{GDBN} just prints the result class @samp{^exit}.
29246 Please note that @samp{^exit} is printed immediately, but it might
29247 take some time for @value{GDBN} to actually exit. During that time, @value{GDBN}
29248 performs necessary cleanups, including killing programs being debugged
29249 or disconnecting from debug hardware, so the frontend should wait till
29250 @value{GDBN} exits and should only forcibly kill @value{GDBN} if it
29251 fails to exit in reasonable time.
29253 @subheading A Bad Command
29255 Here's what happens if you pass a non-existent command:
29259 <- ^error,msg="Undefined MI command: rubbish"
29264 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
29265 @node GDB/MI Command Description Format
29266 @section @sc{gdb/mi} Command Description Format
29268 The remaining sections describe blocks of commands. Each block of
29269 commands is laid out in a fashion similar to this section.
29271 @subheading Motivation
29273 The motivation for this collection of commands.
29275 @subheading Introduction
29277 A brief introduction to this collection of commands as a whole.
29279 @subheading Commands
29281 For each command in the block, the following is described:
29283 @subsubheading Synopsis
29286 -command @var{args}@dots{}
29289 @subsubheading Result
29291 @subsubheading @value{GDBN} Command
29293 The corresponding @value{GDBN} CLI command(s), if any.
29295 @subsubheading Example
29297 Example(s) formatted for readability. Some of the described commands have
29298 not been implemented yet and these are labeled N.A.@: (not available).
29301 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
29302 @node GDB/MI Breakpoint Commands
29303 @section @sc{gdb/mi} Breakpoint Commands
29305 @cindex breakpoint commands for @sc{gdb/mi}
29306 @cindex @sc{gdb/mi}, breakpoint commands
29307 This section documents @sc{gdb/mi} commands for manipulating
29310 @subheading The @code{-break-after} Command
29311 @findex -break-after
29313 @subsubheading Synopsis
29316 -break-after @var{number} @var{count}
29319 The breakpoint number @var{number} is not in effect until it has been
29320 hit @var{count} times. To see how this is reflected in the output of
29321 the @samp{-break-list} command, see the description of the
29322 @samp{-break-list} command below.
29324 @subsubheading @value{GDBN} Command
29326 The corresponding @value{GDBN} command is @samp{ignore}.
29328 @subsubheading Example
29333 ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
29334 enabled="y",addr="0x000100d0",func="main",file="hello.c",
29335 fullname="/home/foo/hello.c",line="5",thread-groups=["i1"],
29343 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
29344 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
29345 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
29346 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
29347 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
29348 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
29349 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
29350 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
29351 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
29352 line="5",thread-groups=["i1"],times="0",ignore="3"@}]@}
29357 @subheading The @code{-break-catch} Command
29358 @findex -break-catch
29361 @subheading The @code{-break-commands} Command
29362 @findex -break-commands
29364 @subsubheading Synopsis
29367 -break-commands @var{number} [ @var{command1} ... @var{commandN} ]
29370 Specifies the CLI commands that should be executed when breakpoint
29371 @var{number} is hit. The parameters @var{command1} to @var{commandN}
29372 are the commands. If no command is specified, any previously-set
29373 commands are cleared. @xref{Break Commands}. Typical use of this
29374 functionality is tracing a program, that is, printing of values of
29375 some variables whenever breakpoint is hit and then continuing.
29377 @subsubheading @value{GDBN} Command
29379 The corresponding @value{GDBN} command is @samp{commands}.
29381 @subsubheading Example
29386 ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
29387 enabled="y",addr="0x000100d0",func="main",file="hello.c",
29388 fullname="/home/foo/hello.c",line="5",thread-groups=["i1"],
29391 -break-commands 1 "print v" "continue"
29396 @subheading The @code{-break-condition} Command
29397 @findex -break-condition
29399 @subsubheading Synopsis
29402 -break-condition @var{number} @var{expr}
29405 Breakpoint @var{number} will stop the program only if the condition in
29406 @var{expr} is true. The condition becomes part of the
29407 @samp{-break-list} output (see the description of the @samp{-break-list}
29410 @subsubheading @value{GDBN} Command
29412 The corresponding @value{GDBN} command is @samp{condition}.
29414 @subsubheading Example
29418 -break-condition 1 1
29422 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
29423 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
29424 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
29425 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
29426 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
29427 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
29428 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
29429 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
29430 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
29431 line="5",cond="1",thread-groups=["i1"],times="0",ignore="3"@}]@}
29435 @subheading The @code{-break-delete} Command
29436 @findex -break-delete
29438 @subsubheading Synopsis
29441 -break-delete ( @var{breakpoint} )+
29444 Delete the breakpoint(s) whose number(s) are specified in the argument
29445 list. This is obviously reflected in the breakpoint list.
29447 @subsubheading @value{GDBN} Command
29449 The corresponding @value{GDBN} command is @samp{delete}.
29451 @subsubheading Example
29459 ^done,BreakpointTable=@{nr_rows="0",nr_cols="6",
29460 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
29461 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
29462 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
29463 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
29464 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
29465 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
29470 @subheading The @code{-break-disable} Command
29471 @findex -break-disable
29473 @subsubheading Synopsis
29476 -break-disable ( @var{breakpoint} )+
29479 Disable the named @var{breakpoint}(s). The field @samp{enabled} in the
29480 break list is now set to @samp{n} for the named @var{breakpoint}(s).
29482 @subsubheading @value{GDBN} Command
29484 The corresponding @value{GDBN} command is @samp{disable}.
29486 @subsubheading Example
29494 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
29495 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
29496 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
29497 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
29498 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
29499 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
29500 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
29501 body=[bkpt=@{number="2",type="breakpoint",disp="keep",enabled="n",
29502 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
29503 line="5",thread-groups=["i1"],times="0"@}]@}
29507 @subheading The @code{-break-enable} Command
29508 @findex -break-enable
29510 @subsubheading Synopsis
29513 -break-enable ( @var{breakpoint} )+
29516 Enable (previously disabled) @var{breakpoint}(s).
29518 @subsubheading @value{GDBN} Command
29520 The corresponding @value{GDBN} command is @samp{enable}.
29522 @subsubheading Example
29530 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
29531 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
29532 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
29533 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
29534 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
29535 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
29536 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
29537 body=[bkpt=@{number="2",type="breakpoint",disp="keep",enabled="y",
29538 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
29539 line="5",thread-groups=["i1"],times="0"@}]@}
29543 @subheading The @code{-break-info} Command
29544 @findex -break-info
29546 @subsubheading Synopsis
29549 -break-info @var{breakpoint}
29553 Get information about a single breakpoint.
29555 The result is a table of breakpoints. @xref{GDB/MI Breakpoint
29556 Information}, for details on the format of each breakpoint in the
29559 @subsubheading @value{GDBN} Command
29561 The corresponding @value{GDBN} command is @samp{info break @var{breakpoint}}.
29563 @subsubheading Example
29566 @subheading The @code{-break-insert} Command
29567 @findex -break-insert
29568 @anchor{-break-insert}
29570 @subsubheading Synopsis
29573 -break-insert [ -t ] [ -h ] [ -f ] [ -d ] [ -a ]
29574 [ -c @var{condition} ] [ -i @var{ignore-count} ]
29575 [ -p @var{thread-id} ] [ @var{location} ]
29579 If specified, @var{location}, can be one of:
29582 @item linespec location
29583 A linespec location. @xref{Linespec Locations}.
29585 @item explicit location
29586 An explicit location. @sc{gdb/mi} explicit locations are
29587 analogous to the CLI's explicit locations using the option names
29588 listed below. @xref{Explicit Locations}.
29591 @item --source @var{filename}
29592 The source file name of the location. This option requires the use
29593 of either @samp{--function} or @samp{--line}.
29595 @item --function @var{function}
29596 The name of a function or method.
29598 @item --label @var{label}
29599 The name of a label.
29601 @item --line @var{lineoffset}
29602 An absolute or relative line offset from the start of the location.
29605 @item address location
29606 An address location, *@var{address}. @xref{Address Locations}.
29610 The possible optional parameters of this command are:
29614 Insert a temporary breakpoint.
29616 Insert a hardware breakpoint.
29618 If @var{location} cannot be parsed (for example if it
29619 refers to unknown files or functions), create a pending
29620 breakpoint. Without this flag, @value{GDBN} will report
29621 an error, and won't create a breakpoint, if @var{location}
29624 Create a disabled breakpoint.
29626 Create a tracepoint. @xref{Tracepoints}. When this parameter
29627 is used together with @samp{-h}, a fast tracepoint is created.
29628 @item -c @var{condition}
29629 Make the breakpoint conditional on @var{condition}.
29630 @item -i @var{ignore-count}
29631 Initialize the @var{ignore-count}.
29632 @item -p @var{thread-id}
29633 Restrict the breakpoint to the thread with the specified global
29637 @subsubheading Result
29639 @xref{GDB/MI Breakpoint Information}, for details on the format of the
29640 resulting breakpoint.
29642 Note: this format is open to change.
29643 @c An out-of-band breakpoint instead of part of the result?
29645 @subsubheading @value{GDBN} Command
29647 The corresponding @value{GDBN} commands are @samp{break}, @samp{tbreak},
29648 @samp{hbreak}, and @samp{thbreak}. @c and @samp{rbreak}.
29650 @subsubheading Example
29655 ^done,bkpt=@{number="1",addr="0x0001072c",file="recursive2.c",
29656 fullname="/home/foo/recursive2.c,line="4",thread-groups=["i1"],
29659 -break-insert -t foo
29660 ^done,bkpt=@{number="2",addr="0x00010774",file="recursive2.c",
29661 fullname="/home/foo/recursive2.c,line="11",thread-groups=["i1"],
29665 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
29666 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
29667 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
29668 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
29669 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
29670 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
29671 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
29672 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
29673 addr="0x0001072c", func="main",file="recursive2.c",
29674 fullname="/home/foo/recursive2.c,"line="4",thread-groups=["i1"],
29676 bkpt=@{number="2",type="breakpoint",disp="del",enabled="y",
29677 addr="0x00010774",func="foo",file="recursive2.c",
29678 fullname="/home/foo/recursive2.c",line="11",thread-groups=["i1"],
29681 @c -break-insert -r foo.*
29682 @c ~int foo(int, int);
29683 @c ^done,bkpt=@{number="3",addr="0x00010774",file="recursive2.c,
29684 @c "fullname="/home/foo/recursive2.c",line="11",thread-groups=["i1"],
29689 @subheading The @code{-dprintf-insert} Command
29690 @findex -dprintf-insert
29692 @subsubheading Synopsis
29695 -dprintf-insert [ -t ] [ -f ] [ -d ]
29696 [ -c @var{condition} ] [ -i @var{ignore-count} ]
29697 [ -p @var{thread-id} ] [ @var{location} ] [ @var{format} ]
29702 If supplied, @var{location} may be specified the same way as for
29703 the @code{-break-insert} command. @xref{-break-insert}.
29705 The possible optional parameters of this command are:
29709 Insert a temporary breakpoint.
29711 If @var{location} cannot be parsed (for example, if it
29712 refers to unknown files or functions), create a pending
29713 breakpoint. Without this flag, @value{GDBN} will report
29714 an error, and won't create a breakpoint, if @var{location}
29717 Create a disabled breakpoint.
29718 @item -c @var{condition}
29719 Make the breakpoint conditional on @var{condition}.
29720 @item -i @var{ignore-count}
29721 Set the ignore count of the breakpoint (@pxref{Conditions, ignore count})
29722 to @var{ignore-count}.
29723 @item -p @var{thread-id}
29724 Restrict the breakpoint to the thread with the specified global
29728 @subsubheading Result
29730 @xref{GDB/MI Breakpoint Information}, for details on the format of the
29731 resulting breakpoint.
29733 @c An out-of-band breakpoint instead of part of the result?
29735 @subsubheading @value{GDBN} Command
29737 The corresponding @value{GDBN} command is @samp{dprintf}.
29739 @subsubheading Example
29743 4-dprintf-insert foo "At foo entry\n"
29744 4^done,bkpt=@{number="1",type="dprintf",disp="keep",enabled="y",
29745 addr="0x000000000040061b",func="foo",file="mi-dprintf.c",
29746 fullname="mi-dprintf.c",line="25",thread-groups=["i1"],
29747 times="0",script=@{"printf \"At foo entry\\n\"","continue"@},
29748 original-location="foo"@}
29750 5-dprintf-insert 26 "arg=%d, g=%d\n" arg g
29751 5^done,bkpt=@{number="2",type="dprintf",disp="keep",enabled="y",
29752 addr="0x000000000040062a",func="foo",file="mi-dprintf.c",
29753 fullname="mi-dprintf.c",line="26",thread-groups=["i1"],
29754 times="0",script=@{"printf \"arg=%d, g=%d\\n\", arg, g","continue"@},
29755 original-location="mi-dprintf.c:26"@}
29759 @subheading The @code{-break-list} Command
29760 @findex -break-list
29762 @subsubheading Synopsis
29768 Displays the list of inserted breakpoints, showing the following fields:
29772 number of the breakpoint
29774 type of the breakpoint: @samp{breakpoint} or @samp{watchpoint}
29776 should the breakpoint be deleted or disabled when it is hit: @samp{keep}
29779 is the breakpoint enabled or no: @samp{y} or @samp{n}
29781 memory location at which the breakpoint is set
29783 logical location of the breakpoint, expressed by function name, file
29785 @item Thread-groups
29786 list of thread groups to which this breakpoint applies
29788 number of times the breakpoint has been hit
29791 If there are no breakpoints or watchpoints, the @code{BreakpointTable}
29792 @code{body} field is an empty list.
29794 @subsubheading @value{GDBN} Command
29796 The corresponding @value{GDBN} command is @samp{info break}.
29798 @subsubheading Example
29803 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
29804 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
29805 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
29806 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
29807 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
29808 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
29809 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
29810 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
29811 addr="0x000100d0",func="main",file="hello.c",line="5",thread-groups=["i1"],
29813 bkpt=@{number="2",type="breakpoint",disp="keep",enabled="y",
29814 addr="0x00010114",func="foo",file="hello.c",fullname="/home/foo/hello.c",
29815 line="13",thread-groups=["i1"],times="0"@}]@}
29819 Here's an example of the result when there are no breakpoints:
29824 ^done,BreakpointTable=@{nr_rows="0",nr_cols="6",
29825 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
29826 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
29827 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
29828 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
29829 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
29830 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
29835 @subheading The @code{-break-passcount} Command
29836 @findex -break-passcount
29838 @subsubheading Synopsis
29841 -break-passcount @var{tracepoint-number} @var{passcount}
29844 Set the passcount for tracepoint @var{tracepoint-number} to
29845 @var{passcount}. If the breakpoint referred to by @var{tracepoint-number}
29846 is not a tracepoint, error is emitted. This corresponds to CLI
29847 command @samp{passcount}.
29849 @subheading The @code{-break-watch} Command
29850 @findex -break-watch
29852 @subsubheading Synopsis
29855 -break-watch [ -a | -r ]
29858 Create a watchpoint. With the @samp{-a} option it will create an
29859 @dfn{access} watchpoint, i.e., a watchpoint that triggers either on a
29860 read from or on a write to the memory location. With the @samp{-r}
29861 option, the watchpoint created is a @dfn{read} watchpoint, i.e., it will
29862 trigger only when the memory location is accessed for reading. Without
29863 either of the options, the watchpoint created is a regular watchpoint,
29864 i.e., it will trigger when the memory location is accessed for writing.
29865 @xref{Set Watchpoints, , Setting Watchpoints}.
29867 Note that @samp{-break-list} will report a single list of watchpoints and
29868 breakpoints inserted.
29870 @subsubheading @value{GDBN} Command
29872 The corresponding @value{GDBN} commands are @samp{watch}, @samp{awatch}, and
29875 @subsubheading Example
29877 Setting a watchpoint on a variable in the @code{main} function:
29882 ^done,wpt=@{number="2",exp="x"@}
29887 *stopped,reason="watchpoint-trigger",wpt=@{number="2",exp="x"@},
29888 value=@{old="-268439212",new="55"@},
29889 frame=@{func="main",args=[],file="recursive2.c",
29890 fullname="/home/foo/bar/recursive2.c",line="5",arch="i386:x86_64"@}
29894 Setting a watchpoint on a variable local to a function. @value{GDBN} will stop
29895 the program execution twice: first for the variable changing value, then
29896 for the watchpoint going out of scope.
29901 ^done,wpt=@{number="5",exp="C"@}
29906 *stopped,reason="watchpoint-trigger",
29907 wpt=@{number="5",exp="C"@},value=@{old="-276895068",new="3"@},
29908 frame=@{func="callee4",args=[],
29909 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
29910 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="13",
29911 arch="i386:x86_64"@}
29916 *stopped,reason="watchpoint-scope",wpnum="5",
29917 frame=@{func="callee3",args=[@{name="strarg",
29918 value="0x11940 \"A string argument.\""@}],
29919 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
29920 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="18",
29921 arch="i386:x86_64"@}
29925 Listing breakpoints and watchpoints, at different points in the program
29926 execution. Note that once the watchpoint goes out of scope, it is
29932 ^done,wpt=@{number="2",exp="C"@}
29935 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
29936 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
29937 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
29938 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
29939 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
29940 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
29941 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
29942 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
29943 addr="0x00010734",func="callee4",
29944 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
29945 fullname="/home/foo/devo/gdb/testsuite/gdb.mi/basics.c"line="8",thread-groups=["i1"],
29947 bkpt=@{number="2",type="watchpoint",disp="keep",
29948 enabled="y",addr="",what="C",thread-groups=["i1"],times="0"@}]@}
29953 *stopped,reason="watchpoint-trigger",wpt=@{number="2",exp="C"@},
29954 value=@{old="-276895068",new="3"@},
29955 frame=@{func="callee4",args=[],
29956 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
29957 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="13",
29958 arch="i386:x86_64"@}
29961 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
29962 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
29963 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
29964 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
29965 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
29966 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
29967 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
29968 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
29969 addr="0x00010734",func="callee4",
29970 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
29971 fullname="/home/foo/devo/gdb/testsuite/gdb.mi/basics.c",line="8",thread-groups=["i1"],
29973 bkpt=@{number="2",type="watchpoint",disp="keep",
29974 enabled="y",addr="",what="C",thread-groups=["i1"],times="-5"@}]@}
29978 ^done,reason="watchpoint-scope",wpnum="2",
29979 frame=@{func="callee3",args=[@{name="strarg",
29980 value="0x11940 \"A string argument.\""@}],
29981 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
29982 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="18",
29983 arch="i386:x86_64"@}
29986 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
29987 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
29988 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
29989 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
29990 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
29991 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
29992 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
29993 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
29994 addr="0x00010734",func="callee4",
29995 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
29996 fullname="/home/foo/devo/gdb/testsuite/gdb.mi/basics.c",line="8",
29997 thread-groups=["i1"],times="1"@}]@}
30002 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
30003 @node GDB/MI Catchpoint Commands
30004 @section @sc{gdb/mi} Catchpoint Commands
30006 This section documents @sc{gdb/mi} commands for manipulating
30010 * Shared Library GDB/MI Catchpoint Commands::
30011 * Ada Exception GDB/MI Catchpoint Commands::
30012 * C++ Exception GDB/MI Catchpoint Commands::
30015 @node Shared Library GDB/MI Catchpoint Commands
30016 @subsection Shared Library @sc{gdb/mi} Catchpoints
30018 @subheading The @code{-catch-load} Command
30019 @findex -catch-load
30021 @subsubheading Synopsis
30024 -catch-load [ -t ] [ -d ] @var{regexp}
30027 Add a catchpoint for library load events. If the @samp{-t} option is used,
30028 the catchpoint is a temporary one (@pxref{Set Breaks, ,Setting
30029 Breakpoints}). If the @samp{-d} option is used, the catchpoint is created
30030 in a disabled state. The @samp{regexp} argument is a regular
30031 expression used to match the name of the loaded library.
30034 @subsubheading @value{GDBN} Command
30036 The corresponding @value{GDBN} command is @samp{catch load}.
30038 @subsubheading Example
30041 -catch-load -t foo.so
30042 ^done,bkpt=@{number="1",type="catchpoint",disp="del",enabled="y",
30043 what="load of library matching foo.so",catch-type="load",times="0"@}
30048 @subheading The @code{-catch-unload} Command
30049 @findex -catch-unload
30051 @subsubheading Synopsis
30054 -catch-unload [ -t ] [ -d ] @var{regexp}
30057 Add a catchpoint for library unload events. If the @samp{-t} option is
30058 used, the catchpoint is a temporary one (@pxref{Set Breaks, ,Setting
30059 Breakpoints}). If the @samp{-d} option is used, the catchpoint is
30060 created in a disabled state. The @samp{regexp} argument is a regular
30061 expression used to match the name of the unloaded library.
30063 @subsubheading @value{GDBN} Command
30065 The corresponding @value{GDBN} command is @samp{catch unload}.
30067 @subsubheading Example
30070 -catch-unload -d bar.so
30071 ^done,bkpt=@{number="2",type="catchpoint",disp="keep",enabled="n",
30072 what="load of library matching bar.so",catch-type="unload",times="0"@}
30076 @node Ada Exception GDB/MI Catchpoint Commands
30077 @subsection Ada Exception @sc{gdb/mi} Catchpoints
30079 The following @sc{gdb/mi} commands can be used to create catchpoints
30080 that stop the execution when Ada exceptions are being raised.
30082 @subheading The @code{-catch-assert} Command
30083 @findex -catch-assert
30085 @subsubheading Synopsis
30088 -catch-assert [ -c @var{condition}] [ -d ] [ -t ]
30091 Add a catchpoint for failed Ada assertions.
30093 The possible optional parameters for this command are:
30096 @item -c @var{condition}
30097 Make the catchpoint conditional on @var{condition}.
30099 Create a disabled catchpoint.
30101 Create a temporary catchpoint.
30104 @subsubheading @value{GDBN} Command
30106 The corresponding @value{GDBN} command is @samp{catch assert}.
30108 @subsubheading Example
30112 ^done,bkptno="5",bkpt=@{number="5",type="breakpoint",disp="keep",
30113 enabled="y",addr="0x0000000000404888",what="failed Ada assertions",
30114 thread-groups=["i1"],times="0",
30115 original-location="__gnat_debug_raise_assert_failure"@}
30119 @subheading The @code{-catch-exception} Command
30120 @findex -catch-exception
30122 @subsubheading Synopsis
30125 -catch-exception [ -c @var{condition}] [ -d ] [ -e @var{exception-name} ]
30129 Add a catchpoint stopping when Ada exceptions are raised.
30130 By default, the command stops the program when any Ada exception
30131 gets raised. But it is also possible, by using some of the
30132 optional parameters described below, to create more selective
30135 The possible optional parameters for this command are:
30138 @item -c @var{condition}
30139 Make the catchpoint conditional on @var{condition}.
30141 Create a disabled catchpoint.
30142 @item -e @var{exception-name}
30143 Only stop when @var{exception-name} is raised. This option cannot
30144 be used combined with @samp{-u}.
30146 Create a temporary catchpoint.
30148 Stop only when an unhandled exception gets raised. This option
30149 cannot be used combined with @samp{-e}.
30152 @subsubheading @value{GDBN} Command
30154 The corresponding @value{GDBN} commands are @samp{catch exception}
30155 and @samp{catch exception unhandled}.
30157 @subsubheading Example
30160 -catch-exception -e Program_Error
30161 ^done,bkptno="4",bkpt=@{number="4",type="breakpoint",disp="keep",
30162 enabled="y",addr="0x0000000000404874",
30163 what="`Program_Error' Ada exception", thread-groups=["i1"],
30164 times="0",original-location="__gnat_debug_raise_exception"@}
30168 @subheading The @code{-catch-handlers} Command
30169 @findex -catch-handlers
30171 @subsubheading Synopsis
30174 -catch-handlers [ -c @var{condition}] [ -d ] [ -e @var{exception-name} ]
30178 Add a catchpoint stopping when Ada exceptions are handled.
30179 By default, the command stops the program when any Ada exception
30180 gets handled. But it is also possible, by using some of the
30181 optional parameters described below, to create more selective
30184 The possible optional parameters for this command are:
30187 @item -c @var{condition}
30188 Make the catchpoint conditional on @var{condition}.
30190 Create a disabled catchpoint.
30191 @item -e @var{exception-name}
30192 Only stop when @var{exception-name} is handled.
30194 Create a temporary catchpoint.
30197 @subsubheading @value{GDBN} Command
30199 The corresponding @value{GDBN} command is @samp{catch handlers}.
30201 @subsubheading Example
30204 -catch-handlers -e Constraint_Error
30205 ^done,bkptno="4",bkpt=@{number="4",type="breakpoint",disp="keep",
30206 enabled="y",addr="0x0000000000402f68",
30207 what="`Constraint_Error' Ada exception handlers",thread-groups=["i1"],
30208 times="0",original-location="__gnat_begin_handler"@}
30212 @node C++ Exception GDB/MI Catchpoint Commands
30213 @subsection C@t{++} Exception @sc{gdb/mi} Catchpoints
30215 The following @sc{gdb/mi} commands can be used to create catchpoints
30216 that stop the execution when C@t{++} exceptions are being throw, rethrown,
30219 @subheading The @code{-catch-throw} Command
30220 @findex -catch-throw
30222 @subsubheading Synopsis
30225 -catch-throw [ -t ] [ -r @var{regexp}]
30228 Stop when the debuggee throws a C@t{++} exception. If @var{regexp} is
30229 given, then only exceptions whose type matches the regular expression
30232 If @samp{-t} is given, then the catchpoint is enabled only for one
30233 stop, the catchpoint is automatically deleted after stopping once for
30236 @subsubheading @value{GDBN} Command
30238 The corresponding @value{GDBN} commands are @samp{catch throw}
30239 and @samp{tcatch throw} (@pxref{Set Catchpoints}).
30241 @subsubheading Example
30244 -catch-throw -r exception_type
30245 ^done,bkpt=@{number="1",type="catchpoint",disp="keep",enabled="y",
30246 what="exception throw",catch-type="throw",
30247 thread-groups=["i1"],
30248 regexp="exception_type",times="0"@}
30254 ~"Catchpoint 1 (exception thrown), 0x00007ffff7ae00ed
30255 in __cxa_throw () from /lib64/libstdc++.so.6\n"
30256 *stopped,bkptno="1",reason="breakpoint-hit",disp="keep",
30257 frame=@{addr="0x00007ffff7ae00ed",func="__cxa_throw",
30258 args=[],from="/lib64/libstdc++.so.6",arch="i386:x86-64"@},
30259 thread-id="1",stopped-threads="all",core="6"
30263 @subheading The @code{-catch-rethrow} Command
30264 @findex -catch-rethrow
30266 @subsubheading Synopsis
30269 -catch-rethrow [ -t ] [ -r @var{regexp}]
30272 Stop when a C@t{++} exception is re-thrown. If @var{regexp} is given,
30273 then only exceptions whose type matches the regular expression will be
30276 If @samp{-t} is given, then the catchpoint is enabled only for one
30277 stop, the catchpoint is automatically deleted after the first event is
30280 @subsubheading @value{GDBN} Command
30282 The corresponding @value{GDBN} commands are @samp{catch rethrow}
30283 and @samp{tcatch rethrow} (@pxref{Set Catchpoints}).
30285 @subsubheading Example
30288 -catch-rethrow -r exception_type
30289 ^done,bkpt=@{number="1",type="catchpoint",disp="keep",enabled="y",
30290 what="exception rethrow",catch-type="rethrow",
30291 thread-groups=["i1"],
30292 regexp="exception_type",times="0"@}
30298 ~"Catchpoint 1 (exception rethrown), 0x00007ffff7ae00ed
30299 in __cxa_rethrow () from /lib64/libstdc++.so.6\n"
30300 *stopped,bkptno="1",reason="breakpoint-hit",disp="keep",
30301 frame=@{addr="0x00007ffff7ae00ed",func="__cxa_rethrow",
30302 args=[],from="/lib64/libstdc++.so.6",arch="i386:x86-64"@},
30303 thread-id="1",stopped-threads="all",core="6"
30307 @subheading The @code{-catch-catch} Command
30308 @findex -catch-catch
30310 @subsubheading Synopsis
30313 -catch-catch [ -t ] [ -r @var{regexp}]
30316 Stop when the debuggee catches a C@t{++} exception. If @var{regexp}
30317 is given, then only exceptions whose type matches the regular
30318 expression will be caught.
30320 If @samp{-t} is given, then the catchpoint is enabled only for one
30321 stop, the catchpoint is automatically deleted after the first event is
30324 @subsubheading @value{GDBN} Command
30326 The corresponding @value{GDBN} commands are @samp{catch catch}
30327 and @samp{tcatch catch} (@pxref{Set Catchpoints}).
30329 @subsubheading Example
30332 -catch-catch -r exception_type
30333 ^done,bkpt=@{number="1",type="catchpoint",disp="keep",enabled="y",
30334 what="exception catch",catch-type="catch",
30335 thread-groups=["i1"],
30336 regexp="exception_type",times="0"@}
30342 ~"Catchpoint 1 (exception caught), 0x00007ffff7ae00ed
30343 in __cxa_begin_catch () from /lib64/libstdc++.so.6\n"
30344 *stopped,bkptno="1",reason="breakpoint-hit",disp="keep",
30345 frame=@{addr="0x00007ffff7ae00ed",func="__cxa_begin_catch",
30346 args=[],from="/lib64/libstdc++.so.6",arch="i386:x86-64"@},
30347 thread-id="1",stopped-threads="all",core="6"
30351 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
30352 @node GDB/MI Program Context
30353 @section @sc{gdb/mi} Program Context
30355 @subheading The @code{-exec-arguments} Command
30356 @findex -exec-arguments
30359 @subsubheading Synopsis
30362 -exec-arguments @var{args}
30365 Set the inferior program arguments, to be used in the next
30368 @subsubheading @value{GDBN} Command
30370 The corresponding @value{GDBN} command is @samp{set args}.
30372 @subsubheading Example
30376 -exec-arguments -v word
30383 @subheading The @code{-exec-show-arguments} Command
30384 @findex -exec-show-arguments
30386 @subsubheading Synopsis
30389 -exec-show-arguments
30392 Print the arguments of the program.
30394 @subsubheading @value{GDBN} Command
30396 The corresponding @value{GDBN} command is @samp{show args}.
30398 @subsubheading Example
30403 @subheading The @code{-environment-cd} Command
30404 @findex -environment-cd
30406 @subsubheading Synopsis
30409 -environment-cd @var{pathdir}
30412 Set @value{GDBN}'s working directory.
30414 @subsubheading @value{GDBN} Command
30416 The corresponding @value{GDBN} command is @samp{cd}.
30418 @subsubheading Example
30422 -environment-cd /kwikemart/marge/ezannoni/flathead-dev/devo/gdb
30428 @subheading The @code{-environment-directory} Command
30429 @findex -environment-directory
30431 @subsubheading Synopsis
30434 -environment-directory [ -r ] [ @var{pathdir} ]+
30437 Add directories @var{pathdir} to beginning of search path for source files.
30438 If the @samp{-r} option is used, the search path is reset to the default
30439 search path. If directories @var{pathdir} are supplied in addition to the
30440 @samp{-r} option, the search path is first reset and then addition
30442 Multiple directories may be specified, separated by blanks. Specifying
30443 multiple directories in a single command
30444 results in the directories added to the beginning of the
30445 search path in the same order they were presented in the command.
30446 If blanks are needed as
30447 part of a directory name, double-quotes should be used around
30448 the name. In the command output, the path will show up separated
30449 by the system directory-separator character. The directory-separator
30450 character must not be used
30451 in any directory name.
30452 If no directories are specified, the current search path is displayed.
30454 @subsubheading @value{GDBN} Command
30456 The corresponding @value{GDBN} command is @samp{dir}.
30458 @subsubheading Example
30462 -environment-directory /kwikemart/marge/ezannoni/flathead-dev/devo/gdb
30463 ^done,source-path="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb:$cdir:$cwd"
30465 -environment-directory ""
30466 ^done,source-path="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb:$cdir:$cwd"
30468 -environment-directory -r /home/jjohnstn/src/gdb /usr/src
30469 ^done,source-path="/home/jjohnstn/src/gdb:/usr/src:$cdir:$cwd"
30471 -environment-directory -r
30472 ^done,source-path="$cdir:$cwd"
30477 @subheading The @code{-environment-path} Command
30478 @findex -environment-path
30480 @subsubheading Synopsis
30483 -environment-path [ -r ] [ @var{pathdir} ]+
30486 Add directories @var{pathdir} to beginning of search path for object files.
30487 If the @samp{-r} option is used, the search path is reset to the original
30488 search path that existed at gdb start-up. If directories @var{pathdir} are
30489 supplied in addition to the
30490 @samp{-r} option, the search path is first reset and then addition
30492 Multiple directories may be specified, separated by blanks. Specifying
30493 multiple directories in a single command
30494 results in the directories added to the beginning of the
30495 search path in the same order they were presented in the command.
30496 If blanks are needed as
30497 part of a directory name, double-quotes should be used around
30498 the name. In the command output, the path will show up separated
30499 by the system directory-separator character. The directory-separator
30500 character must not be used
30501 in any directory name.
30502 If no directories are specified, the current path is displayed.
30505 @subsubheading @value{GDBN} Command
30507 The corresponding @value{GDBN} command is @samp{path}.
30509 @subsubheading Example
30514 ^done,path="/usr/bin"
30516 -environment-path /kwikemart/marge/ezannoni/flathead-dev/ppc-eabi/gdb /bin
30517 ^done,path="/kwikemart/marge/ezannoni/flathead-dev/ppc-eabi/gdb:/bin:/usr/bin"
30519 -environment-path -r /usr/local/bin
30520 ^done,path="/usr/local/bin:/usr/bin"
30525 @subheading The @code{-environment-pwd} Command
30526 @findex -environment-pwd
30528 @subsubheading Synopsis
30534 Show the current working directory.
30536 @subsubheading @value{GDBN} Command
30538 The corresponding @value{GDBN} command is @samp{pwd}.
30540 @subsubheading Example
30545 ^done,cwd="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb"
30549 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
30550 @node GDB/MI Thread Commands
30551 @section @sc{gdb/mi} Thread Commands
30554 @subheading The @code{-thread-info} Command
30555 @findex -thread-info
30557 @subsubheading Synopsis
30560 -thread-info [ @var{thread-id} ]
30563 Reports information about either a specific thread, if the
30564 @var{thread-id} parameter is present, or about all threads.
30565 @var{thread-id} is the thread's global thread ID. When printing
30566 information about all threads, also reports the global ID of the
30569 @subsubheading @value{GDBN} Command
30571 The @samp{info thread} command prints the same information
30574 @subsubheading Result
30576 The result contains the following attributes:
30580 A list of threads. The format of the elements of the list is described in
30581 @ref{GDB/MI Thread Information}.
30583 @item current-thread-id
30584 The global id of the currently selected thread. This field is omitted if there
30585 is no selected thread (for example, when the selected inferior is not running,
30586 and therefore has no threads) or if a @var{thread-id} argument was passed to
30591 @subsubheading Example
30596 @{id="2",target-id="Thread 0xb7e14b90 (LWP 21257)",
30597 frame=@{level="0",addr="0xffffe410",func="__kernel_vsyscall",
30598 args=[]@},state="running"@},
30599 @{id="1",target-id="Thread 0xb7e156b0 (LWP 21254)",
30600 frame=@{level="0",addr="0x0804891f",func="foo",
30601 args=[@{name="i",value="10"@}],
30602 file="/tmp/a.c",fullname="/tmp/a.c",line="158",arch="i386:x86_64"@},
30603 state="running"@}],
30604 current-thread-id="1"
30608 @subheading The @code{-thread-list-ids} Command
30609 @findex -thread-list-ids
30611 @subsubheading Synopsis
30617 Produces a list of the currently known global @value{GDBN} thread ids.
30618 At the end of the list it also prints the total number of such
30621 This command is retained for historical reasons, the
30622 @code{-thread-info} command should be used instead.
30624 @subsubheading @value{GDBN} Command
30626 Part of @samp{info threads} supplies the same information.
30628 @subsubheading Example
30633 ^done,thread-ids=@{thread-id="3",thread-id="2",thread-id="1"@},
30634 current-thread-id="1",number-of-threads="3"
30639 @subheading The @code{-thread-select} Command
30640 @findex -thread-select
30642 @subsubheading Synopsis
30645 -thread-select @var{thread-id}
30648 Make thread with global thread number @var{thread-id} the current
30649 thread. It prints the number of the new current thread, and the
30650 topmost frame for that thread.
30652 This command is deprecated in favor of explicitly using the
30653 @samp{--thread} option to each command.
30655 @subsubheading @value{GDBN} Command
30657 The corresponding @value{GDBN} command is @samp{thread}.
30659 @subsubheading Example
30666 *stopped,reason="end-stepping-range",thread-id="2",line="187",
30667 file="../../../devo/gdb/testsuite/gdb.threads/linux-dp.c"
30671 thread-ids=@{thread-id="3",thread-id="2",thread-id="1"@},
30672 number-of-threads="3"
30675 ^done,new-thread-id="3",
30676 frame=@{level="0",func="vprintf",
30677 args=[@{name="format",value="0x8048e9c \"%*s%c %d %c\\n\""@},
30678 @{name="arg",value="0x2"@}],file="vprintf.c",line="31",arch="i386:x86_64"@}
30682 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
30683 @node GDB/MI Ada Tasking Commands
30684 @section @sc{gdb/mi} Ada Tasking Commands
30686 @subheading The @code{-ada-task-info} Command
30687 @findex -ada-task-info
30689 @subsubheading Synopsis
30692 -ada-task-info [ @var{task-id} ]
30695 Reports information about either a specific Ada task, if the
30696 @var{task-id} parameter is present, or about all Ada tasks.
30698 @subsubheading @value{GDBN} Command
30700 The @samp{info tasks} command prints the same information
30701 about all Ada tasks (@pxref{Ada Tasks}).
30703 @subsubheading Result
30705 The result is a table of Ada tasks. The following columns are
30706 defined for each Ada task:
30710 This field exists only for the current thread. It has the value @samp{*}.
30713 The identifier that @value{GDBN} uses to refer to the Ada task.
30716 The identifier that the target uses to refer to the Ada task.
30719 The global thread identifier of the thread corresponding to the Ada
30722 This field should always exist, as Ada tasks are always implemented
30723 on top of a thread. But if @value{GDBN} cannot find this corresponding
30724 thread for any reason, the field is omitted.
30727 This field exists only when the task was created by another task.
30728 In this case, it provides the ID of the parent task.
30731 The base priority of the task.
30734 The current state of the task. For a detailed description of the
30735 possible states, see @ref{Ada Tasks}.
30738 The name of the task.
30742 @subsubheading Example
30746 ^done,tasks=@{nr_rows="3",nr_cols="8",
30747 hdr=[@{width="1",alignment="-1",col_name="current",colhdr=""@},
30748 @{width="3",alignment="1",col_name="id",colhdr="ID"@},
30749 @{width="9",alignment="1",col_name="task-id",colhdr="TID"@},
30750 @{width="4",alignment="1",col_name="thread-id",colhdr=""@},
30751 @{width="4",alignment="1",col_name="parent-id",colhdr="P-ID"@},
30752 @{width="3",alignment="1",col_name="priority",colhdr="Pri"@},
30753 @{width="22",alignment="-1",col_name="state",colhdr="State"@},
30754 @{width="1",alignment="2",col_name="name",colhdr="Name"@}],
30755 body=[@{current="*",id="1",task-id=" 644010",thread-id="1",priority="48",
30756 state="Child Termination Wait",name="main_task"@}]@}
30760 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
30761 @node GDB/MI Program Execution
30762 @section @sc{gdb/mi} Program Execution
30764 These are the asynchronous commands which generate the out-of-band
30765 record @samp{*stopped}. Currently @value{GDBN} only really executes
30766 asynchronously with remote targets and this interaction is mimicked in
30769 @subheading The @code{-exec-continue} Command
30770 @findex -exec-continue
30772 @subsubheading Synopsis
30775 -exec-continue [--reverse] [--all|--thread-group N]
30778 Resumes the execution of the inferior program, which will continue
30779 to execute until it reaches a debugger stop event. If the
30780 @samp{--reverse} option is specified, execution resumes in reverse until
30781 it reaches a stop event. Stop events may include
30784 breakpoints or watchpoints
30786 signals or exceptions
30788 the end of the process (or its beginning under @samp{--reverse})
30790 the end or beginning of a replay log if one is being used.
30792 In all-stop mode (@pxref{All-Stop
30793 Mode}), may resume only one thread, or all threads, depending on the
30794 value of the @samp{scheduler-locking} variable. If @samp{--all} is
30795 specified, all threads (in all inferiors) will be resumed. The @samp{--all} option is
30796 ignored in all-stop mode. If the @samp{--thread-group} options is
30797 specified, then all threads in that thread group are resumed.
30799 @subsubheading @value{GDBN} Command
30801 The corresponding @value{GDBN} corresponding is @samp{continue}.
30803 @subsubheading Example
30810 *stopped,reason="breakpoint-hit",disp="keep",bkptno="2",frame=@{
30811 func="foo",args=[],file="hello.c",fullname="/home/foo/bar/hello.c",
30812 line="13",arch="i386:x86_64"@}
30817 @subheading The @code{-exec-finish} Command
30818 @findex -exec-finish
30820 @subsubheading Synopsis
30823 -exec-finish [--reverse]
30826 Resumes the execution of the inferior program until the current
30827 function is exited. Displays the results returned by the function.
30828 If the @samp{--reverse} option is specified, resumes the reverse
30829 execution of the inferior program until the point where current
30830 function was called.
30832 @subsubheading @value{GDBN} Command
30834 The corresponding @value{GDBN} command is @samp{finish}.
30836 @subsubheading Example
30838 Function returning @code{void}.
30845 *stopped,reason="function-finished",frame=@{func="main",args=[],
30846 file="hello.c",fullname="/home/foo/bar/hello.c",line="7",arch="i386:x86_64"@}
30850 Function returning other than @code{void}. The name of the internal
30851 @value{GDBN} variable storing the result is printed, together with the
30858 *stopped,reason="function-finished",frame=@{addr="0x000107b0",func="foo",
30859 args=[@{name="a",value="1"],@{name="b",value="9"@}@},
30860 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
30861 arch="i386:x86_64"@},
30862 gdb-result-var="$1",return-value="0"
30867 @subheading The @code{-exec-interrupt} Command
30868 @findex -exec-interrupt
30870 @subsubheading Synopsis
30873 -exec-interrupt [--all|--thread-group N]
30876 Interrupts the background execution of the target. Note how the token
30877 associated with the stop message is the one for the execution command
30878 that has been interrupted. The token for the interrupt itself only
30879 appears in the @samp{^done} output. If the user is trying to
30880 interrupt a non-running program, an error message will be printed.
30882 Note that when asynchronous execution is enabled, this command is
30883 asynchronous just like other execution commands. That is, first the
30884 @samp{^done} response will be printed, and the target stop will be
30885 reported after that using the @samp{*stopped} notification.
30887 In non-stop mode, only the context thread is interrupted by default.
30888 All threads (in all inferiors) will be interrupted if the
30889 @samp{--all} option is specified. If the @samp{--thread-group}
30890 option is specified, all threads in that group will be interrupted.
30892 @subsubheading @value{GDBN} Command
30894 The corresponding @value{GDBN} command is @samp{interrupt}.
30896 @subsubheading Example
30907 111*stopped,signal-name="SIGINT",signal-meaning="Interrupt",
30908 frame=@{addr="0x00010140",func="foo",args=[],file="try.c",
30909 fullname="/home/foo/bar/try.c",line="13",arch="i386:x86_64"@}
30914 ^error,msg="mi_cmd_exec_interrupt: Inferior not executing."
30918 @subheading The @code{-exec-jump} Command
30921 @subsubheading Synopsis
30924 -exec-jump @var{location}
30927 Resumes execution of the inferior program at the location specified by
30928 parameter. @xref{Specify Location}, for a description of the
30929 different forms of @var{location}.
30931 @subsubheading @value{GDBN} Command
30933 The corresponding @value{GDBN} command is @samp{jump}.
30935 @subsubheading Example
30938 -exec-jump foo.c:10
30939 *running,thread-id="all"
30944 @subheading The @code{-exec-next} Command
30947 @subsubheading Synopsis
30950 -exec-next [--reverse]
30953 Resumes execution of the inferior program, stopping when the beginning
30954 of the next source line is reached.
30956 If the @samp{--reverse} option is specified, resumes reverse execution
30957 of the inferior program, stopping at the beginning of the previous
30958 source line. If you issue this command on the first line of a
30959 function, it will take you back to the caller of that function, to the
30960 source line where the function was called.
30963 @subsubheading @value{GDBN} Command
30965 The corresponding @value{GDBN} command is @samp{next}.
30967 @subsubheading Example
30973 *stopped,reason="end-stepping-range",line="8",file="hello.c"
30978 @subheading The @code{-exec-next-instruction} Command
30979 @findex -exec-next-instruction
30981 @subsubheading Synopsis
30984 -exec-next-instruction [--reverse]
30987 Executes one machine instruction. If the instruction is a function
30988 call, continues until the function returns. If the program stops at an
30989 instruction in the middle of a source line, the address will be
30992 If the @samp{--reverse} option is specified, resumes reverse execution
30993 of the inferior program, stopping at the previous instruction. If the
30994 previously executed instruction was a return from another function,
30995 it will continue to execute in reverse until the call to that function
30996 (from the current stack frame) is reached.
30998 @subsubheading @value{GDBN} Command
31000 The corresponding @value{GDBN} command is @samp{nexti}.
31002 @subsubheading Example
31006 -exec-next-instruction
31010 *stopped,reason="end-stepping-range",
31011 addr="0x000100d4",line="5",file="hello.c"
31016 @subheading The @code{-exec-return} Command
31017 @findex -exec-return
31019 @subsubheading Synopsis
31025 Makes current function return immediately. Doesn't execute the inferior.
31026 Displays the new current frame.
31028 @subsubheading @value{GDBN} Command
31030 The corresponding @value{GDBN} command is @samp{return}.
31032 @subsubheading Example
31036 200-break-insert callee4
31037 200^done,bkpt=@{number="1",addr="0x00010734",
31038 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",line="8"@}
31043 000*stopped,reason="breakpoint-hit",disp="keep",bkptno="1",
31044 frame=@{func="callee4",args=[],
31045 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
31046 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="8",
31047 arch="i386:x86_64"@}
31053 111^done,frame=@{level="0",func="callee3",
31054 args=[@{name="strarg",
31055 value="0x11940 \"A string argument.\""@}],
31056 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
31057 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="18",
31058 arch="i386:x86_64"@}
31063 @subheading The @code{-exec-run} Command
31066 @subsubheading Synopsis
31069 -exec-run [ --all | --thread-group N ] [ --start ]
31072 Starts execution of the inferior from the beginning. The inferior
31073 executes until either a breakpoint is encountered or the program
31074 exits. In the latter case the output will include an exit code, if
31075 the program has exited exceptionally.
31077 When neither the @samp{--all} nor the @samp{--thread-group} option
31078 is specified, the current inferior is started. If the
31079 @samp{--thread-group} option is specified, it should refer to a thread
31080 group of type @samp{process}, and that thread group will be started.
31081 If the @samp{--all} option is specified, then all inferiors will be started.
31083 Using the @samp{--start} option instructs the debugger to stop
31084 the execution at the start of the inferior's main subprogram,
31085 following the same behavior as the @code{start} command
31086 (@pxref{Starting}).
31088 @subsubheading @value{GDBN} Command
31090 The corresponding @value{GDBN} command is @samp{run}.
31092 @subsubheading Examples
31097 ^done,bkpt=@{number="1",addr="0x0001072c",file="recursive2.c",line="4"@}
31102 *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",
31103 frame=@{func="main",args=[],file="recursive2.c",
31104 fullname="/home/foo/bar/recursive2.c",line="4",arch="i386:x86_64"@}
31109 Program exited normally:
31117 *stopped,reason="exited-normally"
31122 Program exited exceptionally:
31130 *stopped,reason="exited",exit-code="01"
31134 Another way the program can terminate is if it receives a signal such as
31135 @code{SIGINT}. In this case, @sc{gdb/mi} displays this:
31139 *stopped,reason="exited-signalled",signal-name="SIGINT",
31140 signal-meaning="Interrupt"
31144 @c @subheading -exec-signal
31147 @subheading The @code{-exec-step} Command
31150 @subsubheading Synopsis
31153 -exec-step [--reverse]
31156 Resumes execution of the inferior program, stopping when the beginning
31157 of the next source line is reached, if the next source line is not a
31158 function call. If it is, stop at the first instruction of the called
31159 function. If the @samp{--reverse} option is specified, resumes reverse
31160 execution of the inferior program, stopping at the beginning of the
31161 previously executed source line.
31163 @subsubheading @value{GDBN} Command
31165 The corresponding @value{GDBN} command is @samp{step}.
31167 @subsubheading Example
31169 Stepping into a function:
31175 *stopped,reason="end-stepping-range",
31176 frame=@{func="foo",args=[@{name="a",value="10"@},
31177 @{name="b",value="0"@}],file="recursive2.c",
31178 fullname="/home/foo/bar/recursive2.c",line="11",arch="i386:x86_64"@}
31188 *stopped,reason="end-stepping-range",line="14",file="recursive2.c"
31193 @subheading The @code{-exec-step-instruction} Command
31194 @findex -exec-step-instruction
31196 @subsubheading Synopsis
31199 -exec-step-instruction [--reverse]
31202 Resumes the inferior which executes one machine instruction. If the
31203 @samp{--reverse} option is specified, resumes reverse execution of the
31204 inferior program, stopping at the previously executed instruction.
31205 The output, once @value{GDBN} has stopped, will vary depending on
31206 whether we have stopped in the middle of a source line or not. In the
31207 former case, the address at which the program stopped will be printed
31210 @subsubheading @value{GDBN} Command
31212 The corresponding @value{GDBN} command is @samp{stepi}.
31214 @subsubheading Example
31218 -exec-step-instruction
31222 *stopped,reason="end-stepping-range",
31223 frame=@{func="foo",args=[],file="try.c",
31224 fullname="/home/foo/bar/try.c",line="10",arch="i386:x86_64"@}
31226 -exec-step-instruction
31230 *stopped,reason="end-stepping-range",
31231 frame=@{addr="0x000100f4",func="foo",args=[],file="try.c",
31232 fullname="/home/foo/bar/try.c",line="10",arch="i386:x86_64"@}
31237 @subheading The @code{-exec-until} Command
31238 @findex -exec-until
31240 @subsubheading Synopsis
31243 -exec-until [ @var{location} ]
31246 Executes the inferior until the @var{location} specified in the
31247 argument is reached. If there is no argument, the inferior executes
31248 until a source line greater than the current one is reached. The
31249 reason for stopping in this case will be @samp{location-reached}.
31251 @subsubheading @value{GDBN} Command
31253 The corresponding @value{GDBN} command is @samp{until}.
31255 @subsubheading Example
31259 -exec-until recursive2.c:6
31263 *stopped,reason="location-reached",frame=@{func="main",args=[],
31264 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="6",
31265 arch="i386:x86_64"@}
31270 @subheading -file-clear
31271 Is this going away????
31274 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
31275 @node GDB/MI Stack Manipulation
31276 @section @sc{gdb/mi} Stack Manipulation Commands
31278 @subheading The @code{-enable-frame-filters} Command
31279 @findex -enable-frame-filters
31282 -enable-frame-filters
31285 @value{GDBN} allows Python-based frame filters to affect the output of
31286 the MI commands relating to stack traces. As there is no way to
31287 implement this in a fully backward-compatible way, a front end must
31288 request that this functionality be enabled.
31290 Once enabled, this feature cannot be disabled.
31292 Note that if Python support has not been compiled into @value{GDBN},
31293 this command will still succeed (and do nothing).
31295 @subheading The @code{-stack-info-frame} Command
31296 @findex -stack-info-frame
31298 @subsubheading Synopsis
31304 Get info on the selected frame.
31306 @subsubheading @value{GDBN} Command
31308 The corresponding @value{GDBN} command is @samp{info frame} or @samp{frame}
31309 (without arguments).
31311 @subsubheading Example
31316 ^done,frame=@{level="1",addr="0x0001076c",func="callee3",
31317 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
31318 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="17",
31319 arch="i386:x86_64"@}
31323 @subheading The @code{-stack-info-depth} Command
31324 @findex -stack-info-depth
31326 @subsubheading Synopsis
31329 -stack-info-depth [ @var{max-depth} ]
31332 Return the depth of the stack. If the integer argument @var{max-depth}
31333 is specified, do not count beyond @var{max-depth} frames.
31335 @subsubheading @value{GDBN} Command
31337 There's no equivalent @value{GDBN} command.
31339 @subsubheading Example
31341 For a stack with frame levels 0 through 11:
31348 -stack-info-depth 4
31351 -stack-info-depth 12
31354 -stack-info-depth 11
31357 -stack-info-depth 13
31362 @anchor{-stack-list-arguments}
31363 @subheading The @code{-stack-list-arguments} Command
31364 @findex -stack-list-arguments
31366 @subsubheading Synopsis
31369 -stack-list-arguments [ --no-frame-filters ] [ --skip-unavailable ] @var{print-values}
31370 [ @var{low-frame} @var{high-frame} ]
31373 Display a list of the arguments for the frames between @var{low-frame}
31374 and @var{high-frame} (inclusive). If @var{low-frame} and
31375 @var{high-frame} are not provided, list the arguments for the whole
31376 call stack. If the two arguments are equal, show the single frame
31377 at the corresponding level. It is an error if @var{low-frame} is
31378 larger than the actual number of frames. On the other hand,
31379 @var{high-frame} may be larger than the actual number of frames, in
31380 which case only existing frames will be returned.
31382 If @var{print-values} is 0 or @code{--no-values}, print only the names of
31383 the variables; if it is 1 or @code{--all-values}, print also their
31384 values; and if it is 2 or @code{--simple-values}, print the name,
31385 type and value for simple data types, and the name and type for arrays,
31386 structures and unions. If the option @code{--no-frame-filters} is
31387 supplied, then Python frame filters will not be executed.
31389 If the @code{--skip-unavailable} option is specified, arguments that
31390 are not available are not listed. Partially available arguments
31391 are still displayed, however.
31393 Use of this command to obtain arguments in a single frame is
31394 deprecated in favor of the @samp{-stack-list-variables} command.
31396 @subsubheading @value{GDBN} Command
31398 @value{GDBN} does not have an equivalent command. @code{gdbtk} has a
31399 @samp{gdb_get_args} command which partially overlaps with the
31400 functionality of @samp{-stack-list-arguments}.
31402 @subsubheading Example
31409 frame=@{level="0",addr="0x00010734",func="callee4",
31410 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
31411 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="8",
31412 arch="i386:x86_64"@},
31413 frame=@{level="1",addr="0x0001076c",func="callee3",
31414 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
31415 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="17",
31416 arch="i386:x86_64"@},
31417 frame=@{level="2",addr="0x0001078c",func="callee2",
31418 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
31419 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="22",
31420 arch="i386:x86_64"@},
31421 frame=@{level="3",addr="0x000107b4",func="callee1",
31422 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
31423 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="27",
31424 arch="i386:x86_64"@},
31425 frame=@{level="4",addr="0x000107e0",func="main",
31426 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
31427 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="32",
31428 arch="i386:x86_64"@}]
31430 -stack-list-arguments 0
31433 frame=@{level="0",args=[]@},
31434 frame=@{level="1",args=[name="strarg"]@},
31435 frame=@{level="2",args=[name="intarg",name="strarg"]@},
31436 frame=@{level="3",args=[name="intarg",name="strarg",name="fltarg"]@},
31437 frame=@{level="4",args=[]@}]
31439 -stack-list-arguments 1
31442 frame=@{level="0",args=[]@},
31444 args=[@{name="strarg",value="0x11940 \"A string argument.\""@}]@},
31445 frame=@{level="2",args=[
31446 @{name="intarg",value="2"@},
31447 @{name="strarg",value="0x11940 \"A string argument.\""@}]@},
31448 @{frame=@{level="3",args=[
31449 @{name="intarg",value="2"@},
31450 @{name="strarg",value="0x11940 \"A string argument.\""@},
31451 @{name="fltarg",value="3.5"@}]@},
31452 frame=@{level="4",args=[]@}]
31454 -stack-list-arguments 0 2 2
31455 ^done,stack-args=[frame=@{level="2",args=[name="intarg",name="strarg"]@}]
31457 -stack-list-arguments 1 2 2
31458 ^done,stack-args=[frame=@{level="2",
31459 args=[@{name="intarg",value="2"@},
31460 @{name="strarg",value="0x11940 \"A string argument.\""@}]@}]
31464 @c @subheading -stack-list-exception-handlers
31467 @anchor{-stack-list-frames}
31468 @subheading The @code{-stack-list-frames} Command
31469 @findex -stack-list-frames
31471 @subsubheading Synopsis
31474 -stack-list-frames [ --no-frame-filters @var{low-frame} @var{high-frame} ]
31477 List the frames currently on the stack. For each frame it displays the
31482 The frame number, 0 being the topmost frame, i.e., the innermost function.
31484 The @code{$pc} value for that frame.
31488 File name of the source file where the function lives.
31489 @item @var{fullname}
31490 The full file name of the source file where the function lives.
31492 Line number corresponding to the @code{$pc}.
31494 The shared library where this function is defined. This is only given
31495 if the frame's function is not known.
31497 Frame's architecture.
31500 If invoked without arguments, this command prints a backtrace for the
31501 whole stack. If given two integer arguments, it shows the frames whose
31502 levels are between the two arguments (inclusive). If the two arguments
31503 are equal, it shows the single frame at the corresponding level. It is
31504 an error if @var{low-frame} is larger than the actual number of
31505 frames. On the other hand, @var{high-frame} may be larger than the
31506 actual number of frames, in which case only existing frames will be
31507 returned. If the option @code{--no-frame-filters} is supplied, then
31508 Python frame filters will not be executed.
31510 @subsubheading @value{GDBN} Command
31512 The corresponding @value{GDBN} commands are @samp{backtrace} and @samp{where}.
31514 @subsubheading Example
31516 Full stack backtrace:
31522 [frame=@{level="0",addr="0x0001076c",func="foo",
31523 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="11",
31524 arch="i386:x86_64"@},
31525 frame=@{level="1",addr="0x000107a4",func="foo",
31526 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
31527 arch="i386:x86_64"@},
31528 frame=@{level="2",addr="0x000107a4",func="foo",
31529 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
31530 arch="i386:x86_64"@},
31531 frame=@{level="3",addr="0x000107a4",func="foo",
31532 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
31533 arch="i386:x86_64"@},
31534 frame=@{level="4",addr="0x000107a4",func="foo",
31535 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
31536 arch="i386:x86_64"@},
31537 frame=@{level="5",addr="0x000107a4",func="foo",
31538 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
31539 arch="i386:x86_64"@},
31540 frame=@{level="6",addr="0x000107a4",func="foo",
31541 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
31542 arch="i386:x86_64"@},
31543 frame=@{level="7",addr="0x000107a4",func="foo",
31544 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
31545 arch="i386:x86_64"@},
31546 frame=@{level="8",addr="0x000107a4",func="foo",
31547 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
31548 arch="i386:x86_64"@},
31549 frame=@{level="9",addr="0x000107a4",func="foo",
31550 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
31551 arch="i386:x86_64"@},
31552 frame=@{level="10",addr="0x000107a4",func="foo",
31553 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
31554 arch="i386:x86_64"@},
31555 frame=@{level="11",addr="0x00010738",func="main",
31556 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="4",
31557 arch="i386:x86_64"@}]
31561 Show frames between @var{low_frame} and @var{high_frame}:
31565 -stack-list-frames 3 5
31567 [frame=@{level="3",addr="0x000107a4",func="foo",
31568 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
31569 arch="i386:x86_64"@},
31570 frame=@{level="4",addr="0x000107a4",func="foo",
31571 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
31572 arch="i386:x86_64"@},
31573 frame=@{level="5",addr="0x000107a4",func="foo",
31574 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
31575 arch="i386:x86_64"@}]
31579 Show a single frame:
31583 -stack-list-frames 3 3
31585 [frame=@{level="3",addr="0x000107a4",func="foo",
31586 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
31587 arch="i386:x86_64"@}]
31592 @subheading The @code{-stack-list-locals} Command
31593 @findex -stack-list-locals
31594 @anchor{-stack-list-locals}
31596 @subsubheading Synopsis
31599 -stack-list-locals [ --no-frame-filters ] [ --skip-unavailable ] @var{print-values}
31602 Display the local variable names for the selected frame. If
31603 @var{print-values} is 0 or @code{--no-values}, print only the names of
31604 the variables; if it is 1 or @code{--all-values}, print also their
31605 values; and if it is 2 or @code{--simple-values}, print the name,
31606 type and value for simple data types, and the name and type for arrays,
31607 structures and unions. In this last case, a frontend can immediately
31608 display the value of simple data types and create variable objects for
31609 other data types when the user wishes to explore their values in
31610 more detail. If the option @code{--no-frame-filters} is supplied, then
31611 Python frame filters will not be executed.
31613 If the @code{--skip-unavailable} option is specified, local variables
31614 that are not available are not listed. Partially available local
31615 variables are still displayed, however.
31617 This command is deprecated in favor of the
31618 @samp{-stack-list-variables} command.
31620 @subsubheading @value{GDBN} Command
31622 @samp{info locals} in @value{GDBN}, @samp{gdb_get_locals} in @code{gdbtk}.
31624 @subsubheading Example
31628 -stack-list-locals 0
31629 ^done,locals=[name="A",name="B",name="C"]
31631 -stack-list-locals --all-values
31632 ^done,locals=[@{name="A",value="1"@},@{name="B",value="2"@},
31633 @{name="C",value="@{1, 2, 3@}"@}]
31634 -stack-list-locals --simple-values
31635 ^done,locals=[@{name="A",type="int",value="1"@},
31636 @{name="B",type="int",value="2"@},@{name="C",type="int [3]"@}]
31640 @anchor{-stack-list-variables}
31641 @subheading The @code{-stack-list-variables} Command
31642 @findex -stack-list-variables
31644 @subsubheading Synopsis
31647 -stack-list-variables [ --no-frame-filters ] [ --skip-unavailable ] @var{print-values}
31650 Display the names of local variables and function arguments for the selected frame. If
31651 @var{print-values} is 0 or @code{--no-values}, print only the names of
31652 the variables; if it is 1 or @code{--all-values}, print also their
31653 values; and if it is 2 or @code{--simple-values}, print the name,
31654 type and value for simple data types, and the name and type for arrays,
31655 structures and unions. If the option @code{--no-frame-filters} is
31656 supplied, then Python frame filters will not be executed.
31658 If the @code{--skip-unavailable} option is specified, local variables
31659 and arguments that are not available are not listed. Partially
31660 available arguments and local variables are still displayed, however.
31662 @subsubheading Example
31666 -stack-list-variables --thread 1 --frame 0 --all-values
31667 ^done,variables=[@{name="x",value="11"@},@{name="s",value="@{a = 1, b = 2@}"@}]
31672 @subheading The @code{-stack-select-frame} Command
31673 @findex -stack-select-frame
31675 @subsubheading Synopsis
31678 -stack-select-frame @var{framenum}
31681 Change the selected frame. Select a different frame @var{framenum} on
31684 This command in deprecated in favor of passing the @samp{--frame}
31685 option to every command.
31687 @subsubheading @value{GDBN} Command
31689 The corresponding @value{GDBN} commands are @samp{frame}, @samp{up},
31690 @samp{down}, @samp{select-frame}, @samp{up-silent}, and @samp{down-silent}.
31692 @subsubheading Example
31696 -stack-select-frame 2
31701 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
31702 @node GDB/MI Variable Objects
31703 @section @sc{gdb/mi} Variable Objects
31707 @subheading Motivation for Variable Objects in @sc{gdb/mi}
31709 For the implementation of a variable debugger window (locals, watched
31710 expressions, etc.), we are proposing the adaptation of the existing code
31711 used by @code{Insight}.
31713 The two main reasons for that are:
31717 It has been proven in practice (it is already on its second generation).
31720 It will shorten development time (needless to say how important it is
31724 The original interface was designed to be used by Tcl code, so it was
31725 slightly changed so it could be used through @sc{gdb/mi}. This section
31726 describes the @sc{gdb/mi} operations that will be available and gives some
31727 hints about their use.
31729 @emph{Note}: In addition to the set of operations described here, we
31730 expect the @sc{gui} implementation of a variable window to require, at
31731 least, the following operations:
31734 @item @code{-gdb-show} @code{output-radix}
31735 @item @code{-stack-list-arguments}
31736 @item @code{-stack-list-locals}
31737 @item @code{-stack-select-frame}
31742 @subheading Introduction to Variable Objects
31744 @cindex variable objects in @sc{gdb/mi}
31746 Variable objects are "object-oriented" MI interface for examining and
31747 changing values of expressions. Unlike some other MI interfaces that
31748 work with expressions, variable objects are specifically designed for
31749 simple and efficient presentation in the frontend. A variable object
31750 is identified by string name. When a variable object is created, the
31751 frontend specifies the expression for that variable object. The
31752 expression can be a simple variable, or it can be an arbitrary complex
31753 expression, and can even involve CPU registers. After creating a
31754 variable object, the frontend can invoke other variable object
31755 operations---for example to obtain or change the value of a variable
31756 object, or to change display format.
31758 Variable objects have hierarchical tree structure. Any variable object
31759 that corresponds to a composite type, such as structure in C, has
31760 a number of child variable objects, for example corresponding to each
31761 element of a structure. A child variable object can itself have
31762 children, recursively. Recursion ends when we reach
31763 leaf variable objects, which always have built-in types. Child variable
31764 objects are created only by explicit request, so if a frontend
31765 is not interested in the children of a particular variable object, no
31766 child will be created.
31768 For a leaf variable object it is possible to obtain its value as a
31769 string, or set the value from a string. String value can be also
31770 obtained for a non-leaf variable object, but it's generally a string
31771 that only indicates the type of the object, and does not list its
31772 contents. Assignment to a non-leaf variable object is not allowed.
31774 A frontend does not need to read the values of all variable objects each time
31775 the program stops. Instead, MI provides an update command that lists all
31776 variable objects whose values has changed since the last update
31777 operation. This considerably reduces the amount of data that must
31778 be transferred to the frontend. As noted above, children variable
31779 objects are created on demand, and only leaf variable objects have a
31780 real value. As result, gdb will read target memory only for leaf
31781 variables that frontend has created.
31783 The automatic update is not always desirable. For example, a frontend
31784 might want to keep a value of some expression for future reference,
31785 and never update it. For another example, fetching memory is
31786 relatively slow for embedded targets, so a frontend might want
31787 to disable automatic update for the variables that are either not
31788 visible on the screen, or ``closed''. This is possible using so
31789 called ``frozen variable objects''. Such variable objects are never
31790 implicitly updated.
31792 Variable objects can be either @dfn{fixed} or @dfn{floating}. For the
31793 fixed variable object, the expression is parsed when the variable
31794 object is created, including associating identifiers to specific
31795 variables. The meaning of expression never changes. For a floating
31796 variable object the values of variables whose names appear in the
31797 expressions are re-evaluated every time in the context of the current
31798 frame. Consider this example:
31803 struct work_state state;
31810 If a fixed variable object for the @code{state} variable is created in
31811 this function, and we enter the recursive call, the variable
31812 object will report the value of @code{state} in the top-level
31813 @code{do_work} invocation. On the other hand, a floating variable
31814 object will report the value of @code{state} in the current frame.
31816 If an expression specified when creating a fixed variable object
31817 refers to a local variable, the variable object becomes bound to the
31818 thread and frame in which the variable object is created. When such
31819 variable object is updated, @value{GDBN} makes sure that the
31820 thread/frame combination the variable object is bound to still exists,
31821 and re-evaluates the variable object in context of that thread/frame.
31823 The following is the complete set of @sc{gdb/mi} operations defined to
31824 access this functionality:
31826 @multitable @columnfractions .4 .6
31827 @item @strong{Operation}
31828 @tab @strong{Description}
31830 @item @code{-enable-pretty-printing}
31831 @tab enable Python-based pretty-printing
31832 @item @code{-var-create}
31833 @tab create a variable object
31834 @item @code{-var-delete}
31835 @tab delete the variable object and/or its children
31836 @item @code{-var-set-format}
31837 @tab set the display format of this variable
31838 @item @code{-var-show-format}
31839 @tab show the display format of this variable
31840 @item @code{-var-info-num-children}
31841 @tab tells how many children this object has
31842 @item @code{-var-list-children}
31843 @tab return a list of the object's children
31844 @item @code{-var-info-type}
31845 @tab show the type of this variable object
31846 @item @code{-var-info-expression}
31847 @tab print parent-relative expression that this variable object represents
31848 @item @code{-var-info-path-expression}
31849 @tab print full expression that this variable object represents
31850 @item @code{-var-show-attributes}
31851 @tab is this variable editable? does it exist here?
31852 @item @code{-var-evaluate-expression}
31853 @tab get the value of this variable
31854 @item @code{-var-assign}
31855 @tab set the value of this variable
31856 @item @code{-var-update}
31857 @tab update the variable and its children
31858 @item @code{-var-set-frozen}
31859 @tab set frozeness attribute
31860 @item @code{-var-set-update-range}
31861 @tab set range of children to display on update
31864 In the next subsection we describe each operation in detail and suggest
31865 how it can be used.
31867 @subheading Description And Use of Operations on Variable Objects
31869 @subheading The @code{-enable-pretty-printing} Command
31870 @findex -enable-pretty-printing
31873 -enable-pretty-printing
31876 @value{GDBN} allows Python-based visualizers to affect the output of the
31877 MI variable object commands. However, because there was no way to
31878 implement this in a fully backward-compatible way, a front end must
31879 request that this functionality be enabled.
31881 Once enabled, this feature cannot be disabled.
31883 Note that if Python support has not been compiled into @value{GDBN},
31884 this command will still succeed (and do nothing).
31886 This feature is currently (as of @value{GDBN} 7.0) experimental, and
31887 may work differently in future versions of @value{GDBN}.
31889 @subheading The @code{-var-create} Command
31890 @findex -var-create
31892 @subsubheading Synopsis
31895 -var-create @{@var{name} | "-"@}
31896 @{@var{frame-addr} | "*" | "@@"@} @var{expression}
31899 This operation creates a variable object, which allows the monitoring of
31900 a variable, the result of an expression, a memory cell or a CPU
31903 The @var{name} parameter is the string by which the object can be
31904 referenced. It must be unique. If @samp{-} is specified, the varobj
31905 system will generate a string ``varNNNNNN'' automatically. It will be
31906 unique provided that one does not specify @var{name} of that format.
31907 The command fails if a duplicate name is found.
31909 The frame under which the expression should be evaluated can be
31910 specified by @var{frame-addr}. A @samp{*} indicates that the current
31911 frame should be used. A @samp{@@} indicates that a floating variable
31912 object must be created.
31914 @var{expression} is any expression valid on the current language set (must not
31915 begin with a @samp{*}), or one of the following:
31919 @samp{*@var{addr}}, where @var{addr} is the address of a memory cell
31922 @samp{*@var{addr}-@var{addr}} --- a memory address range (TBD)
31925 @samp{$@var{regname}} --- a CPU register name
31928 @cindex dynamic varobj
31929 A varobj's contents may be provided by a Python-based pretty-printer. In this
31930 case the varobj is known as a @dfn{dynamic varobj}. Dynamic varobjs
31931 have slightly different semantics in some cases. If the
31932 @code{-enable-pretty-printing} command is not sent, then @value{GDBN}
31933 will never create a dynamic varobj. This ensures backward
31934 compatibility for existing clients.
31936 @subsubheading Result
31938 This operation returns attributes of the newly-created varobj. These
31943 The name of the varobj.
31946 The number of children of the varobj. This number is not necessarily
31947 reliable for a dynamic varobj. Instead, you must examine the
31948 @samp{has_more} attribute.
31951 The varobj's scalar value. For a varobj whose type is some sort of
31952 aggregate (e.g., a @code{struct}), or for a dynamic varobj, this value
31953 will not be interesting.
31956 The varobj's type. This is a string representation of the type, as
31957 would be printed by the @value{GDBN} CLI. If @samp{print object}
31958 (@pxref{Print Settings, set print object}) is set to @code{on}, the
31959 @emph{actual} (derived) type of the object is shown rather than the
31960 @emph{declared} one.
31963 If a variable object is bound to a specific thread, then this is the
31964 thread's global identifier.
31967 For a dynamic varobj, this indicates whether there appear to be any
31968 children available. For a non-dynamic varobj, this will be 0.
31971 This attribute will be present and have the value @samp{1} if the
31972 varobj is a dynamic varobj. If the varobj is not a dynamic varobj,
31973 then this attribute will not be present.
31976 A dynamic varobj can supply a display hint to the front end. The
31977 value comes directly from the Python pretty-printer object's
31978 @code{display_hint} method. @xref{Pretty Printing API}.
31981 Typical output will look like this:
31984 name="@var{name}",numchild="@var{N}",type="@var{type}",thread-id="@var{M}",
31985 has_more="@var{has_more}"
31989 @subheading The @code{-var-delete} Command
31990 @findex -var-delete
31992 @subsubheading Synopsis
31995 -var-delete [ -c ] @var{name}
31998 Deletes a previously created variable object and all of its children.
31999 With the @samp{-c} option, just deletes the children.
32001 Returns an error if the object @var{name} is not found.
32004 @subheading The @code{-var-set-format} Command
32005 @findex -var-set-format
32007 @subsubheading Synopsis
32010 -var-set-format @var{name} @var{format-spec}
32013 Sets the output format for the value of the object @var{name} to be
32016 @anchor{-var-set-format}
32017 The syntax for the @var{format-spec} is as follows:
32020 @var{format-spec} @expansion{}
32021 @{binary | decimal | hexadecimal | octal | natural | zero-hexadecimal@}
32024 The natural format is the default format choosen automatically
32025 based on the variable type (like decimal for an @code{int}, hex
32026 for pointers, etc.).
32028 The zero-hexadecimal format has a representation similar to hexadecimal
32029 but with padding zeroes to the left of the value. For example, a 32-bit
32030 hexadecimal value of 0x1234 would be represented as 0x00001234 in the
32031 zero-hexadecimal format.
32033 For a variable with children, the format is set only on the
32034 variable itself, and the children are not affected.
32036 @subheading The @code{-var-show-format} Command
32037 @findex -var-show-format
32039 @subsubheading Synopsis
32042 -var-show-format @var{name}
32045 Returns the format used to display the value of the object @var{name}.
32048 @var{format} @expansion{}
32053 @subheading The @code{-var-info-num-children} Command
32054 @findex -var-info-num-children
32056 @subsubheading Synopsis
32059 -var-info-num-children @var{name}
32062 Returns the number of children of a variable object @var{name}:
32068 Note that this number is not completely reliable for a dynamic varobj.
32069 It will return the current number of children, but more children may
32073 @subheading The @code{-var-list-children} Command
32074 @findex -var-list-children
32076 @subsubheading Synopsis
32079 -var-list-children [@var{print-values}] @var{name} [@var{from} @var{to}]
32081 @anchor{-var-list-children}
32083 Return a list of the children of the specified variable object and
32084 create variable objects for them, if they do not already exist. With
32085 a single argument or if @var{print-values} has a value of 0 or
32086 @code{--no-values}, print only the names of the variables; if
32087 @var{print-values} is 1 or @code{--all-values}, also print their
32088 values; and if it is 2 or @code{--simple-values} print the name and
32089 value for simple data types and just the name for arrays, structures
32092 @var{from} and @var{to}, if specified, indicate the range of children
32093 to report. If @var{from} or @var{to} is less than zero, the range is
32094 reset and all children will be reported. Otherwise, children starting
32095 at @var{from} (zero-based) and up to and excluding @var{to} will be
32098 If a child range is requested, it will only affect the current call to
32099 @code{-var-list-children}, but not future calls to @code{-var-update}.
32100 For this, you must instead use @code{-var-set-update-range}. The
32101 intent of this approach is to enable a front end to implement any
32102 update approach it likes; for example, scrolling a view may cause the
32103 front end to request more children with @code{-var-list-children}, and
32104 then the front end could call @code{-var-set-update-range} with a
32105 different range to ensure that future updates are restricted to just
32108 For each child the following results are returned:
32113 Name of the variable object created for this child.
32116 The expression to be shown to the user by the front end to designate this child.
32117 For example this may be the name of a structure member.
32119 For a dynamic varobj, this value cannot be used to form an
32120 expression. There is no way to do this at all with a dynamic varobj.
32122 For C/C@t{++} structures there are several pseudo children returned to
32123 designate access qualifiers. For these pseudo children @var{exp} is
32124 @samp{public}, @samp{private}, or @samp{protected}. In this case the
32125 type and value are not present.
32127 A dynamic varobj will not report the access qualifying
32128 pseudo-children, regardless of the language. This information is not
32129 available at all with a dynamic varobj.
32132 Number of children this child has. For a dynamic varobj, this will be
32136 The type of the child. If @samp{print object}
32137 (@pxref{Print Settings, set print object}) is set to @code{on}, the
32138 @emph{actual} (derived) type of the object is shown rather than the
32139 @emph{declared} one.
32142 If values were requested, this is the value.
32145 If this variable object is associated with a thread, this is the
32146 thread's global thread id. Otherwise this result is not present.
32149 If the variable object is frozen, this variable will be present with a value of 1.
32152 A dynamic varobj can supply a display hint to the front end. The
32153 value comes directly from the Python pretty-printer object's
32154 @code{display_hint} method. @xref{Pretty Printing API}.
32157 This attribute will be present and have the value @samp{1} if the
32158 varobj is a dynamic varobj. If the varobj is not a dynamic varobj,
32159 then this attribute will not be present.
32163 The result may have its own attributes:
32167 A dynamic varobj can supply a display hint to the front end. The
32168 value comes directly from the Python pretty-printer object's
32169 @code{display_hint} method. @xref{Pretty Printing API}.
32172 This is an integer attribute which is nonzero if there are children
32173 remaining after the end of the selected range.
32176 @subsubheading Example
32180 -var-list-children n
32181 ^done,numchild=@var{n},children=[child=@{name=@var{name},exp=@var{exp},
32182 numchild=@var{n},type=@var{type}@},@r{(repeats N times)}]
32184 -var-list-children --all-values n
32185 ^done,numchild=@var{n},children=[child=@{name=@var{name},exp=@var{exp},
32186 numchild=@var{n},value=@var{value},type=@var{type}@},@r{(repeats N times)}]
32190 @subheading The @code{-var-info-type} Command
32191 @findex -var-info-type
32193 @subsubheading Synopsis
32196 -var-info-type @var{name}
32199 Returns the type of the specified variable @var{name}. The type is
32200 returned as a string in the same format as it is output by the
32204 type=@var{typename}
32208 @subheading The @code{-var-info-expression} Command
32209 @findex -var-info-expression
32211 @subsubheading Synopsis
32214 -var-info-expression @var{name}
32217 Returns a string that is suitable for presenting this
32218 variable object in user interface. The string is generally
32219 not valid expression in the current language, and cannot be evaluated.
32221 For example, if @code{a} is an array, and variable object
32222 @code{A} was created for @code{a}, then we'll get this output:
32225 (gdb) -var-info-expression A.1
32226 ^done,lang="C",exp="1"
32230 Here, the value of @code{lang} is the language name, which can be
32231 found in @ref{Supported Languages}.
32233 Note that the output of the @code{-var-list-children} command also
32234 includes those expressions, so the @code{-var-info-expression} command
32237 @subheading The @code{-var-info-path-expression} Command
32238 @findex -var-info-path-expression
32240 @subsubheading Synopsis
32243 -var-info-path-expression @var{name}
32246 Returns an expression that can be evaluated in the current
32247 context and will yield the same value that a variable object has.
32248 Compare this with the @code{-var-info-expression} command, which
32249 result can be used only for UI presentation. Typical use of
32250 the @code{-var-info-path-expression} command is creating a
32251 watchpoint from a variable object.
32253 This command is currently not valid for children of a dynamic varobj,
32254 and will give an error when invoked on one.
32256 For example, suppose @code{C} is a C@t{++} class, derived from class
32257 @code{Base}, and that the @code{Base} class has a member called
32258 @code{m_size}. Assume a variable @code{c} is has the type of
32259 @code{C} and a variable object @code{C} was created for variable
32260 @code{c}. Then, we'll get this output:
32262 (gdb) -var-info-path-expression C.Base.public.m_size
32263 ^done,path_expr=((Base)c).m_size)
32266 @subheading The @code{-var-show-attributes} Command
32267 @findex -var-show-attributes
32269 @subsubheading Synopsis
32272 -var-show-attributes @var{name}
32275 List attributes of the specified variable object @var{name}:
32278 status=@var{attr} [ ( ,@var{attr} )* ]
32282 where @var{attr} is @code{@{ @{ editable | noneditable @} | TBD @}}.
32284 @subheading The @code{-var-evaluate-expression} Command
32285 @findex -var-evaluate-expression
32287 @subsubheading Synopsis
32290 -var-evaluate-expression [-f @var{format-spec}] @var{name}
32293 Evaluates the expression that is represented by the specified variable
32294 object and returns its value as a string. The format of the string
32295 can be specified with the @samp{-f} option. The possible values of
32296 this option are the same as for @code{-var-set-format}
32297 (@pxref{-var-set-format}). If the @samp{-f} option is not specified,
32298 the current display format will be used. The current display format
32299 can be changed using the @code{-var-set-format} command.
32305 Note that one must invoke @code{-var-list-children} for a variable
32306 before the value of a child variable can be evaluated.
32308 @subheading The @code{-var-assign} Command
32309 @findex -var-assign
32311 @subsubheading Synopsis
32314 -var-assign @var{name} @var{expression}
32317 Assigns the value of @var{expression} to the variable object specified
32318 by @var{name}. The object must be @samp{editable}. If the variable's
32319 value is altered by the assign, the variable will show up in any
32320 subsequent @code{-var-update} list.
32322 @subsubheading Example
32330 ^done,changelist=[@{name="var1",in_scope="true",type_changed="false"@}]
32334 @subheading The @code{-var-update} Command
32335 @findex -var-update
32337 @subsubheading Synopsis
32340 -var-update [@var{print-values}] @{@var{name} | "*"@}
32343 Reevaluate the expressions corresponding to the variable object
32344 @var{name} and all its direct and indirect children, and return the
32345 list of variable objects whose values have changed; @var{name} must
32346 be a root variable object. Here, ``changed'' means that the result of
32347 @code{-var-evaluate-expression} before and after the
32348 @code{-var-update} is different. If @samp{*} is used as the variable
32349 object names, all existing variable objects are updated, except
32350 for frozen ones (@pxref{-var-set-frozen}). The option
32351 @var{print-values} determines whether both names and values, or just
32352 names are printed. The possible values of this option are the same
32353 as for @code{-var-list-children} (@pxref{-var-list-children}). It is
32354 recommended to use the @samp{--all-values} option, to reduce the
32355 number of MI commands needed on each program stop.
32357 With the @samp{*} parameter, if a variable object is bound to a
32358 currently running thread, it will not be updated, without any
32361 If @code{-var-set-update-range} was previously used on a varobj, then
32362 only the selected range of children will be reported.
32364 @code{-var-update} reports all the changed varobjs in a tuple named
32367 Each item in the change list is itself a tuple holding:
32371 The name of the varobj.
32374 If values were requested for this update, then this field will be
32375 present and will hold the value of the varobj.
32378 @anchor{-var-update}
32379 This field is a string which may take one of three values:
32383 The variable object's current value is valid.
32386 The variable object does not currently hold a valid value but it may
32387 hold one in the future if its associated expression comes back into
32391 The variable object no longer holds a valid value.
32392 This can occur when the executable file being debugged has changed,
32393 either through recompilation or by using the @value{GDBN} @code{file}
32394 command. The front end should normally choose to delete these variable
32398 In the future new values may be added to this list so the front should
32399 be prepared for this possibility. @xref{GDB/MI Development and Front Ends, ,@sc{GDB/MI} Development and Front Ends}.
32402 This is only present if the varobj is still valid. If the type
32403 changed, then this will be the string @samp{true}; otherwise it will
32406 When a varobj's type changes, its children are also likely to have
32407 become incorrect. Therefore, the varobj's children are automatically
32408 deleted when this attribute is @samp{true}. Also, the varobj's update
32409 range, when set using the @code{-var-set-update-range} command, is
32413 If the varobj's type changed, then this field will be present and will
32416 @item new_num_children
32417 For a dynamic varobj, if the number of children changed, or if the
32418 type changed, this will be the new number of children.
32420 The @samp{numchild} field in other varobj responses is generally not
32421 valid for a dynamic varobj -- it will show the number of children that
32422 @value{GDBN} knows about, but because dynamic varobjs lazily
32423 instantiate their children, this will not reflect the number of
32424 children which may be available.
32426 The @samp{new_num_children} attribute only reports changes to the
32427 number of children known by @value{GDBN}. This is the only way to
32428 detect whether an update has removed children (which necessarily can
32429 only happen at the end of the update range).
32432 The display hint, if any.
32435 This is an integer value, which will be 1 if there are more children
32436 available outside the varobj's update range.
32439 This attribute will be present and have the value @samp{1} if the
32440 varobj is a dynamic varobj. If the varobj is not a dynamic varobj,
32441 then this attribute will not be present.
32444 If new children were added to a dynamic varobj within the selected
32445 update range (as set by @code{-var-set-update-range}), then they will
32446 be listed in this attribute.
32449 @subsubheading Example
32456 -var-update --all-values var1
32457 ^done,changelist=[@{name="var1",value="3",in_scope="true",
32458 type_changed="false"@}]
32462 @subheading The @code{-var-set-frozen} Command
32463 @findex -var-set-frozen
32464 @anchor{-var-set-frozen}
32466 @subsubheading Synopsis
32469 -var-set-frozen @var{name} @var{flag}
32472 Set the frozenness flag on the variable object @var{name}. The
32473 @var{flag} parameter should be either @samp{1} to make the variable
32474 frozen or @samp{0} to make it unfrozen. If a variable object is
32475 frozen, then neither itself, nor any of its children, are
32476 implicitly updated by @code{-var-update} of
32477 a parent variable or by @code{-var-update *}. Only
32478 @code{-var-update} of the variable itself will update its value and
32479 values of its children. After a variable object is unfrozen, it is
32480 implicitly updated by all subsequent @code{-var-update} operations.
32481 Unfreezing a variable does not update it, only subsequent
32482 @code{-var-update} does.
32484 @subsubheading Example
32488 -var-set-frozen V 1
32493 @subheading The @code{-var-set-update-range} command
32494 @findex -var-set-update-range
32495 @anchor{-var-set-update-range}
32497 @subsubheading Synopsis
32500 -var-set-update-range @var{name} @var{from} @var{to}
32503 Set the range of children to be returned by future invocations of
32504 @code{-var-update}.
32506 @var{from} and @var{to} indicate the range of children to report. If
32507 @var{from} or @var{to} is less than zero, the range is reset and all
32508 children will be reported. Otherwise, children starting at @var{from}
32509 (zero-based) and up to and excluding @var{to} will be reported.
32511 @subsubheading Example
32515 -var-set-update-range V 1 2
32519 @subheading The @code{-var-set-visualizer} command
32520 @findex -var-set-visualizer
32521 @anchor{-var-set-visualizer}
32523 @subsubheading Synopsis
32526 -var-set-visualizer @var{name} @var{visualizer}
32529 Set a visualizer for the variable object @var{name}.
32531 @var{visualizer} is the visualizer to use. The special value
32532 @samp{None} means to disable any visualizer in use.
32534 If not @samp{None}, @var{visualizer} must be a Python expression.
32535 This expression must evaluate to a callable object which accepts a
32536 single argument. @value{GDBN} will call this object with the value of
32537 the varobj @var{name} as an argument (this is done so that the same
32538 Python pretty-printing code can be used for both the CLI and MI).
32539 When called, this object must return an object which conforms to the
32540 pretty-printing interface (@pxref{Pretty Printing API}).
32542 The pre-defined function @code{gdb.default_visualizer} may be used to
32543 select a visualizer by following the built-in process
32544 (@pxref{Selecting Pretty-Printers}). This is done automatically when
32545 a varobj is created, and so ordinarily is not needed.
32547 This feature is only available if Python support is enabled. The MI
32548 command @code{-list-features} (@pxref{GDB/MI Support Commands})
32549 can be used to check this.
32551 @subsubheading Example
32553 Resetting the visualizer:
32557 -var-set-visualizer V None
32561 Reselecting the default (type-based) visualizer:
32565 -var-set-visualizer V gdb.default_visualizer
32569 Suppose @code{SomeClass} is a visualizer class. A lambda expression
32570 can be used to instantiate this class for a varobj:
32574 -var-set-visualizer V "lambda val: SomeClass()"
32578 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
32579 @node GDB/MI Data Manipulation
32580 @section @sc{gdb/mi} Data Manipulation
32582 @cindex data manipulation, in @sc{gdb/mi}
32583 @cindex @sc{gdb/mi}, data manipulation
32584 This section describes the @sc{gdb/mi} commands that manipulate data:
32585 examine memory and registers, evaluate expressions, etc.
32587 For details about what an addressable memory unit is,
32588 @pxref{addressable memory unit}.
32590 @c REMOVED FROM THE INTERFACE.
32591 @c @subheading -data-assign
32592 @c Change the value of a program variable. Plenty of side effects.
32593 @c @subsubheading GDB Command
32595 @c @subsubheading Example
32598 @subheading The @code{-data-disassemble} Command
32599 @findex -data-disassemble
32601 @subsubheading Synopsis
32605 [ -s @var{start-addr} -e @var{end-addr} ]
32606 | [ -a @var{addr} ]
32607 | [ -f @var{filename} -l @var{linenum} [ -n @var{lines} ] ]
32615 @item @var{start-addr}
32616 is the beginning address (or @code{$pc})
32617 @item @var{end-addr}
32620 is an address anywhere within (or the name of) the function to
32621 disassemble. If an address is specified, the whole function
32622 surrounding that address will be disassembled. If a name is
32623 specified, the whole function with that name will be disassembled.
32624 @item @var{filename}
32625 is the name of the file to disassemble
32626 @item @var{linenum}
32627 is the line number to disassemble around
32629 is the number of disassembly lines to be produced. If it is -1,
32630 the whole function will be disassembled, in case no @var{end-addr} is
32631 specified. If @var{end-addr} is specified as a non-zero value, and
32632 @var{lines} is lower than the number of disassembly lines between
32633 @var{start-addr} and @var{end-addr}, only @var{lines} lines are
32634 displayed; if @var{lines} is higher than the number of lines between
32635 @var{start-addr} and @var{end-addr}, only the lines up to @var{end-addr}
32640 @item 0 disassembly only
32641 @item 1 mixed source and disassembly (deprecated)
32642 @item 2 disassembly with raw opcodes
32643 @item 3 mixed source and disassembly with raw opcodes (deprecated)
32644 @item 4 mixed source and disassembly
32645 @item 5 mixed source and disassembly with raw opcodes
32648 Modes 1 and 3 are deprecated. The output is ``source centric''
32649 which hasn't proved useful in practice.
32650 @xref{Machine Code}, for a discussion of the difference between
32651 @code{/m} and @code{/s} output of the @code{disassemble} command.
32654 @subsubheading Result
32656 The result of the @code{-data-disassemble} command will be a list named
32657 @samp{asm_insns}, the contents of this list depend on the @var{mode}
32658 used with the @code{-data-disassemble} command.
32660 For modes 0 and 2 the @samp{asm_insns} list contains tuples with the
32665 The address at which this instruction was disassembled.
32668 The name of the function this instruction is within.
32671 The decimal offset in bytes from the start of @samp{func-name}.
32674 The text disassembly for this @samp{address}.
32677 This field is only present for modes 2, 3 and 5. This contains the raw opcode
32678 bytes for the @samp{inst} field.
32682 For modes 1, 3, 4 and 5 the @samp{asm_insns} list contains tuples named
32683 @samp{src_and_asm_line}, each of which has the following fields:
32687 The line number within @samp{file}.
32690 The file name from the compilation unit. This might be an absolute
32691 file name or a relative file name depending on the compile command
32695 Absolute file name of @samp{file}. It is converted to a canonical form
32696 using the source file search path
32697 (@pxref{Source Path, ,Specifying Source Directories})
32698 and after resolving all the symbolic links.
32700 If the source file is not found this field will contain the path as
32701 present in the debug information.
32703 @item line_asm_insn
32704 This is a list of tuples containing the disassembly for @samp{line} in
32705 @samp{file}. The fields of each tuple are the same as for
32706 @code{-data-disassemble} in @var{mode} 0 and 2, so @samp{address},
32707 @samp{func-name}, @samp{offset}, @samp{inst}, and optionally
32712 Note that whatever included in the @samp{inst} field, is not
32713 manipulated directly by @sc{gdb/mi}, i.e., it is not possible to
32716 @subsubheading @value{GDBN} Command
32718 The corresponding @value{GDBN} command is @samp{disassemble}.
32720 @subsubheading Example
32722 Disassemble from the current value of @code{$pc} to @code{$pc + 20}:
32726 -data-disassemble -s $pc -e "$pc + 20" -- 0
32729 @{address="0x000107c0",func-name="main",offset="4",
32730 inst="mov 2, %o0"@},
32731 @{address="0x000107c4",func-name="main",offset="8",
32732 inst="sethi %hi(0x11800), %o2"@},
32733 @{address="0x000107c8",func-name="main",offset="12",
32734 inst="or %o2, 0x140, %o1\t! 0x11940 <_lib_version+8>"@},
32735 @{address="0x000107cc",func-name="main",offset="16",
32736 inst="sethi %hi(0x11800), %o2"@},
32737 @{address="0x000107d0",func-name="main",offset="20",
32738 inst="or %o2, 0x168, %o4\t! 0x11968 <_lib_version+48>"@}]
32742 Disassemble the whole @code{main} function. Line 32 is part of
32746 -data-disassemble -f basics.c -l 32 -- 0
32748 @{address="0x000107bc",func-name="main",offset="0",
32749 inst="save %sp, -112, %sp"@},
32750 @{address="0x000107c0",func-name="main",offset="4",
32751 inst="mov 2, %o0"@},
32752 @{address="0x000107c4",func-name="main",offset="8",
32753 inst="sethi %hi(0x11800), %o2"@},
32755 @{address="0x0001081c",func-name="main",offset="96",inst="ret "@},
32756 @{address="0x00010820",func-name="main",offset="100",inst="restore "@}]
32760 Disassemble 3 instructions from the start of @code{main}:
32764 -data-disassemble -f basics.c -l 32 -n 3 -- 0
32766 @{address="0x000107bc",func-name="main",offset="0",
32767 inst="save %sp, -112, %sp"@},
32768 @{address="0x000107c0",func-name="main",offset="4",
32769 inst="mov 2, %o0"@},
32770 @{address="0x000107c4",func-name="main",offset="8",
32771 inst="sethi %hi(0x11800), %o2"@}]
32775 Disassemble 3 instructions from the start of @code{main} in mixed mode:
32779 -data-disassemble -f basics.c -l 32 -n 3 -- 1
32781 src_and_asm_line=@{line="31",
32782 file="../../../src/gdb/testsuite/gdb.mi/basics.c",
32783 fullname="/absolute/path/to/src/gdb/testsuite/gdb.mi/basics.c",
32784 line_asm_insn=[@{address="0x000107bc",
32785 func-name="main",offset="0",inst="save %sp, -112, %sp"@}]@},
32786 src_and_asm_line=@{line="32",
32787 file="../../../src/gdb/testsuite/gdb.mi/basics.c",
32788 fullname="/absolute/path/to/src/gdb/testsuite/gdb.mi/basics.c",
32789 line_asm_insn=[@{address="0x000107c0",
32790 func-name="main",offset="4",inst="mov 2, %o0"@},
32791 @{address="0x000107c4",func-name="main",offset="8",
32792 inst="sethi %hi(0x11800), %o2"@}]@}]
32797 @subheading The @code{-data-evaluate-expression} Command
32798 @findex -data-evaluate-expression
32800 @subsubheading Synopsis
32803 -data-evaluate-expression @var{expr}
32806 Evaluate @var{expr} as an expression. The expression could contain an
32807 inferior function call. The function call will execute synchronously.
32808 If the expression contains spaces, it must be enclosed in double quotes.
32810 @subsubheading @value{GDBN} Command
32812 The corresponding @value{GDBN} commands are @samp{print}, @samp{output}, and
32813 @samp{call}. In @code{gdbtk} only, there's a corresponding
32814 @samp{gdb_eval} command.
32816 @subsubheading Example
32818 In the following example, the numbers that precede the commands are the
32819 @dfn{tokens} described in @ref{GDB/MI Command Syntax, ,@sc{gdb/mi}
32820 Command Syntax}. Notice how @sc{gdb/mi} returns the same tokens in its
32824 211-data-evaluate-expression A
32827 311-data-evaluate-expression &A
32828 311^done,value="0xefffeb7c"
32830 411-data-evaluate-expression A+3
32833 511-data-evaluate-expression "A + 3"
32839 @subheading The @code{-data-list-changed-registers} Command
32840 @findex -data-list-changed-registers
32842 @subsubheading Synopsis
32845 -data-list-changed-registers
32848 Display a list of the registers that have changed.
32850 @subsubheading @value{GDBN} Command
32852 @value{GDBN} doesn't have a direct analog for this command; @code{gdbtk}
32853 has the corresponding command @samp{gdb_changed_register_list}.
32855 @subsubheading Example
32857 On a PPC MBX board:
32865 *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",frame=@{
32866 func="main",args=[],file="try.c",fullname="/home/foo/bar/try.c",
32867 line="5",arch="powerpc"@}
32869 -data-list-changed-registers
32870 ^done,changed-registers=["0","1","2","4","5","6","7","8","9",
32871 "10","11","13","14","15","16","17","18","19","20","21","22","23",
32872 "24","25","26","27","28","30","31","64","65","66","67","69"]
32877 @subheading The @code{-data-list-register-names} Command
32878 @findex -data-list-register-names
32880 @subsubheading Synopsis
32883 -data-list-register-names [ ( @var{regno} )+ ]
32886 Show a list of register names for the current target. If no arguments
32887 are given, it shows a list of the names of all the registers. If
32888 integer numbers are given as arguments, it will print a list of the
32889 names of the registers corresponding to the arguments. To ensure
32890 consistency between a register name and its number, the output list may
32891 include empty register names.
32893 @subsubheading @value{GDBN} Command
32895 @value{GDBN} does not have a command which corresponds to
32896 @samp{-data-list-register-names}. In @code{gdbtk} there is a
32897 corresponding command @samp{gdb_regnames}.
32899 @subsubheading Example
32901 For the PPC MBX board:
32904 -data-list-register-names
32905 ^done,register-names=["r0","r1","r2","r3","r4","r5","r6","r7",
32906 "r8","r9","r10","r11","r12","r13","r14","r15","r16","r17","r18",
32907 "r19","r20","r21","r22","r23","r24","r25","r26","r27","r28","r29",
32908 "r30","r31","f0","f1","f2","f3","f4","f5","f6","f7","f8","f9",
32909 "f10","f11","f12","f13","f14","f15","f16","f17","f18","f19","f20",
32910 "f21","f22","f23","f24","f25","f26","f27","f28","f29","f30","f31",
32911 "", "pc","ps","cr","lr","ctr","xer"]
32913 -data-list-register-names 1 2 3
32914 ^done,register-names=["r1","r2","r3"]
32918 @subheading The @code{-data-list-register-values} Command
32919 @findex -data-list-register-values
32921 @subsubheading Synopsis
32924 -data-list-register-values
32925 [ @code{--skip-unavailable} ] @var{fmt} [ ( @var{regno} )*]
32928 Display the registers' contents. The format according to which the
32929 registers' contents are to be returned is given by @var{fmt}, followed
32930 by an optional list of numbers specifying the registers to display. A
32931 missing list of numbers indicates that the contents of all the
32932 registers must be returned. The @code{--skip-unavailable} option
32933 indicates that only the available registers are to be returned.
32935 Allowed formats for @var{fmt} are:
32952 @subsubheading @value{GDBN} Command
32954 The corresponding @value{GDBN} commands are @samp{info reg}, @samp{info
32955 all-reg}, and (in @code{gdbtk}) @samp{gdb_fetch_registers}.
32957 @subsubheading Example
32959 For a PPC MBX board (note: line breaks are for readability only, they
32960 don't appear in the actual output):
32964 -data-list-register-values r 64 65
32965 ^done,register-values=[@{number="64",value="0xfe00a300"@},
32966 @{number="65",value="0x00029002"@}]
32968 -data-list-register-values x
32969 ^done,register-values=[@{number="0",value="0xfe0043c8"@},
32970 @{number="1",value="0x3fff88"@},@{number="2",value="0xfffffffe"@},
32971 @{number="3",value="0x0"@},@{number="4",value="0xa"@},
32972 @{number="5",value="0x3fff68"@},@{number="6",value="0x3fff58"@},
32973 @{number="7",value="0xfe011e98"@},@{number="8",value="0x2"@},
32974 @{number="9",value="0xfa202820"@},@{number="10",value="0xfa202808"@},
32975 @{number="11",value="0x1"@},@{number="12",value="0x0"@},
32976 @{number="13",value="0x4544"@},@{number="14",value="0xffdfffff"@},
32977 @{number="15",value="0xffffffff"@},@{number="16",value="0xfffffeff"@},
32978 @{number="17",value="0xefffffed"@},@{number="18",value="0xfffffffe"@},
32979 @{number="19",value="0xffffffff"@},@{number="20",value="0xffffffff"@},
32980 @{number="21",value="0xffffffff"@},@{number="22",value="0xfffffff7"@},
32981 @{number="23",value="0xffffffff"@},@{number="24",value="0xffffffff"@},
32982 @{number="25",value="0xffffffff"@},@{number="26",value="0xfffffffb"@},
32983 @{number="27",value="0xffffffff"@},@{number="28",value="0xf7bfffff"@},
32984 @{number="29",value="0x0"@},@{number="30",value="0xfe010000"@},
32985 @{number="31",value="0x0"@},@{number="32",value="0x0"@},
32986 @{number="33",value="0x0"@},@{number="34",value="0x0"@},
32987 @{number="35",value="0x0"@},@{number="36",value="0x0"@},
32988 @{number="37",value="0x0"@},@{number="38",value="0x0"@},
32989 @{number="39",value="0x0"@},@{number="40",value="0x0"@},
32990 @{number="41",value="0x0"@},@{number="42",value="0x0"@},
32991 @{number="43",value="0x0"@},@{number="44",value="0x0"@},
32992 @{number="45",value="0x0"@},@{number="46",value="0x0"@},
32993 @{number="47",value="0x0"@},@{number="48",value="0x0"@},
32994 @{number="49",value="0x0"@},@{number="50",value="0x0"@},
32995 @{number="51",value="0x0"@},@{number="52",value="0x0"@},
32996 @{number="53",value="0x0"@},@{number="54",value="0x0"@},
32997 @{number="55",value="0x0"@},@{number="56",value="0x0"@},
32998 @{number="57",value="0x0"@},@{number="58",value="0x0"@},
32999 @{number="59",value="0x0"@},@{number="60",value="0x0"@},
33000 @{number="61",value="0x0"@},@{number="62",value="0x0"@},
33001 @{number="63",value="0x0"@},@{number="64",value="0xfe00a300"@},
33002 @{number="65",value="0x29002"@},@{number="66",value="0x202f04b5"@},
33003 @{number="67",value="0xfe0043b0"@},@{number="68",value="0xfe00b3e4"@},
33004 @{number="69",value="0x20002b03"@}]
33009 @subheading The @code{-data-read-memory} Command
33010 @findex -data-read-memory
33012 This command is deprecated, use @code{-data-read-memory-bytes} instead.
33014 @subsubheading Synopsis
33017 -data-read-memory [ -o @var{byte-offset} ]
33018 @var{address} @var{word-format} @var{word-size}
33019 @var{nr-rows} @var{nr-cols} [ @var{aschar} ]
33026 @item @var{address}
33027 An expression specifying the address of the first memory word to be
33028 read. Complex expressions containing embedded white space should be
33029 quoted using the C convention.
33031 @item @var{word-format}
33032 The format to be used to print the memory words. The notation is the
33033 same as for @value{GDBN}'s @code{print} command (@pxref{Output Formats,
33036 @item @var{word-size}
33037 The size of each memory word in bytes.
33039 @item @var{nr-rows}
33040 The number of rows in the output table.
33042 @item @var{nr-cols}
33043 The number of columns in the output table.
33046 If present, indicates that each row should include an @sc{ascii} dump. The
33047 value of @var{aschar} is used as a padding character when a byte is not a
33048 member of the printable @sc{ascii} character set (printable @sc{ascii}
33049 characters are those whose code is between 32 and 126, inclusively).
33051 @item @var{byte-offset}
33052 An offset to add to the @var{address} before fetching memory.
33055 This command displays memory contents as a table of @var{nr-rows} by
33056 @var{nr-cols} words, each word being @var{word-size} bytes. In total,
33057 @code{@var{nr-rows} * @var{nr-cols} * @var{word-size}} bytes are read
33058 (returned as @samp{total-bytes}). Should less than the requested number
33059 of bytes be returned by the target, the missing words are identified
33060 using @samp{N/A}. The number of bytes read from the target is returned
33061 in @samp{nr-bytes} and the starting address used to read memory in
33064 The address of the next/previous row or page is available in
33065 @samp{next-row} and @samp{prev-row}, @samp{next-page} and
33068 @subsubheading @value{GDBN} Command
33070 The corresponding @value{GDBN} command is @samp{x}. @code{gdbtk} has
33071 @samp{gdb_get_mem} memory read command.
33073 @subsubheading Example
33075 Read six bytes of memory starting at @code{bytes+6} but then offset by
33076 @code{-6} bytes. Format as three rows of two columns. One byte per
33077 word. Display each word in hex.
33081 9-data-read-memory -o -6 -- bytes+6 x 1 3 2
33082 9^done,addr="0x00001390",nr-bytes="6",total-bytes="6",
33083 next-row="0x00001396",prev-row="0x0000138e",next-page="0x00001396",
33084 prev-page="0x0000138a",memory=[
33085 @{addr="0x00001390",data=["0x00","0x01"]@},
33086 @{addr="0x00001392",data=["0x02","0x03"]@},
33087 @{addr="0x00001394",data=["0x04","0x05"]@}]
33091 Read two bytes of memory starting at address @code{shorts + 64} and
33092 display as a single word formatted in decimal.
33096 5-data-read-memory shorts+64 d 2 1 1
33097 5^done,addr="0x00001510",nr-bytes="2",total-bytes="2",
33098 next-row="0x00001512",prev-row="0x0000150e",
33099 next-page="0x00001512",prev-page="0x0000150e",memory=[
33100 @{addr="0x00001510",data=["128"]@}]
33104 Read thirty two bytes of memory starting at @code{bytes+16} and format
33105 as eight rows of four columns. Include a string encoding with @samp{x}
33106 used as the non-printable character.
33110 4-data-read-memory bytes+16 x 1 8 4 x
33111 4^done,addr="0x000013a0",nr-bytes="32",total-bytes="32",
33112 next-row="0x000013c0",prev-row="0x0000139c",
33113 next-page="0x000013c0",prev-page="0x00001380",memory=[
33114 @{addr="0x000013a0",data=["0x10","0x11","0x12","0x13"],ascii="xxxx"@},
33115 @{addr="0x000013a4",data=["0x14","0x15","0x16","0x17"],ascii="xxxx"@},
33116 @{addr="0x000013a8",data=["0x18","0x19","0x1a","0x1b"],ascii="xxxx"@},
33117 @{addr="0x000013ac",data=["0x1c","0x1d","0x1e","0x1f"],ascii="xxxx"@},
33118 @{addr="0x000013b0",data=["0x20","0x21","0x22","0x23"],ascii=" !\"#"@},
33119 @{addr="0x000013b4",data=["0x24","0x25","0x26","0x27"],ascii="$%&'"@},
33120 @{addr="0x000013b8",data=["0x28","0x29","0x2a","0x2b"],ascii="()*+"@},
33121 @{addr="0x000013bc",data=["0x2c","0x2d","0x2e","0x2f"],ascii=",-./"@}]
33125 @subheading The @code{-data-read-memory-bytes} Command
33126 @findex -data-read-memory-bytes
33128 @subsubheading Synopsis
33131 -data-read-memory-bytes [ -o @var{offset} ]
33132 @var{address} @var{count}
33139 @item @var{address}
33140 An expression specifying the address of the first addressable memory unit
33141 to be read. Complex expressions containing embedded white space should be
33142 quoted using the C convention.
33145 The number of addressable memory units to read. This should be an integer
33149 The offset relative to @var{address} at which to start reading. This
33150 should be an integer literal. This option is provided so that a frontend
33151 is not required to first evaluate address and then perform address
33152 arithmetics itself.
33156 This command attempts to read all accessible memory regions in the
33157 specified range. First, all regions marked as unreadable in the memory
33158 map (if one is defined) will be skipped. @xref{Memory Region
33159 Attributes}. Second, @value{GDBN} will attempt to read the remaining
33160 regions. For each one, if reading full region results in an errors,
33161 @value{GDBN} will try to read a subset of the region.
33163 In general, every single memory unit in the region may be readable or not,
33164 and the only way to read every readable unit is to try a read at
33165 every address, which is not practical. Therefore, @value{GDBN} will
33166 attempt to read all accessible memory units at either beginning or the end
33167 of the region, using a binary division scheme. This heuristic works
33168 well for reading accross a memory map boundary. Note that if a region
33169 has a readable range that is neither at the beginning or the end,
33170 @value{GDBN} will not read it.
33172 The result record (@pxref{GDB/MI Result Records}) that is output of
33173 the command includes a field named @samp{memory} whose content is a
33174 list of tuples. Each tuple represent a successfully read memory block
33175 and has the following fields:
33179 The start address of the memory block, as hexadecimal literal.
33182 The end address of the memory block, as hexadecimal literal.
33185 The offset of the memory block, as hexadecimal literal, relative to
33186 the start address passed to @code{-data-read-memory-bytes}.
33189 The contents of the memory block, in hex.
33195 @subsubheading @value{GDBN} Command
33197 The corresponding @value{GDBN} command is @samp{x}.
33199 @subsubheading Example
33203 -data-read-memory-bytes &a 10
33204 ^done,memory=[@{begin="0xbffff154",offset="0x00000000",
33206 contents="01000000020000000300"@}]
33211 @subheading The @code{-data-write-memory-bytes} Command
33212 @findex -data-write-memory-bytes
33214 @subsubheading Synopsis
33217 -data-write-memory-bytes @var{address} @var{contents}
33218 -data-write-memory-bytes @var{address} @var{contents} @r{[}@var{count}@r{]}
33225 @item @var{address}
33226 An expression specifying the address of the first addressable memory unit
33227 to be written. Complex expressions containing embedded white space should
33228 be quoted using the C convention.
33230 @item @var{contents}
33231 The hex-encoded data to write. It is an error if @var{contents} does
33232 not represent an integral number of addressable memory units.
33235 Optional argument indicating the number of addressable memory units to be
33236 written. If @var{count} is greater than @var{contents}' length,
33237 @value{GDBN} will repeatedly write @var{contents} until it fills
33238 @var{count} memory units.
33242 @subsubheading @value{GDBN} Command
33244 There's no corresponding @value{GDBN} command.
33246 @subsubheading Example
33250 -data-write-memory-bytes &a "aabbccdd"
33257 -data-write-memory-bytes &a "aabbccdd" 16e
33262 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
33263 @node GDB/MI Tracepoint Commands
33264 @section @sc{gdb/mi} Tracepoint Commands
33266 The commands defined in this section implement MI support for
33267 tracepoints. For detailed introduction, see @ref{Tracepoints}.
33269 @subheading The @code{-trace-find} Command
33270 @findex -trace-find
33272 @subsubheading Synopsis
33275 -trace-find @var{mode} [@var{parameters}@dots{}]
33278 Find a trace frame using criteria defined by @var{mode} and
33279 @var{parameters}. The following table lists permissible
33280 modes and their parameters. For details of operation, see @ref{tfind}.
33285 No parameters are required. Stops examining trace frames.
33288 An integer is required as parameter. Selects tracepoint frame with
33291 @item tracepoint-number
33292 An integer is required as parameter. Finds next
33293 trace frame that corresponds to tracepoint with the specified number.
33296 An address is required as parameter. Finds
33297 next trace frame that corresponds to any tracepoint at the specified
33300 @item pc-inside-range
33301 Two addresses are required as parameters. Finds next trace
33302 frame that corresponds to a tracepoint at an address inside the
33303 specified range. Both bounds are considered to be inside the range.
33305 @item pc-outside-range
33306 Two addresses are required as parameters. Finds
33307 next trace frame that corresponds to a tracepoint at an address outside
33308 the specified range. Both bounds are considered to be inside the range.
33311 Line specification is required as parameter. @xref{Specify Location}.
33312 Finds next trace frame that corresponds to a tracepoint at
33313 the specified location.
33317 If @samp{none} was passed as @var{mode}, the response does not
33318 have fields. Otherwise, the response may have the following fields:
33322 This field has either @samp{0} or @samp{1} as the value, depending
33323 on whether a matching tracepoint was found.
33326 The index of the found traceframe. This field is present iff
33327 the @samp{found} field has value of @samp{1}.
33330 The index of the found tracepoint. This field is present iff
33331 the @samp{found} field has value of @samp{1}.
33334 The information about the frame corresponding to the found trace
33335 frame. This field is present only if a trace frame was found.
33336 @xref{GDB/MI Frame Information}, for description of this field.
33340 @subsubheading @value{GDBN} Command
33342 The corresponding @value{GDBN} command is @samp{tfind}.
33344 @subheading -trace-define-variable
33345 @findex -trace-define-variable
33347 @subsubheading Synopsis
33350 -trace-define-variable @var{name} [ @var{value} ]
33353 Create trace variable @var{name} if it does not exist. If
33354 @var{value} is specified, sets the initial value of the specified
33355 trace variable to that value. Note that the @var{name} should start
33356 with the @samp{$} character.
33358 @subsubheading @value{GDBN} Command
33360 The corresponding @value{GDBN} command is @samp{tvariable}.
33362 @subheading The @code{-trace-frame-collected} Command
33363 @findex -trace-frame-collected
33365 @subsubheading Synopsis
33368 -trace-frame-collected
33369 [--var-print-values @var{var_pval}]
33370 [--comp-print-values @var{comp_pval}]
33371 [--registers-format @var{regformat}]
33372 [--memory-contents]
33375 This command returns the set of collected objects, register names,
33376 trace state variable names, memory ranges and computed expressions
33377 that have been collected at a particular trace frame. The optional
33378 parameters to the command affect the output format in different ways.
33379 See the output description table below for more details.
33381 The reported names can be used in the normal manner to create
33382 varobjs and inspect the objects themselves. The items returned by
33383 this command are categorized so that it is clear which is a variable,
33384 which is a register, which is a trace state variable, which is a
33385 memory range and which is a computed expression.
33387 For instance, if the actions were
33389 collect myVar, myArray[myIndex], myObj.field, myPtr->field, myCount + 2
33390 collect *(int*)0xaf02bef0@@40
33394 the object collected in its entirety would be @code{myVar}. The
33395 object @code{myArray} would be partially collected, because only the
33396 element at index @code{myIndex} would be collected. The remaining
33397 objects would be computed expressions.
33399 An example output would be:
33403 -trace-frame-collected
33405 explicit-variables=[@{name="myVar",value="1"@}],
33406 computed-expressions=[@{name="myArray[myIndex]",value="0"@},
33407 @{name="myObj.field",value="0"@},
33408 @{name="myPtr->field",value="1"@},
33409 @{name="myCount + 2",value="3"@},
33410 @{name="$tvar1 + 1",value="43970027"@}],
33411 registers=[@{number="0",value="0x7fe2c6e79ec8"@},
33412 @{number="1",value="0x0"@},
33413 @{number="2",value="0x4"@},
33415 @{number="125",value="0x0"@}],
33416 tvars=[@{name="$tvar1",current="43970026"@}],
33417 memory=[@{address="0x0000000000602264",length="4"@},
33418 @{address="0x0000000000615bc0",length="4"@}]
33425 @item explicit-variables
33426 The set of objects that have been collected in their entirety (as
33427 opposed to collecting just a few elements of an array or a few struct
33428 members). For each object, its name and value are printed.
33429 The @code{--var-print-values} option affects how or whether the value
33430 field is output. If @var{var_pval} is 0, then print only the names;
33431 if it is 1, print also their values; and if it is 2, print the name,
33432 type and value for simple data types, and the name and type for
33433 arrays, structures and unions.
33435 @item computed-expressions
33436 The set of computed expressions that have been collected at the
33437 current trace frame. The @code{--comp-print-values} option affects
33438 this set like the @code{--var-print-values} option affects the
33439 @code{explicit-variables} set. See above.
33442 The registers that have been collected at the current trace frame.
33443 For each register collected, the name and current value are returned.
33444 The value is formatted according to the @code{--registers-format}
33445 option. See the @command{-data-list-register-values} command for a
33446 list of the allowed formats. The default is @samp{x}.
33449 The trace state variables that have been collected at the current
33450 trace frame. For each trace state variable collected, the name and
33451 current value are returned.
33454 The set of memory ranges that have been collected at the current trace
33455 frame. Its content is a list of tuples. Each tuple represents a
33456 collected memory range and has the following fields:
33460 The start address of the memory range, as hexadecimal literal.
33463 The length of the memory range, as decimal literal.
33466 The contents of the memory block, in hex. This field is only present
33467 if the @code{--memory-contents} option is specified.
33473 @subsubheading @value{GDBN} Command
33475 There is no corresponding @value{GDBN} command.
33477 @subsubheading Example
33479 @subheading -trace-list-variables
33480 @findex -trace-list-variables
33482 @subsubheading Synopsis
33485 -trace-list-variables
33488 Return a table of all defined trace variables. Each element of the
33489 table has the following fields:
33493 The name of the trace variable. This field is always present.
33496 The initial value. This is a 64-bit signed integer. This
33497 field is always present.
33500 The value the trace variable has at the moment. This is a 64-bit
33501 signed integer. This field is absent iff current value is
33502 not defined, for example if the trace was never run, or is
33507 @subsubheading @value{GDBN} Command
33509 The corresponding @value{GDBN} command is @samp{tvariables}.
33511 @subsubheading Example
33515 -trace-list-variables
33516 ^done,trace-variables=@{nr_rows="1",nr_cols="3",
33517 hdr=[@{width="15",alignment="-1",col_name="name",colhdr="Name"@},
33518 @{width="11",alignment="-1",col_name="initial",colhdr="Initial"@},
33519 @{width="11",alignment="-1",col_name="current",colhdr="Current"@}],
33520 body=[variable=@{name="$trace_timestamp",initial="0"@}
33521 variable=@{name="$foo",initial="10",current="15"@}]@}
33525 @subheading -trace-save
33526 @findex -trace-save
33528 @subsubheading Synopsis
33531 -trace-save [ -r ] [ -ctf ] @var{filename}
33534 Saves the collected trace data to @var{filename}. Without the
33535 @samp{-r} option, the data is downloaded from the target and saved
33536 in a local file. With the @samp{-r} option the target is asked
33537 to perform the save.
33539 By default, this command will save the trace in the tfile format. You can
33540 supply the optional @samp{-ctf} argument to save it the CTF format. See
33541 @ref{Trace Files} for more information about CTF.
33543 @subsubheading @value{GDBN} Command
33545 The corresponding @value{GDBN} command is @samp{tsave}.
33548 @subheading -trace-start
33549 @findex -trace-start
33551 @subsubheading Synopsis
33557 Starts a tracing experiment. The result of this command does not
33560 @subsubheading @value{GDBN} Command
33562 The corresponding @value{GDBN} command is @samp{tstart}.
33564 @subheading -trace-status
33565 @findex -trace-status
33567 @subsubheading Synopsis
33573 Obtains the status of a tracing experiment. The result may include
33574 the following fields:
33579 May have a value of either @samp{0}, when no tracing operations are
33580 supported, @samp{1}, when all tracing operations are supported, or
33581 @samp{file} when examining trace file. In the latter case, examining
33582 of trace frame is possible but new tracing experiement cannot be
33583 started. This field is always present.
33586 May have a value of either @samp{0} or @samp{1} depending on whether
33587 tracing experiement is in progress on target. This field is present
33588 if @samp{supported} field is not @samp{0}.
33591 Report the reason why the tracing was stopped last time. This field
33592 may be absent iff tracing was never stopped on target yet. The
33593 value of @samp{request} means the tracing was stopped as result of
33594 the @code{-trace-stop} command. The value of @samp{overflow} means
33595 the tracing buffer is full. The value of @samp{disconnection} means
33596 tracing was automatically stopped when @value{GDBN} has disconnected.
33597 The value of @samp{passcount} means tracing was stopped when a
33598 tracepoint was passed a maximal number of times for that tracepoint.
33599 This field is present if @samp{supported} field is not @samp{0}.
33601 @item stopping-tracepoint
33602 The number of tracepoint whose passcount as exceeded. This field is
33603 present iff the @samp{stop-reason} field has the value of
33607 @itemx frames-created
33608 The @samp{frames} field is a count of the total number of trace frames
33609 in the trace buffer, while @samp{frames-created} is the total created
33610 during the run, including ones that were discarded, such as when a
33611 circular trace buffer filled up. Both fields are optional.
33615 These fields tell the current size of the tracing buffer and the
33616 remaining space. These fields are optional.
33619 The value of the circular trace buffer flag. @code{1} means that the
33620 trace buffer is circular and old trace frames will be discarded if
33621 necessary to make room, @code{0} means that the trace buffer is linear
33625 The value of the disconnected tracing flag. @code{1} means that
33626 tracing will continue after @value{GDBN} disconnects, @code{0} means
33627 that the trace run will stop.
33630 The filename of the trace file being examined. This field is
33631 optional, and only present when examining a trace file.
33635 @subsubheading @value{GDBN} Command
33637 The corresponding @value{GDBN} command is @samp{tstatus}.
33639 @subheading -trace-stop
33640 @findex -trace-stop
33642 @subsubheading Synopsis
33648 Stops a tracing experiment. The result of this command has the same
33649 fields as @code{-trace-status}, except that the @samp{supported} and
33650 @samp{running} fields are not output.
33652 @subsubheading @value{GDBN} Command
33654 The corresponding @value{GDBN} command is @samp{tstop}.
33657 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
33658 @node GDB/MI Symbol Query
33659 @section @sc{gdb/mi} Symbol Query Commands
33663 @subheading The @code{-symbol-info-address} Command
33664 @findex -symbol-info-address
33666 @subsubheading Synopsis
33669 -symbol-info-address @var{symbol}
33672 Describe where @var{symbol} is stored.
33674 @subsubheading @value{GDBN} Command
33676 The corresponding @value{GDBN} command is @samp{info address}.
33678 @subsubheading Example
33682 @subheading The @code{-symbol-info-file} Command
33683 @findex -symbol-info-file
33685 @subsubheading Synopsis
33691 Show the file for the symbol.
33693 @subsubheading @value{GDBN} Command
33695 There's no equivalent @value{GDBN} command. @code{gdbtk} has
33696 @samp{gdb_find_file}.
33698 @subsubheading Example
33702 @subheading The @code{-symbol-info-function} Command
33703 @findex -symbol-info-function
33705 @subsubheading Synopsis
33708 -symbol-info-function
33711 Show which function the symbol lives in.
33713 @subsubheading @value{GDBN} Command
33715 @samp{gdb_get_function} in @code{gdbtk}.
33717 @subsubheading Example
33721 @subheading The @code{-symbol-info-line} Command
33722 @findex -symbol-info-line
33724 @subsubheading Synopsis
33730 Show the core addresses of the code for a source line.
33732 @subsubheading @value{GDBN} Command
33734 The corresponding @value{GDBN} command is @samp{info line}.
33735 @code{gdbtk} has the @samp{gdb_get_line} and @samp{gdb_get_file} commands.
33737 @subsubheading Example
33741 @subheading The @code{-symbol-info-symbol} Command
33742 @findex -symbol-info-symbol
33744 @subsubheading Synopsis
33747 -symbol-info-symbol @var{addr}
33750 Describe what symbol is at location @var{addr}.
33752 @subsubheading @value{GDBN} Command
33754 The corresponding @value{GDBN} command is @samp{info symbol}.
33756 @subsubheading Example
33760 @subheading The @code{-symbol-list-functions} Command
33761 @findex -symbol-list-functions
33763 @subsubheading Synopsis
33766 -symbol-list-functions
33769 List the functions in the executable.
33771 @subsubheading @value{GDBN} Command
33773 @samp{info functions} in @value{GDBN}, @samp{gdb_listfunc} and
33774 @samp{gdb_search} in @code{gdbtk}.
33776 @subsubheading Example
33781 @subheading The @code{-symbol-list-lines} Command
33782 @findex -symbol-list-lines
33784 @subsubheading Synopsis
33787 -symbol-list-lines @var{filename}
33790 Print the list of lines that contain code and their associated program
33791 addresses for the given source filename. The entries are sorted in
33792 ascending PC order.
33794 @subsubheading @value{GDBN} Command
33796 There is no corresponding @value{GDBN} command.
33798 @subsubheading Example
33801 -symbol-list-lines basics.c
33802 ^done,lines=[@{pc="0x08048554",line="7"@},@{pc="0x0804855a",line="8"@}]
33808 @subheading The @code{-symbol-list-types} Command
33809 @findex -symbol-list-types
33811 @subsubheading Synopsis
33817 List all the type names.
33819 @subsubheading @value{GDBN} Command
33821 The corresponding commands are @samp{info types} in @value{GDBN},
33822 @samp{gdb_search} in @code{gdbtk}.
33824 @subsubheading Example
33828 @subheading The @code{-symbol-list-variables} Command
33829 @findex -symbol-list-variables
33831 @subsubheading Synopsis
33834 -symbol-list-variables
33837 List all the global and static variable names.
33839 @subsubheading @value{GDBN} Command
33841 @samp{info variables} in @value{GDBN}, @samp{gdb_search} in @code{gdbtk}.
33843 @subsubheading Example
33847 @subheading The @code{-symbol-locate} Command
33848 @findex -symbol-locate
33850 @subsubheading Synopsis
33856 @subsubheading @value{GDBN} Command
33858 @samp{gdb_loc} in @code{gdbtk}.
33860 @subsubheading Example
33864 @subheading The @code{-symbol-type} Command
33865 @findex -symbol-type
33867 @subsubheading Synopsis
33870 -symbol-type @var{variable}
33873 Show type of @var{variable}.
33875 @subsubheading @value{GDBN} Command
33877 The corresponding @value{GDBN} command is @samp{ptype}, @code{gdbtk} has
33878 @samp{gdb_obj_variable}.
33880 @subsubheading Example
33885 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
33886 @node GDB/MI File Commands
33887 @section @sc{gdb/mi} File Commands
33889 This section describes the GDB/MI commands to specify executable file names
33890 and to read in and obtain symbol table information.
33892 @subheading The @code{-file-exec-and-symbols} Command
33893 @findex -file-exec-and-symbols
33895 @subsubheading Synopsis
33898 -file-exec-and-symbols @var{file}
33901 Specify the executable file to be debugged. This file is the one from
33902 which the symbol table is also read. If no file is specified, the
33903 command clears the executable and symbol information. If breakpoints
33904 are set when using this command with no arguments, @value{GDBN} will produce
33905 error messages. Otherwise, no output is produced, except a completion
33908 @subsubheading @value{GDBN} Command
33910 The corresponding @value{GDBN} command is @samp{file}.
33912 @subsubheading Example
33916 -file-exec-and-symbols /kwikemart/marge/ezannoni/TRUNK/mbx/hello.mbx
33922 @subheading The @code{-file-exec-file} Command
33923 @findex -file-exec-file
33925 @subsubheading Synopsis
33928 -file-exec-file @var{file}
33931 Specify the executable file to be debugged. Unlike
33932 @samp{-file-exec-and-symbols}, the symbol table is @emph{not} read
33933 from this file. If used without argument, @value{GDBN} clears the information
33934 about the executable file. No output is produced, except a completion
33937 @subsubheading @value{GDBN} Command
33939 The corresponding @value{GDBN} command is @samp{exec-file}.
33941 @subsubheading Example
33945 -file-exec-file /kwikemart/marge/ezannoni/TRUNK/mbx/hello.mbx
33952 @subheading The @code{-file-list-exec-sections} Command
33953 @findex -file-list-exec-sections
33955 @subsubheading Synopsis
33958 -file-list-exec-sections
33961 List the sections of the current executable file.
33963 @subsubheading @value{GDBN} Command
33965 The @value{GDBN} command @samp{info file} shows, among the rest, the same
33966 information as this command. @code{gdbtk} has a corresponding command
33967 @samp{gdb_load_info}.
33969 @subsubheading Example
33974 @subheading The @code{-file-list-exec-source-file} Command
33975 @findex -file-list-exec-source-file
33977 @subsubheading Synopsis
33980 -file-list-exec-source-file
33983 List the line number, the current source file, and the absolute path
33984 to the current source file for the current executable. The macro
33985 information field has a value of @samp{1} or @samp{0} depending on
33986 whether or not the file includes preprocessor macro information.
33988 @subsubheading @value{GDBN} Command
33990 The @value{GDBN} equivalent is @samp{info source}
33992 @subsubheading Example
33996 123-file-list-exec-source-file
33997 123^done,line="1",file="foo.c",fullname="/home/bar/foo.c,macro-info="1"
34002 @subheading The @code{-file-list-exec-source-files} Command
34003 @findex -file-list-exec-source-files
34005 @subsubheading Synopsis
34008 -file-list-exec-source-files
34011 List the source files for the current executable.
34013 It will always output both the filename and fullname (absolute file
34014 name) of a source file.
34016 @subsubheading @value{GDBN} Command
34018 The @value{GDBN} equivalent is @samp{info sources}.
34019 @code{gdbtk} has an analogous command @samp{gdb_listfiles}.
34021 @subsubheading Example
34024 -file-list-exec-source-files
34026 @{file=foo.c,fullname=/home/foo.c@},
34027 @{file=/home/bar.c,fullname=/home/bar.c@},
34028 @{file=gdb_could_not_find_fullpath.c@}]
34032 @subheading The @code{-file-list-shared-libraries} Command
34033 @findex -file-list-shared-libraries
34035 @subsubheading Synopsis
34038 -file-list-shared-libraries [ @var{regexp} ]
34041 List the shared libraries in the program.
34042 With a regular expression @var{regexp}, only those libraries whose
34043 names match @var{regexp} are listed.
34045 @subsubheading @value{GDBN} Command
34047 The corresponding @value{GDBN} command is @samp{info shared}. The fields
34048 have a similar meaning to the @code{=library-loaded} notification.
34049 The @code{ranges} field specifies the multiple segments belonging to this
34050 library. Each range has the following fields:
34054 The address defining the inclusive lower bound of the segment.
34056 The address defining the exclusive upper bound of the segment.
34059 @subsubheading Example
34062 -file-list-exec-source-files
34063 ^done,shared-libraries=[
34064 @{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"@}]@},
34065 @{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"@}]@}]
34071 @subheading The @code{-file-list-symbol-files} Command
34072 @findex -file-list-symbol-files
34074 @subsubheading Synopsis
34077 -file-list-symbol-files
34082 @subsubheading @value{GDBN} Command
34084 The corresponding @value{GDBN} command is @samp{info file} (part of it).
34086 @subsubheading Example
34091 @subheading The @code{-file-symbol-file} Command
34092 @findex -file-symbol-file
34094 @subsubheading Synopsis
34097 -file-symbol-file @var{file}
34100 Read symbol table info from the specified @var{file} argument. When
34101 used without arguments, clears @value{GDBN}'s symbol table info. No output is
34102 produced, except for a completion notification.
34104 @subsubheading @value{GDBN} Command
34106 The corresponding @value{GDBN} command is @samp{symbol-file}.
34108 @subsubheading Example
34112 -file-symbol-file /kwikemart/marge/ezannoni/TRUNK/mbx/hello.mbx
34118 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
34119 @node GDB/MI Memory Overlay Commands
34120 @section @sc{gdb/mi} Memory Overlay Commands
34122 The memory overlay commands are not implemented.
34124 @c @subheading -overlay-auto
34126 @c @subheading -overlay-list-mapping-state
34128 @c @subheading -overlay-list-overlays
34130 @c @subheading -overlay-map
34132 @c @subheading -overlay-off
34134 @c @subheading -overlay-on
34136 @c @subheading -overlay-unmap
34138 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
34139 @node GDB/MI Signal Handling Commands
34140 @section @sc{gdb/mi} Signal Handling Commands
34142 Signal handling commands are not implemented.
34144 @c @subheading -signal-handle
34146 @c @subheading -signal-list-handle-actions
34148 @c @subheading -signal-list-signal-types
34152 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
34153 @node GDB/MI Target Manipulation
34154 @section @sc{gdb/mi} Target Manipulation Commands
34157 @subheading The @code{-target-attach} Command
34158 @findex -target-attach
34160 @subsubheading Synopsis
34163 -target-attach @var{pid} | @var{gid} | @var{file}
34166 Attach to a process @var{pid} or a file @var{file} outside of
34167 @value{GDBN}, or a thread group @var{gid}. If attaching to a thread
34168 group, the id previously returned by
34169 @samp{-list-thread-groups --available} must be used.
34171 @subsubheading @value{GDBN} Command
34173 The corresponding @value{GDBN} command is @samp{attach}.
34175 @subsubheading Example
34179 =thread-created,id="1"
34180 *stopped,thread-id="1",frame=@{addr="0xb7f7e410",func="bar",args=[]@}
34186 @subheading The @code{-target-compare-sections} Command
34187 @findex -target-compare-sections
34189 @subsubheading Synopsis
34192 -target-compare-sections [ @var{section} ]
34195 Compare data of section @var{section} on target to the exec file.
34196 Without the argument, all sections are compared.
34198 @subsubheading @value{GDBN} Command
34200 The @value{GDBN} equivalent is @samp{compare-sections}.
34202 @subsubheading Example
34207 @subheading The @code{-target-detach} Command
34208 @findex -target-detach
34210 @subsubheading Synopsis
34213 -target-detach [ @var{pid} | @var{gid} ]
34216 Detach from the remote target which normally resumes its execution.
34217 If either @var{pid} or @var{gid} is specified, detaches from either
34218 the specified process, or specified thread group. There's no output.
34220 @subsubheading @value{GDBN} Command
34222 The corresponding @value{GDBN} command is @samp{detach}.
34224 @subsubheading Example
34234 @subheading The @code{-target-disconnect} Command
34235 @findex -target-disconnect
34237 @subsubheading Synopsis
34243 Disconnect from the remote target. There's no output and the target is
34244 generally not resumed.
34246 @subsubheading @value{GDBN} Command
34248 The corresponding @value{GDBN} command is @samp{disconnect}.
34250 @subsubheading Example
34260 @subheading The @code{-target-download} Command
34261 @findex -target-download
34263 @subsubheading Synopsis
34269 Loads the executable onto the remote target.
34270 It prints out an update message every half second, which includes the fields:
34274 The name of the section.
34276 The size of what has been sent so far for that section.
34278 The size of the section.
34280 The total size of what was sent so far (the current and the previous sections).
34282 The size of the overall executable to download.
34286 Each message is sent as status record (@pxref{GDB/MI Output Syntax, ,
34287 @sc{gdb/mi} Output Syntax}).
34289 In addition, it prints the name and size of the sections, as they are
34290 downloaded. These messages include the following fields:
34294 The name of the section.
34296 The size of the section.
34298 The size of the overall executable to download.
34302 At the end, a summary is printed.
34304 @subsubheading @value{GDBN} Command
34306 The corresponding @value{GDBN} command is @samp{load}.
34308 @subsubheading Example
34310 Note: each status message appears on a single line. Here the messages
34311 have been broken down so that they can fit onto a page.
34316 +download,@{section=".text",section-size="6668",total-size="9880"@}
34317 +download,@{section=".text",section-sent="512",section-size="6668",
34318 total-sent="512",total-size="9880"@}
34319 +download,@{section=".text",section-sent="1024",section-size="6668",
34320 total-sent="1024",total-size="9880"@}
34321 +download,@{section=".text",section-sent="1536",section-size="6668",
34322 total-sent="1536",total-size="9880"@}
34323 +download,@{section=".text",section-sent="2048",section-size="6668",
34324 total-sent="2048",total-size="9880"@}
34325 +download,@{section=".text",section-sent="2560",section-size="6668",
34326 total-sent="2560",total-size="9880"@}
34327 +download,@{section=".text",section-sent="3072",section-size="6668",
34328 total-sent="3072",total-size="9880"@}
34329 +download,@{section=".text",section-sent="3584",section-size="6668",
34330 total-sent="3584",total-size="9880"@}
34331 +download,@{section=".text",section-sent="4096",section-size="6668",
34332 total-sent="4096",total-size="9880"@}
34333 +download,@{section=".text",section-sent="4608",section-size="6668",
34334 total-sent="4608",total-size="9880"@}
34335 +download,@{section=".text",section-sent="5120",section-size="6668",
34336 total-sent="5120",total-size="9880"@}
34337 +download,@{section=".text",section-sent="5632",section-size="6668",
34338 total-sent="5632",total-size="9880"@}
34339 +download,@{section=".text",section-sent="6144",section-size="6668",
34340 total-sent="6144",total-size="9880"@}
34341 +download,@{section=".text",section-sent="6656",section-size="6668",
34342 total-sent="6656",total-size="9880"@}
34343 +download,@{section=".init",section-size="28",total-size="9880"@}
34344 +download,@{section=".fini",section-size="28",total-size="9880"@}
34345 +download,@{section=".data",section-size="3156",total-size="9880"@}
34346 +download,@{section=".data",section-sent="512",section-size="3156",
34347 total-sent="7236",total-size="9880"@}
34348 +download,@{section=".data",section-sent="1024",section-size="3156",
34349 total-sent="7748",total-size="9880"@}
34350 +download,@{section=".data",section-sent="1536",section-size="3156",
34351 total-sent="8260",total-size="9880"@}
34352 +download,@{section=".data",section-sent="2048",section-size="3156",
34353 total-sent="8772",total-size="9880"@}
34354 +download,@{section=".data",section-sent="2560",section-size="3156",
34355 total-sent="9284",total-size="9880"@}
34356 +download,@{section=".data",section-sent="3072",section-size="3156",
34357 total-sent="9796",total-size="9880"@}
34358 ^done,address="0x10004",load-size="9880",transfer-rate="6586",
34365 @subheading The @code{-target-exec-status} Command
34366 @findex -target-exec-status
34368 @subsubheading Synopsis
34371 -target-exec-status
34374 Provide information on the state of the target (whether it is running or
34375 not, for instance).
34377 @subsubheading @value{GDBN} Command
34379 There's no equivalent @value{GDBN} command.
34381 @subsubheading Example
34385 @subheading The @code{-target-list-available-targets} Command
34386 @findex -target-list-available-targets
34388 @subsubheading Synopsis
34391 -target-list-available-targets
34394 List the possible targets to connect to.
34396 @subsubheading @value{GDBN} Command
34398 The corresponding @value{GDBN} command is @samp{help target}.
34400 @subsubheading Example
34404 @subheading The @code{-target-list-current-targets} Command
34405 @findex -target-list-current-targets
34407 @subsubheading Synopsis
34410 -target-list-current-targets
34413 Describe the current target.
34415 @subsubheading @value{GDBN} Command
34417 The corresponding information is printed by @samp{info file} (among
34420 @subsubheading Example
34424 @subheading The @code{-target-list-parameters} Command
34425 @findex -target-list-parameters
34427 @subsubheading Synopsis
34430 -target-list-parameters
34436 @subsubheading @value{GDBN} Command
34440 @subsubheading Example
34443 @subheading The @code{-target-flash-erase} Command
34444 @findex -target-flash-erase
34446 @subsubheading Synopsis
34449 -target-flash-erase
34452 Erases all known flash memory regions on the target.
34454 The corresponding @value{GDBN} command is @samp{flash-erase}.
34456 The output is a list of flash regions that have been erased, with starting
34457 addresses and memory region sizes.
34461 -target-flash-erase
34462 ^done,erased-regions=@{address="0x0",size="0x40000"@}
34466 @subheading The @code{-target-select} Command
34467 @findex -target-select
34469 @subsubheading Synopsis
34472 -target-select @var{type} @var{parameters @dots{}}
34475 Connect @value{GDBN} to the remote target. This command takes two args:
34479 The type of target, for instance @samp{remote}, etc.
34480 @item @var{parameters}
34481 Device names, host names and the like. @xref{Target Commands, ,
34482 Commands for Managing Targets}, for more details.
34485 The output is a connection notification, followed by the address at
34486 which the target program is, in the following form:
34489 ^connected,addr="@var{address}",func="@var{function name}",
34490 args=[@var{arg list}]
34493 @subsubheading @value{GDBN} Command
34495 The corresponding @value{GDBN} command is @samp{target}.
34497 @subsubheading Example
34501 -target-select remote /dev/ttya
34502 ^connected,addr="0xfe00a300",func="??",args=[]
34506 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
34507 @node GDB/MI File Transfer Commands
34508 @section @sc{gdb/mi} File Transfer Commands
34511 @subheading The @code{-target-file-put} Command
34512 @findex -target-file-put
34514 @subsubheading Synopsis
34517 -target-file-put @var{hostfile} @var{targetfile}
34520 Copy file @var{hostfile} from the host system (the machine running
34521 @value{GDBN}) to @var{targetfile} on the target system.
34523 @subsubheading @value{GDBN} Command
34525 The corresponding @value{GDBN} command is @samp{remote put}.
34527 @subsubheading Example
34531 -target-file-put localfile remotefile
34537 @subheading The @code{-target-file-get} Command
34538 @findex -target-file-get
34540 @subsubheading Synopsis
34543 -target-file-get @var{targetfile} @var{hostfile}
34546 Copy file @var{targetfile} from the target system to @var{hostfile}
34547 on the host system.
34549 @subsubheading @value{GDBN} Command
34551 The corresponding @value{GDBN} command is @samp{remote get}.
34553 @subsubheading Example
34557 -target-file-get remotefile localfile
34563 @subheading The @code{-target-file-delete} Command
34564 @findex -target-file-delete
34566 @subsubheading Synopsis
34569 -target-file-delete @var{targetfile}
34572 Delete @var{targetfile} from the target system.
34574 @subsubheading @value{GDBN} Command
34576 The corresponding @value{GDBN} command is @samp{remote delete}.
34578 @subsubheading Example
34582 -target-file-delete remotefile
34588 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
34589 @node GDB/MI Ada Exceptions Commands
34590 @section Ada Exceptions @sc{gdb/mi} Commands
34592 @subheading The @code{-info-ada-exceptions} Command
34593 @findex -info-ada-exceptions
34595 @subsubheading Synopsis
34598 -info-ada-exceptions [ @var{regexp}]
34601 List all Ada exceptions defined within the program being debugged.
34602 With a regular expression @var{regexp}, only those exceptions whose
34603 names match @var{regexp} are listed.
34605 @subsubheading @value{GDBN} Command
34607 The corresponding @value{GDBN} command is @samp{info exceptions}.
34609 @subsubheading Result
34611 The result is a table of Ada exceptions. The following columns are
34612 defined for each exception:
34616 The name of the exception.
34619 The address of the exception.
34623 @subsubheading Example
34626 -info-ada-exceptions aint
34627 ^done,ada-exceptions=@{nr_rows="2",nr_cols="2",
34628 hdr=[@{width="1",alignment="-1",col_name="name",colhdr="Name"@},
34629 @{width="1",alignment="-1",col_name="address",colhdr="Address"@}],
34630 body=[@{name="constraint_error",address="0x0000000000613da0"@},
34631 @{name="const.aint_global_e",address="0x0000000000613b00"@}]@}
34634 @subheading Catching Ada Exceptions
34636 The commands describing how to ask @value{GDBN} to stop when a program
34637 raises an exception are described at @ref{Ada Exception GDB/MI
34638 Catchpoint Commands}.
34641 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
34642 @node GDB/MI Support Commands
34643 @section @sc{gdb/mi} Support Commands
34645 Since new commands and features get regularly added to @sc{gdb/mi},
34646 some commands are available to help front-ends query the debugger
34647 about support for these capabilities. Similarly, it is also possible
34648 to query @value{GDBN} about target support of certain features.
34650 @subheading The @code{-info-gdb-mi-command} Command
34651 @cindex @code{-info-gdb-mi-command}
34652 @findex -info-gdb-mi-command
34654 @subsubheading Synopsis
34657 -info-gdb-mi-command @var{cmd_name}
34660 Query support for the @sc{gdb/mi} command named @var{cmd_name}.
34662 Note that the dash (@code{-}) starting all @sc{gdb/mi} commands
34663 is technically not part of the command name (@pxref{GDB/MI Input
34664 Syntax}), and thus should be omitted in @var{cmd_name}. However,
34665 for ease of use, this command also accepts the form with the leading
34668 @subsubheading @value{GDBN} Command
34670 There is no corresponding @value{GDBN} command.
34672 @subsubheading Result
34674 The result is a tuple. There is currently only one field:
34678 This field is equal to @code{"true"} if the @sc{gdb/mi} command exists,
34679 @code{"false"} otherwise.
34683 @subsubheading Example
34685 Here is an example where the @sc{gdb/mi} command does not exist:
34688 -info-gdb-mi-command unsupported-command
34689 ^done,command=@{exists="false"@}
34693 And here is an example where the @sc{gdb/mi} command is known
34697 -info-gdb-mi-command symbol-list-lines
34698 ^done,command=@{exists="true"@}
34701 @subheading The @code{-list-features} Command
34702 @findex -list-features
34703 @cindex supported @sc{gdb/mi} features, list
34705 Returns a list of particular features of the MI protocol that
34706 this version of gdb implements. A feature can be a command,
34707 or a new field in an output of some command, or even an
34708 important bugfix. While a frontend can sometimes detect presence
34709 of a feature at runtime, it is easier to perform detection at debugger
34712 The command returns a list of strings, with each string naming an
34713 available feature. Each returned string is just a name, it does not
34714 have any internal structure. The list of possible feature names
34720 (gdb) -list-features
34721 ^done,result=["feature1","feature2"]
34724 The current list of features is:
34727 @item frozen-varobjs
34728 Indicates support for the @code{-var-set-frozen} command, as well
34729 as possible presense of the @code{frozen} field in the output
34730 of @code{-varobj-create}.
34731 @item pending-breakpoints
34732 Indicates support for the @option{-f} option to the @code{-break-insert}
34735 Indicates Python scripting support, Python-based
34736 pretty-printing commands, and possible presence of the
34737 @samp{display_hint} field in the output of @code{-var-list-children}
34739 Indicates support for the @code{-thread-info} command.
34740 @item data-read-memory-bytes
34741 Indicates support for the @code{-data-read-memory-bytes} and the
34742 @code{-data-write-memory-bytes} commands.
34743 @item breakpoint-notifications
34744 Indicates that changes to breakpoints and breakpoints created via the
34745 CLI will be announced via async records.
34746 @item ada-task-info
34747 Indicates support for the @code{-ada-task-info} command.
34748 @item language-option
34749 Indicates that all @sc{gdb/mi} commands accept the @option{--language}
34750 option (@pxref{Context management}).
34751 @item info-gdb-mi-command
34752 Indicates support for the @code{-info-gdb-mi-command} command.
34753 @item undefined-command-error-code
34754 Indicates support for the "undefined-command" error code in error result
34755 records, produced when trying to execute an undefined @sc{gdb/mi} command
34756 (@pxref{GDB/MI Result Records}).
34757 @item exec-run-start-option
34758 Indicates that the @code{-exec-run} command supports the @option{--start}
34759 option (@pxref{GDB/MI Program Execution}).
34760 @item data-disassemble-a-option
34761 Indicates that the @code{-data-disassemble} command supports the @option{-a}
34762 option (@pxref{GDB/MI Data Manipulation}).
34765 @subheading The @code{-list-target-features} Command
34766 @findex -list-target-features
34768 Returns a list of particular features that are supported by the
34769 target. Those features affect the permitted MI commands, but
34770 unlike the features reported by the @code{-list-features} command, the
34771 features depend on which target GDB is using at the moment. Whenever
34772 a target can change, due to commands such as @code{-target-select},
34773 @code{-target-attach} or @code{-exec-run}, the list of target features
34774 may change, and the frontend should obtain it again.
34778 (gdb) -list-target-features
34779 ^done,result=["async"]
34782 The current list of features is:
34786 Indicates that the target is capable of asynchronous command
34787 execution, which means that @value{GDBN} will accept further commands
34788 while the target is running.
34791 Indicates that the target is capable of reverse execution.
34792 @xref{Reverse Execution}, for more information.
34796 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
34797 @node GDB/MI Miscellaneous Commands
34798 @section Miscellaneous @sc{gdb/mi} Commands
34800 @c @subheading -gdb-complete
34802 @subheading The @code{-gdb-exit} Command
34805 @subsubheading Synopsis
34811 Exit @value{GDBN} immediately.
34813 @subsubheading @value{GDBN} Command
34815 Approximately corresponds to @samp{quit}.
34817 @subsubheading Example
34827 @subheading The @code{-exec-abort} Command
34828 @findex -exec-abort
34830 @subsubheading Synopsis
34836 Kill the inferior running program.
34838 @subsubheading @value{GDBN} Command
34840 The corresponding @value{GDBN} command is @samp{kill}.
34842 @subsubheading Example
34847 @subheading The @code{-gdb-set} Command
34850 @subsubheading Synopsis
34856 Set an internal @value{GDBN} variable.
34857 @c IS THIS A DOLLAR VARIABLE? OR SOMETHING LIKE ANNOTATE ?????
34859 @subsubheading @value{GDBN} Command
34861 The corresponding @value{GDBN} command is @samp{set}.
34863 @subsubheading Example
34873 @subheading The @code{-gdb-show} Command
34876 @subsubheading Synopsis
34882 Show the current value of a @value{GDBN} variable.
34884 @subsubheading @value{GDBN} Command
34886 The corresponding @value{GDBN} command is @samp{show}.
34888 @subsubheading Example
34897 @c @subheading -gdb-source
34900 @subheading The @code{-gdb-version} Command
34901 @findex -gdb-version
34903 @subsubheading Synopsis
34909 Show version information for @value{GDBN}. Used mostly in testing.
34911 @subsubheading @value{GDBN} Command
34913 The @value{GDBN} equivalent is @samp{show version}. @value{GDBN} by
34914 default shows this information when you start an interactive session.
34916 @subsubheading Example
34918 @c This example modifies the actual output from GDB to avoid overfull
34924 ~Copyright 2000 Free Software Foundation, Inc.
34925 ~GDB is free software, covered by the GNU General Public License, and
34926 ~you are welcome to change it and/or distribute copies of it under
34927 ~ certain conditions.
34928 ~Type "show copying" to see the conditions.
34929 ~There is absolutely no warranty for GDB. Type "show warranty" for
34931 ~This GDB was configured as
34932 "--host=sparc-sun-solaris2.5.1 --target=ppc-eabi".
34937 @subheading The @code{-list-thread-groups} Command
34938 @findex -list-thread-groups
34940 @subheading Synopsis
34943 -list-thread-groups [ --available ] [ --recurse 1 ] [ @var{group} ... ]
34946 Lists thread groups (@pxref{Thread groups}). When a single thread
34947 group is passed as the argument, lists the children of that group.
34948 When several thread group are passed, lists information about those
34949 thread groups. Without any parameters, lists information about all
34950 top-level thread groups.
34952 Normally, thread groups that are being debugged are reported.
34953 With the @samp{--available} option, @value{GDBN} reports thread groups
34954 available on the target.
34956 The output of this command may have either a @samp{threads} result or
34957 a @samp{groups} result. The @samp{thread} result has a list of tuples
34958 as value, with each tuple describing a thread (@pxref{GDB/MI Thread
34959 Information}). The @samp{groups} result has a list of tuples as value,
34960 each tuple describing a thread group. If top-level groups are
34961 requested (that is, no parameter is passed), or when several groups
34962 are passed, the output always has a @samp{groups} result. The format
34963 of the @samp{group} result is described below.
34965 To reduce the number of roundtrips it's possible to list thread groups
34966 together with their children, by passing the @samp{--recurse} option
34967 and the recursion depth. Presently, only recursion depth of 1 is
34968 permitted. If this option is present, then every reported thread group
34969 will also include its children, either as @samp{group} or
34970 @samp{threads} field.
34972 In general, any combination of option and parameters is permitted, with
34973 the following caveats:
34977 When a single thread group is passed, the output will typically
34978 be the @samp{threads} result. Because threads may not contain
34979 anything, the @samp{recurse} option will be ignored.
34982 When the @samp{--available} option is passed, limited information may
34983 be available. In particular, the list of threads of a process might
34984 be inaccessible. Further, specifying specific thread groups might
34985 not give any performance advantage over listing all thread groups.
34986 The frontend should assume that @samp{-list-thread-groups --available}
34987 is always an expensive operation and cache the results.
34991 The @samp{groups} result is a list of tuples, where each tuple may
34992 have the following fields:
34996 Identifier of the thread group. This field is always present.
34997 The identifier is an opaque string; frontends should not try to
34998 convert it to an integer, even though it might look like one.
35001 The type of the thread group. At present, only @samp{process} is a
35005 The target-specific process identifier. This field is only present
35006 for thread groups of type @samp{process} and only if the process exists.
35009 The exit code of this group's last exited thread, formatted in octal.
35010 This field is only present for thread groups of type @samp{process} and
35011 only if the process is not running.
35014 The number of children this thread group has. This field may be
35015 absent for an available thread group.
35018 This field has a list of tuples as value, each tuple describing a
35019 thread. It may be present if the @samp{--recurse} option is
35020 specified, and it's actually possible to obtain the threads.
35023 This field is a list of integers, each identifying a core that one
35024 thread of the group is running on. This field may be absent if
35025 such information is not available.
35028 The name of the executable file that corresponds to this thread group.
35029 The field is only present for thread groups of type @samp{process},
35030 and only if there is a corresponding executable file.
35034 @subheading Example
35038 -list-thread-groups
35039 ^done,groups=[@{id="17",type="process",pid="yyy",num_children="2"@}]
35040 -list-thread-groups 17
35041 ^done,threads=[@{id="2",target-id="Thread 0xb7e14b90 (LWP 21257)",
35042 frame=@{level="0",addr="0xffffe410",func="__kernel_vsyscall",args=[]@},state="running"@},
35043 @{id="1",target-id="Thread 0xb7e156b0 (LWP 21254)",
35044 frame=@{level="0",addr="0x0804891f",func="foo",args=[@{name="i",value="10"@}],
35045 file="/tmp/a.c",fullname="/tmp/a.c",line="158",arch="i386:x86_64"@},state="running"@}]]
35046 -list-thread-groups --available
35047 ^done,groups=[@{id="17",type="process",pid="yyy",num_children="2",cores=[1,2]@}]
35048 -list-thread-groups --available --recurse 1
35049 ^done,groups=[@{id="17", types="process",pid="yyy",num_children="2",cores=[1,2],
35050 threads=[@{id="1",target-id="Thread 0xb7e14b90",cores=[1]@},
35051 @{id="2",target-id="Thread 0xb7e14b90",cores=[2]@}]@},..]
35052 -list-thread-groups --available --recurse 1 17 18
35053 ^done,groups=[@{id="17", types="process",pid="yyy",num_children="2",cores=[1,2],
35054 threads=[@{id="1",target-id="Thread 0xb7e14b90",cores=[1]@},
35055 @{id="2",target-id="Thread 0xb7e14b90",cores=[2]@}]@},...]
35058 @subheading The @code{-info-os} Command
35061 @subsubheading Synopsis
35064 -info-os [ @var{type} ]
35067 If no argument is supplied, the command returns a table of available
35068 operating-system-specific information types. If one of these types is
35069 supplied as an argument @var{type}, then the command returns a table
35070 of data of that type.
35072 The types of information available depend on the target operating
35075 @subsubheading @value{GDBN} Command
35077 The corresponding @value{GDBN} command is @samp{info os}.
35079 @subsubheading Example
35081 When run on a @sc{gnu}/Linux system, the output will look something
35087 ^done,OSDataTable=@{nr_rows="10",nr_cols="3",
35088 hdr=[@{width="10",alignment="-1",col_name="col0",colhdr="Type"@},
35089 @{width="10",alignment="-1",col_name="col1",colhdr="Description"@},
35090 @{width="10",alignment="-1",col_name="col2",colhdr="Title"@}],
35091 body=[item=@{col0="cpus",col1="Listing of all cpus/cores on the system",
35093 item=@{col0="files",col1="Listing of all file descriptors",
35094 col2="File descriptors"@},
35095 item=@{col0="modules",col1="Listing of all loaded kernel modules",
35096 col2="Kernel modules"@},
35097 item=@{col0="msg",col1="Listing of all message queues",
35098 col2="Message queues"@},
35099 item=@{col0="processes",col1="Listing of all processes",
35100 col2="Processes"@},
35101 item=@{col0="procgroups",col1="Listing of all process groups",
35102 col2="Process groups"@},
35103 item=@{col0="semaphores",col1="Listing of all semaphores",
35104 col2="Semaphores"@},
35105 item=@{col0="shm",col1="Listing of all shared-memory regions",
35106 col2="Shared-memory regions"@},
35107 item=@{col0="sockets",col1="Listing of all internet-domain sockets",
35109 item=@{col0="threads",col1="Listing of all threads",
35113 ^done,OSDataTable=@{nr_rows="190",nr_cols="4",
35114 hdr=[@{width="10",alignment="-1",col_name="col0",colhdr="pid"@},
35115 @{width="10",alignment="-1",col_name="col1",colhdr="user"@},
35116 @{width="10",alignment="-1",col_name="col2",colhdr="command"@},
35117 @{width="10",alignment="-1",col_name="col3",colhdr="cores"@}],
35118 body=[item=@{col0="1",col1="root",col2="/sbin/init",col3="0"@},
35119 item=@{col0="2",col1="root",col2="[kthreadd]",col3="1"@},
35120 item=@{col0="3",col1="root",col2="[ksoftirqd/0]",col3="0"@},
35122 item=@{col0="26446",col1="stan",col2="bash",col3="0"@},
35123 item=@{col0="28152",col1="stan",col2="bash",col3="1"@}]@}
35127 (Note that the MI output here includes a @code{"Title"} column that
35128 does not appear in command-line @code{info os}; this column is useful
35129 for MI clients that want to enumerate the types of data, such as in a
35130 popup menu, but is needless clutter on the command line, and
35131 @code{info os} omits it.)
35133 @subheading The @code{-add-inferior} Command
35134 @findex -add-inferior
35136 @subheading Synopsis
35142 Creates a new inferior (@pxref{Inferiors and Programs}). The created
35143 inferior is not associated with any executable. Such association may
35144 be established with the @samp{-file-exec-and-symbols} command
35145 (@pxref{GDB/MI File Commands}). The command response has a single
35146 field, @samp{inferior}, whose value is the identifier of the
35147 thread group corresponding to the new inferior.
35149 @subheading Example
35154 ^done,inferior="i3"
35157 @subheading The @code{-interpreter-exec} Command
35158 @findex -interpreter-exec
35160 @subheading Synopsis
35163 -interpreter-exec @var{interpreter} @var{command}
35165 @anchor{-interpreter-exec}
35167 Execute the specified @var{command} in the given @var{interpreter}.
35169 @subheading @value{GDBN} Command
35171 The corresponding @value{GDBN} command is @samp{interpreter-exec}.
35173 @subheading Example
35177 -interpreter-exec console "break main"
35178 &"During symbol reading, couldn't parse type; debugger out of date?.\n"
35179 &"During symbol reading, bad structure-type format.\n"
35180 ~"Breakpoint 1 at 0x8074fc6: file ../../src/gdb/main.c, line 743.\n"
35185 @subheading The @code{-inferior-tty-set} Command
35186 @findex -inferior-tty-set
35188 @subheading Synopsis
35191 -inferior-tty-set /dev/pts/1
35194 Set terminal for future runs of the program being debugged.
35196 @subheading @value{GDBN} Command
35198 The corresponding @value{GDBN} command is @samp{set inferior-tty} /dev/pts/1.
35200 @subheading Example
35204 -inferior-tty-set /dev/pts/1
35209 @subheading The @code{-inferior-tty-show} Command
35210 @findex -inferior-tty-show
35212 @subheading Synopsis
35218 Show terminal for future runs of program being debugged.
35220 @subheading @value{GDBN} Command
35222 The corresponding @value{GDBN} command is @samp{show inferior-tty}.
35224 @subheading Example
35228 -inferior-tty-set /dev/pts/1
35232 ^done,inferior_tty_terminal="/dev/pts/1"
35236 @subheading The @code{-enable-timings} Command
35237 @findex -enable-timings
35239 @subheading Synopsis
35242 -enable-timings [yes | no]
35245 Toggle the printing of the wallclock, user and system times for an MI
35246 command as a field in its output. This command is to help frontend
35247 developers optimize the performance of their code. No argument is
35248 equivalent to @samp{yes}.
35250 @subheading @value{GDBN} Command
35254 @subheading Example
35262 ^done,bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
35263 addr="0x080484ed",func="main",file="myprog.c",
35264 fullname="/home/nickrob/myprog.c",line="73",thread-groups=["i1"],
35266 time=@{wallclock="0.05185",user="0.00800",system="0.00000"@}
35274 *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",thread-id="0",
35275 frame=@{addr="0x080484ed",func="main",args=[@{name="argc",value="1"@},
35276 @{name="argv",value="0xbfb60364"@}],file="myprog.c",
35277 fullname="/home/nickrob/myprog.c",line="73",arch="i386:x86_64"@}
35281 @subheading The @code{-complete} Command
35284 @subheading Synopsis
35287 -complete @var{command}
35290 Show a list of completions for partially typed CLI @var{command}.
35292 This command is intended for @sc{gdb/mi} frontends that cannot use two separate
35293 CLI and MI channels --- for example: because of lack of PTYs like on Windows or
35294 because @value{GDBN} is used remotely via a SSH connection.
35298 The result consists of two or three fields:
35302 This field contains the completed @var{command}. If @var{command}
35303 has no known completions, this field is omitted.
35306 This field contains a (possibly empty) array of matches. It is always present.
35308 @item max_completions_reached
35309 This field contains @code{1} if number of known completions is above
35310 @code{max-completions} limit (@pxref{Completion}), otherwise it contains
35311 @code{0}. It is always present.
35315 @subheading @value{GDBN} Command
35317 The corresponding @value{GDBN} command is @samp{complete}.
35319 @subheading Example
35324 ^done,completion="break",
35325 matches=["break","break-range"],
35326 max_completions_reached="0"
35329 ^done,completion="b ma",
35330 matches=["b madvise","b main"],max_completions_reached="0"
35332 -complete "b push_b"
35333 ^done,completion="b push_back(",
35335 "b A::push_back(void*)",
35336 "b std::string::push_back(char)",
35337 "b std::vector<int, std::allocator<int> >::push_back(int&&)"],
35338 max_completions_reached="0"
35340 -complete "nonexist"
35341 ^done,matches=[],max_completions_reached="0"
35347 @chapter @value{GDBN} Annotations
35349 This chapter describes annotations in @value{GDBN}. Annotations were
35350 designed to interface @value{GDBN} to graphical user interfaces or other
35351 similar programs which want to interact with @value{GDBN} at a
35352 relatively high level.
35354 The annotation mechanism has largely been superseded by @sc{gdb/mi}
35358 This is Edition @value{EDITION}, @value{DATE}.
35362 * Annotations Overview:: What annotations are; the general syntax.
35363 * Server Prefix:: Issuing a command without affecting user state.
35364 * Prompting:: Annotations marking @value{GDBN}'s need for input.
35365 * Errors:: Annotations for error messages.
35366 * Invalidation:: Some annotations describe things now invalid.
35367 * Annotations for Running::
35368 Whether the program is running, how it stopped, etc.
35369 * Source Annotations:: Annotations describing source code.
35372 @node Annotations Overview
35373 @section What is an Annotation?
35374 @cindex annotations
35376 Annotations start with a newline character, two @samp{control-z}
35377 characters, and the name of the annotation. If there is no additional
35378 information associated with this annotation, the name of the annotation
35379 is followed immediately by a newline. If there is additional
35380 information, the name of the annotation is followed by a space, the
35381 additional information, and a newline. The additional information
35382 cannot contain newline characters.
35384 Any output not beginning with a newline and two @samp{control-z}
35385 characters denotes literal output from @value{GDBN}. Currently there is
35386 no need for @value{GDBN} to output a newline followed by two
35387 @samp{control-z} characters, but if there was such a need, the
35388 annotations could be extended with an @samp{escape} annotation which
35389 means those three characters as output.
35391 The annotation @var{level}, which is specified using the
35392 @option{--annotate} command line option (@pxref{Mode Options}), controls
35393 how much information @value{GDBN} prints together with its prompt,
35394 values of expressions, source lines, and other types of output. Level 0
35395 is for no annotations, level 1 is for use when @value{GDBN} is run as a
35396 subprocess of @sc{gnu} Emacs, level 3 is the maximum annotation suitable
35397 for programs that control @value{GDBN}, and level 2 annotations have
35398 been made obsolete (@pxref{Limitations, , Limitations of the Annotation
35399 Interface, annotate, GDB's Obsolete Annotations}).
35402 @kindex set annotate
35403 @item set annotate @var{level}
35404 The @value{GDBN} command @code{set annotate} sets the level of
35405 annotations to the specified @var{level}.
35407 @item show annotate
35408 @kindex show annotate
35409 Show the current annotation level.
35412 This chapter describes level 3 annotations.
35414 A simple example of starting up @value{GDBN} with annotations is:
35417 $ @kbd{gdb --annotate=3}
35419 Copyright 2003 Free Software Foundation, Inc.
35420 GDB is free software, covered by the GNU General Public License,
35421 and you are welcome to change it and/or distribute copies of it
35422 under certain conditions.
35423 Type "show copying" to see the conditions.
35424 There is absolutely no warranty for GDB. Type "show warranty"
35426 This GDB was configured as "i386-pc-linux-gnu"
35437 Here @samp{quit} is input to @value{GDBN}; the rest is output from
35438 @value{GDBN}. The three lines beginning @samp{^Z^Z} (where @samp{^Z}
35439 denotes a @samp{control-z} character) are annotations; the rest is
35440 output from @value{GDBN}.
35442 @node Server Prefix
35443 @section The Server Prefix
35444 @cindex server prefix
35446 If you prefix a command with @samp{server } then it will not affect
35447 the command history, nor will it affect @value{GDBN}'s notion of which
35448 command to repeat if @key{RET} is pressed on a line by itself. This
35449 means that commands can be run behind a user's back by a front-end in
35450 a transparent manner.
35452 The @code{server } prefix does not affect the recording of values into
35453 the value history; to print a value without recording it into the
35454 value history, use the @code{output} command instead of the
35455 @code{print} command.
35457 Using this prefix also disables confirmation requests
35458 (@pxref{confirmation requests}).
35461 @section Annotation for @value{GDBN} Input
35463 @cindex annotations for prompts
35464 When @value{GDBN} prompts for input, it annotates this fact so it is possible
35465 to know when to send output, when the output from a given command is
35468 Different kinds of input each have a different @dfn{input type}. Each
35469 input type has three annotations: a @code{pre-} annotation, which
35470 denotes the beginning of any prompt which is being output, a plain
35471 annotation, which denotes the end of the prompt, and then a @code{post-}
35472 annotation which denotes the end of any echo which may (or may not) be
35473 associated with the input. For example, the @code{prompt} input type
35474 features the following annotations:
35482 The input types are
35485 @findex pre-prompt annotation
35486 @findex prompt annotation
35487 @findex post-prompt annotation
35489 When @value{GDBN} is prompting for a command (the main @value{GDBN} prompt).
35491 @findex pre-commands annotation
35492 @findex commands annotation
35493 @findex post-commands annotation
35495 When @value{GDBN} prompts for a set of commands, like in the @code{commands}
35496 command. The annotations are repeated for each command which is input.
35498 @findex pre-overload-choice annotation
35499 @findex overload-choice annotation
35500 @findex post-overload-choice annotation
35501 @item overload-choice
35502 When @value{GDBN} wants the user to select between various overloaded functions.
35504 @findex pre-query annotation
35505 @findex query annotation
35506 @findex post-query annotation
35508 When @value{GDBN} wants the user to confirm a potentially dangerous operation.
35510 @findex pre-prompt-for-continue annotation
35511 @findex prompt-for-continue annotation
35512 @findex post-prompt-for-continue annotation
35513 @item prompt-for-continue
35514 When @value{GDBN} is asking the user to press return to continue. Note: Don't
35515 expect this to work well; instead use @code{set height 0} to disable
35516 prompting. This is because the counting of lines is buggy in the
35517 presence of annotations.
35522 @cindex annotations for errors, warnings and interrupts
35524 @findex quit annotation
35529 This annotation occurs right before @value{GDBN} responds to an interrupt.
35531 @findex error annotation
35536 This annotation occurs right before @value{GDBN} responds to an error.
35538 Quit and error annotations indicate that any annotations which @value{GDBN} was
35539 in the middle of may end abruptly. For example, if a
35540 @code{value-history-begin} annotation is followed by a @code{error}, one
35541 cannot expect to receive the matching @code{value-history-end}. One
35542 cannot expect not to receive it either, however; an error annotation
35543 does not necessarily mean that @value{GDBN} is immediately returning all the way
35546 @findex error-begin annotation
35547 A quit or error annotation may be preceded by
35553 Any output between that and the quit or error annotation is the error
35556 Warning messages are not yet annotated.
35557 @c If we want to change that, need to fix warning(), type_error(),
35558 @c range_error(), and possibly other places.
35561 @section Invalidation Notices
35563 @cindex annotations for invalidation messages
35564 The following annotations say that certain pieces of state may have
35568 @findex frames-invalid annotation
35569 @item ^Z^Zframes-invalid
35571 The frames (for example, output from the @code{backtrace} command) may
35574 @findex breakpoints-invalid annotation
35575 @item ^Z^Zbreakpoints-invalid
35577 The breakpoints may have changed. For example, the user just added or
35578 deleted a breakpoint.
35581 @node Annotations for Running
35582 @section Running the Program
35583 @cindex annotations for running programs
35585 @findex starting annotation
35586 @findex stopping annotation
35587 When the program starts executing due to a @value{GDBN} command such as
35588 @code{step} or @code{continue},
35594 is output. When the program stops,
35600 is output. Before the @code{stopped} annotation, a variety of
35601 annotations describe how the program stopped.
35604 @findex exited annotation
35605 @item ^Z^Zexited @var{exit-status}
35606 The program exited, and @var{exit-status} is the exit status (zero for
35607 successful exit, otherwise nonzero).
35609 @findex signalled annotation
35610 @findex signal-name annotation
35611 @findex signal-name-end annotation
35612 @findex signal-string annotation
35613 @findex signal-string-end annotation
35614 @item ^Z^Zsignalled
35615 The program exited with a signal. After the @code{^Z^Zsignalled}, the
35616 annotation continues:
35622 ^Z^Zsignal-name-end
35626 ^Z^Zsignal-string-end
35631 where @var{name} is the name of the signal, such as @code{SIGILL} or
35632 @code{SIGSEGV}, and @var{string} is the explanation of the signal, such
35633 as @code{Illegal Instruction} or @code{Segmentation fault}. The arguments
35634 @var{intro-text}, @var{middle-text}, and @var{end-text} are for the
35635 user's benefit and have no particular format.
35637 @findex signal annotation
35639 The syntax of this annotation is just like @code{signalled}, but @value{GDBN} is
35640 just saying that the program received the signal, not that it was
35641 terminated with it.
35643 @findex breakpoint annotation
35644 @item ^Z^Zbreakpoint @var{number}
35645 The program hit breakpoint number @var{number}.
35647 @findex watchpoint annotation
35648 @item ^Z^Zwatchpoint @var{number}
35649 The program hit watchpoint number @var{number}.
35652 @node Source Annotations
35653 @section Displaying Source
35654 @cindex annotations for source display
35656 @findex source annotation
35657 The following annotation is used instead of displaying source code:
35660 ^Z^Zsource @var{filename}:@var{line}:@var{character}:@var{middle}:@var{addr}
35663 where @var{filename} is an absolute file name indicating which source
35664 file, @var{line} is the line number within that file (where 1 is the
35665 first line in the file), @var{character} is the character position
35666 within the file (where 0 is the first character in the file) (for most
35667 debug formats this will necessarily point to the beginning of a line),
35668 @var{middle} is @samp{middle} if @var{addr} is in the middle of the
35669 line, or @samp{beg} if @var{addr} is at the beginning of the line, and
35670 @var{addr} is the address in the target program associated with the
35671 source which is being displayed. The @var{addr} is in the form @samp{0x}
35672 followed by one or more lowercase hex digits (note that this does not
35673 depend on the language).
35675 @node JIT Interface
35676 @chapter JIT Compilation Interface
35677 @cindex just-in-time compilation
35678 @cindex JIT compilation interface
35680 This chapter documents @value{GDBN}'s @dfn{just-in-time} (JIT) compilation
35681 interface. A JIT compiler is a program or library that generates native
35682 executable code at runtime and executes it, usually in order to achieve good
35683 performance while maintaining platform independence.
35685 Programs that use JIT compilation are normally difficult to debug because
35686 portions of their code are generated at runtime, instead of being loaded from
35687 object files, which is where @value{GDBN} normally finds the program's symbols
35688 and debug information. In order to debug programs that use JIT compilation,
35689 @value{GDBN} has an interface that allows the program to register in-memory
35690 symbol files with @value{GDBN} at runtime.
35692 If you are using @value{GDBN} to debug a program that uses this interface, then
35693 it should work transparently so long as you have not stripped the binary. If
35694 you are developing a JIT compiler, then the interface is documented in the rest
35695 of this chapter. At this time, the only known client of this interface is the
35698 Broadly speaking, the JIT interface mirrors the dynamic loader interface. The
35699 JIT compiler communicates with @value{GDBN} by writing data into a global
35700 variable and calling a fuction at a well-known symbol. When @value{GDBN}
35701 attaches, it reads a linked list of symbol files from the global variable to
35702 find existing code, and puts a breakpoint in the function so that it can find
35703 out about additional code.
35706 * Declarations:: Relevant C struct declarations
35707 * Registering Code:: Steps to register code
35708 * Unregistering Code:: Steps to unregister code
35709 * Custom Debug Info:: Emit debug information in a custom format
35713 @section JIT Declarations
35715 These are the relevant struct declarations that a C program should include to
35716 implement the interface:
35726 struct jit_code_entry
35728 struct jit_code_entry *next_entry;
35729 struct jit_code_entry *prev_entry;
35730 const char *symfile_addr;
35731 uint64_t symfile_size;
35734 struct jit_descriptor
35737 /* This type should be jit_actions_t, but we use uint32_t
35738 to be explicit about the bitwidth. */
35739 uint32_t action_flag;
35740 struct jit_code_entry *relevant_entry;
35741 struct jit_code_entry *first_entry;
35744 /* GDB puts a breakpoint in this function. */
35745 void __attribute__((noinline)) __jit_debug_register_code() @{ @};
35747 /* Make sure to specify the version statically, because the
35748 debugger may check the version before we can set it. */
35749 struct jit_descriptor __jit_debug_descriptor = @{ 1, 0, 0, 0 @};
35752 If the JIT is multi-threaded, then it is important that the JIT synchronize any
35753 modifications to this global data properly, which can easily be done by putting
35754 a global mutex around modifications to these structures.
35756 @node Registering Code
35757 @section Registering Code
35759 To register code with @value{GDBN}, the JIT should follow this protocol:
35763 Generate an object file in memory with symbols and other desired debug
35764 information. The file must include the virtual addresses of the sections.
35767 Create a code entry for the file, which gives the start and size of the symbol
35771 Add it to the linked list in the JIT descriptor.
35774 Point the relevant_entry field of the descriptor at the entry.
35777 Set @code{action_flag} to @code{JIT_REGISTER} and call
35778 @code{__jit_debug_register_code}.
35781 When @value{GDBN} is attached and the breakpoint fires, @value{GDBN} uses the
35782 @code{relevant_entry} pointer so it doesn't have to walk the list looking for
35783 new code. However, the linked list must still be maintained in order to allow
35784 @value{GDBN} to attach to a running process and still find the symbol files.
35786 @node Unregistering Code
35787 @section Unregistering Code
35789 If code is freed, then the JIT should use the following protocol:
35793 Remove the code entry corresponding to the code from the linked list.
35796 Point the @code{relevant_entry} field of the descriptor at the code entry.
35799 Set @code{action_flag} to @code{JIT_UNREGISTER} and call
35800 @code{__jit_debug_register_code}.
35803 If the JIT frees or recompiles code without unregistering it, then @value{GDBN}
35804 and the JIT will leak the memory used for the associated symbol files.
35806 @node Custom Debug Info
35807 @section Custom Debug Info
35808 @cindex custom JIT debug info
35809 @cindex JIT debug info reader
35811 Generating debug information in platform-native file formats (like ELF
35812 or COFF) may be an overkill for JIT compilers; especially if all the
35813 debug info is used for is displaying a meaningful backtrace. The
35814 issue can be resolved by having the JIT writers decide on a debug info
35815 format and also provide a reader that parses the debug info generated
35816 by the JIT compiler. This section gives a brief overview on writing
35817 such a parser. More specific details can be found in the source file
35818 @file{gdb/jit-reader.in}, which is also installed as a header at
35819 @file{@var{includedir}/gdb/jit-reader.h} for easy inclusion.
35821 The reader is implemented as a shared object (so this functionality is
35822 not available on platforms which don't allow loading shared objects at
35823 runtime). Two @value{GDBN} commands, @code{jit-reader-load} and
35824 @code{jit-reader-unload} are provided, to be used to load and unload
35825 the readers from a preconfigured directory. Once loaded, the shared
35826 object is used the parse the debug information emitted by the JIT
35830 * Using JIT Debug Info Readers:: How to use supplied readers correctly
35831 * Writing JIT Debug Info Readers:: Creating a debug-info reader
35834 @node Using JIT Debug Info Readers
35835 @subsection Using JIT Debug Info Readers
35836 @kindex jit-reader-load
35837 @kindex jit-reader-unload
35839 Readers can be loaded and unloaded using the @code{jit-reader-load}
35840 and @code{jit-reader-unload} commands.
35843 @item jit-reader-load @var{reader}
35844 Load the JIT reader named @var{reader}, which is a shared
35845 object specified as either an absolute or a relative file name. In
35846 the latter case, @value{GDBN} will try to load the reader from a
35847 pre-configured directory, usually @file{@var{libdir}/gdb/} on a UNIX
35848 system (here @var{libdir} is the system library directory, often
35849 @file{/usr/local/lib}).
35851 Only one reader can be active at a time; trying to load a second
35852 reader when one is already loaded will result in @value{GDBN}
35853 reporting an error. A new JIT reader can be loaded by first unloading
35854 the current one using @code{jit-reader-unload} and then invoking
35855 @code{jit-reader-load}.
35857 @item jit-reader-unload
35858 Unload the currently loaded JIT reader.
35862 @node Writing JIT Debug Info Readers
35863 @subsection Writing JIT Debug Info Readers
35864 @cindex writing JIT debug info readers
35866 As mentioned, a reader is essentially a shared object conforming to a
35867 certain ABI. This ABI is described in @file{jit-reader.h}.
35869 @file{jit-reader.h} defines the structures, macros and functions
35870 required to write a reader. It is installed (along with
35871 @value{GDBN}), in @file{@var{includedir}/gdb} where @var{includedir} is
35872 the system include directory.
35874 Readers need to be released under a GPL compatible license. A reader
35875 can be declared as released under such a license by placing the macro
35876 @code{GDB_DECLARE_GPL_COMPATIBLE_READER} in a source file.
35878 The entry point for readers is the symbol @code{gdb_init_reader},
35879 which is expected to be a function with the prototype
35881 @findex gdb_init_reader
35883 extern struct gdb_reader_funcs *gdb_init_reader (void);
35886 @cindex @code{struct gdb_reader_funcs}
35888 @code{struct gdb_reader_funcs} contains a set of pointers to callback
35889 functions. These functions are executed to read the debug info
35890 generated by the JIT compiler (@code{read}), to unwind stack frames
35891 (@code{unwind}) and to create canonical frame IDs
35892 (@code{get_Frame_id}). It also has a callback that is called when the
35893 reader is being unloaded (@code{destroy}). The struct looks like this
35896 struct gdb_reader_funcs
35898 /* Must be set to GDB_READER_INTERFACE_VERSION. */
35899 int reader_version;
35901 /* For use by the reader. */
35904 gdb_read_debug_info *read;
35905 gdb_unwind_frame *unwind;
35906 gdb_get_frame_id *get_frame_id;
35907 gdb_destroy_reader *destroy;
35911 @cindex @code{struct gdb_symbol_callbacks}
35912 @cindex @code{struct gdb_unwind_callbacks}
35914 The callbacks are provided with another set of callbacks by
35915 @value{GDBN} to do their job. For @code{read}, these callbacks are
35916 passed in a @code{struct gdb_symbol_callbacks} and for @code{unwind}
35917 and @code{get_frame_id}, in a @code{struct gdb_unwind_callbacks}.
35918 @code{struct gdb_symbol_callbacks} has callbacks to create new object
35919 files and new symbol tables inside those object files. @code{struct
35920 gdb_unwind_callbacks} has callbacks to read registers off the current
35921 frame and to write out the values of the registers in the previous
35922 frame. Both have a callback (@code{target_read}) to read bytes off the
35923 target's address space.
35925 @node In-Process Agent
35926 @chapter In-Process Agent
35927 @cindex debugging agent
35928 The traditional debugging model is conceptually low-speed, but works fine,
35929 because most bugs can be reproduced in debugging-mode execution. However,
35930 as multi-core or many-core processors are becoming mainstream, and
35931 multi-threaded programs become more and more popular, there should be more
35932 and more bugs that only manifest themselves at normal-mode execution, for
35933 example, thread races, because debugger's interference with the program's
35934 timing may conceal the bugs. On the other hand, in some applications,
35935 it is not feasible for the debugger to interrupt the program's execution
35936 long enough for the developer to learn anything helpful about its behavior.
35937 If the program's correctness depends on its real-time behavior, delays
35938 introduced by a debugger might cause the program to fail, even when the
35939 code itself is correct. It is useful to be able to observe the program's
35940 behavior without interrupting it.
35942 Therefore, traditional debugging model is too intrusive to reproduce
35943 some bugs. In order to reduce the interference with the program, we can
35944 reduce the number of operations performed by debugger. The
35945 @dfn{In-Process Agent}, a shared library, is running within the same
35946 process with inferior, and is able to perform some debugging operations
35947 itself. As a result, debugger is only involved when necessary, and
35948 performance of debugging can be improved accordingly. Note that
35949 interference with program can be reduced but can't be removed completely,
35950 because the in-process agent will still stop or slow down the program.
35952 The in-process agent can interpret and execute Agent Expressions
35953 (@pxref{Agent Expressions}) during performing debugging operations. The
35954 agent expressions can be used for different purposes, such as collecting
35955 data in tracepoints, and condition evaluation in breakpoints.
35957 @anchor{Control Agent}
35958 You can control whether the in-process agent is used as an aid for
35959 debugging with the following commands:
35962 @kindex set agent on
35964 Causes the in-process agent to perform some operations on behalf of the
35965 debugger. Just which operations requested by the user will be done
35966 by the in-process agent depends on the its capabilities. For example,
35967 if you request to evaluate breakpoint conditions in the in-process agent,
35968 and the in-process agent has such capability as well, then breakpoint
35969 conditions will be evaluated in the in-process agent.
35971 @kindex set agent off
35972 @item set agent off
35973 Disables execution of debugging operations by the in-process agent. All
35974 of the operations will be performed by @value{GDBN}.
35978 Display the current setting of execution of debugging operations by
35979 the in-process agent.
35983 * In-Process Agent Protocol::
35986 @node In-Process Agent Protocol
35987 @section In-Process Agent Protocol
35988 @cindex in-process agent protocol
35990 The in-process agent is able to communicate with both @value{GDBN} and
35991 GDBserver (@pxref{In-Process Agent}). This section documents the protocol
35992 used for communications between @value{GDBN} or GDBserver and the IPA.
35993 In general, @value{GDBN} or GDBserver sends commands
35994 (@pxref{IPA Protocol Commands}) and data to in-process agent, and then
35995 in-process agent replies back with the return result of the command, or
35996 some other information. The data sent to in-process agent is composed
35997 of primitive data types, such as 4-byte or 8-byte type, and composite
35998 types, which are called objects (@pxref{IPA Protocol Objects}).
36001 * IPA Protocol Objects::
36002 * IPA Protocol Commands::
36005 @node IPA Protocol Objects
36006 @subsection IPA Protocol Objects
36007 @cindex ipa protocol objects
36009 The commands sent to and results received from agent may contain some
36010 complex data types called @dfn{objects}.
36012 The in-process agent is running on the same machine with @value{GDBN}
36013 or GDBserver, so it doesn't have to handle as much differences between
36014 two ends as remote protocol (@pxref{Remote Protocol}) tries to handle.
36015 However, there are still some differences of two ends in two processes:
36019 word size. On some 64-bit machines, @value{GDBN} or GDBserver can be
36020 compiled as a 64-bit executable, while in-process agent is a 32-bit one.
36022 ABI. Some machines may have multiple types of ABI, @value{GDBN} or
36023 GDBserver is compiled with one, and in-process agent is compiled with
36027 Here are the IPA Protocol Objects:
36031 agent expression object. It represents an agent expression
36032 (@pxref{Agent Expressions}).
36033 @anchor{agent expression object}
36035 tracepoint action object. It represents a tracepoint action
36036 (@pxref{Tracepoint Actions,,Tracepoint Action Lists}) to collect registers,
36037 memory, static trace data and to evaluate expression.
36038 @anchor{tracepoint action object}
36040 tracepoint object. It represents a tracepoint (@pxref{Tracepoints}).
36041 @anchor{tracepoint object}
36045 The following table describes important attributes of each IPA protocol
36048 @multitable @columnfractions .30 .20 .50
36049 @headitem Name @tab Size @tab Description
36050 @item @emph{agent expression object} @tab @tab
36051 @item length @tab 4 @tab length of bytes code
36052 @item byte code @tab @var{length} @tab contents of byte code
36053 @item @emph{tracepoint action for collecting memory} @tab @tab
36054 @item 'M' @tab 1 @tab type of tracepoint action
36055 @item addr @tab 8 @tab if @var{basereg} is @samp{-1}, @var{addr} is the
36056 address of the lowest byte to collect, otherwise @var{addr} is the offset
36057 of @var{basereg} for memory collecting.
36058 @item len @tab 8 @tab length of memory for collecting
36059 @item basereg @tab 4 @tab the register number containing the starting
36060 memory address for collecting.
36061 @item @emph{tracepoint action for collecting registers} @tab @tab
36062 @item 'R' @tab 1 @tab type of tracepoint action
36063 @item @emph{tracepoint action for collecting static trace data} @tab @tab
36064 @item 'L' @tab 1 @tab type of tracepoint action
36065 @item @emph{tracepoint action for expression evaluation} @tab @tab
36066 @item 'X' @tab 1 @tab type of tracepoint action
36067 @item agent expression @tab length of @tab @ref{agent expression object}
36068 @item @emph{tracepoint object} @tab @tab
36069 @item number @tab 4 @tab number of tracepoint
36070 @item address @tab 8 @tab address of tracepoint inserted on
36071 @item type @tab 4 @tab type of tracepoint
36072 @item enabled @tab 1 @tab enable or disable of tracepoint
36073 @item step_count @tab 8 @tab step
36074 @item pass_count @tab 8 @tab pass
36075 @item numactions @tab 4 @tab number of tracepoint actions
36076 @item hit count @tab 8 @tab hit count
36077 @item trace frame usage @tab 8 @tab trace frame usage
36078 @item compiled_cond @tab 8 @tab compiled condition
36079 @item orig_size @tab 8 @tab orig size
36080 @item condition @tab 4 if condition is NULL otherwise length of
36081 @ref{agent expression object}
36082 @tab zero if condition is NULL, otherwise is
36083 @ref{agent expression object}
36084 @item actions @tab variable
36085 @tab numactions number of @ref{tracepoint action object}
36088 @node IPA Protocol Commands
36089 @subsection IPA Protocol Commands
36090 @cindex ipa protocol commands
36092 The spaces in each command are delimiters to ease reading this commands
36093 specification. They don't exist in real commands.
36097 @item FastTrace:@var{tracepoint_object} @var{gdb_jump_pad_head}
36098 Installs a new fast tracepoint described by @var{tracepoint_object}
36099 (@pxref{tracepoint object}). The @var{gdb_jump_pad_head}, 8-byte long, is the
36100 head of @dfn{jumppad}, which is used to jump to data collection routine
36105 @item OK @var{target_address} @var{gdb_jump_pad_head} @var{fjump_size} @var{fjump}
36106 @var{target_address} is address of tracepoint in the inferior.
36107 The @var{gdb_jump_pad_head} is updated head of jumppad. Both of
36108 @var{target_address} and @var{gdb_jump_pad_head} are 8-byte long.
36109 The @var{fjump} contains a sequence of instructions jump to jumppad entry.
36110 The @var{fjump_size}, 4-byte long, is the size of @var{fjump}.
36117 Closes the in-process agent. This command is sent when @value{GDBN} or GDBserver
36118 is about to kill inferiors.
36126 @item probe_marker_at:@var{address}
36127 Asks in-process agent to probe the marker at @var{address}.
36134 @item unprobe_marker_at:@var{address}
36135 Asks in-process agent to unprobe the marker at @var{address}.
36139 @chapter Reporting Bugs in @value{GDBN}
36140 @cindex bugs in @value{GDBN}
36141 @cindex reporting bugs in @value{GDBN}
36143 Your bug reports play an essential role in making @value{GDBN} reliable.
36145 Reporting a bug may help you by bringing a solution to your problem, or it
36146 may not. But in any case the principal function of a bug report is to help
36147 the entire community by making the next version of @value{GDBN} work better. Bug
36148 reports are your contribution to the maintenance of @value{GDBN}.
36150 In order for a bug report to serve its purpose, you must include the
36151 information that enables us to fix the bug.
36154 * Bug Criteria:: Have you found a bug?
36155 * Bug Reporting:: How to report bugs
36159 @section Have You Found a Bug?
36160 @cindex bug criteria
36162 If you are not sure whether you have found a bug, here are some guidelines:
36165 @cindex fatal signal
36166 @cindex debugger crash
36167 @cindex crash of debugger
36169 If the debugger gets a fatal signal, for any input whatever, that is a
36170 @value{GDBN} bug. Reliable debuggers never crash.
36172 @cindex error on valid input
36174 If @value{GDBN} produces an error message for valid input, that is a
36175 bug. (Note that if you're cross debugging, the problem may also be
36176 somewhere in the connection to the target.)
36178 @cindex invalid input
36180 If @value{GDBN} does not produce an error message for invalid input,
36181 that is a bug. However, you should note that your idea of
36182 ``invalid input'' might be our idea of ``an extension'' or ``support
36183 for traditional practice''.
36186 If you are an experienced user of debugging tools, your suggestions
36187 for improvement of @value{GDBN} are welcome in any case.
36190 @node Bug Reporting
36191 @section How to Report Bugs
36192 @cindex bug reports
36193 @cindex @value{GDBN} bugs, reporting
36195 A number of companies and individuals offer support for @sc{gnu} products.
36196 If you obtained @value{GDBN} from a support organization, we recommend you
36197 contact that organization first.
36199 You can find contact information for many support companies and
36200 individuals in the file @file{etc/SERVICE} in the @sc{gnu} Emacs
36202 @c should add a web page ref...
36205 @ifset BUGURL_DEFAULT
36206 In any event, we also recommend that you submit bug reports for
36207 @value{GDBN}. The preferred method is to submit them directly using
36208 @uref{http://www.gnu.org/software/gdb/bugs/, @value{GDBN}'s Bugs web
36209 page}. Alternatively, the @email{bug-gdb@@gnu.org, e-mail gateway} can
36212 @strong{Do not send bug reports to @samp{info-gdb}, or to
36213 @samp{help-gdb}, or to any newsgroups.} Most users of @value{GDBN} do
36214 not want to receive bug reports. Those that do have arranged to receive
36217 The mailing list @samp{bug-gdb} has a newsgroup @samp{gnu.gdb.bug} which
36218 serves as a repeater. The mailing list and the newsgroup carry exactly
36219 the same messages. Often people think of posting bug reports to the
36220 newsgroup instead of mailing them. This appears to work, but it has one
36221 problem which can be crucial: a newsgroup posting often lacks a mail
36222 path back to the sender. Thus, if we need to ask for more information,
36223 we may be unable to reach you. For this reason, it is better to send
36224 bug reports to the mailing list.
36226 @ifclear BUGURL_DEFAULT
36227 In any event, we also recommend that you submit bug reports for
36228 @value{GDBN} to @value{BUGURL}.
36232 The fundamental principle of reporting bugs usefully is this:
36233 @strong{report all the facts}. If you are not sure whether to state a
36234 fact or leave it out, state it!
36236 Often people omit facts because they think they know what causes the
36237 problem and assume that some details do not matter. Thus, you might
36238 assume that the name of the variable you use in an example does not matter.
36239 Well, probably it does not, but one cannot be sure. Perhaps the bug is a
36240 stray memory reference which happens to fetch from the location where that
36241 name is stored in memory; perhaps, if the name were different, the contents
36242 of that location would fool the debugger into doing the right thing despite
36243 the bug. Play it safe and give a specific, complete example. That is the
36244 easiest thing for you to do, and the most helpful.
36246 Keep in mind that the purpose of a bug report is to enable us to fix the
36247 bug. It may be that the bug has been reported previously, but neither
36248 you nor we can know that unless your bug report is complete and
36251 Sometimes people give a few sketchy facts and ask, ``Does this ring a
36252 bell?'' Those bug reports are useless, and we urge everyone to
36253 @emph{refuse to respond to them} except to chide the sender to report
36256 To enable us to fix the bug, you should include all these things:
36260 The version of @value{GDBN}. @value{GDBN} announces it if you start
36261 with no arguments; you can also print it at any time using @code{show
36264 Without this, we will not know whether there is any point in looking for
36265 the bug in the current version of @value{GDBN}.
36268 The type of machine you are using, and the operating system name and
36272 The details of the @value{GDBN} build-time configuration.
36273 @value{GDBN} shows these details if you invoke it with the
36274 @option{--configuration} command-line option, or if you type
36275 @code{show configuration} at @value{GDBN}'s prompt.
36278 What compiler (and its version) was used to compile @value{GDBN}---e.g.@:
36279 ``@value{GCC}--2.8.1''.
36282 What compiler (and its version) was used to compile the program you are
36283 debugging---e.g.@: ``@value{GCC}--2.8.1'', or ``HP92453-01 A.10.32.03 HP
36284 C Compiler''. For @value{NGCC}, you can say @kbd{@value{GCC} --version}
36285 to get this information; for other compilers, see the documentation for
36289 The command arguments you gave the compiler to compile your example and
36290 observe the bug. For example, did you use @samp{-O}? To guarantee
36291 you will not omit something important, list them all. A copy of the
36292 Makefile (or the output from make) is sufficient.
36294 If we were to try to guess the arguments, we would probably guess wrong
36295 and then we might not encounter the bug.
36298 A complete input script, and all necessary source files, that will
36302 A description of what behavior you observe that you believe is
36303 incorrect. For example, ``It gets a fatal signal.''
36305 Of course, if the bug is that @value{GDBN} gets a fatal signal, then we
36306 will certainly notice it. But if the bug is incorrect output, we might
36307 not notice unless it is glaringly wrong. You might as well not give us
36308 a chance to make a mistake.
36310 Even if the problem you experience is a fatal signal, you should still
36311 say so explicitly. Suppose something strange is going on, such as, your
36312 copy of @value{GDBN} is out of synch, or you have encountered a bug in
36313 the C library on your system. (This has happened!) Your copy might
36314 crash and ours would not. If you told us to expect a crash, then when
36315 ours fails to crash, we would know that the bug was not happening for
36316 us. If you had not told us to expect a crash, then we would not be able
36317 to draw any conclusion from our observations.
36320 @cindex recording a session script
36321 To collect all this information, you can use a session recording program
36322 such as @command{script}, which is available on many Unix systems.
36323 Just run your @value{GDBN} session inside @command{script} and then
36324 include the @file{typescript} file with your bug report.
36326 Another way to record a @value{GDBN} session is to run @value{GDBN}
36327 inside Emacs and then save the entire buffer to a file.
36330 If you wish to suggest changes to the @value{GDBN} source, send us context
36331 diffs. If you even discuss something in the @value{GDBN} source, refer to
36332 it by context, not by line number.
36334 The line numbers in our development sources will not match those in your
36335 sources. Your line numbers would convey no useful information to us.
36339 Here are some things that are not necessary:
36343 A description of the envelope of the bug.
36345 Often people who encounter a bug spend a lot of time investigating
36346 which changes to the input file will make the bug go away and which
36347 changes will not affect it.
36349 This is often time consuming and not very useful, because the way we
36350 will find the bug is by running a single example under the debugger
36351 with breakpoints, not by pure deduction from a series of examples.
36352 We recommend that you save your time for something else.
36354 Of course, if you can find a simpler example to report @emph{instead}
36355 of the original one, that is a convenience for us. Errors in the
36356 output will be easier to spot, running under the debugger will take
36357 less time, and so on.
36359 However, simplification is not vital; if you do not want to do this,
36360 report the bug anyway and send us the entire test case you used.
36363 A patch for the bug.
36365 A patch for the bug does help us if it is a good one. But do not omit
36366 the necessary information, such as the test case, on the assumption that
36367 a patch is all we need. We might see problems with your patch and decide
36368 to fix the problem another way, or we might not understand it at all.
36370 Sometimes with a program as complicated as @value{GDBN} it is very hard to
36371 construct an example that will make the program follow a certain path
36372 through the code. If you do not send us the example, we will not be able
36373 to construct one, so we will not be able to verify that the bug is fixed.
36375 And if we cannot understand what bug you are trying to fix, or why your
36376 patch should be an improvement, we will not install it. A test case will
36377 help us to understand.
36380 A guess about what the bug is or what it depends on.
36382 Such guesses are usually wrong. Even we cannot guess right about such
36383 things without first using the debugger to find the facts.
36386 @c The readline documentation is distributed with the readline code
36387 @c and consists of the two following files:
36390 @c Use -I with makeinfo to point to the appropriate directory,
36391 @c environment var TEXINPUTS with TeX.
36392 @ifclear SYSTEM_READLINE
36393 @include rluser.texi
36394 @include hsuser.texi
36398 @appendix In Memoriam
36400 The @value{GDBN} project mourns the loss of the following long-time
36405 Fred was a long-standing contributor to @value{GDBN} (1991-2006), and
36406 to Free Software in general. Outside of @value{GDBN}, he was known in
36407 the Amiga world for his series of Fish Disks, and the GeekGadget project.
36409 @item Michael Snyder
36410 Michael was one of the Global Maintainers of the @value{GDBN} project,
36411 with contributions recorded as early as 1996, until 2011. In addition
36412 to his day to day participation, he was a large driving force behind
36413 adding Reverse Debugging to @value{GDBN}.
36416 Beyond their technical contributions to the project, they were also
36417 enjoyable members of the Free Software Community. We will miss them.
36419 @node Formatting Documentation
36420 @appendix Formatting Documentation
36422 @cindex @value{GDBN} reference card
36423 @cindex reference card
36424 The @value{GDBN} 4 release includes an already-formatted reference card, ready
36425 for printing with PostScript or Ghostscript, in the @file{gdb}
36426 subdirectory of the main source directory@footnote{In
36427 @file{gdb-@value{GDBVN}/gdb/refcard.ps} of the version @value{GDBVN}
36428 release.}. If you can use PostScript or Ghostscript with your printer,
36429 you can print the reference card immediately with @file{refcard.ps}.
36431 The release also includes the source for the reference card. You
36432 can format it, using @TeX{}, by typing:
36438 The @value{GDBN} reference card is designed to print in @dfn{landscape}
36439 mode on US ``letter'' size paper;
36440 that is, on a sheet 11 inches wide by 8.5 inches
36441 high. You will need to specify this form of printing as an option to
36442 your @sc{dvi} output program.
36444 @cindex documentation
36446 All the documentation for @value{GDBN} comes as part of the machine-readable
36447 distribution. The documentation is written in Texinfo format, which is
36448 a documentation system that uses a single source file to produce both
36449 on-line information and a printed manual. You can use one of the Info
36450 formatting commands to create the on-line version of the documentation
36451 and @TeX{} (or @code{texi2roff}) to typeset the printed version.
36453 @value{GDBN} includes an already formatted copy of the on-line Info
36454 version of this manual in the @file{gdb} subdirectory. The main Info
36455 file is @file{gdb-@value{GDBVN}/gdb/gdb.info}, and it refers to
36456 subordinate files matching @samp{gdb.info*} in the same directory. If
36457 necessary, you can print out these files, or read them with any editor;
36458 but they are easier to read using the @code{info} subsystem in @sc{gnu}
36459 Emacs or the standalone @code{info} program, available as part of the
36460 @sc{gnu} Texinfo distribution.
36462 If you want to format these Info files yourself, you need one of the
36463 Info formatting programs, such as @code{texinfo-format-buffer} or
36466 If you have @code{makeinfo} installed, and are in the top level
36467 @value{GDBN} source directory (@file{gdb-@value{GDBVN}}, in the case of
36468 version @value{GDBVN}), you can make the Info file by typing:
36475 If you want to typeset and print copies of this manual, you need @TeX{},
36476 a program to print its @sc{dvi} output files, and @file{texinfo.tex}, the
36477 Texinfo definitions file.
36479 @TeX{} is a typesetting program; it does not print files directly, but
36480 produces output files called @sc{dvi} files. To print a typeset
36481 document, you need a program to print @sc{dvi} files. If your system
36482 has @TeX{} installed, chances are it has such a program. The precise
36483 command to use depends on your system; @kbd{lpr -d} is common; another
36484 (for PostScript devices) is @kbd{dvips}. The @sc{dvi} print command may
36485 require a file name without any extension or a @samp{.dvi} extension.
36487 @TeX{} also requires a macro definitions file called
36488 @file{texinfo.tex}. This file tells @TeX{} how to typeset a document
36489 written in Texinfo format. On its own, @TeX{} cannot either read or
36490 typeset a Texinfo file. @file{texinfo.tex} is distributed with GDB
36491 and is located in the @file{gdb-@var{version-number}/texinfo}
36494 If you have @TeX{} and a @sc{dvi} printer program installed, you can
36495 typeset and print this manual. First switch to the @file{gdb}
36496 subdirectory of the main source directory (for example, to
36497 @file{gdb-@value{GDBVN}/gdb}) and type:
36503 Then give @file{gdb.dvi} to your @sc{dvi} printing program.
36505 @node Installing GDB
36506 @appendix Installing @value{GDBN}
36507 @cindex installation
36510 * Requirements:: Requirements for building @value{GDBN}
36511 * Running Configure:: Invoking the @value{GDBN} @file{configure} script
36512 * Separate Objdir:: Compiling @value{GDBN} in another directory
36513 * Config Names:: Specifying names for hosts and targets
36514 * Configure Options:: Summary of options for configure
36515 * System-wide configuration:: Having a system-wide init file
36519 @section Requirements for Building @value{GDBN}
36520 @cindex building @value{GDBN}, requirements for
36522 Building @value{GDBN} requires various tools and packages to be available.
36523 Other packages will be used only if they are found.
36525 @heading Tools/Packages Necessary for Building @value{GDBN}
36527 @item C@t{++}11 compiler
36528 @value{GDBN} is written in C@t{++}11. It should be buildable with any
36529 recent C@t{++}11 compiler, e.g.@: GCC.
36532 @value{GDBN}'s build system relies on features only found in the GNU
36533 make program. Other variants of @code{make} will not work.
36536 @heading Tools/Packages Optional for Building @value{GDBN}
36540 @value{GDBN} can use the Expat XML parsing library. This library may be
36541 included with your operating system distribution; if it is not, you
36542 can get the latest version from @url{http://expat.sourceforge.net}.
36543 The @file{configure} script will search for this library in several
36544 standard locations; if it is installed in an unusual path, you can
36545 use the @option{--with-libexpat-prefix} option to specify its location.
36551 Remote protocol memory maps (@pxref{Memory Map Format})
36553 Target descriptions (@pxref{Target Descriptions})
36555 Remote shared library lists (@xref{Library List Format},
36556 or alternatively @pxref{Library List Format for SVR4 Targets})
36558 MS-Windows shared libraries (@pxref{Shared Libraries})
36560 Traceframe info (@pxref{Traceframe Info Format})
36562 Branch trace (@pxref{Branch Trace Format},
36563 @pxref{Branch Trace Configuration Format})
36567 @value{GDBN} can be scripted using GNU Guile. @xref{Guile}. By
36568 default, @value{GDBN} will be compiled if the Guile libraries are
36569 installed and are found by @file{configure}. You can use the
36570 @code{--with-guile} option to request Guile, and pass either the Guile
36571 version number or the file name of the relevant @code{pkg-config}
36572 program to choose a particular version of Guile.
36575 @value{GDBN}'s features related to character sets (@pxref{Character
36576 Sets}) require a functioning @code{iconv} implementation. If you are
36577 on a GNU system, then this is provided by the GNU C Library. Some
36578 other systems also provide a working @code{iconv}.
36580 If @value{GDBN} is using the @code{iconv} program which is installed
36581 in a non-standard place, you will need to tell @value{GDBN} where to
36582 find it. This is done with @option{--with-iconv-bin} which specifies
36583 the directory that contains the @code{iconv} program. This program is
36584 run in order to make a list of the available character sets.
36586 On systems without @code{iconv}, you can install GNU Libiconv. If
36587 Libiconv is installed in a standard place, @value{GDBN} will
36588 automatically use it if it is needed. If you have previously
36589 installed Libiconv in a non-standard place, you can use the
36590 @option{--with-libiconv-prefix} option to @file{configure}.
36592 @value{GDBN}'s top-level @file{configure} and @file{Makefile} will
36593 arrange to build Libiconv if a directory named @file{libiconv} appears
36594 in the top-most source directory. If Libiconv is built this way, and
36595 if the operating system does not provide a suitable @code{iconv}
36596 implementation, then the just-built library will automatically be used
36597 by @value{GDBN}. One easy way to set this up is to download GNU
36598 Libiconv, unpack it inside the top-level directory of the @value{GDBN}
36599 source tree, and then rename the directory holding the Libiconv source
36600 code to @samp{libiconv}.
36603 @value{GDBN} can support debugging sections that are compressed with
36604 the LZMA library. @xref{MiniDebugInfo}. If this library is not
36605 included with your operating system, you can find it in the xz package
36606 at @url{http://tukaani.org/xz/}. If the LZMA library is available in
36607 the usual place, then the @file{configure} script will use it
36608 automatically. If it is installed in an unusual path, you can use the
36609 @option{--with-lzma-prefix} option to specify its location.
36613 @value{GDBN} can use the GNU MPFR multiple-precision floating-point
36614 library. This library may be included with your operating system
36615 distribution; if it is not, you can get the latest version from
36616 @url{http://www.mpfr.org}. The @file{configure} script will search
36617 for this library in several standard locations; if it is installed
36618 in an unusual path, you can use the @option{--with-libmpfr-prefix}
36619 option to specify its location.
36621 GNU MPFR is used to emulate target floating-point arithmetic during
36622 expression evaluation when the target uses different floating-point
36623 formats than the host. If GNU MPFR it is not available, @value{GDBN}
36624 will fall back to using host floating-point arithmetic.
36627 @value{GDBN} can be scripted using Python language. @xref{Python}.
36628 By default, @value{GDBN} will be compiled if the Python libraries are
36629 installed and are found by @file{configure}. You can use the
36630 @code{--with-python} option to request Python, and pass either the
36631 file name of the relevant @code{python} executable, or the name of the
36632 directory in which Python is installed, to choose a particular
36633 installation of Python.
36636 @cindex compressed debug sections
36637 @value{GDBN} will use the @samp{zlib} library, if available, to read
36638 compressed debug sections. Some linkers, such as GNU gold, are capable
36639 of producing binaries with compressed debug sections. If @value{GDBN}
36640 is compiled with @samp{zlib}, it will be able to read the debug
36641 information in such binaries.
36643 The @samp{zlib} library is likely included with your operating system
36644 distribution; if it is not, you can get the latest version from
36645 @url{http://zlib.net}.
36648 @node Running Configure
36649 @section Invoking the @value{GDBN} @file{configure} Script
36650 @cindex configuring @value{GDBN}
36651 @value{GDBN} comes with a @file{configure} script that automates the process
36652 of preparing @value{GDBN} for installation; you can then use @code{make} to
36653 build the @code{gdb} program.
36655 @c irrelevant in info file; it's as current as the code it lives with.
36656 @footnote{If you have a more recent version of @value{GDBN} than @value{GDBVN},
36657 look at the @file{README} file in the sources; we may have improved the
36658 installation procedures since publishing this manual.}
36661 The @value{GDBN} distribution includes all the source code you need for
36662 @value{GDBN} in a single directory, whose name is usually composed by
36663 appending the version number to @samp{gdb}.
36665 For example, the @value{GDBN} version @value{GDBVN} distribution is in the
36666 @file{gdb-@value{GDBVN}} directory. That directory contains:
36669 @item gdb-@value{GDBVN}/configure @r{(and supporting files)}
36670 script for configuring @value{GDBN} and all its supporting libraries
36672 @item gdb-@value{GDBVN}/gdb
36673 the source specific to @value{GDBN} itself
36675 @item gdb-@value{GDBVN}/bfd
36676 source for the Binary File Descriptor library
36678 @item gdb-@value{GDBVN}/include
36679 @sc{gnu} include files
36681 @item gdb-@value{GDBVN}/libiberty
36682 source for the @samp{-liberty} free software library
36684 @item gdb-@value{GDBVN}/opcodes
36685 source for the library of opcode tables and disassemblers
36687 @item gdb-@value{GDBVN}/readline
36688 source for the @sc{gnu} command-line interface
36691 There may be other subdirectories as well.
36693 The simplest way to configure and build @value{GDBN} is to run @file{configure}
36694 from the @file{gdb-@var{version-number}} source directory, which in
36695 this example is the @file{gdb-@value{GDBVN}} directory.
36697 First switch to the @file{gdb-@var{version-number}} source directory
36698 if you are not already in it; then run @file{configure}. Pass the
36699 identifier for the platform on which @value{GDBN} will run as an
36705 cd gdb-@value{GDBVN}
36710 Running @samp{configure} and then running @code{make} builds the
36711 included supporting libraries, then @code{gdb} itself. The configured
36712 source files, and the binaries, are left in the corresponding source
36716 @file{configure} is a Bourne-shell (@code{/bin/sh}) script; if your
36717 system does not recognize this automatically when you run a different
36718 shell, you may need to run @code{sh} on it explicitly:
36724 You should run the @file{configure} script from the top directory in the
36725 source tree, the @file{gdb-@var{version-number}} directory. If you run
36726 @file{configure} from one of the subdirectories, you will configure only
36727 that subdirectory. That is usually not what you want. In particular,
36728 if you run the first @file{configure} from the @file{gdb} subdirectory
36729 of the @file{gdb-@var{version-number}} directory, you will omit the
36730 configuration of @file{bfd}, @file{readline}, and other sibling
36731 directories of the @file{gdb} subdirectory. This leads to build errors
36732 about missing include files such as @file{bfd/bfd.h}.
36734 You can install @code{@value{GDBN}} anywhere. The best way to do this
36735 is to pass the @code{--prefix} option to @code{configure}, and then
36736 install it with @code{make install}.
36738 @node Separate Objdir
36739 @section Compiling @value{GDBN} in Another Directory
36741 If you want to run @value{GDBN} versions for several host or target machines,
36742 you need a different @code{gdb} compiled for each combination of
36743 host and target. @file{configure} is designed to make this easy by
36744 allowing you to generate each configuration in a separate subdirectory,
36745 rather than in the source directory. If your @code{make} program
36746 handles the @samp{VPATH} feature (@sc{gnu} @code{make} does), running
36747 @code{make} in each of these directories builds the @code{gdb}
36748 program specified there.
36750 To build @code{gdb} in a separate directory, run @file{configure}
36751 with the @samp{--srcdir} option to specify where to find the source.
36752 (You also need to specify a path to find @file{configure}
36753 itself from your working directory. If the path to @file{configure}
36754 would be the same as the argument to @samp{--srcdir}, you can leave out
36755 the @samp{--srcdir} option; it is assumed.)
36757 For example, with version @value{GDBVN}, you can build @value{GDBN} in a
36758 separate directory for a Sun 4 like this:
36762 cd gdb-@value{GDBVN}
36765 ../gdb-@value{GDBVN}/configure
36770 When @file{configure} builds a configuration using a remote source
36771 directory, it creates a tree for the binaries with the same structure
36772 (and using the same names) as the tree under the source directory. In
36773 the example, you'd find the Sun 4 library @file{libiberty.a} in the
36774 directory @file{gdb-sun4/libiberty}, and @value{GDBN} itself in
36775 @file{gdb-sun4/gdb}.
36777 Make sure that your path to the @file{configure} script has just one
36778 instance of @file{gdb} in it. If your path to @file{configure} looks
36779 like @file{../gdb-@value{GDBVN}/gdb/configure}, you are configuring only
36780 one subdirectory of @value{GDBN}, not the whole package. This leads to
36781 build errors about missing include files such as @file{bfd/bfd.h}.
36783 One popular reason to build several @value{GDBN} configurations in separate
36784 directories is to configure @value{GDBN} for cross-compiling (where
36785 @value{GDBN} runs on one machine---the @dfn{host}---while debugging
36786 programs that run on another machine---the @dfn{target}).
36787 You specify a cross-debugging target by
36788 giving the @samp{--target=@var{target}} option to @file{configure}.
36790 When you run @code{make} to build a program or library, you must run
36791 it in a configured directory---whatever directory you were in when you
36792 called @file{configure} (or one of its subdirectories).
36794 The @code{Makefile} that @file{configure} generates in each source
36795 directory also runs recursively. If you type @code{make} in a source
36796 directory such as @file{gdb-@value{GDBVN}} (or in a separate configured
36797 directory configured with @samp{--srcdir=@var{dirname}/gdb-@value{GDBVN}}), you
36798 will build all the required libraries, and then build GDB.
36800 When you have multiple hosts or targets configured in separate
36801 directories, you can run @code{make} on them in parallel (for example,
36802 if they are NFS-mounted on each of the hosts); they will not interfere
36806 @section Specifying Names for Hosts and Targets
36808 The specifications used for hosts and targets in the @file{configure}
36809 script are based on a three-part naming scheme, but some short predefined
36810 aliases are also supported. The full naming scheme encodes three pieces
36811 of information in the following pattern:
36814 @var{architecture}-@var{vendor}-@var{os}
36817 For example, you can use the alias @code{sun4} as a @var{host} argument,
36818 or as the value for @var{target} in a @code{--target=@var{target}}
36819 option. The equivalent full name is @samp{sparc-sun-sunos4}.
36821 The @file{configure} script accompanying @value{GDBN} does not provide
36822 any query facility to list all supported host and target names or
36823 aliases. @file{configure} calls the Bourne shell script
36824 @code{config.sub} to map abbreviations to full names; you can read the
36825 script, if you wish, or you can use it to test your guesses on
36826 abbreviations---for example:
36829 % sh config.sub i386-linux
36831 % sh config.sub alpha-linux
36832 alpha-unknown-linux-gnu
36833 % sh config.sub hp9k700
36835 % sh config.sub sun4
36836 sparc-sun-sunos4.1.1
36837 % sh config.sub sun3
36838 m68k-sun-sunos4.1.1
36839 % sh config.sub i986v
36840 Invalid configuration `i986v': machine `i986v' not recognized
36844 @code{config.sub} is also distributed in the @value{GDBN} source
36845 directory (@file{gdb-@value{GDBVN}}, for version @value{GDBVN}).
36847 @node Configure Options
36848 @section @file{configure} Options
36850 Here is a summary of the @file{configure} options and arguments that
36851 are most often useful for building @value{GDBN}. @file{configure}
36852 also has several other options not listed here. @inforef{Running
36853 configure scripts,,autoconf.info}, for a full
36854 explanation of @file{configure}.
36857 configure @r{[}--help@r{]}
36858 @r{[}--prefix=@var{dir}@r{]}
36859 @r{[}--exec-prefix=@var{dir}@r{]}
36860 @r{[}--srcdir=@var{dirname}@r{]}
36861 @r{[}--target=@var{target}@r{]}
36865 You may introduce options with a single @samp{-} rather than
36866 @samp{--} if you prefer; but you may abbreviate option names if you use
36871 Display a quick summary of how to invoke @file{configure}.
36873 @item --prefix=@var{dir}
36874 Configure the source to install programs and files under directory
36877 @item --exec-prefix=@var{dir}
36878 Configure the source to install programs under directory
36881 @c avoid splitting the warning from the explanation:
36883 @item --srcdir=@var{dirname}
36884 Use this option to make configurations in directories separate from the
36885 @value{GDBN} source directories. Among other things, you can use this to
36886 build (or maintain) several configurations simultaneously, in separate
36887 directories. @file{configure} writes configuration-specific files in
36888 the current directory, but arranges for them to use the source in the
36889 directory @var{dirname}. @file{configure} creates directories under
36890 the working directory in parallel to the source directories below
36893 @item --target=@var{target}
36894 Configure @value{GDBN} for cross-debugging programs running on the specified
36895 @var{target}. Without this option, @value{GDBN} is configured to debug
36896 programs that run on the same machine (@var{host}) as @value{GDBN} itself.
36898 There is no convenient way to generate a list of all available
36899 targets. Also see the @code{--enable-targets} option, below.
36902 There are many other options that are specific to @value{GDBN}. This
36903 lists just the most common ones; there are some very specialized
36904 options not described here.
36907 @item --enable-targets=@r{[}@var{target}@r{]}@dots{}
36908 @itemx --enable-targets=all
36909 Configure @value{GDBN} for cross-debugging programs running on the
36910 specified list of targets. The special value @samp{all} configures
36911 @value{GDBN} for debugging programs running on any target it supports.
36913 @item --with-gdb-datadir=@var{path}
36914 Set the @value{GDBN}-specific data directory. @value{GDBN} will look
36915 here for certain supporting files or scripts. This defaults to the
36916 @file{gdb} subdirectory of @samp{datadi} (which can be set using
36919 @item --with-relocated-sources=@var{dir}
36920 Sets up the default source path substitution rule so that directory
36921 names recorded in debug information will be automatically adjusted for
36922 any directory under @var{dir}. @var{dir} should be a subdirectory of
36923 @value{GDBN}'s configured prefix, the one mentioned in the
36924 @code{--prefix} or @code{--exec-prefix} options to configure. This
36925 option is useful if GDB is supposed to be moved to a different place
36928 @item --enable-64-bit-bfd
36929 Enable 64-bit support in BFD on 32-bit hosts.
36931 @item --disable-gdbmi
36932 Build @value{GDBN} without the GDB/MI machine interface
36936 Build @value{GDBN} with the text-mode full-screen user interface
36937 (TUI). Requires a curses library (ncurses and cursesX are also
36940 @item --with-curses
36941 Use the curses library instead of the termcap library, for text-mode
36942 terminal operations.
36944 @item --with-libunwind-ia64
36945 Use the libunwind library for unwinding function call stack on ia64
36946 target platforms. See http://www.nongnu.org/libunwind/index.html for
36949 @item --with-system-readline
36950 Use the readline library installed on the host, rather than the
36951 library supplied as part of @value{GDBN}. Readline 7 or newer is
36952 required; this is enforced by the build system.
36954 @item --with-system-zlib
36955 Use the zlib library installed on the host, rather than the library
36956 supplied as part of @value{GDBN}.
36959 Build @value{GDBN} with Expat, a library for XML parsing. (Done by
36960 default if libexpat is installed and found at configure time.) This
36961 library is used to read XML files supplied with @value{GDBN}. If it
36962 is unavailable, some features, such as remote protocol memory maps,
36963 target descriptions, and shared library lists, that are based on XML
36964 files, will not be available in @value{GDBN}. If your host does not
36965 have libexpat installed, you can get the latest version from
36966 `http://expat.sourceforge.net'.
36968 @item --with-libiconv-prefix@r{[}=@var{dir}@r{]}
36970 Build @value{GDBN} with GNU libiconv, a character set encoding
36971 conversion library. This is not done by default, as on GNU systems
36972 the @code{iconv} that is built in to the C library is sufficient. If
36973 your host does not have a working @code{iconv}, you can get the latest
36974 version of GNU iconv from `https://www.gnu.org/software/libiconv/'.
36976 @value{GDBN}'s build system also supports building GNU libiconv as
36977 part of the overall build. @xref{Requirements}.
36980 Build @value{GDBN} with LZMA, a compression library. (Done by default
36981 if liblzma is installed and found at configure time.) LZMA is used by
36982 @value{GDBN}'s "mini debuginfo" feature, which is only useful on
36983 platforms using the ELF object file format. If your host does not
36984 have liblzma installed, you can get the latest version from
36985 `https://tukaani.org/xz/'.
36988 Build @value{GDBN} with GNU MPFR, a library for multiple-precision
36989 floating-point computation with correct rounding. (Done by default if
36990 GNU MPFR is installed and found at configure time.) This library is
36991 used to emulate target floating-point arithmetic during expression
36992 evaluation when the target uses different floating-point formats than
36993 the host. If GNU MPFR is not available, @value{GDBN} will fall back
36994 to using host floating-point arithmetic. If your host does not have
36995 GNU MPFR installed, you can get the latest version from
36996 `http://www.mpfr.org'.
36998 @item --with-python@r{[}=@var{python}@r{]}
36999 Build @value{GDBN} with Python scripting support. (Done by default if
37000 libpython is present and found at configure time.) Python makes
37001 @value{GDBN} scripting much more powerful than the restricted CLI
37002 scripting language. If your host does not have Python installed, you
37003 can find it on `http://www.python.org/download/'. The oldest version
37004 of Python supported by GDB is 2.6. The optional argument @var{python}
37005 is used to find the Python headers and libraries. It can be either
37006 the name of a Python executable, or the name of the directory in which
37007 Python is installed.
37009 @item --with-guile[=GUILE]'
37010 Build @value{GDBN} with GNU Guile scripting support. (Done by default
37011 if libguile is present and found at configure time.) If your host
37012 does not have Guile installed, you can find it at
37013 `https://www.gnu.org/software/guile/'. The optional argument GUILE
37014 can be a version number, which will cause @code{configure} to try to
37015 use that version of Guile; or the file name of a @code{pkg-config}
37016 executable, which will be queried to find the information needed to
37017 compile and link against Guile.
37019 @item --without-included-regex
37020 Don't use the regex library included with @value{GDBN} (as part of the
37021 libiberty library). This is the default on hosts with version 2 of
37024 @item --with-sysroot=@var{dir}
37025 Use @var{dir} as the default system root directory for libraries whose
37026 file names begin with @file{/lib}' or @file{/usr/lib'}. (The value of
37027 @var{dir} can be modified at run time by using the @command{set
37028 sysroot} command.) If @var{dir} is under the @value{GDBN} configured
37029 prefix (set with @code{--prefix} or @code{--exec-prefix options}, the
37030 default system root will be automatically adjusted if and when
37031 @value{GDBN} is moved to a different location.
37033 @item --with-system-gdbinit=@var{file}
37034 Configure @value{GDBN} to automatically load a system-wide init file.
37035 @var{file} should be an absolute file name. If @var{file} is in a
37036 directory under the configured prefix, and @value{GDBN} is moved to
37037 another location after being built, the location of the system-wide
37038 init file will be adjusted accordingly.
37040 @item --enable-build-warnings
37041 When building the @value{GDBN} sources, ask the compiler to warn about
37042 any code which looks even vaguely suspicious. It passes many
37043 different warning flags, depending on the exact version of the
37044 compiler you are using.
37046 @item --enable-werror
37047 Treat compiler warnings as werrors. It adds the @code{-Werror} flag
37048 to the compiler, which will fail the compilation if the compiler
37049 outputs any warning messages.
37051 @item --enable-ubsan
37052 Enable the GCC undefined behavior sanitizer. This is disabled by
37053 default, but passing @code{--enable-ubsan=yes} or
37054 @code{--enable-ubsan=auto} to @code{configure} will enable it. The
37055 undefined behavior sanitizer checks for C@t{++} undefined behavior.
37056 It has a performance cost, so if you are looking at @value{GDBN}'s
37057 performance, you should disable it. The undefined behavior sanitizer
37058 was first introduced in GCC 4.9.
37061 @node System-wide configuration
37062 @section System-wide configuration and settings
37063 @cindex system-wide init file
37065 @value{GDBN} can be configured to have a system-wide init file;
37066 this file will be read and executed at startup (@pxref{Startup, , What
37067 @value{GDBN} does during startup}).
37069 Here is the corresponding configure option:
37072 @item --with-system-gdbinit=@var{file}
37073 Specify that the default location of the system-wide init file is
37077 If @value{GDBN} has been configured with the option @option{--prefix=$prefix},
37078 it may be subject to relocation. Two possible cases:
37082 If the default location of this init file contains @file{$prefix},
37083 it will be subject to relocation. Suppose that the configure options
37084 are @option{--prefix=$prefix --with-system-gdbinit=$prefix/etc/gdbinit};
37085 if @value{GDBN} is moved from @file{$prefix} to @file{$install}, the system
37086 init file is looked for as @file{$install/etc/gdbinit} instead of
37087 @file{$prefix/etc/gdbinit}.
37090 By contrast, if the default location does not contain the prefix,
37091 it will not be relocated. E.g.@: if @value{GDBN} has been configured with
37092 @option{--prefix=/usr/local --with-system-gdbinit=/usr/share/gdb/gdbinit},
37093 then @value{GDBN} will always look for @file{/usr/share/gdb/gdbinit},
37094 wherever @value{GDBN} is installed.
37097 If the configured location of the system-wide init file (as given by the
37098 @option{--with-system-gdbinit} option at configure time) is in the
37099 data-directory (as specified by @option{--with-gdb-datadir} at configure
37100 time) or in one of its subdirectories, then @value{GDBN} will look for the
37101 system-wide init file in the directory specified by the
37102 @option{--data-directory} command-line option.
37103 Note that the system-wide init file is only read once, during @value{GDBN}
37104 initialization. If the data-directory is changed after @value{GDBN} has
37105 started with the @code{set data-directory} command, the file will not be
37109 * System-wide Configuration Scripts:: Installed System-wide Configuration Scripts
37112 @node System-wide Configuration Scripts
37113 @subsection Installed System-wide Configuration Scripts
37114 @cindex system-wide configuration scripts
37116 The @file{system-gdbinit} directory, located inside the data-directory
37117 (as specified by @option{--with-gdb-datadir} at configure time) contains
37118 a number of scripts which can be used as system-wide init files. To
37119 automatically source those scripts at startup, @value{GDBN} should be
37120 configured with @option{--with-system-gdbinit}. Otherwise, any user
37121 should be able to source them by hand as needed.
37123 The following scripts are currently available:
37126 @item @file{elinos.py}
37128 @cindex ELinOS system-wide configuration script
37129 This script is useful when debugging a program on an ELinOS target.
37130 It takes advantage of the environment variables defined in a standard
37131 ELinOS environment in order to determine the location of the system
37132 shared libraries, and then sets the @samp{solib-absolute-prefix}
37133 and @samp{solib-search-path} variables appropriately.
37135 @item @file{wrs-linux.py}
37136 @pindex wrs-linux.py
37137 @cindex Wind River Linux system-wide configuration script
37138 This script is useful when debugging a program on a target running
37139 Wind River Linux. It expects the @env{ENV_PREFIX} to be set to
37140 the host-side sysroot used by the target system.
37144 @node Maintenance Commands
37145 @appendix Maintenance Commands
37146 @cindex maintenance commands
37147 @cindex internal commands
37149 In addition to commands intended for @value{GDBN} users, @value{GDBN}
37150 includes a number of commands intended for @value{GDBN} developers,
37151 that are not documented elsewhere in this manual. These commands are
37152 provided here for reference. (For commands that turn on debugging
37153 messages, see @ref{Debugging Output}.)
37156 @kindex maint agent
37157 @kindex maint agent-eval
37158 @item maint agent @r{[}-at @var{location}@r{,}@r{]} @var{expression}
37159 @itemx maint agent-eval @r{[}-at @var{location}@r{,}@r{]} @var{expression}
37160 Translate the given @var{expression} into remote agent bytecodes.
37161 This command is useful for debugging the Agent Expression mechanism
37162 (@pxref{Agent Expressions}). The @samp{agent} version produces an
37163 expression useful for data collection, such as by tracepoints, while
37164 @samp{maint agent-eval} produces an expression that evaluates directly
37165 to a result. For instance, a collection expression for @code{globa +
37166 globb} will include bytecodes to record four bytes of memory at each
37167 of the addresses of @code{globa} and @code{globb}, while discarding
37168 the result of the addition, while an evaluation expression will do the
37169 addition and return the sum.
37170 If @code{-at} is given, generate remote agent bytecode for @var{location}.
37171 If not, generate remote agent bytecode for current frame PC address.
37173 @kindex maint agent-printf
37174 @item maint agent-printf @var{format},@var{expr},...
37175 Translate the given format string and list of argument expressions
37176 into remote agent bytecodes and display them as a disassembled list.
37177 This command is useful for debugging the agent version of dynamic
37178 printf (@pxref{Dynamic Printf}).
37180 @kindex maint info breakpoints
37181 @item @anchor{maint info breakpoints}maint info breakpoints
37182 Using the same format as @samp{info breakpoints}, display both the
37183 breakpoints you've set explicitly, and those @value{GDBN} is using for
37184 internal purposes. Internal breakpoints are shown with negative
37185 breakpoint numbers. The type column identifies what kind of breakpoint
37190 Normal, explicitly set breakpoint.
37193 Normal, explicitly set watchpoint.
37196 Internal breakpoint, used to handle correctly stepping through
37197 @code{longjmp} calls.
37199 @item longjmp resume
37200 Internal breakpoint at the target of a @code{longjmp}.
37203 Temporary internal breakpoint used by the @value{GDBN} @code{until} command.
37206 Temporary internal breakpoint used by the @value{GDBN} @code{finish} command.
37209 Shared library events.
37213 @kindex maint info btrace
37214 @item maint info btrace
37215 Pint information about raw branch tracing data.
37217 @kindex maint btrace packet-history
37218 @item maint btrace packet-history
37219 Print the raw branch trace packets that are used to compute the
37220 execution history for the @samp{record btrace} command. Both the
37221 information and the format in which it is printed depend on the btrace
37226 For the BTS recording format, print a list of blocks of sequential
37227 code. For each block, the following information is printed:
37231 Newer blocks have higher numbers. The oldest block has number zero.
37232 @item Lowest @samp{PC}
37233 @item Highest @samp{PC}
37237 For the Intel Processor Trace recording format, print a list of
37238 Intel Processor Trace packets. For each packet, the following
37239 information is printed:
37242 @item Packet number
37243 Newer packets have higher numbers. The oldest packet has number zero.
37245 The packet's offset in the trace stream.
37246 @item Packet opcode and payload
37250 @kindex maint btrace clear-packet-history
37251 @item maint btrace clear-packet-history
37252 Discards the cached packet history printed by the @samp{maint btrace
37253 packet-history} command. The history will be computed again when
37256 @kindex maint btrace clear
37257 @item maint btrace clear
37258 Discard the branch trace data. The data will be fetched anew and the
37259 branch trace will be recomputed when needed.
37261 This implicitly truncates the branch trace to a single branch trace
37262 buffer. When updating branch trace incrementally, the branch trace
37263 available to @value{GDBN} may be bigger than a single branch trace
37266 @kindex maint set btrace pt skip-pad
37267 @item maint set btrace pt skip-pad
37268 @kindex maint show btrace pt skip-pad
37269 @item maint show btrace pt skip-pad
37270 Control whether @value{GDBN} will skip PAD packets when computing the
37273 @kindex set displaced-stepping
37274 @kindex show displaced-stepping
37275 @cindex displaced stepping support
37276 @cindex out-of-line single-stepping
37277 @item set displaced-stepping
37278 @itemx show displaced-stepping
37279 Control whether or not @value{GDBN} will do @dfn{displaced stepping}
37280 if the target supports it. Displaced stepping is a way to single-step
37281 over breakpoints without removing them from the inferior, by executing
37282 an out-of-line copy of the instruction that was originally at the
37283 breakpoint location. It is also known as out-of-line single-stepping.
37286 @item set displaced-stepping on
37287 If the target architecture supports it, @value{GDBN} will use
37288 displaced stepping to step over breakpoints.
37290 @item set displaced-stepping off
37291 @value{GDBN} will not use displaced stepping to step over breakpoints,
37292 even if such is supported by the target architecture.
37294 @cindex non-stop mode, and @samp{set displaced-stepping}
37295 @item set displaced-stepping auto
37296 This is the default mode. @value{GDBN} will use displaced stepping
37297 only if non-stop mode is active (@pxref{Non-Stop Mode}) and the target
37298 architecture supports displaced stepping.
37301 @kindex maint check-psymtabs
37302 @item maint check-psymtabs
37303 Check the consistency of currently expanded psymtabs versus symtabs.
37304 Use this to check, for example, whether a symbol is in one but not the other.
37306 @kindex maint check-symtabs
37307 @item maint check-symtabs
37308 Check the consistency of currently expanded symtabs.
37310 @kindex maint expand-symtabs
37311 @item maint expand-symtabs [@var{regexp}]
37312 Expand symbol tables.
37313 If @var{regexp} is specified, only expand symbol tables for file
37314 names matching @var{regexp}.
37316 @kindex maint set catch-demangler-crashes
37317 @kindex maint show catch-demangler-crashes
37318 @cindex demangler crashes
37319 @item maint set catch-demangler-crashes [on|off]
37320 @itemx maint show catch-demangler-crashes
37321 Control whether @value{GDBN} should attempt to catch crashes in the
37322 symbol name demangler. The default is to attempt to catch crashes.
37323 If enabled, the first time a crash is caught, a core file is created,
37324 the offending symbol is displayed and the user is presented with the
37325 option to terminate the current session.
37327 @kindex maint cplus first_component
37328 @item maint cplus first_component @var{name}
37329 Print the first C@t{++} class/namespace component of @var{name}.
37331 @kindex maint cplus namespace
37332 @item maint cplus namespace
37333 Print the list of possible C@t{++} namespaces.
37335 @kindex maint deprecate
37336 @kindex maint undeprecate
37337 @cindex deprecated commands
37338 @item maint deprecate @var{command} @r{[}@var{replacement}@r{]}
37339 @itemx maint undeprecate @var{command}
37340 Deprecate or undeprecate the named @var{command}. Deprecated commands
37341 cause @value{GDBN} to issue a warning when you use them. The optional
37342 argument @var{replacement} says which newer command should be used in
37343 favor of the deprecated one; if it is given, @value{GDBN} will mention
37344 the replacement as part of the warning.
37346 @kindex maint dump-me
37347 @item maint dump-me
37348 @cindex @code{SIGQUIT} signal, dump core of @value{GDBN}
37349 Cause a fatal signal in the debugger and force it to dump its core.
37350 This is supported only on systems which support aborting a program
37351 with the @code{SIGQUIT} signal.
37353 @kindex maint internal-error
37354 @kindex maint internal-warning
37355 @kindex maint demangler-warning
37356 @cindex demangler crashes
37357 @item maint internal-error @r{[}@var{message-text}@r{]}
37358 @itemx maint internal-warning @r{[}@var{message-text}@r{]}
37359 @itemx maint demangler-warning @r{[}@var{message-text}@r{]}
37361 Cause @value{GDBN} to call the internal function @code{internal_error},
37362 @code{internal_warning} or @code{demangler_warning} and hence behave
37363 as though an internal problem has been detected. In addition to
37364 reporting the internal problem, these functions give the user the
37365 opportunity to either quit @value{GDBN} or (for @code{internal_error}
37366 and @code{internal_warning}) create a core file of the current
37367 @value{GDBN} session.
37369 These commands take an optional parameter @var{message-text} that is
37370 used as the text of the error or warning message.
37372 Here's an example of using @code{internal-error}:
37375 (@value{GDBP}) @kbd{maint internal-error testing, 1, 2}
37376 @dots{}/maint.c:121: internal-error: testing, 1, 2
37377 A problem internal to GDB has been detected. Further
37378 debugging may prove unreliable.
37379 Quit this debugging session? (y or n) @kbd{n}
37380 Create a core file? (y or n) @kbd{n}
37384 @cindex @value{GDBN} internal error
37385 @cindex internal errors, control of @value{GDBN} behavior
37386 @cindex demangler crashes
37388 @kindex maint set internal-error
37389 @kindex maint show internal-error
37390 @kindex maint set internal-warning
37391 @kindex maint show internal-warning
37392 @kindex maint set demangler-warning
37393 @kindex maint show demangler-warning
37394 @item maint set internal-error @var{action} [ask|yes|no]
37395 @itemx maint show internal-error @var{action}
37396 @itemx maint set internal-warning @var{action} [ask|yes|no]
37397 @itemx maint show internal-warning @var{action}
37398 @itemx maint set demangler-warning @var{action} [ask|yes|no]
37399 @itemx maint show demangler-warning @var{action}
37400 When @value{GDBN} reports an internal problem (error or warning) it
37401 gives the user the opportunity to both quit @value{GDBN} and create a
37402 core file of the current @value{GDBN} session. These commands let you
37403 override the default behaviour for each particular @var{action},
37404 described in the table below.
37408 You can specify that @value{GDBN} should always (yes) or never (no)
37409 quit. The default is to ask the user what to do.
37412 You can specify that @value{GDBN} should always (yes) or never (no)
37413 create a core file. The default is to ask the user what to do. Note
37414 that there is no @code{corefile} option for @code{demangler-warning}:
37415 demangler warnings always create a core file and this cannot be
37419 @kindex maint packet
37420 @item maint packet @var{text}
37421 If @value{GDBN} is talking to an inferior via the serial protocol,
37422 then this command sends the string @var{text} to the inferior, and
37423 displays the response packet. @value{GDBN} supplies the initial
37424 @samp{$} character, the terminating @samp{#} character, and the
37427 @kindex maint print architecture
37428 @item maint print architecture @r{[}@var{file}@r{]}
37429 Print the entire architecture configuration. The optional argument
37430 @var{file} names the file where the output goes.
37432 @kindex maint print c-tdesc @r{[}@var{file}@r{]}
37433 @item maint print c-tdesc
37434 Print the target description (@pxref{Target Descriptions}) as
37435 a C source file. By default, the target description is for the current
37436 target, but if the optional argument @var{file} is provided, that file
37437 is used to produce the description. The @var{file} should be an XML
37438 document, of the form described in @ref{Target Description Format}.
37439 The created source file is built into @value{GDBN} when @value{GDBN} is
37440 built again. This command is used by developers after they add or
37441 modify XML target descriptions.
37443 @kindex maint check xml-descriptions
37444 @item maint check xml-descriptions @var{dir}
37445 Check that the target descriptions dynamically created by @value{GDBN}
37446 equal the descriptions created from XML files found in @var{dir}.
37448 @anchor{maint check libthread-db}
37449 @kindex maint check libthread-db
37450 @item maint check libthread-db
37451 Run integrity checks on the current inferior's thread debugging
37452 library. This exercises all @code{libthread_db} functionality used by
37453 @value{GDBN} on GNU/Linux systems, and by extension also exercises the
37454 @code{proc_service} functions provided by @value{GDBN} that
37455 @code{libthread_db} uses. Note that parts of the test may be skipped
37456 on some platforms when debugging core files.
37458 @kindex maint print dummy-frames
37459 @item maint print dummy-frames
37460 Prints the contents of @value{GDBN}'s internal dummy-frame stack.
37463 (@value{GDBP}) @kbd{b add}
37465 (@value{GDBP}) @kbd{print add(2,3)}
37466 Breakpoint 2, add (a=2, b=3) at @dots{}
37468 The program being debugged stopped while in a function called from GDB.
37470 (@value{GDBP}) @kbd{maint print dummy-frames}
37471 0xa8206d8: id=@{stack=0xbfffe734,code=0xbfffe73f,!special@}, ptid=process 9353
37475 Takes an optional file parameter.
37477 @kindex maint print registers
37478 @kindex maint print raw-registers
37479 @kindex maint print cooked-registers
37480 @kindex maint print register-groups
37481 @kindex maint print remote-registers
37482 @item maint print registers @r{[}@var{file}@r{]}
37483 @itemx maint print raw-registers @r{[}@var{file}@r{]}
37484 @itemx maint print cooked-registers @r{[}@var{file}@r{]}
37485 @itemx maint print register-groups @r{[}@var{file}@r{]}
37486 @itemx maint print remote-registers @r{[}@var{file}@r{]}
37487 Print @value{GDBN}'s internal register data structures.
37489 The command @code{maint print raw-registers} includes the contents of
37490 the raw register cache; the command @code{maint print
37491 cooked-registers} includes the (cooked) value of all registers,
37492 including registers which aren't available on the target nor visible
37493 to user; the command @code{maint print register-groups} includes the
37494 groups that each register is a member of; and the command @code{maint
37495 print remote-registers} includes the remote target's register numbers
37496 and offsets in the `G' packets.
37498 These commands take an optional parameter, a file name to which to
37499 write the information.
37501 @kindex maint print reggroups
37502 @item maint print reggroups @r{[}@var{file}@r{]}
37503 Print @value{GDBN}'s internal register group data structures. The
37504 optional argument @var{file} tells to what file to write the
37507 The register groups info looks like this:
37510 (@value{GDBP}) @kbd{maint print reggroups}
37523 This command forces @value{GDBN} to flush its internal register cache.
37525 @kindex maint print objfiles
37526 @cindex info for known object files
37527 @item maint print objfiles @r{[}@var{regexp}@r{]}
37528 Print a dump of all known object files.
37529 If @var{regexp} is specified, only print object files whose names
37530 match @var{regexp}. For each object file, this command prints its name,
37531 address in memory, and all of its psymtabs and symtabs.
37533 @kindex maint print user-registers
37534 @cindex user registers
37535 @item maint print user-registers
37536 List all currently available @dfn{user registers}. User registers
37537 typically provide alternate names for actual hardware registers. They
37538 include the four ``standard'' registers @code{$fp}, @code{$pc},
37539 @code{$sp}, and @code{$ps}. @xref{standard registers}. User
37540 registers can be used in expressions in the same way as the canonical
37541 register names, but only the latter are listed by the @code{info
37542 registers} and @code{maint print registers} commands.
37544 @kindex maint print section-scripts
37545 @cindex info for known .debug_gdb_scripts-loaded scripts
37546 @item maint print section-scripts [@var{regexp}]
37547 Print a dump of scripts specified in the @code{.debug_gdb_section} section.
37548 If @var{regexp} is specified, only print scripts loaded by object files
37549 matching @var{regexp}.
37550 For each script, this command prints its name as specified in the objfile,
37551 and the full path if known.
37552 @xref{dotdebug_gdb_scripts section}.
37554 @kindex maint print statistics
37555 @cindex bcache statistics
37556 @item maint print statistics
37557 This command prints, for each object file in the program, various data
37558 about that object file followed by the byte cache (@dfn{bcache})
37559 statistics for the object file. The objfile data includes the number
37560 of minimal, partial, full, and stabs symbols, the number of types
37561 defined by the objfile, the number of as yet unexpanded psym tables,
37562 the number of line tables and string tables, and the amount of memory
37563 used by the various tables. The bcache statistics include the counts,
37564 sizes, and counts of duplicates of all and unique objects, max,
37565 average, and median entry size, total memory used and its overhead and
37566 savings, and various measures of the hash table size and chain
37569 @kindex maint print target-stack
37570 @cindex target stack description
37571 @item maint print target-stack
37572 A @dfn{target} is an interface between the debugger and a particular
37573 kind of file or process. Targets can be stacked in @dfn{strata},
37574 so that more than one target can potentially respond to a request.
37575 In particular, memory accesses will walk down the stack of targets
37576 until they find a target that is interested in handling that particular
37579 This command prints a short description of each layer that was pushed on
37580 the @dfn{target stack}, starting from the top layer down to the bottom one.
37582 @kindex maint print type
37583 @cindex type chain of a data type
37584 @item maint print type @var{expr}
37585 Print the type chain for a type specified by @var{expr}. The argument
37586 can be either a type name or a symbol. If it is a symbol, the type of
37587 that symbol is described. The type chain produced by this command is
37588 a recursive definition of the data type as stored in @value{GDBN}'s
37589 data structures, including its flags and contained types.
37591 @kindex maint selftest
37593 @item maint selftest @r{[}@var{filter}@r{]}
37594 Run any self tests that were compiled in to @value{GDBN}. This will
37595 print a message showing how many tests were run, and how many failed.
37596 If a @var{filter} is passed, only the tests with @var{filter} in their
37599 @kindex maint info selftests
37601 @item maint info selftests
37602 List the selftests compiled in to @value{GDBN}.
37604 @kindex maint set dwarf always-disassemble
37605 @kindex maint show dwarf always-disassemble
37606 @item maint set dwarf always-disassemble
37607 @item maint show dwarf always-disassemble
37608 Control the behavior of @code{info address} when using DWARF debugging
37611 The default is @code{off}, which means that @value{GDBN} should try to
37612 describe a variable's location in an easily readable format. When
37613 @code{on}, @value{GDBN} will instead display the DWARF location
37614 expression in an assembly-like format. Note that some locations are
37615 too complex for @value{GDBN} to describe simply; in this case you will
37616 always see the disassembly form.
37618 Here is an example of the resulting disassembly:
37621 (gdb) info addr argc
37622 Symbol "argc" is a complex DWARF expression:
37626 For more information on these expressions, see
37627 @uref{http://www.dwarfstd.org/, the DWARF standard}.
37629 @kindex maint set dwarf max-cache-age
37630 @kindex maint show dwarf max-cache-age
37631 @item maint set dwarf max-cache-age
37632 @itemx maint show dwarf max-cache-age
37633 Control the DWARF compilation unit cache.
37635 @cindex DWARF compilation units cache
37636 In object files with inter-compilation-unit references, such as those
37637 produced by the GCC option @samp{-feliminate-dwarf2-dups}, the DWARF
37638 reader needs to frequently refer to previously read compilation units.
37639 This setting controls how long a compilation unit will remain in the
37640 cache if it is not referenced. A higher limit means that cached
37641 compilation units will be stored in memory longer, and more total
37642 memory will be used. Setting it to zero disables caching, which will
37643 slow down @value{GDBN} startup, but reduce memory consumption.
37645 @kindex maint set dwarf unwinders
37646 @kindex maint show dwarf unwinders
37647 @item maint set dwarf unwinders
37648 @itemx maint show dwarf unwinders
37649 Control use of the DWARF frame unwinders.
37651 @cindex DWARF frame unwinders
37652 Many targets that support DWARF debugging use @value{GDBN}'s DWARF
37653 frame unwinders to build the backtrace. Many of these targets will
37654 also have a second mechanism for building the backtrace for use in
37655 cases where DWARF information is not available, this second mechanism
37656 is often an analysis of a function's prologue.
37658 In order to extend testing coverage of the second level stack
37659 unwinding mechanisms it is helpful to be able to disable the DWARF
37660 stack unwinders, this can be done with this switch.
37662 In normal use of @value{GDBN} disabling the DWARF unwinders is not
37663 advisable, there are cases that are better handled through DWARF than
37664 prologue analysis, and the debug experience is likely to be better
37665 with the DWARF frame unwinders enabled.
37667 If DWARF frame unwinders are not supported for a particular target
37668 architecture, then enabling this flag does not cause them to be used.
37669 @kindex maint set profile
37670 @kindex maint show profile
37671 @cindex profiling GDB
37672 @item maint set profile
37673 @itemx maint show profile
37674 Control profiling of @value{GDBN}.
37676 Profiling will be disabled until you use the @samp{maint set profile}
37677 command to enable it. When you enable profiling, the system will begin
37678 collecting timing and execution count data; when you disable profiling or
37679 exit @value{GDBN}, the results will be written to a log file. Remember that
37680 if you use profiling, @value{GDBN} will overwrite the profiling log file
37681 (often called @file{gmon.out}). If you have a record of important profiling
37682 data in a @file{gmon.out} file, be sure to move it to a safe location.
37684 Configuring with @samp{--enable-profiling} arranges for @value{GDBN} to be
37685 compiled with the @samp{-pg} compiler option.
37687 @kindex maint set show-debug-regs
37688 @kindex maint show show-debug-regs
37689 @cindex hardware debug registers
37690 @item maint set show-debug-regs
37691 @itemx maint show show-debug-regs
37692 Control whether to show variables that mirror the hardware debug
37693 registers. Use @code{on} to enable, @code{off} to disable. If
37694 enabled, the debug registers values are shown when @value{GDBN} inserts or
37695 removes a hardware breakpoint or watchpoint, and when the inferior
37696 triggers a hardware-assisted breakpoint or watchpoint.
37698 @kindex maint set show-all-tib
37699 @kindex maint show show-all-tib
37700 @item maint set show-all-tib
37701 @itemx maint show show-all-tib
37702 Control whether to show all non zero areas within a 1k block starting
37703 at thread local base, when using the @samp{info w32 thread-information-block}
37706 @kindex maint set target-async
37707 @kindex maint show target-async
37708 @item maint set target-async
37709 @itemx maint show target-async
37710 This controls whether @value{GDBN} targets operate in synchronous or
37711 asynchronous mode (@pxref{Background Execution}). Normally the
37712 default is asynchronous, if it is available; but this can be changed
37713 to more easily debug problems occurring only in synchronous mode.
37715 @kindex maint set target-non-stop @var{mode} [on|off|auto]
37716 @kindex maint show target-non-stop
37717 @item maint set target-non-stop
37718 @itemx maint show target-non-stop
37720 This controls whether @value{GDBN} targets always operate in non-stop
37721 mode even if @code{set non-stop} is @code{off} (@pxref{Non-Stop
37722 Mode}). The default is @code{auto}, meaning non-stop mode is enabled
37723 if supported by the target.
37726 @item maint set target-non-stop auto
37727 This is the default mode. @value{GDBN} controls the target in
37728 non-stop mode if the target supports it.
37730 @item maint set target-non-stop on
37731 @value{GDBN} controls the target in non-stop mode even if the target
37732 does not indicate support.
37734 @item maint set target-non-stop off
37735 @value{GDBN} does not control the target in non-stop mode even if the
37736 target supports it.
37739 @kindex maint set per-command
37740 @kindex maint show per-command
37741 @item maint set per-command
37742 @itemx maint show per-command
37743 @cindex resources used by commands
37745 @value{GDBN} can display the resources used by each command.
37746 This is useful in debugging performance problems.
37749 @item maint set per-command space [on|off]
37750 @itemx maint show per-command space
37751 Enable or disable the printing of the memory used by GDB for each command.
37752 If enabled, @value{GDBN} will display how much memory each command
37753 took, following the command's own output.
37754 This can also be requested by invoking @value{GDBN} with the
37755 @option{--statistics} command-line switch (@pxref{Mode Options}).
37757 @item maint set per-command time [on|off]
37758 @itemx maint show per-command time
37759 Enable or disable the printing of the execution time of @value{GDBN}
37761 If enabled, @value{GDBN} will display how much time it
37762 took to execute each command, following the command's own output.
37763 Both CPU time and wallclock time are printed.
37764 Printing both is useful when trying to determine whether the cost is
37765 CPU or, e.g., disk/network latency.
37766 Note that the CPU time printed is for @value{GDBN} only, it does not include
37767 the execution time of the inferior because there's no mechanism currently
37768 to compute how much time was spent by @value{GDBN} and how much time was
37769 spent by the program been debugged.
37770 This can also be requested by invoking @value{GDBN} with the
37771 @option{--statistics} command-line switch (@pxref{Mode Options}).
37773 @item maint set per-command symtab [on|off]
37774 @itemx maint show per-command symtab
37775 Enable or disable the printing of basic symbol table statistics
37777 If enabled, @value{GDBN} will display the following information:
37781 number of symbol tables
37783 number of primary symbol tables
37785 number of blocks in the blockvector
37789 @kindex maint set check-libthread-db
37790 @kindex maint show check-libthread-db
37791 @item maint set check-libthread-db [on|off]
37792 @itemx maint show check-libthread-db
37793 Control whether @value{GDBN} should run integrity checks on inferior
37794 specific thread debugging libraries as they are loaded. The default
37795 is not to perform such checks. If any check fails @value{GDBN} will
37796 unload the library and continue searching for a suitable candidate as
37797 described in @ref{set libthread-db-search-path}. For more information
37798 about the tests, see @ref{maint check libthread-db}.
37800 @kindex maint space
37801 @cindex memory used by commands
37802 @item maint space @var{value}
37803 An alias for @code{maint set per-command space}.
37804 A non-zero value enables it, zero disables it.
37807 @cindex time of command execution
37808 @item maint time @var{value}
37809 An alias for @code{maint set per-command time}.
37810 A non-zero value enables it, zero disables it.
37812 @kindex maint translate-address
37813 @item maint translate-address @r{[}@var{section}@r{]} @var{addr}
37814 Find the symbol stored at the location specified by the address
37815 @var{addr} and an optional section name @var{section}. If found,
37816 @value{GDBN} prints the name of the closest symbol and an offset from
37817 the symbol's location to the specified address. This is similar to
37818 the @code{info address} command (@pxref{Symbols}), except that this
37819 command also allows to find symbols in other sections.
37821 If section was not specified, the section in which the symbol was found
37822 is also printed. For dynamically linked executables, the name of
37823 executable or shared library containing the symbol is printed as well.
37825 @kindex maint test-options
37826 @item maint test-options require-delimiter
37827 @itemx maint test-options unknown-is-error
37828 @itemx maint test-options unknown-is-operand
37829 These commands are used by the testsuite to validate the command
37830 options framework. The @code{require-delimiter} variant requires a
37831 double-dash delimiter to indicate end of options. The
37832 @code{unknown-is-error} and @code{unknown-is-operand} do not. The
37833 @code{unknown-is-error} variant throws an error on unknown option,
37834 while @code{unknown-is-operand} treats unknown options as the start of
37835 the command's operands. When run, the commands output the result of
37836 the processed options. When completed, the commands store the
37837 internal result of completion in a variable exposed by the @code{maint
37838 show test-options-completion-result} command.
37840 @kindex maint show test-options-completion-result
37841 @item maint show test-options-completion-result
37842 Shows the result of completing the @code{maint test-options}
37843 subcommands. This is used by the testsuite to validate completion
37844 support in the command options framework.
37846 @kindex maint set test-settings
37847 @kindex maint show test-settings
37848 @item maint set test-settings @var{kind}
37849 @itemx maint show test-settings @var{kind}
37850 These are representative commands for each @var{kind} of setting type
37851 @value{GDBN} supports. They are used by the testsuite for exercising
37852 the settings infrastructure.
37855 @item maint with @var{setting} [@var{value}] [-- @var{command}]
37856 Like the @code{with} command, but works with @code{maintenance set}
37857 variables. This is used by the testsuite to exercise the @code{with}
37858 command's infrastructure.
37862 The following command is useful for non-interactive invocations of
37863 @value{GDBN}, such as in the test suite.
37866 @item set watchdog @var{nsec}
37867 @kindex set watchdog
37868 @cindex watchdog timer
37869 @cindex timeout for commands
37870 Set the maximum number of seconds @value{GDBN} will wait for the
37871 target operation to finish. If this time expires, @value{GDBN}
37872 reports and error and the command is aborted.
37874 @item show watchdog
37875 Show the current setting of the target wait timeout.
37878 @node Remote Protocol
37879 @appendix @value{GDBN} Remote Serial Protocol
37884 * Stop Reply Packets::
37885 * General Query Packets::
37886 * Architecture-Specific Protocol Details::
37887 * Tracepoint Packets::
37888 * Host I/O Packets::
37890 * Notification Packets::
37891 * Remote Non-Stop::
37892 * Packet Acknowledgment::
37894 * File-I/O Remote Protocol Extension::
37895 * Library List Format::
37896 * Library List Format for SVR4 Targets::
37897 * Memory Map Format::
37898 * Thread List Format::
37899 * Traceframe Info Format::
37900 * Branch Trace Format::
37901 * Branch Trace Configuration Format::
37907 There may be occasions when you need to know something about the
37908 protocol---for example, if there is only one serial port to your target
37909 machine, you might want your program to do something special if it
37910 recognizes a packet meant for @value{GDBN}.
37912 In the examples below, @samp{->} and @samp{<-} are used to indicate
37913 transmitted and received data, respectively.
37915 @cindex protocol, @value{GDBN} remote serial
37916 @cindex serial protocol, @value{GDBN} remote
37917 @cindex remote serial protocol
37918 All @value{GDBN} commands and responses (other than acknowledgments
37919 and notifications, see @ref{Notification Packets}) are sent as a
37920 @var{packet}. A @var{packet} is introduced with the character
37921 @samp{$}, the actual @var{packet-data}, and the terminating character
37922 @samp{#} followed by a two-digit @var{checksum}:
37925 @code{$}@var{packet-data}@code{#}@var{checksum}
37929 @cindex checksum, for @value{GDBN} remote
37931 The two-digit @var{checksum} is computed as the modulo 256 sum of all
37932 characters between the leading @samp{$} and the trailing @samp{#} (an
37933 eight bit unsigned checksum).
37935 Implementors should note that prior to @value{GDBN} 5.0 the protocol
37936 specification also included an optional two-digit @var{sequence-id}:
37939 @code{$}@var{sequence-id}@code{:}@var{packet-data}@code{#}@var{checksum}
37942 @cindex sequence-id, for @value{GDBN} remote
37944 That @var{sequence-id} was appended to the acknowledgment. @value{GDBN}
37945 has never output @var{sequence-id}s. Stubs that handle packets added
37946 since @value{GDBN} 5.0 must not accept @var{sequence-id}.
37948 When either the host or the target machine receives a packet, the first
37949 response expected is an acknowledgment: either @samp{+} (to indicate
37950 the package was received correctly) or @samp{-} (to request
37954 -> @code{$}@var{packet-data}@code{#}@var{checksum}
37959 The @samp{+}/@samp{-} acknowledgments can be disabled
37960 once a connection is established.
37961 @xref{Packet Acknowledgment}, for details.
37963 The host (@value{GDBN}) sends @var{command}s, and the target (the
37964 debugging stub incorporated in your program) sends a @var{response}. In
37965 the case of step and continue @var{command}s, the response is only sent
37966 when the operation has completed, and the target has again stopped all
37967 threads in all attached processes. This is the default all-stop mode
37968 behavior, but the remote protocol also supports @value{GDBN}'s non-stop
37969 execution mode; see @ref{Remote Non-Stop}, for details.
37971 @var{packet-data} consists of a sequence of characters with the
37972 exception of @samp{#} and @samp{$} (see @samp{X} packet for additional
37975 @cindex remote protocol, field separator
37976 Fields within the packet should be separated using @samp{,} @samp{;} or
37977 @samp{:}. Except where otherwise noted all numbers are represented in
37978 @sc{hex} with leading zeros suppressed.
37980 Implementors should note that prior to @value{GDBN} 5.0, the character
37981 @samp{:} could not appear as the third character in a packet (as it
37982 would potentially conflict with the @var{sequence-id}).
37984 @cindex remote protocol, binary data
37985 @anchor{Binary Data}
37986 Binary data in most packets is encoded either as two hexadecimal
37987 digits per byte of binary data. This allowed the traditional remote
37988 protocol to work over connections which were only seven-bit clean.
37989 Some packets designed more recently assume an eight-bit clean
37990 connection, and use a more efficient encoding to send and receive
37993 The binary data representation uses @code{7d} (@sc{ascii} @samp{@}})
37994 as an escape character. Any escaped byte is transmitted as the escape
37995 character followed by the original character XORed with @code{0x20}.
37996 For example, the byte @code{0x7d} would be transmitted as the two
37997 bytes @code{0x7d 0x5d}. The bytes @code{0x23} (@sc{ascii} @samp{#}),
37998 @code{0x24} (@sc{ascii} @samp{$}), and @code{0x7d} (@sc{ascii}
37999 @samp{@}}) must always be escaped. Responses sent by the stub
38000 must also escape @code{0x2a} (@sc{ascii} @samp{*}), so that it
38001 is not interpreted as the start of a run-length encoded sequence
38004 Response @var{data} can be run-length encoded to save space.
38005 Run-length encoding replaces runs of identical characters with one
38006 instance of the repeated character, followed by a @samp{*} and a
38007 repeat count. The repeat count is itself sent encoded, to avoid
38008 binary characters in @var{data}: a value of @var{n} is sent as
38009 @code{@var{n}+29}. For a repeat count greater or equal to 3, this
38010 produces a printable @sc{ascii} character, e.g.@: a space (@sc{ascii}
38011 code 32) for a repeat count of 3. (This is because run-length
38012 encoding starts to win for counts 3 or more.) Thus, for example,
38013 @samp{0* } is a run-length encoding of ``0000'': the space character
38014 after @samp{*} means repeat the leading @code{0} @w{@code{32 - 29 =
38017 The printable characters @samp{#} and @samp{$} or with a numeric value
38018 greater than 126 must not be used. Runs of six repeats (@samp{#}) or
38019 seven repeats (@samp{$}) can be expanded using a repeat count of only
38020 five (@samp{"}). For example, @samp{00000000} can be encoded as
38023 The error response returned for some packets includes a two character
38024 error number. That number is not well defined.
38026 @cindex empty response, for unsupported packets
38027 For any @var{command} not supported by the stub, an empty response
38028 (@samp{$#00}) should be returned. That way it is possible to extend the
38029 protocol. A newer @value{GDBN} can tell if a packet is supported based
38032 At a minimum, a stub is required to support the @samp{g} and @samp{G}
38033 commands for register access, and the @samp{m} and @samp{M} commands
38034 for memory access. Stubs that only control single-threaded targets
38035 can implement run control with the @samp{c} (continue), and @samp{s}
38036 (step) commands. Stubs that support multi-threading targets should
38037 support the @samp{vCont} command. All other commands are optional.
38042 The following table provides a complete list of all currently defined
38043 @var{command}s and their corresponding response @var{data}.
38044 @xref{File-I/O Remote Protocol Extension}, for details about the File
38045 I/O extension of the remote protocol.
38047 Each packet's description has a template showing the packet's overall
38048 syntax, followed by an explanation of the packet's meaning. We
38049 include spaces in some of the templates for clarity; these are not
38050 part of the packet's syntax. No @value{GDBN} packet uses spaces to
38051 separate its components. For example, a template like @samp{foo
38052 @var{bar} @var{baz}} describes a packet beginning with the three ASCII
38053 bytes @samp{foo}, followed by a @var{bar}, followed directly by a
38054 @var{baz}. @value{GDBN} does not transmit a space character between the
38055 @samp{foo} and the @var{bar}, or between the @var{bar} and the
38058 @cindex @var{thread-id}, in remote protocol
38059 @anchor{thread-id syntax}
38060 Several packets and replies include a @var{thread-id} field to identify
38061 a thread. Normally these are positive numbers with a target-specific
38062 interpretation, formatted as big-endian hex strings. A @var{thread-id}
38063 can also be a literal @samp{-1} to indicate all threads, or @samp{0} to
38066 In addition, the remote protocol supports a multiprocess feature in
38067 which the @var{thread-id} syntax is extended to optionally include both
38068 process and thread ID fields, as @samp{p@var{pid}.@var{tid}}.
38069 The @var{pid} (process) and @var{tid} (thread) components each have the
38070 format described above: a positive number with target-specific
38071 interpretation formatted as a big-endian hex string, literal @samp{-1}
38072 to indicate all processes or threads (respectively), or @samp{0} to
38073 indicate an arbitrary process or thread. Specifying just a process, as
38074 @samp{p@var{pid}}, is equivalent to @samp{p@var{pid}.-1}. It is an
38075 error to specify all processes but a specific thread, such as
38076 @samp{p-1.@var{tid}}. Note that the @samp{p} prefix is @emph{not} used
38077 for those packets and replies explicitly documented to include a process
38078 ID, rather than a @var{thread-id}.
38080 The multiprocess @var{thread-id} syntax extensions are only used if both
38081 @value{GDBN} and the stub report support for the @samp{multiprocess}
38082 feature using @samp{qSupported}. @xref{multiprocess extensions}, for
38085 Note that all packet forms beginning with an upper- or lower-case
38086 letter, other than those described here, are reserved for future use.
38088 Here are the packet descriptions.
38093 @cindex @samp{!} packet
38094 @anchor{extended mode}
38095 Enable extended mode. In extended mode, the remote server is made
38096 persistent. The @samp{R} packet is used to restart the program being
38102 The remote target both supports and has enabled extended mode.
38106 @cindex @samp{?} packet
38108 Indicate the reason the target halted. The reply is the same as for
38109 step and continue. This packet has a special interpretation when the
38110 target is in non-stop mode; see @ref{Remote Non-Stop}.
38113 @xref{Stop Reply Packets}, for the reply specifications.
38115 @item A @var{arglen},@var{argnum},@var{arg},@dots{}
38116 @cindex @samp{A} packet
38117 Initialized @code{argv[]} array passed into program. @var{arglen}
38118 specifies the number of bytes in the hex encoded byte stream
38119 @var{arg}. See @code{gdbserver} for more details.
38124 The arguments were set.
38130 @cindex @samp{b} packet
38131 (Don't use this packet; its behavior is not well-defined.)
38132 Change the serial line speed to @var{baud}.
38134 JTC: @emph{When does the transport layer state change? When it's
38135 received, or after the ACK is transmitted. In either case, there are
38136 problems if the command or the acknowledgment packet is dropped.}
38138 Stan: @emph{If people really wanted to add something like this, and get
38139 it working for the first time, they ought to modify ser-unix.c to send
38140 some kind of out-of-band message to a specially-setup stub and have the
38141 switch happen "in between" packets, so that from remote protocol's point
38142 of view, nothing actually happened.}
38144 @item B @var{addr},@var{mode}
38145 @cindex @samp{B} packet
38146 Set (@var{mode} is @samp{S}) or clear (@var{mode} is @samp{C}) a
38147 breakpoint at @var{addr}.
38149 Don't use this packet. Use the @samp{Z} and @samp{z} packets instead
38150 (@pxref{insert breakpoint or watchpoint packet}).
38152 @cindex @samp{bc} packet
38155 Backward continue. Execute the target system in reverse. No parameter.
38156 @xref{Reverse Execution}, for more information.
38159 @xref{Stop Reply Packets}, for the reply specifications.
38161 @cindex @samp{bs} packet
38164 Backward single step. Execute one instruction in reverse. No parameter.
38165 @xref{Reverse Execution}, for more information.
38168 @xref{Stop Reply Packets}, for the reply specifications.
38170 @item c @r{[}@var{addr}@r{]}
38171 @cindex @samp{c} packet
38172 Continue at @var{addr}, which is the address to resume. If @var{addr}
38173 is omitted, resume at current address.
38175 This packet is deprecated for multi-threading support. @xref{vCont
38179 @xref{Stop Reply Packets}, for the reply specifications.
38181 @item C @var{sig}@r{[};@var{addr}@r{]}
38182 @cindex @samp{C} packet
38183 Continue with signal @var{sig} (hex signal number). If
38184 @samp{;@var{addr}} is omitted, resume at same address.
38186 This packet is deprecated for multi-threading support. @xref{vCont
38190 @xref{Stop Reply Packets}, for the reply specifications.
38193 @cindex @samp{d} packet
38196 Don't use this packet; instead, define a general set packet
38197 (@pxref{General Query Packets}).
38201 @cindex @samp{D} packet
38202 The first form of the packet is used to detach @value{GDBN} from the
38203 remote system. It is sent to the remote target
38204 before @value{GDBN} disconnects via the @code{detach} command.
38206 The second form, including a process ID, is used when multiprocess
38207 protocol extensions are enabled (@pxref{multiprocess extensions}), to
38208 detach only a specific process. The @var{pid} is specified as a
38209 big-endian hex string.
38219 @item F @var{RC},@var{EE},@var{CF};@var{XX}
38220 @cindex @samp{F} packet
38221 A reply from @value{GDBN} to an @samp{F} packet sent by the target.
38222 This is part of the File-I/O protocol extension. @xref{File-I/O
38223 Remote Protocol Extension}, for the specification.
38226 @anchor{read registers packet}
38227 @cindex @samp{g} packet
38228 Read general registers.
38232 @item @var{XX@dots{}}
38233 Each byte of register data is described by two hex digits. The bytes
38234 with the register are transmitted in target byte order. The size of
38235 each register and their position within the @samp{g} packet are
38236 determined by the @value{GDBN} internal gdbarch functions
38237 @code{DEPRECATED_REGISTER_RAW_SIZE} and @code{gdbarch_register_name}.
38239 When reading registers from a trace frame (@pxref{Analyze Collected
38240 Data,,Using the Collected Data}), the stub may also return a string of
38241 literal @samp{x}'s in place of the register data digits, to indicate
38242 that the corresponding register has not been collected, thus its value
38243 is unavailable. For example, for an architecture with 4 registers of
38244 4 bytes each, the following reply indicates to @value{GDBN} that
38245 registers 0 and 2 have not been collected, while registers 1 and 3
38246 have been collected, and both have zero value:
38250 <- @code{xxxxxxxx00000000xxxxxxxx00000000}
38257 @item G @var{XX@dots{}}
38258 @cindex @samp{G} packet
38259 Write general registers. @xref{read registers packet}, for a
38260 description of the @var{XX@dots{}} data.
38270 @item H @var{op} @var{thread-id}
38271 @cindex @samp{H} packet
38272 Set thread for subsequent operations (@samp{m}, @samp{M}, @samp{g},
38273 @samp{G}, et.al.). Depending on the operation to be performed, @var{op}
38274 should be @samp{c} for step and continue operations (note that this
38275 is deprecated, supporting the @samp{vCont} command is a better
38276 option), and @samp{g} for other operations. The thread designator
38277 @var{thread-id} has the format and interpretation described in
38278 @ref{thread-id syntax}.
38289 @c 'H': How restrictive (or permissive) is the thread model. If a
38290 @c thread is selected and stopped, are other threads allowed
38291 @c to continue to execute? As I mentioned above, I think the
38292 @c semantics of each command when a thread is selected must be
38293 @c described. For example:
38295 @c 'g': If the stub supports threads and a specific thread is
38296 @c selected, returns the register block from that thread;
38297 @c otherwise returns current registers.
38299 @c 'G' If the stub supports threads and a specific thread is
38300 @c selected, sets the registers of the register block of
38301 @c that thread; otherwise sets current registers.
38303 @item i @r{[}@var{addr}@r{[},@var{nnn}@r{]]}
38304 @anchor{cycle step packet}
38305 @cindex @samp{i} packet
38306 Step the remote target by a single clock cycle. If @samp{,@var{nnn}} is
38307 present, cycle step @var{nnn} cycles. If @var{addr} is present, cycle
38308 step starting at that address.
38311 @cindex @samp{I} packet
38312 Signal, then cycle step. @xref{step with signal packet}. @xref{cycle
38316 @cindex @samp{k} packet
38319 The exact effect of this packet is not specified.
38321 For a bare-metal target, it may power cycle or reset the target
38322 system. For that reason, the @samp{k} packet has no reply.
38324 For a single-process target, it may kill that process if possible.
38326 A multiple-process target may choose to kill just one process, or all
38327 that are under @value{GDBN}'s control. For more precise control, use
38328 the vKill packet (@pxref{vKill packet}).
38330 If the target system immediately closes the connection in response to
38331 @samp{k}, @value{GDBN} does not consider the lack of packet
38332 acknowledgment to be an error, and assumes the kill was successful.
38334 If connected using @kbd{target extended-remote}, and the target does
38335 not close the connection in response to a kill request, @value{GDBN}
38336 probes the target state as if a new connection was opened
38337 (@pxref{? packet}).
38339 @item m @var{addr},@var{length}
38340 @cindex @samp{m} packet
38341 Read @var{length} addressable memory units starting at address @var{addr}
38342 (@pxref{addressable memory unit}). Note that @var{addr} may not be aligned to
38343 any particular boundary.
38345 The stub need not use any particular size or alignment when gathering
38346 data from memory for the response; even if @var{addr} is word-aligned
38347 and @var{length} is a multiple of the word size, the stub is free to
38348 use byte accesses, or not. For this reason, this packet may not be
38349 suitable for accessing memory-mapped I/O devices.
38350 @cindex alignment of remote memory accesses
38351 @cindex size of remote memory accesses
38352 @cindex memory, alignment and size of remote accesses
38356 @item @var{XX@dots{}}
38357 Memory contents; each byte is transmitted as a two-digit hexadecimal number.
38358 The reply may contain fewer addressable memory units than requested if the
38359 server was able to read only part of the region of memory.
38364 @item M @var{addr},@var{length}:@var{XX@dots{}}
38365 @cindex @samp{M} packet
38366 Write @var{length} addressable memory units starting at address @var{addr}
38367 (@pxref{addressable memory unit}). The data is given by @var{XX@dots{}}; each
38368 byte is transmitted as a two-digit hexadecimal number.
38375 for an error (this includes the case where only part of the data was
38380 @cindex @samp{p} packet
38381 Read the value of register @var{n}; @var{n} is in hex.
38382 @xref{read registers packet}, for a description of how the returned
38383 register value is encoded.
38387 @item @var{XX@dots{}}
38388 the register's value
38392 Indicating an unrecognized @var{query}.
38395 @item P @var{n@dots{}}=@var{r@dots{}}
38396 @anchor{write register packet}
38397 @cindex @samp{P} packet
38398 Write register @var{n@dots{}} with value @var{r@dots{}}. The register
38399 number @var{n} is in hexadecimal, and @var{r@dots{}} contains two hex
38400 digits for each byte in the register (target byte order).
38410 @item q @var{name} @var{params}@dots{}
38411 @itemx Q @var{name} @var{params}@dots{}
38412 @cindex @samp{q} packet
38413 @cindex @samp{Q} packet
38414 General query (@samp{q}) and set (@samp{Q}). These packets are
38415 described fully in @ref{General Query Packets}.
38418 @cindex @samp{r} packet
38419 Reset the entire system.
38421 Don't use this packet; use the @samp{R} packet instead.
38424 @cindex @samp{R} packet
38425 Restart the program being debugged. The @var{XX}, while needed, is ignored.
38426 This packet is only available in extended mode (@pxref{extended mode}).
38428 The @samp{R} packet has no reply.
38430 @item s @r{[}@var{addr}@r{]}
38431 @cindex @samp{s} packet
38432 Single step, resuming at @var{addr}. If
38433 @var{addr} is omitted, resume at same address.
38435 This packet is deprecated for multi-threading support. @xref{vCont
38439 @xref{Stop Reply Packets}, for the reply specifications.
38441 @item S @var{sig}@r{[};@var{addr}@r{]}
38442 @anchor{step with signal packet}
38443 @cindex @samp{S} packet
38444 Step with signal. This is analogous to the @samp{C} packet, but
38445 requests a single-step, rather than a normal resumption of execution.
38447 This packet is deprecated for multi-threading support. @xref{vCont
38451 @xref{Stop Reply Packets}, for the reply specifications.
38453 @item t @var{addr}:@var{PP},@var{MM}
38454 @cindex @samp{t} packet
38455 Search backwards starting at address @var{addr} for a match with pattern
38456 @var{PP} and mask @var{MM}, both of which are are 4 byte long.
38457 There must be at least 3 digits in @var{addr}.
38459 @item T @var{thread-id}
38460 @cindex @samp{T} packet
38461 Find out if the thread @var{thread-id} is alive. @xref{thread-id syntax}.
38466 thread is still alive
38472 Packets starting with @samp{v} are identified by a multi-letter name,
38473 up to the first @samp{;} or @samp{?} (or the end of the packet).
38475 @item vAttach;@var{pid}
38476 @cindex @samp{vAttach} packet
38477 Attach to a new process with the specified process ID @var{pid}.
38478 The process ID is a
38479 hexadecimal integer identifying the process. In all-stop mode, all
38480 threads in the attached process are stopped; in non-stop mode, it may be
38481 attached without being stopped if that is supported by the target.
38483 @c In non-stop mode, on a successful vAttach, the stub should set the
38484 @c current thread to a thread of the newly-attached process. After
38485 @c attaching, GDB queries for the attached process's thread ID with qC.
38486 @c Also note that, from a user perspective, whether or not the
38487 @c target is stopped on attach in non-stop mode depends on whether you
38488 @c use the foreground or background version of the attach command, not
38489 @c on what vAttach does; GDB does the right thing with respect to either
38490 @c stopping or restarting threads.
38492 This packet is only available in extended mode (@pxref{extended mode}).
38498 @item @r{Any stop packet}
38499 for success in all-stop mode (@pxref{Stop Reply Packets})
38501 for success in non-stop mode (@pxref{Remote Non-Stop})
38504 @item vCont@r{[};@var{action}@r{[}:@var{thread-id}@r{]]}@dots{}
38505 @cindex @samp{vCont} packet
38506 @anchor{vCont packet}
38507 Resume the inferior, specifying different actions for each thread.
38509 For each inferior thread, the leftmost action with a matching
38510 @var{thread-id} is applied. Threads that don't match any action
38511 remain in their current state. Thread IDs are specified using the
38512 syntax described in @ref{thread-id syntax}. If multiprocess
38513 extensions (@pxref{multiprocess extensions}) are supported, actions
38514 can be specified to match all threads in a process by using the
38515 @samp{p@var{pid}.-1} form of the @var{thread-id}. An action with no
38516 @var{thread-id} matches all threads. Specifying no actions is an
38519 Currently supported actions are:
38525 Continue with signal @var{sig}. The signal @var{sig} should be two hex digits.
38529 Step with signal @var{sig}. The signal @var{sig} should be two hex digits.
38532 @item r @var{start},@var{end}
38533 Step once, and then keep stepping as long as the thread stops at
38534 addresses between @var{start} (inclusive) and @var{end} (exclusive).
38535 The remote stub reports a stop reply when either the thread goes out
38536 of the range or is stopped due to an unrelated reason, such as hitting
38537 a breakpoint. @xref{range stepping}.
38539 If the range is empty (@var{start} == @var{end}), then the action
38540 becomes equivalent to the @samp{s} action. In other words,
38541 single-step once, and report the stop (even if the stepped instruction
38542 jumps to @var{start}).
38544 (A stop reply may be sent at any point even if the PC is still within
38545 the stepping range; for example, it is valid to implement this packet
38546 in a degenerate way as a single instruction step operation.)
38550 The optional argument @var{addr} normally associated with the
38551 @samp{c}, @samp{C}, @samp{s}, and @samp{S} packets is
38552 not supported in @samp{vCont}.
38554 The @samp{t} action is only relevant in non-stop mode
38555 (@pxref{Remote Non-Stop}) and may be ignored by the stub otherwise.
38556 A stop reply should be generated for any affected thread not already stopped.
38557 When a thread is stopped by means of a @samp{t} action,
38558 the corresponding stop reply should indicate that the thread has stopped with
38559 signal @samp{0}, regardless of whether the target uses some other signal
38560 as an implementation detail.
38562 The server must ignore @samp{c}, @samp{C}, @samp{s}, @samp{S}, and
38563 @samp{r} actions for threads that are already running. Conversely,
38564 the server must ignore @samp{t} actions for threads that are already
38567 @emph{Note:} In non-stop mode, a thread is considered running until
38568 @value{GDBN} acknowleges an asynchronous stop notification for it with
38569 the @samp{vStopped} packet (@pxref{Remote Non-Stop}).
38571 The stub must support @samp{vCont} if it reports support for
38572 multiprocess extensions (@pxref{multiprocess extensions}).
38575 @xref{Stop Reply Packets}, for the reply specifications.
38578 @cindex @samp{vCont?} packet
38579 Request a list of actions supported by the @samp{vCont} packet.
38583 @item vCont@r{[};@var{action}@dots{}@r{]}
38584 The @samp{vCont} packet is supported. Each @var{action} is a supported
38585 command in the @samp{vCont} packet.
38587 The @samp{vCont} packet is not supported.
38590 @anchor{vCtrlC packet}
38592 @cindex @samp{vCtrlC} packet
38593 Interrupt remote target as if a control-C was pressed on the remote
38594 terminal. This is the equivalent to reacting to the @code{^C}
38595 (@samp{\003}, the control-C character) character in all-stop mode
38596 while the target is running, except this works in non-stop mode.
38597 @xref{interrupting remote targets}, for more info on the all-stop
38608 @item vFile:@var{operation}:@var{parameter}@dots{}
38609 @cindex @samp{vFile} packet
38610 Perform a file operation on the target system. For details,
38611 see @ref{Host I/O Packets}.
38613 @item vFlashErase:@var{addr},@var{length}
38614 @cindex @samp{vFlashErase} packet
38615 Direct the stub to erase @var{length} bytes of flash starting at
38616 @var{addr}. The region may enclose any number of flash blocks, but
38617 its start and end must fall on block boundaries, as indicated by the
38618 flash block size appearing in the memory map (@pxref{Memory Map
38619 Format}). @value{GDBN} groups flash memory programming operations
38620 together, and sends a @samp{vFlashDone} request after each group; the
38621 stub is allowed to delay erase operation until the @samp{vFlashDone}
38622 packet is received.
38632 @item vFlashWrite:@var{addr}:@var{XX@dots{}}
38633 @cindex @samp{vFlashWrite} packet
38634 Direct the stub to write data to flash address @var{addr}. The data
38635 is passed in binary form using the same encoding as for the @samp{X}
38636 packet (@pxref{Binary Data}). The memory ranges specified by
38637 @samp{vFlashWrite} packets preceding a @samp{vFlashDone} packet must
38638 not overlap, and must appear in order of increasing addresses
38639 (although @samp{vFlashErase} packets for higher addresses may already
38640 have been received; the ordering is guaranteed only between
38641 @samp{vFlashWrite} packets). If a packet writes to an address that was
38642 neither erased by a preceding @samp{vFlashErase} packet nor by some other
38643 target-specific method, the results are unpredictable.
38651 for vFlashWrite addressing non-flash memory
38657 @cindex @samp{vFlashDone} packet
38658 Indicate to the stub that flash programming operation is finished.
38659 The stub is permitted to delay or batch the effects of a group of
38660 @samp{vFlashErase} and @samp{vFlashWrite} packets until a
38661 @samp{vFlashDone} packet is received. The contents of the affected
38662 regions of flash memory are unpredictable until the @samp{vFlashDone}
38663 request is completed.
38665 @item vKill;@var{pid}
38666 @cindex @samp{vKill} packet
38667 @anchor{vKill packet}
38668 Kill the process with the specified process ID @var{pid}, which is a
38669 hexadecimal integer identifying the process. This packet is used in
38670 preference to @samp{k} when multiprocess protocol extensions are
38671 supported; see @ref{multiprocess extensions}.
38681 @item vMustReplyEmpty
38682 @cindex @samp{vMustReplyEmpty} packet
38683 The correct reply to an unknown @samp{v} packet is to return the empty
38684 string, however, some older versions of @command{gdbserver} would
38685 incorrectly return @samp{OK} for unknown @samp{v} packets.
38687 The @samp{vMustReplyEmpty} is used as a feature test to check how
38688 @command{gdbserver} handles unknown packets, it is important that this
38689 packet be handled in the same way as other unknown @samp{v} packets.
38690 If this packet is handled differently to other unknown @samp{v}
38691 packets then it is possile that @value{GDBN} may run into problems in
38692 other areas, specifically around use of @samp{vFile:setfs:}.
38694 @item vRun;@var{filename}@r{[};@var{argument}@r{]}@dots{}
38695 @cindex @samp{vRun} packet
38696 Run the program @var{filename}, passing it each @var{argument} on its
38697 command line. The file and arguments are hex-encoded strings. If
38698 @var{filename} is an empty string, the stub may use a default program
38699 (e.g.@: the last program run). The program is created in the stopped
38702 @c FIXME: What about non-stop mode?
38704 This packet is only available in extended mode (@pxref{extended mode}).
38710 @item @r{Any stop packet}
38711 for success (@pxref{Stop Reply Packets})
38715 @cindex @samp{vStopped} packet
38716 @xref{Notification Packets}.
38718 @item X @var{addr},@var{length}:@var{XX@dots{}}
38720 @cindex @samp{X} packet
38721 Write data to memory, where the data is transmitted in binary.
38722 Memory is specified by its address @var{addr} and number of addressable memory
38723 units @var{length} (@pxref{addressable memory unit});
38724 @samp{@var{XX}@dots{}} is binary data (@pxref{Binary Data}).
38734 @item z @var{type},@var{addr},@var{kind}
38735 @itemx Z @var{type},@var{addr},@var{kind}
38736 @anchor{insert breakpoint or watchpoint packet}
38737 @cindex @samp{z} packet
38738 @cindex @samp{Z} packets
38739 Insert (@samp{Z}) or remove (@samp{z}) a @var{type} breakpoint or
38740 watchpoint starting at address @var{address} of kind @var{kind}.
38742 Each breakpoint and watchpoint packet @var{type} is documented
38745 @emph{Implementation notes: A remote target shall return an empty string
38746 for an unrecognized breakpoint or watchpoint packet @var{type}. A
38747 remote target shall support either both or neither of a given
38748 @samp{Z@var{type}@dots{}} and @samp{z@var{type}@dots{}} packet pair. To
38749 avoid potential problems with duplicate packets, the operations should
38750 be implemented in an idempotent way.}
38752 @item z0,@var{addr},@var{kind}
38753 @itemx Z0,@var{addr},@var{kind}@r{[};@var{cond_list}@dots{}@r{]}@r{[};cmds:@var{persist},@var{cmd_list}@dots{}@r{]}
38754 @cindex @samp{z0} packet
38755 @cindex @samp{Z0} packet
38756 Insert (@samp{Z0}) or remove (@samp{z0}) a software breakpoint at address
38757 @var{addr} of type @var{kind}.
38759 A software breakpoint is implemented by replacing the instruction at
38760 @var{addr} with a software breakpoint or trap instruction. The
38761 @var{kind} is target-specific and typically indicates the size of the
38762 breakpoint in bytes that should be inserted. E.g., the @sc{arm} and
38763 @sc{mips} can insert either a 2 or 4 byte breakpoint. Some
38764 architectures have additional meanings for @var{kind}
38765 (@pxref{Architecture-Specific Protocol Details}); if no
38766 architecture-specific value is being used, it should be @samp{0}.
38767 @var{kind} is hex-encoded. @var{cond_list} is an optional list of
38768 conditional expressions in bytecode form that should be evaluated on
38769 the target's side. These are the conditions that should be taken into
38770 consideration when deciding if the breakpoint trigger should be
38771 reported back to @value{GDBN}.
38773 See also the @samp{swbreak} stop reason (@pxref{swbreak stop reason})
38774 for how to best report a software breakpoint event to @value{GDBN}.
38776 The @var{cond_list} parameter is comprised of a series of expressions,
38777 concatenated without separators. Each expression has the following form:
38781 @item X @var{len},@var{expr}
38782 @var{len} is the length of the bytecode expression and @var{expr} is the
38783 actual conditional expression in bytecode form.
38787 The optional @var{cmd_list} parameter introduces commands that may be
38788 run on the target, rather than being reported back to @value{GDBN}.
38789 The parameter starts with a numeric flag @var{persist}; if the flag is
38790 nonzero, then the breakpoint may remain active and the commands
38791 continue to be run even when @value{GDBN} disconnects from the target.
38792 Following this flag is a series of expressions concatenated with no
38793 separators. Each expression has the following form:
38797 @item X @var{len},@var{expr}
38798 @var{len} is the length of the bytecode expression and @var{expr} is the
38799 actual commands expression in bytecode form.
38803 @emph{Implementation note: It is possible for a target to copy or move
38804 code that contains software breakpoints (e.g., when implementing
38805 overlays). The behavior of this packet, in the presence of such a
38806 target, is not defined.}
38818 @item z1,@var{addr},@var{kind}
38819 @itemx Z1,@var{addr},@var{kind}@r{[};@var{cond_list}@dots{}@r{]}@r{[};cmds:@var{persist},@var{cmd_list}@dots{}@r{]}
38820 @cindex @samp{z1} packet
38821 @cindex @samp{Z1} packet
38822 Insert (@samp{Z1}) or remove (@samp{z1}) a hardware breakpoint at
38823 address @var{addr}.
38825 A hardware breakpoint is implemented using a mechanism that is not
38826 dependent on being able to modify the target's memory. The
38827 @var{kind}, @var{cond_list}, and @var{cmd_list} arguments have the
38828 same meaning as in @samp{Z0} packets.
38830 @emph{Implementation note: A hardware breakpoint is not affected by code
38843 @item z2,@var{addr},@var{kind}
38844 @itemx Z2,@var{addr},@var{kind}
38845 @cindex @samp{z2} packet
38846 @cindex @samp{Z2} packet
38847 Insert (@samp{Z2}) or remove (@samp{z2}) a write watchpoint at @var{addr}.
38848 The number of bytes to watch is specified by @var{kind}.
38860 @item z3,@var{addr},@var{kind}
38861 @itemx Z3,@var{addr},@var{kind}
38862 @cindex @samp{z3} packet
38863 @cindex @samp{Z3} packet
38864 Insert (@samp{Z3}) or remove (@samp{z3}) a read watchpoint at @var{addr}.
38865 The number of bytes to watch is specified by @var{kind}.
38877 @item z4,@var{addr},@var{kind}
38878 @itemx Z4,@var{addr},@var{kind}
38879 @cindex @samp{z4} packet
38880 @cindex @samp{Z4} packet
38881 Insert (@samp{Z4}) or remove (@samp{z4}) an access watchpoint at @var{addr}.
38882 The number of bytes to watch is specified by @var{kind}.
38896 @node Stop Reply Packets
38897 @section Stop Reply Packets
38898 @cindex stop reply packets
38900 The @samp{C}, @samp{c}, @samp{S}, @samp{s}, @samp{vCont},
38901 @samp{vAttach}, @samp{vRun}, @samp{vStopped}, and @samp{?} packets can
38902 receive any of the below as a reply. Except for @samp{?}
38903 and @samp{vStopped}, that reply is only returned
38904 when the target halts. In the below the exact meaning of @dfn{signal
38905 number} is defined by the header @file{include/gdb/signals.h} in the
38906 @value{GDBN} source code.
38908 In non-stop mode, the server will simply reply @samp{OK} to commands
38909 such as @samp{vCont}; any stop will be the subject of a future
38910 notification. @xref{Remote Non-Stop}.
38912 As in the description of request packets, we include spaces in the
38913 reply templates for clarity; these are not part of the reply packet's
38914 syntax. No @value{GDBN} stop reply packet uses spaces to separate its
38920 The program received signal number @var{AA} (a two-digit hexadecimal
38921 number). This is equivalent to a @samp{T} response with no
38922 @var{n}:@var{r} pairs.
38924 @item T @var{AA} @var{n1}:@var{r1};@var{n2}:@var{r2};@dots{}
38925 @cindex @samp{T} packet reply
38926 The program received signal number @var{AA} (a two-digit hexadecimal
38927 number). This is equivalent to an @samp{S} response, except that the
38928 @samp{@var{n}:@var{r}} pairs can carry values of important registers
38929 and other information directly in the stop reply packet, reducing
38930 round-trip latency. Single-step and breakpoint traps are reported
38931 this way. Each @samp{@var{n}:@var{r}} pair is interpreted as follows:
38935 If @var{n} is a hexadecimal number, it is a register number, and the
38936 corresponding @var{r} gives that register's value. The data @var{r} is a
38937 series of bytes in target byte order, with each byte given by a
38938 two-digit hex number.
38941 If @var{n} is @samp{thread}, then @var{r} is the @var{thread-id} of
38942 the stopped thread, as specified in @ref{thread-id syntax}.
38945 If @var{n} is @samp{core}, then @var{r} is the hexadecimal number of
38946 the core on which the stop event was detected.
38949 If @var{n} is a recognized @dfn{stop reason}, it describes a more
38950 specific event that stopped the target. The currently defined stop
38951 reasons are listed below. The @var{aa} should be @samp{05}, the trap
38952 signal. At most one stop reason should be present.
38955 Otherwise, @value{GDBN} should ignore this @samp{@var{n}:@var{r}} pair
38956 and go on to the next; this allows us to extend the protocol in the
38960 The currently defined stop reasons are:
38966 The packet indicates a watchpoint hit, and @var{r} is the data address, in
38969 @item syscall_entry
38970 @itemx syscall_return
38971 The packet indicates a syscall entry or return, and @var{r} is the
38972 syscall number, in hex.
38974 @cindex shared library events, remote reply
38976 The packet indicates that the loaded libraries have changed.
38977 @value{GDBN} should use @samp{qXfer:libraries:read} to fetch a new
38978 list of loaded libraries. The @var{r} part is ignored.
38980 @cindex replay log events, remote reply
38982 The packet indicates that the target cannot continue replaying
38983 logged execution events, because it has reached the end (or the
38984 beginning when executing backward) of the log. The value of @var{r}
38985 will be either @samp{begin} or @samp{end}. @xref{Reverse Execution},
38986 for more information.
38989 @anchor{swbreak stop reason}
38990 The packet indicates a software breakpoint instruction was executed,
38991 irrespective of whether it was @value{GDBN} that planted the
38992 breakpoint or the breakpoint is hardcoded in the program. The @var{r}
38993 part must be left empty.
38995 On some architectures, such as x86, at the architecture level, when a
38996 breakpoint instruction executes the program counter points at the
38997 breakpoint address plus an offset. On such targets, the stub is
38998 responsible for adjusting the PC to point back at the breakpoint
39001 This packet should not be sent by default; older @value{GDBN} versions
39002 did not support it. @value{GDBN} requests it, by supplying an
39003 appropriate @samp{qSupported} feature (@pxref{qSupported}). The
39004 remote stub must also supply the appropriate @samp{qSupported} feature
39005 indicating support.
39007 This packet is required for correct non-stop mode operation.
39010 The packet indicates the target stopped for a hardware breakpoint.
39011 The @var{r} part must be left empty.
39013 The same remarks about @samp{qSupported} and non-stop mode above
39016 @cindex fork events, remote reply
39018 The packet indicates that @code{fork} was called, and @var{r}
39019 is the thread ID of the new child process. Refer to
39020 @ref{thread-id syntax} for the format of the @var{thread-id}
39021 field. This packet is only applicable to targets that support
39024 This packet should not be sent by default; older @value{GDBN} versions
39025 did not support it. @value{GDBN} requests it, by supplying an
39026 appropriate @samp{qSupported} feature (@pxref{qSupported}). The
39027 remote stub must also supply the appropriate @samp{qSupported} feature
39028 indicating support.
39030 @cindex vfork events, remote reply
39032 The packet indicates that @code{vfork} was called, and @var{r}
39033 is the thread ID of the new child process. Refer to
39034 @ref{thread-id syntax} for the format of the @var{thread-id}
39035 field. This packet is only applicable to targets that support
39038 This packet should not be sent by default; older @value{GDBN} versions
39039 did not support it. @value{GDBN} requests it, by supplying an
39040 appropriate @samp{qSupported} feature (@pxref{qSupported}). The
39041 remote stub must also supply the appropriate @samp{qSupported} feature
39042 indicating support.
39044 @cindex vforkdone events, remote reply
39046 The packet indicates that a child process created by a vfork
39047 has either called @code{exec} or terminated, so that the
39048 address spaces of the parent and child process are no longer
39049 shared. The @var{r} part is ignored. This packet is only
39050 applicable to targets that support vforkdone events.
39052 This packet should not be sent by default; older @value{GDBN} versions
39053 did not support it. @value{GDBN} requests it, by supplying an
39054 appropriate @samp{qSupported} feature (@pxref{qSupported}). The
39055 remote stub must also supply the appropriate @samp{qSupported} feature
39056 indicating support.
39058 @cindex exec events, remote reply
39060 The packet indicates that @code{execve} was called, and @var{r}
39061 is the absolute pathname of the file that was executed, in hex.
39062 This packet is only applicable to targets that support exec events.
39064 This packet should not be sent by default; older @value{GDBN} versions
39065 did not support it. @value{GDBN} requests it, by supplying an
39066 appropriate @samp{qSupported} feature (@pxref{qSupported}). The
39067 remote stub must also supply the appropriate @samp{qSupported} feature
39068 indicating support.
39070 @cindex thread create event, remote reply
39071 @anchor{thread create event}
39073 The packet indicates that the thread was just created. The new thread
39074 is stopped until @value{GDBN} sets it running with a resumption packet
39075 (@pxref{vCont packet}). This packet should not be sent by default;
39076 @value{GDBN} requests it with the @ref{QThreadEvents} packet. See
39077 also the @samp{w} (@pxref{thread exit event}) remote reply below. The
39078 @var{r} part is ignored.
39083 @itemx W @var{AA} ; process:@var{pid}
39084 The process exited, and @var{AA} is the exit status. This is only
39085 applicable to certain targets.
39087 The second form of the response, including the process ID of the
39088 exited process, can be used only when @value{GDBN} has reported
39089 support for multiprocess protocol extensions; see @ref{multiprocess
39090 extensions}. Both @var{AA} and @var{pid} are formatted as big-endian
39094 @itemx X @var{AA} ; process:@var{pid}
39095 The process terminated with signal @var{AA}.
39097 The second form of the response, including the process ID of the
39098 terminated process, can be used only when @value{GDBN} has reported
39099 support for multiprocess protocol extensions; see @ref{multiprocess
39100 extensions}. Both @var{AA} and @var{pid} are formatted as big-endian
39103 @anchor{thread exit event}
39104 @cindex thread exit event, remote reply
39105 @item w @var{AA} ; @var{tid}
39107 The thread exited, and @var{AA} is the exit status. This response
39108 should not be sent by default; @value{GDBN} requests it with the
39109 @ref{QThreadEvents} packet. See also @ref{thread create event} above.
39110 @var{AA} is formatted as a big-endian hex string.
39113 There are no resumed threads left in the target. In other words, even
39114 though the process is alive, the last resumed thread has exited. For
39115 example, say the target process has two threads: thread 1 and thread
39116 2. The client leaves thread 1 stopped, and resumes thread 2, which
39117 subsequently exits. At this point, even though the process is still
39118 alive, and thus no @samp{W} stop reply is sent, no thread is actually
39119 executing either. The @samp{N} stop reply thus informs the client
39120 that it can stop waiting for stop replies. This packet should not be
39121 sent by default; older @value{GDBN} versions did not support it.
39122 @value{GDBN} requests it, by supplying an appropriate
39123 @samp{qSupported} feature (@pxref{qSupported}). The remote stub must
39124 also supply the appropriate @samp{qSupported} feature indicating
39127 @item O @var{XX}@dots{}
39128 @samp{@var{XX}@dots{}} is hex encoding of @sc{ascii} data, to be
39129 written as the program's console output. This can happen at any time
39130 while the program is running and the debugger should continue to wait
39131 for @samp{W}, @samp{T}, etc. This reply is not permitted in non-stop mode.
39133 @item F @var{call-id},@var{parameter}@dots{}
39134 @var{call-id} is the identifier which says which host system call should
39135 be called. This is just the name of the function. Translation into the
39136 correct system call is only applicable as it's defined in @value{GDBN}.
39137 @xref{File-I/O Remote Protocol Extension}, for a list of implemented
39140 @samp{@var{parameter}@dots{}} is a list of parameters as defined for
39141 this very system call.
39143 The target replies with this packet when it expects @value{GDBN} to
39144 call a host system call on behalf of the target. @value{GDBN} replies
39145 with an appropriate @samp{F} packet and keeps up waiting for the next
39146 reply packet from the target. The latest @samp{C}, @samp{c}, @samp{S}
39147 or @samp{s} action is expected to be continued. @xref{File-I/O Remote
39148 Protocol Extension}, for more details.
39152 @node General Query Packets
39153 @section General Query Packets
39154 @cindex remote query requests
39156 Packets starting with @samp{q} are @dfn{general query packets};
39157 packets starting with @samp{Q} are @dfn{general set packets}. General
39158 query and set packets are a semi-unified form for retrieving and
39159 sending information to and from the stub.
39161 The initial letter of a query or set packet is followed by a name
39162 indicating what sort of thing the packet applies to. For example,
39163 @value{GDBN} may use a @samp{qSymbol} packet to exchange symbol
39164 definitions with the stub. These packet names follow some
39169 The name must not contain commas, colons or semicolons.
39171 Most @value{GDBN} query and set packets have a leading upper case
39174 The names of custom vendor packets should use a company prefix, in
39175 lower case, followed by a period. For example, packets designed at
39176 the Acme Corporation might begin with @samp{qacme.foo} (for querying
39177 foos) or @samp{Qacme.bar} (for setting bars).
39180 The name of a query or set packet should be separated from any
39181 parameters by a @samp{:}; the parameters themselves should be
39182 separated by @samp{,} or @samp{;}. Stubs must be careful to match the
39183 full packet name, and check for a separator or the end of the packet,
39184 in case two packet names share a common prefix. New packets should not begin
39185 with @samp{qC}, @samp{qP}, or @samp{qL}@footnote{The @samp{qP} and @samp{qL}
39186 packets predate these conventions, and have arguments without any terminator
39187 for the packet name; we suspect they are in widespread use in places that
39188 are difficult to upgrade. The @samp{qC} packet has no arguments, but some
39189 existing stubs (e.g.@: RedBoot) are known to not check for the end of the
39192 Like the descriptions of the other packets, each description here
39193 has a template showing the packet's overall syntax, followed by an
39194 explanation of the packet's meaning. We include spaces in some of the
39195 templates for clarity; these are not part of the packet's syntax. No
39196 @value{GDBN} packet uses spaces to separate its components.
39198 Here are the currently defined query and set packets:
39204 Turn on or off the agent as a helper to perform some debugging operations
39205 delegated from @value{GDBN} (@pxref{Control Agent}).
39207 @item QAllow:@var{op}:@var{val}@dots{}
39208 @cindex @samp{QAllow} packet
39209 Specify which operations @value{GDBN} expects to request of the
39210 target, as a semicolon-separated list of operation name and value
39211 pairs. Possible values for @var{op} include @samp{WriteReg},
39212 @samp{WriteMem}, @samp{InsertBreak}, @samp{InsertTrace},
39213 @samp{InsertFastTrace}, and @samp{Stop}. @var{val} is either 0,
39214 indicating that @value{GDBN} will not request the operation, or 1,
39215 indicating that it may. (The target can then use this to set up its
39216 own internals optimally, for instance if the debugger never expects to
39217 insert breakpoints, it may not need to install its own trap handler.)
39220 @cindex current thread, remote request
39221 @cindex @samp{qC} packet
39222 Return the current thread ID.
39226 @item QC @var{thread-id}
39227 Where @var{thread-id} is a thread ID as documented in
39228 @ref{thread-id syntax}.
39229 @item @r{(anything else)}
39230 Any other reply implies the old thread ID.
39233 @item qCRC:@var{addr},@var{length}
39234 @cindex CRC of memory block, remote request
39235 @cindex @samp{qCRC} packet
39236 @anchor{qCRC packet}
39237 Compute the CRC checksum of a block of memory using CRC-32 defined in
39238 IEEE 802.3. The CRC is computed byte at a time, taking the most
39239 significant bit of each byte first. The initial pattern code
39240 @code{0xffffffff} is used to ensure leading zeros affect the CRC.
39242 @emph{Note:} This is the same CRC used in validating separate debug
39243 files (@pxref{Separate Debug Files, , Debugging Information in Separate
39244 Files}). However the algorithm is slightly different. When validating
39245 separate debug files, the CRC is computed taking the @emph{least}
39246 significant bit of each byte first, and the final result is inverted to
39247 detect trailing zeros.
39252 An error (such as memory fault)
39253 @item C @var{crc32}
39254 The specified memory region's checksum is @var{crc32}.
39257 @item QDisableRandomization:@var{value}
39258 @cindex disable address space randomization, remote request
39259 @cindex @samp{QDisableRandomization} packet
39260 Some target operating systems will randomize the virtual address space
39261 of the inferior process as a security feature, but provide a feature
39262 to disable such randomization, e.g.@: to allow for a more deterministic
39263 debugging experience. On such systems, this packet with a @var{value}
39264 of 1 directs the target to disable address space randomization for
39265 processes subsequently started via @samp{vRun} packets, while a packet
39266 with a @var{value} of 0 tells the target to enable address space
39269 This packet is only available in extended mode (@pxref{extended mode}).
39274 The request succeeded.
39277 An error occurred. The error number @var{nn} is given as hex digits.
39280 An empty reply indicates that @samp{QDisableRandomization} is not supported
39284 This packet is not probed by default; the remote stub must request it,
39285 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
39286 This should only be done on targets that actually support disabling
39287 address space randomization.
39289 @item QStartupWithShell:@var{value}
39290 @cindex startup with shell, remote request
39291 @cindex @samp{QStartupWithShell} packet
39292 On UNIX-like targets, it is possible to start the inferior using a
39293 shell program. This is the default behavior on both @value{GDBN} and
39294 @command{gdbserver} (@pxref{set startup-with-shell}). This packet is
39295 used to inform @command{gdbserver} whether it should start the
39296 inferior using a shell or not.
39298 If @var{value} is @samp{0}, @command{gdbserver} will not use a shell
39299 to start the inferior. If @var{value} is @samp{1},
39300 @command{gdbserver} will use a shell to start the inferior. All other
39301 values are considered an error.
39303 This packet is only available in extended mode (@pxref{extended
39309 The request succeeded.
39312 An error occurred. The error number @var{nn} is given as hex digits.
39315 This packet is not probed by default; the remote stub must request it,
39316 by supplying an appropriate @samp{qSupported} response
39317 (@pxref{qSupported}). This should only be done on targets that
39318 actually support starting the inferior using a shell.
39320 Use of this packet is controlled by the @code{set startup-with-shell}
39321 command; @pxref{set startup-with-shell}.
39323 @item QEnvironmentHexEncoded:@var{hex-value}
39324 @anchor{QEnvironmentHexEncoded}
39325 @cindex set environment variable, remote request
39326 @cindex @samp{QEnvironmentHexEncoded} packet
39327 On UNIX-like targets, it is possible to set environment variables that
39328 will be passed to the inferior during the startup process. This
39329 packet is used to inform @command{gdbserver} of an environment
39330 variable that has been defined by the user on @value{GDBN} (@pxref{set
39333 The packet is composed by @var{hex-value}, an hex encoded
39334 representation of the @var{name=value} format representing an
39335 environment variable. The name of the environment variable is
39336 represented by @var{name}, and the value to be assigned to the
39337 environment variable is represented by @var{value}. If the variable
39338 has no value (i.e., the value is @code{null}), then @var{value} will
39341 This packet is only available in extended mode (@pxref{extended
39347 The request succeeded.
39350 This packet is not probed by default; the remote stub must request it,
39351 by supplying an appropriate @samp{qSupported} response
39352 (@pxref{qSupported}). This should only be done on targets that
39353 actually support passing environment variables to the starting
39356 This packet is related to the @code{set environment} command;
39357 @pxref{set environment}.
39359 @item QEnvironmentUnset:@var{hex-value}
39360 @anchor{QEnvironmentUnset}
39361 @cindex unset environment variable, remote request
39362 @cindex @samp{QEnvironmentUnset} packet
39363 On UNIX-like targets, it is possible to unset environment variables
39364 before starting the inferior in the remote target. This packet is
39365 used to inform @command{gdbserver} of an environment variable that has
39366 been unset by the user on @value{GDBN} (@pxref{unset environment}).
39368 The packet is composed by @var{hex-value}, an hex encoded
39369 representation of the name of the environment variable to be unset.
39371 This packet is only available in extended mode (@pxref{extended
39377 The request succeeded.
39380 This packet is not probed by default; the remote stub must request it,
39381 by supplying an appropriate @samp{qSupported} response
39382 (@pxref{qSupported}). This should only be done on targets that
39383 actually support passing environment variables to the starting
39386 This packet is related to the @code{unset environment} command;
39387 @pxref{unset environment}.
39389 @item QEnvironmentReset
39390 @anchor{QEnvironmentReset}
39391 @cindex reset environment, remote request
39392 @cindex @samp{QEnvironmentReset} packet
39393 On UNIX-like targets, this packet is used to reset the state of
39394 environment variables in the remote target before starting the
39395 inferior. In this context, reset means unsetting all environment
39396 variables that were previously set by the user (i.e., were not
39397 initially present in the environment). It is sent to
39398 @command{gdbserver} before the @samp{QEnvironmentHexEncoded}
39399 (@pxref{QEnvironmentHexEncoded}) and the @samp{QEnvironmentUnset}
39400 (@pxref{QEnvironmentUnset}) packets.
39402 This packet is only available in extended mode (@pxref{extended
39408 The request succeeded.
39411 This packet is not probed by default; the remote stub must request it,
39412 by supplying an appropriate @samp{qSupported} response
39413 (@pxref{qSupported}). This should only be done on targets that
39414 actually support passing environment variables to the starting
39417 @item QSetWorkingDir:@r{[}@var{directory}@r{]}
39418 @anchor{QSetWorkingDir packet}
39419 @cindex set working directory, remote request
39420 @cindex @samp{QSetWorkingDir} packet
39421 This packet is used to inform the remote server of the intended
39422 current working directory for programs that are going to be executed.
39424 The packet is composed by @var{directory}, an hex encoded
39425 representation of the directory that the remote inferior will use as
39426 its current working directory. If @var{directory} is an empty string,
39427 the remote server should reset the inferior's current working
39428 directory to its original, empty value.
39430 This packet is only available in extended mode (@pxref{extended
39436 The request succeeded.
39440 @itemx qsThreadInfo
39441 @cindex list active threads, remote request
39442 @cindex @samp{qfThreadInfo} packet
39443 @cindex @samp{qsThreadInfo} packet
39444 Obtain a list of all active thread IDs from the target (OS). Since there
39445 may be too many active threads to fit into one reply packet, this query
39446 works iteratively: it may require more than one query/reply sequence to
39447 obtain the entire list of threads. The first query of the sequence will
39448 be the @samp{qfThreadInfo} query; subsequent queries in the
39449 sequence will be the @samp{qsThreadInfo} query.
39451 NOTE: This packet replaces the @samp{qL} query (see below).
39455 @item m @var{thread-id}
39457 @item m @var{thread-id},@var{thread-id}@dots{}
39458 a comma-separated list of thread IDs
39460 (lower case letter @samp{L}) denotes end of list.
39463 In response to each query, the target will reply with a list of one or
39464 more thread IDs, separated by commas.
39465 @value{GDBN} will respond to each reply with a request for more thread
39466 ids (using the @samp{qs} form of the query), until the target responds
39467 with @samp{l} (lower-case ell, for @dfn{last}).
39468 Refer to @ref{thread-id syntax}, for the format of the @var{thread-id}
39471 @emph{Note: @value{GDBN} will send the @code{qfThreadInfo} query during the
39472 initial connection with the remote target, and the very first thread ID
39473 mentioned in the reply will be stopped by @value{GDBN} in a subsequent
39474 message. Therefore, the stub should ensure that the first thread ID in
39475 the @code{qfThreadInfo} reply is suitable for being stopped by @value{GDBN}.}
39477 @item qGetTLSAddr:@var{thread-id},@var{offset},@var{lm}
39478 @cindex get thread-local storage address, remote request
39479 @cindex @samp{qGetTLSAddr} packet
39480 Fetch the address associated with thread local storage specified
39481 by @var{thread-id}, @var{offset}, and @var{lm}.
39483 @var{thread-id} is the thread ID associated with the
39484 thread for which to fetch the TLS address. @xref{thread-id syntax}.
39486 @var{offset} is the (big endian, hex encoded) offset associated with the
39487 thread local variable. (This offset is obtained from the debug
39488 information associated with the variable.)
39490 @var{lm} is the (big endian, hex encoded) OS/ABI-specific encoding of the
39491 load module associated with the thread local storage. For example,
39492 a @sc{gnu}/Linux system will pass the link map address of the shared
39493 object associated with the thread local storage under consideration.
39494 Other operating environments may choose to represent the load module
39495 differently, so the precise meaning of this parameter will vary.
39499 @item @var{XX}@dots{}
39500 Hex encoded (big endian) bytes representing the address of the thread
39501 local storage requested.
39504 An error occurred. The error number @var{nn} is given as hex digits.
39507 An empty reply indicates that @samp{qGetTLSAddr} is not supported by the stub.
39510 @item qGetTIBAddr:@var{thread-id}
39511 @cindex get thread information block address
39512 @cindex @samp{qGetTIBAddr} packet
39513 Fetch address of the Windows OS specific Thread Information Block.
39515 @var{thread-id} is the thread ID associated with the thread.
39519 @item @var{XX}@dots{}
39520 Hex encoded (big endian) bytes representing the linear address of the
39521 thread information block.
39524 An error occured. This means that either the thread was not found, or the
39525 address could not be retrieved.
39528 An empty reply indicates that @samp{qGetTIBAddr} is not supported by the stub.
39531 @item qL @var{startflag} @var{threadcount} @var{nextthread}
39532 Obtain thread information from RTOS. Where: @var{startflag} (one hex
39533 digit) is one to indicate the first query and zero to indicate a
39534 subsequent query; @var{threadcount} (two hex digits) is the maximum
39535 number of threads the response packet can contain; and @var{nextthread}
39536 (eight hex digits), for subsequent queries (@var{startflag} is zero), is
39537 returned in the response as @var{argthread}.
39539 Don't use this packet; use the @samp{qfThreadInfo} query instead (see above).
39543 @item qM @var{count} @var{done} @var{argthread} @var{thread}@dots{}
39544 Where: @var{count} (two hex digits) is the number of threads being
39545 returned; @var{done} (one hex digit) is zero to indicate more threads
39546 and one indicates no further threads; @var{argthreadid} (eight hex
39547 digits) is @var{nextthread} from the request packet; @var{thread}@dots{}
39548 is a sequence of thread IDs, @var{threadid} (eight hex
39549 digits), from the target. See @code{remote.c:parse_threadlist_response()}.
39553 @cindex section offsets, remote request
39554 @cindex @samp{qOffsets} packet
39555 Get section offsets that the target used when relocating the downloaded
39560 @item Text=@var{xxx};Data=@var{yyy}@r{[};Bss=@var{zzz}@r{]}
39561 Relocate the @code{Text} section by @var{xxx} from its original address.
39562 Relocate the @code{Data} section by @var{yyy} from its original address.
39563 If the object file format provides segment information (e.g.@: @sc{elf}
39564 @samp{PT_LOAD} program headers), @value{GDBN} will relocate entire
39565 segments by the supplied offsets.
39567 @emph{Note: while a @code{Bss} offset may be included in the response,
39568 @value{GDBN} ignores this and instead applies the @code{Data} offset
39569 to the @code{Bss} section.}
39571 @item TextSeg=@var{xxx}@r{[};DataSeg=@var{yyy}@r{]}
39572 Relocate the first segment of the object file, which conventionally
39573 contains program code, to a starting address of @var{xxx}. If
39574 @samp{DataSeg} is specified, relocate the second segment, which
39575 conventionally contains modifiable data, to a starting address of
39576 @var{yyy}. @value{GDBN} will report an error if the object file
39577 does not contain segment information, or does not contain at least
39578 as many segments as mentioned in the reply. Extra segments are
39579 kept at fixed offsets relative to the last relocated segment.
39582 @item qP @var{mode} @var{thread-id}
39583 @cindex thread information, remote request
39584 @cindex @samp{qP} packet
39585 Returns information on @var{thread-id}. Where: @var{mode} is a hex
39586 encoded 32 bit mode; @var{thread-id} is a thread ID
39587 (@pxref{thread-id syntax}).
39589 Don't use this packet; use the @samp{qThreadExtraInfo} query instead
39592 Reply: see @code{remote.c:remote_unpack_thread_info_response()}.
39596 @cindex non-stop mode, remote request
39597 @cindex @samp{QNonStop} packet
39599 Enter non-stop (@samp{QNonStop:1}) or all-stop (@samp{QNonStop:0}) mode.
39600 @xref{Remote Non-Stop}, for more information.
39605 The request succeeded.
39608 An error occurred. The error number @var{nn} is given as hex digits.
39611 An empty reply indicates that @samp{QNonStop} is not supported by
39615 This packet is not probed by default; the remote stub must request it,
39616 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
39617 Use of this packet is controlled by the @code{set non-stop} command;
39618 @pxref{Non-Stop Mode}.
39620 @item QCatchSyscalls:1 @r{[};@var{sysno}@r{]}@dots{}
39621 @itemx QCatchSyscalls:0
39622 @cindex catch syscalls from inferior, remote request
39623 @cindex @samp{QCatchSyscalls} packet
39624 @anchor{QCatchSyscalls}
39625 Enable (@samp{QCatchSyscalls:1}) or disable (@samp{QCatchSyscalls:0})
39626 catching syscalls from the inferior process.
39628 For @samp{QCatchSyscalls:1}, each listed syscall @var{sysno} (encoded
39629 in hex) should be reported to @value{GDBN}. If no syscall @var{sysno}
39630 is listed, every system call should be reported.
39632 Note that if a syscall not in the list is reported, @value{GDBN} will
39633 still filter the event according to its own list from all corresponding
39634 @code{catch syscall} commands. However, it is more efficient to only
39635 report the requested syscalls.
39637 Multiple @samp{QCatchSyscalls:1} packets do not combine; any earlier
39638 @samp{QCatchSyscalls:1} list is completely replaced by the new list.
39640 If the inferior process execs, the state of @samp{QCatchSyscalls} is
39641 kept for the new process too. On targets where exec may affect syscall
39642 numbers, for example with exec between 32 and 64-bit processes, the
39643 client should send a new packet with the new syscall list.
39648 The request succeeded.
39651 An error occurred. @var{nn} are hex digits.
39654 An empty reply indicates that @samp{QCatchSyscalls} is not supported by
39658 Use of this packet is controlled by the @code{set remote catch-syscalls}
39659 command (@pxref{Remote Configuration, set remote catch-syscalls}).
39660 This packet is not probed by default; the remote stub must request it,
39661 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
39663 @item QPassSignals: @var{signal} @r{[};@var{signal}@r{]}@dots{}
39664 @cindex pass signals to inferior, remote request
39665 @cindex @samp{QPassSignals} packet
39666 @anchor{QPassSignals}
39667 Each listed @var{signal} should be passed directly to the inferior process.
39668 Signals are numbered identically to continue packets and stop replies
39669 (@pxref{Stop Reply Packets}). Each @var{signal} list item should be
39670 strictly greater than the previous item. These signals do not need to stop
39671 the inferior, or be reported to @value{GDBN}. All other signals should be
39672 reported to @value{GDBN}. Multiple @samp{QPassSignals} packets do not
39673 combine; any earlier @samp{QPassSignals} list is completely replaced by the
39674 new list. This packet improves performance when using @samp{handle
39675 @var{signal} nostop noprint pass}.
39680 The request succeeded.
39683 An error occurred. The error number @var{nn} is given as hex digits.
39686 An empty reply indicates that @samp{QPassSignals} is not supported by
39690 Use of this packet is controlled by the @code{set remote pass-signals}
39691 command (@pxref{Remote Configuration, set remote pass-signals}).
39692 This packet is not probed by default; the remote stub must request it,
39693 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
39695 @item QProgramSignals: @var{signal} @r{[};@var{signal}@r{]}@dots{}
39696 @cindex signals the inferior may see, remote request
39697 @cindex @samp{QProgramSignals} packet
39698 @anchor{QProgramSignals}
39699 Each listed @var{signal} may be delivered to the inferior process.
39700 Others should be silently discarded.
39702 In some cases, the remote stub may need to decide whether to deliver a
39703 signal to the program or not without @value{GDBN} involvement. One
39704 example of that is while detaching --- the program's threads may have
39705 stopped for signals that haven't yet had a chance of being reported to
39706 @value{GDBN}, and so the remote stub can use the signal list specified
39707 by this packet to know whether to deliver or ignore those pending
39710 This does not influence whether to deliver a signal as requested by a
39711 resumption packet (@pxref{vCont packet}).
39713 Signals are numbered identically to continue packets and stop replies
39714 (@pxref{Stop Reply Packets}). Each @var{signal} list item should be
39715 strictly greater than the previous item. Multiple
39716 @samp{QProgramSignals} packets do not combine; any earlier
39717 @samp{QProgramSignals} list is completely replaced by the new list.
39722 The request succeeded.
39725 An error occurred. The error number @var{nn} is given as hex digits.
39728 An empty reply indicates that @samp{QProgramSignals} is not supported
39732 Use of this packet is controlled by the @code{set remote program-signals}
39733 command (@pxref{Remote Configuration, set remote program-signals}).
39734 This packet is not probed by default; the remote stub must request it,
39735 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
39737 @anchor{QThreadEvents}
39738 @item QThreadEvents:1
39739 @itemx QThreadEvents:0
39740 @cindex thread create/exit events, remote request
39741 @cindex @samp{QThreadEvents} packet
39743 Enable (@samp{QThreadEvents:1}) or disable (@samp{QThreadEvents:0})
39744 reporting of thread create and exit events. @xref{thread create
39745 event}, for the reply specifications. For example, this is used in
39746 non-stop mode when @value{GDBN} stops a set of threads and
39747 synchronously waits for the their corresponding stop replies. Without
39748 exit events, if one of the threads exits, @value{GDBN} would hang
39749 forever not knowing that it should no longer expect a stop for that
39750 same thread. @value{GDBN} does not enable this feature unless the
39751 stub reports that it supports it by including @samp{QThreadEvents+} in
39752 its @samp{qSupported} reply.
39757 The request succeeded.
39760 An error occurred. The error number @var{nn} is given as hex digits.
39763 An empty reply indicates that @samp{QThreadEvents} is not supported by
39767 Use of this packet is controlled by the @code{set remote thread-events}
39768 command (@pxref{Remote Configuration, set remote thread-events}).
39770 @item qRcmd,@var{command}
39771 @cindex execute remote command, remote request
39772 @cindex @samp{qRcmd} packet
39773 @var{command} (hex encoded) is passed to the local interpreter for
39774 execution. Invalid commands should be reported using the output
39775 string. Before the final result packet, the target may also respond
39776 with a number of intermediate @samp{O@var{output}} console output
39777 packets. @emph{Implementors should note that providing access to a
39778 stubs's interpreter may have security implications}.
39783 A command response with no output.
39785 A command response with the hex encoded output string @var{OUTPUT}.
39787 Indicate a badly formed request.
39789 An empty reply indicates that @samp{qRcmd} is not recognized.
39792 (Note that the @code{qRcmd} packet's name is separated from the
39793 command by a @samp{,}, not a @samp{:}, contrary to the naming
39794 conventions above. Please don't use this packet as a model for new
39797 @item qSearch:memory:@var{address};@var{length};@var{search-pattern}
39798 @cindex searching memory, in remote debugging
39800 @cindex @samp{qSearch:memory} packet
39802 @cindex @samp{qSearch memory} packet
39803 @anchor{qSearch memory}
39804 Search @var{length} bytes at @var{address} for @var{search-pattern}.
39805 Both @var{address} and @var{length} are encoded in hex;
39806 @var{search-pattern} is a sequence of bytes, also hex encoded.
39811 The pattern was not found.
39813 The pattern was found at @var{address}.
39815 A badly formed request or an error was encountered while searching memory.
39817 An empty reply indicates that @samp{qSearch:memory} is not recognized.
39820 @item QStartNoAckMode
39821 @cindex @samp{QStartNoAckMode} packet
39822 @anchor{QStartNoAckMode}
39823 Request that the remote stub disable the normal @samp{+}/@samp{-}
39824 protocol acknowledgments (@pxref{Packet Acknowledgment}).
39829 The stub has switched to no-acknowledgment mode.
39830 @value{GDBN} acknowledges this reponse,
39831 but neither the stub nor @value{GDBN} shall send or expect further
39832 @samp{+}/@samp{-} acknowledgments in the current connection.
39834 An empty reply indicates that the stub does not support no-acknowledgment mode.
39837 @item qSupported @r{[}:@var{gdbfeature} @r{[};@var{gdbfeature}@r{]}@dots{} @r{]}
39838 @cindex supported packets, remote query
39839 @cindex features of the remote protocol
39840 @cindex @samp{qSupported} packet
39841 @anchor{qSupported}
39842 Tell the remote stub about features supported by @value{GDBN}, and
39843 query the stub for features it supports. This packet allows
39844 @value{GDBN} and the remote stub to take advantage of each others'
39845 features. @samp{qSupported} also consolidates multiple feature probes
39846 at startup, to improve @value{GDBN} performance---a single larger
39847 packet performs better than multiple smaller probe packets on
39848 high-latency links. Some features may enable behavior which must not
39849 be on by default, e.g.@: because it would confuse older clients or
39850 stubs. Other features may describe packets which could be
39851 automatically probed for, but are not. These features must be
39852 reported before @value{GDBN} will use them. This ``default
39853 unsupported'' behavior is not appropriate for all packets, but it
39854 helps to keep the initial connection time under control with new
39855 versions of @value{GDBN} which support increasing numbers of packets.
39859 @item @var{stubfeature} @r{[};@var{stubfeature}@r{]}@dots{}
39860 The stub supports or does not support each returned @var{stubfeature},
39861 depending on the form of each @var{stubfeature} (see below for the
39864 An empty reply indicates that @samp{qSupported} is not recognized,
39865 or that no features needed to be reported to @value{GDBN}.
39868 The allowed forms for each feature (either a @var{gdbfeature} in the
39869 @samp{qSupported} packet, or a @var{stubfeature} in the response)
39873 @item @var{name}=@var{value}
39874 The remote protocol feature @var{name} is supported, and associated
39875 with the specified @var{value}. The format of @var{value} depends
39876 on the feature, but it must not include a semicolon.
39878 The remote protocol feature @var{name} is supported, and does not
39879 need an associated value.
39881 The remote protocol feature @var{name} is not supported.
39883 The remote protocol feature @var{name} may be supported, and
39884 @value{GDBN} should auto-detect support in some other way when it is
39885 needed. This form will not be used for @var{gdbfeature} notifications,
39886 but may be used for @var{stubfeature} responses.
39889 Whenever the stub receives a @samp{qSupported} request, the
39890 supplied set of @value{GDBN} features should override any previous
39891 request. This allows @value{GDBN} to put the stub in a known
39892 state, even if the stub had previously been communicating with
39893 a different version of @value{GDBN}.
39895 The following values of @var{gdbfeature} (for the packet sent by @value{GDBN})
39900 This feature indicates whether @value{GDBN} supports multiprocess
39901 extensions to the remote protocol. @value{GDBN} does not use such
39902 extensions unless the stub also reports that it supports them by
39903 including @samp{multiprocess+} in its @samp{qSupported} reply.
39904 @xref{multiprocess extensions}, for details.
39907 This feature indicates that @value{GDBN} supports the XML target
39908 description. If the stub sees @samp{xmlRegisters=} with target
39909 specific strings separated by a comma, it will report register
39913 This feature indicates whether @value{GDBN} supports the
39914 @samp{qRelocInsn} packet (@pxref{Tracepoint Packets,,Relocate
39915 instruction reply packet}).
39918 This feature indicates whether @value{GDBN} supports the swbreak stop
39919 reason in stop replies. @xref{swbreak stop reason}, for details.
39922 This feature indicates whether @value{GDBN} supports the hwbreak stop
39923 reason in stop replies. @xref{swbreak stop reason}, for details.
39926 This feature indicates whether @value{GDBN} supports fork event
39927 extensions to the remote protocol. @value{GDBN} does not use such
39928 extensions unless the stub also reports that it supports them by
39929 including @samp{fork-events+} in its @samp{qSupported} reply.
39932 This feature indicates whether @value{GDBN} supports vfork event
39933 extensions to the remote protocol. @value{GDBN} does not use such
39934 extensions unless the stub also reports that it supports them by
39935 including @samp{vfork-events+} in its @samp{qSupported} reply.
39938 This feature indicates whether @value{GDBN} supports exec event
39939 extensions to the remote protocol. @value{GDBN} does not use such
39940 extensions unless the stub also reports that it supports them by
39941 including @samp{exec-events+} in its @samp{qSupported} reply.
39943 @item vContSupported
39944 This feature indicates whether @value{GDBN} wants to know the
39945 supported actions in the reply to @samp{vCont?} packet.
39948 Stubs should ignore any unknown values for
39949 @var{gdbfeature}. Any @value{GDBN} which sends a @samp{qSupported}
39950 packet supports receiving packets of unlimited length (earlier
39951 versions of @value{GDBN} may reject overly long responses). Additional values
39952 for @var{gdbfeature} may be defined in the future to let the stub take
39953 advantage of new features in @value{GDBN}, e.g.@: incompatible
39954 improvements in the remote protocol---the @samp{multiprocess} feature is
39955 an example of such a feature. The stub's reply should be independent
39956 of the @var{gdbfeature} entries sent by @value{GDBN}; first @value{GDBN}
39957 describes all the features it supports, and then the stub replies with
39958 all the features it supports.
39960 Similarly, @value{GDBN} will silently ignore unrecognized stub feature
39961 responses, as long as each response uses one of the standard forms.
39963 Some features are flags. A stub which supports a flag feature
39964 should respond with a @samp{+} form response. Other features
39965 require values, and the stub should respond with an @samp{=}
39968 Each feature has a default value, which @value{GDBN} will use if
39969 @samp{qSupported} is not available or if the feature is not mentioned
39970 in the @samp{qSupported} response. The default values are fixed; a
39971 stub is free to omit any feature responses that match the defaults.
39973 Not all features can be probed, but for those which can, the probing
39974 mechanism is useful: in some cases, a stub's internal
39975 architecture may not allow the protocol layer to know some information
39976 about the underlying target in advance. This is especially common in
39977 stubs which may be configured for multiple targets.
39979 These are the currently defined stub features and their properties:
39981 @multitable @columnfractions 0.35 0.2 0.12 0.2
39982 @c NOTE: The first row should be @headitem, but we do not yet require
39983 @c a new enough version of Texinfo (4.7) to use @headitem.
39985 @tab Value Required
39989 @item @samp{PacketSize}
39994 @item @samp{qXfer:auxv:read}
39999 @item @samp{qXfer:btrace:read}
40004 @item @samp{qXfer:btrace-conf:read}
40009 @item @samp{qXfer:exec-file:read}
40014 @item @samp{qXfer:features:read}
40019 @item @samp{qXfer:libraries:read}
40024 @item @samp{qXfer:libraries-svr4:read}
40029 @item @samp{augmented-libraries-svr4-read}
40034 @item @samp{qXfer:memory-map:read}
40039 @item @samp{qXfer:sdata:read}
40044 @item @samp{qXfer:siginfo:read}
40049 @item @samp{qXfer:siginfo:write}
40054 @item @samp{qXfer:threads:read}
40059 @item @samp{qXfer:traceframe-info:read}
40064 @item @samp{qXfer:uib:read}
40069 @item @samp{qXfer:fdpic:read}
40074 @item @samp{Qbtrace:off}
40079 @item @samp{Qbtrace:bts}
40084 @item @samp{Qbtrace:pt}
40089 @item @samp{Qbtrace-conf:bts:size}
40094 @item @samp{Qbtrace-conf:pt:size}
40099 @item @samp{QNonStop}
40104 @item @samp{QCatchSyscalls}
40109 @item @samp{QPassSignals}
40114 @item @samp{QStartNoAckMode}
40119 @item @samp{multiprocess}
40124 @item @samp{ConditionalBreakpoints}
40129 @item @samp{ConditionalTracepoints}
40134 @item @samp{ReverseContinue}
40139 @item @samp{ReverseStep}
40144 @item @samp{TracepointSource}
40149 @item @samp{QAgent}
40154 @item @samp{QAllow}
40159 @item @samp{QDisableRandomization}
40164 @item @samp{EnableDisableTracepoints}
40169 @item @samp{QTBuffer:size}
40174 @item @samp{tracenz}
40179 @item @samp{BreakpointCommands}
40184 @item @samp{swbreak}
40189 @item @samp{hwbreak}
40194 @item @samp{fork-events}
40199 @item @samp{vfork-events}
40204 @item @samp{exec-events}
40209 @item @samp{QThreadEvents}
40214 @item @samp{no-resumed}
40221 These are the currently defined stub features, in more detail:
40224 @cindex packet size, remote protocol
40225 @item PacketSize=@var{bytes}
40226 The remote stub can accept packets up to at least @var{bytes} in
40227 length. @value{GDBN} will send packets up to this size for bulk
40228 transfers, and will never send larger packets. This is a limit on the
40229 data characters in the packet, including the frame and checksum.
40230 There is no trailing NUL byte in a remote protocol packet; if the stub
40231 stores packets in a NUL-terminated format, it should allow an extra
40232 byte in its buffer for the NUL. If this stub feature is not supported,
40233 @value{GDBN} guesses based on the size of the @samp{g} packet response.
40235 @item qXfer:auxv:read
40236 The remote stub understands the @samp{qXfer:auxv:read} packet
40237 (@pxref{qXfer auxiliary vector read}).
40239 @item qXfer:btrace:read
40240 The remote stub understands the @samp{qXfer:btrace:read}
40241 packet (@pxref{qXfer btrace read}).
40243 @item qXfer:btrace-conf:read
40244 The remote stub understands the @samp{qXfer:btrace-conf:read}
40245 packet (@pxref{qXfer btrace-conf read}).
40247 @item qXfer:exec-file:read
40248 The remote stub understands the @samp{qXfer:exec-file:read} packet
40249 (@pxref{qXfer executable filename read}).
40251 @item qXfer:features:read
40252 The remote stub understands the @samp{qXfer:features:read} packet
40253 (@pxref{qXfer target description read}).
40255 @item qXfer:libraries:read
40256 The remote stub understands the @samp{qXfer:libraries:read} packet
40257 (@pxref{qXfer library list read}).
40259 @item qXfer:libraries-svr4:read
40260 The remote stub understands the @samp{qXfer:libraries-svr4:read} packet
40261 (@pxref{qXfer svr4 library list read}).
40263 @item augmented-libraries-svr4-read
40264 The remote stub understands the augmented form of the
40265 @samp{qXfer:libraries-svr4:read} packet
40266 (@pxref{qXfer svr4 library list read}).
40268 @item qXfer:memory-map:read
40269 The remote stub understands the @samp{qXfer:memory-map:read} packet
40270 (@pxref{qXfer memory map read}).
40272 @item qXfer:sdata:read
40273 The remote stub understands the @samp{qXfer:sdata:read} packet
40274 (@pxref{qXfer sdata read}).
40276 @item qXfer:siginfo:read
40277 The remote stub understands the @samp{qXfer:siginfo:read} packet
40278 (@pxref{qXfer siginfo read}).
40280 @item qXfer:siginfo:write
40281 The remote stub understands the @samp{qXfer:siginfo:write} packet
40282 (@pxref{qXfer siginfo write}).
40284 @item qXfer:threads:read
40285 The remote stub understands the @samp{qXfer:threads:read} packet
40286 (@pxref{qXfer threads read}).
40288 @item qXfer:traceframe-info:read
40289 The remote stub understands the @samp{qXfer:traceframe-info:read}
40290 packet (@pxref{qXfer traceframe info read}).
40292 @item qXfer:uib:read
40293 The remote stub understands the @samp{qXfer:uib:read}
40294 packet (@pxref{qXfer unwind info block}).
40296 @item qXfer:fdpic:read
40297 The remote stub understands the @samp{qXfer:fdpic:read}
40298 packet (@pxref{qXfer fdpic loadmap read}).
40301 The remote stub understands the @samp{QNonStop} packet
40302 (@pxref{QNonStop}).
40304 @item QCatchSyscalls
40305 The remote stub understands the @samp{QCatchSyscalls} packet
40306 (@pxref{QCatchSyscalls}).
40309 The remote stub understands the @samp{QPassSignals} packet
40310 (@pxref{QPassSignals}).
40312 @item QStartNoAckMode
40313 The remote stub understands the @samp{QStartNoAckMode} packet and
40314 prefers to operate in no-acknowledgment mode. @xref{Packet Acknowledgment}.
40317 @anchor{multiprocess extensions}
40318 @cindex multiprocess extensions, in remote protocol
40319 The remote stub understands the multiprocess extensions to the remote
40320 protocol syntax. The multiprocess extensions affect the syntax of
40321 thread IDs in both packets and replies (@pxref{thread-id syntax}), and
40322 add process IDs to the @samp{D} packet and @samp{W} and @samp{X}
40323 replies. Note that reporting this feature indicates support for the
40324 syntactic extensions only, not that the stub necessarily supports
40325 debugging of more than one process at a time. The stub must not use
40326 multiprocess extensions in packet replies unless @value{GDBN} has also
40327 indicated it supports them in its @samp{qSupported} request.
40329 @item qXfer:osdata:read
40330 The remote stub understands the @samp{qXfer:osdata:read} packet
40331 ((@pxref{qXfer osdata read}).
40333 @item ConditionalBreakpoints
40334 The target accepts and implements evaluation of conditional expressions
40335 defined for breakpoints. The target will only report breakpoint triggers
40336 when such conditions are true (@pxref{Conditions, ,Break Conditions}).
40338 @item ConditionalTracepoints
40339 The remote stub accepts and implements conditional expressions defined
40340 for tracepoints (@pxref{Tracepoint Conditions}).
40342 @item ReverseContinue
40343 The remote stub accepts and implements the reverse continue packet
40347 The remote stub accepts and implements the reverse step packet
40350 @item TracepointSource
40351 The remote stub understands the @samp{QTDPsrc} packet that supplies
40352 the source form of tracepoint definitions.
40355 The remote stub understands the @samp{QAgent} packet.
40358 The remote stub understands the @samp{QAllow} packet.
40360 @item QDisableRandomization
40361 The remote stub understands the @samp{QDisableRandomization} packet.
40363 @item StaticTracepoint
40364 @cindex static tracepoints, in remote protocol
40365 The remote stub supports static tracepoints.
40367 @item InstallInTrace
40368 @anchor{install tracepoint in tracing}
40369 The remote stub supports installing tracepoint in tracing.
40371 @item EnableDisableTracepoints
40372 The remote stub supports the @samp{QTEnable} (@pxref{QTEnable}) and
40373 @samp{QTDisable} (@pxref{QTDisable}) packets that allow tracepoints
40374 to be enabled and disabled while a trace experiment is running.
40376 @item QTBuffer:size
40377 The remote stub supports the @samp{QTBuffer:size} (@pxref{QTBuffer-size})
40378 packet that allows to change the size of the trace buffer.
40381 @cindex string tracing, in remote protocol
40382 The remote stub supports the @samp{tracenz} bytecode for collecting strings.
40383 See @ref{Bytecode Descriptions} for details about the bytecode.
40385 @item BreakpointCommands
40386 @cindex breakpoint commands, in remote protocol
40387 The remote stub supports running a breakpoint's command list itself,
40388 rather than reporting the hit to @value{GDBN}.
40391 The remote stub understands the @samp{Qbtrace:off} packet.
40394 The remote stub understands the @samp{Qbtrace:bts} packet.
40397 The remote stub understands the @samp{Qbtrace:pt} packet.
40399 @item Qbtrace-conf:bts:size
40400 The remote stub understands the @samp{Qbtrace-conf:bts:size} packet.
40402 @item Qbtrace-conf:pt:size
40403 The remote stub understands the @samp{Qbtrace-conf:pt:size} packet.
40406 The remote stub reports the @samp{swbreak} stop reason for memory
40410 The remote stub reports the @samp{hwbreak} stop reason for hardware
40414 The remote stub reports the @samp{fork} stop reason for fork events.
40417 The remote stub reports the @samp{vfork} stop reason for vfork events
40418 and vforkdone events.
40421 The remote stub reports the @samp{exec} stop reason for exec events.
40423 @item vContSupported
40424 The remote stub reports the supported actions in the reply to
40425 @samp{vCont?} packet.
40427 @item QThreadEvents
40428 The remote stub understands the @samp{QThreadEvents} packet.
40431 The remote stub reports the @samp{N} stop reply.
40436 @cindex symbol lookup, remote request
40437 @cindex @samp{qSymbol} packet
40438 Notify the target that @value{GDBN} is prepared to serve symbol lookup
40439 requests. Accept requests from the target for the values of symbols.
40444 The target does not need to look up any (more) symbols.
40445 @item qSymbol:@var{sym_name}
40446 The target requests the value of symbol @var{sym_name} (hex encoded).
40447 @value{GDBN} may provide the value by using the
40448 @samp{qSymbol:@var{sym_value}:@var{sym_name}} message, described
40452 @item qSymbol:@var{sym_value}:@var{sym_name}
40453 Set the value of @var{sym_name} to @var{sym_value}.
40455 @var{sym_name} (hex encoded) is the name of a symbol whose value the
40456 target has previously requested.
40458 @var{sym_value} (hex) is the value for symbol @var{sym_name}. If
40459 @value{GDBN} cannot supply a value for @var{sym_name}, then this field
40465 The target does not need to look up any (more) symbols.
40466 @item qSymbol:@var{sym_name}
40467 The target requests the value of a new symbol @var{sym_name} (hex
40468 encoded). @value{GDBN} will continue to supply the values of symbols
40469 (if available), until the target ceases to request them.
40474 @itemx QTDisconnected
40481 @itemx qTMinFTPILen
40483 @xref{Tracepoint Packets}.
40485 @item qThreadExtraInfo,@var{thread-id}
40486 @cindex thread attributes info, remote request
40487 @cindex @samp{qThreadExtraInfo} packet
40488 Obtain from the target OS a printable string description of thread
40489 attributes for the thread @var{thread-id}; see @ref{thread-id syntax},
40490 for the forms of @var{thread-id}. This
40491 string may contain anything that the target OS thinks is interesting
40492 for @value{GDBN} to tell the user about the thread. The string is
40493 displayed in @value{GDBN}'s @code{info threads} display. Some
40494 examples of possible thread extra info strings are @samp{Runnable}, or
40495 @samp{Blocked on Mutex}.
40499 @item @var{XX}@dots{}
40500 Where @samp{@var{XX}@dots{}} is a hex encoding of @sc{ascii} data,
40501 comprising the printable string containing the extra information about
40502 the thread's attributes.
40505 (Note that the @code{qThreadExtraInfo} packet's name is separated from
40506 the command by a @samp{,}, not a @samp{:}, contrary to the naming
40507 conventions above. Please don't use this packet as a model for new
40526 @xref{Tracepoint Packets}.
40528 @item qXfer:@var{object}:read:@var{annex}:@var{offset},@var{length}
40529 @cindex read special object, remote request
40530 @cindex @samp{qXfer} packet
40531 @anchor{qXfer read}
40532 Read uninterpreted bytes from the target's special data area
40533 identified by the keyword @var{object}. Request @var{length} bytes
40534 starting at @var{offset} bytes into the data. The content and
40535 encoding of @var{annex} is specific to @var{object}; it can supply
40536 additional details about what data to access.
40541 Data @var{data} (@pxref{Binary Data}) has been read from the
40542 target. There may be more data at a higher address (although
40543 it is permitted to return @samp{m} even for the last valid
40544 block of data, as long as at least one byte of data was read).
40545 It is possible for @var{data} to have fewer bytes than the @var{length} in the
40549 Data @var{data} (@pxref{Binary Data}) has been read from the target.
40550 There is no more data to be read. It is possible for @var{data} to
40551 have fewer bytes than the @var{length} in the request.
40554 The @var{offset} in the request is at the end of the data.
40555 There is no more data to be read.
40558 The request was malformed, or @var{annex} was invalid.
40561 The offset was invalid, or there was an error encountered reading the data.
40562 The @var{nn} part is a hex-encoded @code{errno} value.
40565 An empty reply indicates the @var{object} string was not recognized by
40566 the stub, or that the object does not support reading.
40569 Here are the specific requests of this form defined so far. All the
40570 @samp{qXfer:@var{object}:read:@dots{}} requests use the same reply
40571 formats, listed above.
40574 @item qXfer:auxv:read::@var{offset},@var{length}
40575 @anchor{qXfer auxiliary vector read}
40576 Access the target's @dfn{auxiliary vector}. @xref{OS Information,
40577 auxiliary vector}. Note @var{annex} must be empty.
40579 This packet is not probed by default; the remote stub must request it,
40580 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
40582 @item qXfer:btrace:read:@var{annex}:@var{offset},@var{length}
40583 @anchor{qXfer btrace read}
40585 Return a description of the current branch trace.
40586 @xref{Branch Trace Format}. The annex part of the generic @samp{qXfer}
40587 packet may have one of the following values:
40591 Returns all available branch trace.
40594 Returns all available branch trace if the branch trace changed since
40595 the last read request.
40598 Returns the new branch trace since the last read request. Adds a new
40599 block to the end of the trace that begins at zero and ends at the source
40600 location of the first branch in the trace buffer. This extra block is
40601 used to stitch traces together.
40603 If the trace buffer overflowed, returns an error indicating the overflow.
40606 This packet is not probed by default; the remote stub must request it
40607 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
40609 @item qXfer:btrace-conf:read::@var{offset},@var{length}
40610 @anchor{qXfer btrace-conf read}
40612 Return a description of the current branch trace configuration.
40613 @xref{Branch Trace Configuration Format}.
40615 This packet is not probed by default; the remote stub must request it
40616 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
40618 @item qXfer:exec-file:read:@var{annex}:@var{offset},@var{length}
40619 @anchor{qXfer executable filename read}
40620 Return the full absolute name of the file that was executed to create
40621 a process running on the remote system. The annex specifies the
40622 numeric process ID of the process to query, encoded as a hexadecimal
40623 number. If the annex part is empty the remote stub should return the
40624 filename corresponding to the currently executing process.
40626 This packet is not probed by default; the remote stub must request it,
40627 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
40629 @item qXfer:features:read:@var{annex}:@var{offset},@var{length}
40630 @anchor{qXfer target description read}
40631 Access the @dfn{target description}. @xref{Target Descriptions}. The
40632 annex specifies which XML document to access. The main description is
40633 always loaded from the @samp{target.xml} annex.
40635 This packet is not probed by default; the remote stub must request it,
40636 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
40638 @item qXfer:libraries:read:@var{annex}:@var{offset},@var{length}
40639 @anchor{qXfer library list read}
40640 Access the target's list of loaded libraries. @xref{Library List Format}.
40641 The annex part of the generic @samp{qXfer} packet must be empty
40642 (@pxref{qXfer read}).
40644 Targets which maintain a list of libraries in the program's memory do
40645 not need to implement this packet; it is designed for platforms where
40646 the operating system manages the list of loaded libraries.
40648 This packet is not probed by default; the remote stub must request it,
40649 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
40651 @item qXfer:libraries-svr4:read:@var{annex}:@var{offset},@var{length}
40652 @anchor{qXfer svr4 library list read}
40653 Access the target's list of loaded libraries when the target is an SVR4
40654 platform. @xref{Library List Format for SVR4 Targets}. The annex part
40655 of the generic @samp{qXfer} packet must be empty unless the remote
40656 stub indicated it supports the augmented form of this packet
40657 by supplying an appropriate @samp{qSupported} response
40658 (@pxref{qXfer read}, @ref{qSupported}).
40660 This packet is optional for better performance on SVR4 targets.
40661 @value{GDBN} uses memory read packets to read the SVR4 library list otherwise.
40663 This packet is not probed by default; the remote stub must request it,
40664 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
40666 If the remote stub indicates it supports the augmented form of this
40667 packet then the annex part of the generic @samp{qXfer} packet may
40668 contain a semicolon-separated list of @samp{@var{name}=@var{value}}
40669 arguments. The currently supported arguments are:
40672 @item start=@var{address}
40673 A hexadecimal number specifying the address of the @samp{struct
40674 link_map} to start reading the library list from. If unset or zero
40675 then the first @samp{struct link_map} in the library list will be
40676 chosen as the starting point.
40678 @item prev=@var{address}
40679 A hexadecimal number specifying the address of the @samp{struct
40680 link_map} immediately preceding the @samp{struct link_map}
40681 specified by the @samp{start} argument. If unset or zero then
40682 the remote stub will expect that no @samp{struct link_map}
40683 exists prior to the starting point.
40687 Arguments that are not understood by the remote stub will be silently
40690 @item qXfer:memory-map:read::@var{offset},@var{length}
40691 @anchor{qXfer memory map read}
40692 Access the target's @dfn{memory-map}. @xref{Memory Map Format}. The
40693 annex part of the generic @samp{qXfer} packet must be empty
40694 (@pxref{qXfer read}).
40696 This packet is not probed by default; the remote stub must request it,
40697 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
40699 @item qXfer:sdata:read::@var{offset},@var{length}
40700 @anchor{qXfer sdata read}
40702 Read contents of the extra collected static tracepoint marker
40703 information. The annex part of the generic @samp{qXfer} packet must
40704 be empty (@pxref{qXfer read}). @xref{Tracepoint Actions,,Tracepoint
40707 This packet is not probed by default; the remote stub must request it,
40708 by supplying an appropriate @samp{qSupported} response
40709 (@pxref{qSupported}).
40711 @item qXfer:siginfo:read::@var{offset},@var{length}
40712 @anchor{qXfer siginfo read}
40713 Read contents of the extra signal information on the target
40714 system. The annex part of the generic @samp{qXfer} packet must be
40715 empty (@pxref{qXfer read}).
40717 This packet is not probed by default; the remote stub must request it,
40718 by supplying an appropriate @samp{qSupported} response
40719 (@pxref{qSupported}).
40721 @item qXfer:threads:read::@var{offset},@var{length}
40722 @anchor{qXfer threads read}
40723 Access the list of threads on target. @xref{Thread List Format}. The
40724 annex part of the generic @samp{qXfer} packet must be empty
40725 (@pxref{qXfer read}).
40727 This packet is not probed by default; the remote stub must request it,
40728 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
40730 @item qXfer:traceframe-info:read::@var{offset},@var{length}
40731 @anchor{qXfer traceframe info read}
40733 Return a description of the current traceframe's contents.
40734 @xref{Traceframe Info Format}. The annex part of the generic
40735 @samp{qXfer} packet must be empty (@pxref{qXfer read}).
40737 This packet is not probed by default; the remote stub must request it,
40738 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
40740 @item qXfer:uib:read:@var{pc}:@var{offset},@var{length}
40741 @anchor{qXfer unwind info block}
40743 Return the unwind information block for @var{pc}. This packet is used
40744 on OpenVMS/ia64 to ask the kernel unwind information.
40746 This packet is not probed by default.
40748 @item qXfer:fdpic:read:@var{annex}:@var{offset},@var{length}
40749 @anchor{qXfer fdpic loadmap read}
40750 Read contents of @code{loadmap}s on the target system. The
40751 annex, either @samp{exec} or @samp{interp}, specifies which @code{loadmap},
40752 executable @code{loadmap} or interpreter @code{loadmap} to read.
40754 This packet is not probed by default; the remote stub must request it,
40755 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
40757 @item qXfer:osdata:read::@var{offset},@var{length}
40758 @anchor{qXfer osdata read}
40759 Access the target's @dfn{operating system information}.
40760 @xref{Operating System Information}.
40764 @item qXfer:@var{object}:write:@var{annex}:@var{offset}:@var{data}@dots{}
40765 @cindex write data into object, remote request
40766 @anchor{qXfer write}
40767 Write uninterpreted bytes into the target's special data area
40768 identified by the keyword @var{object}, starting at @var{offset} bytes
40769 into the data. The binary-encoded data (@pxref{Binary Data}) to be
40770 written is given by @var{data}@dots{}. The content and encoding of @var{annex}
40771 is specific to @var{object}; it can supply additional details about what data
40777 @var{nn} (hex encoded) is the number of bytes written.
40778 This may be fewer bytes than supplied in the request.
40781 The request was malformed, or @var{annex} was invalid.
40784 The offset was invalid, or there was an error encountered writing the data.
40785 The @var{nn} part is a hex-encoded @code{errno} value.
40788 An empty reply indicates the @var{object} string was not
40789 recognized by the stub, or that the object does not support writing.
40792 Here are the specific requests of this form defined so far. All the
40793 @samp{qXfer:@var{object}:write:@dots{}} requests use the same reply
40794 formats, listed above.
40797 @item qXfer:siginfo:write::@var{offset}:@var{data}@dots{}
40798 @anchor{qXfer siginfo write}
40799 Write @var{data} to the extra signal information on the target system.
40800 The annex part of the generic @samp{qXfer} packet must be
40801 empty (@pxref{qXfer write}).
40803 This packet is not probed by default; the remote stub must request it,
40804 by supplying an appropriate @samp{qSupported} response
40805 (@pxref{qSupported}).
40808 @item qXfer:@var{object}:@var{operation}:@dots{}
40809 Requests of this form may be added in the future. When a stub does
40810 not recognize the @var{object} keyword, or its support for
40811 @var{object} does not recognize the @var{operation} keyword, the stub
40812 must respond with an empty packet.
40814 @item qAttached:@var{pid}
40815 @cindex query attached, remote request
40816 @cindex @samp{qAttached} packet
40817 Return an indication of whether the remote server attached to an
40818 existing process or created a new process. When the multiprocess
40819 protocol extensions are supported (@pxref{multiprocess extensions}),
40820 @var{pid} is an integer in hexadecimal format identifying the target
40821 process. Otherwise, @value{GDBN} will omit the @var{pid} field and
40822 the query packet will be simplified as @samp{qAttached}.
40824 This query is used, for example, to know whether the remote process
40825 should be detached or killed when a @value{GDBN} session is ended with
40826 the @code{quit} command.
40831 The remote server attached to an existing process.
40833 The remote server created a new process.
40835 A badly formed request or an error was encountered.
40839 Enable branch tracing for the current thread using Branch Trace Store.
40844 Branch tracing has been enabled.
40846 A badly formed request or an error was encountered.
40850 Enable branch tracing for the current thread using Intel Processor Trace.
40855 Branch tracing has been enabled.
40857 A badly formed request or an error was encountered.
40861 Disable branch tracing for the current thread.
40866 Branch tracing has been disabled.
40868 A badly formed request or an error was encountered.
40871 @item Qbtrace-conf:bts:size=@var{value}
40872 Set the requested ring buffer size for new threads that use the
40873 btrace recording method in bts format.
40878 The ring buffer size has been set.
40880 A badly formed request or an error was encountered.
40883 @item Qbtrace-conf:pt:size=@var{value}
40884 Set the requested ring buffer size for new threads that use the
40885 btrace recording method in pt format.
40890 The ring buffer size has been set.
40892 A badly formed request or an error was encountered.
40897 @node Architecture-Specific Protocol Details
40898 @section Architecture-Specific Protocol Details
40900 This section describes how the remote protocol is applied to specific
40901 target architectures. Also see @ref{Standard Target Features}, for
40902 details of XML target descriptions for each architecture.
40905 * ARM-Specific Protocol Details::
40906 * MIPS-Specific Protocol Details::
40909 @node ARM-Specific Protocol Details
40910 @subsection @acronym{ARM}-specific Protocol Details
40913 * ARM Breakpoint Kinds::
40916 @node ARM Breakpoint Kinds
40917 @subsubsection @acronym{ARM} Breakpoint Kinds
40918 @cindex breakpoint kinds, @acronym{ARM}
40920 These breakpoint kinds are defined for the @samp{Z0} and @samp{Z1} packets.
40925 16-bit Thumb mode breakpoint.
40928 32-bit Thumb mode (Thumb-2) breakpoint.
40931 32-bit @acronym{ARM} mode breakpoint.
40935 @node MIPS-Specific Protocol Details
40936 @subsection @acronym{MIPS}-specific Protocol Details
40939 * MIPS Register packet Format::
40940 * MIPS Breakpoint Kinds::
40943 @node MIPS Register packet Format
40944 @subsubsection @acronym{MIPS} Register Packet Format
40945 @cindex register packet format, @acronym{MIPS}
40947 The following @code{g}/@code{G} packets have previously been defined.
40948 In the below, some thirty-two bit registers are transferred as
40949 sixty-four bits. Those registers should be zero/sign extended (which?)
40950 to fill the space allocated. Register bytes are transferred in target
40951 byte order. The two nibbles within a register byte are transferred
40952 most-significant -- least-significant.
40957 All registers are transferred as thirty-two bit quantities in the order:
40958 32 general-purpose; sr; lo; hi; bad; cause; pc; 32 floating-point
40959 registers; fsr; fir; fp.
40962 All registers are transferred as sixty-four bit quantities (including
40963 thirty-two bit registers such as @code{sr}). The ordering is the same
40968 @node MIPS Breakpoint Kinds
40969 @subsubsection @acronym{MIPS} Breakpoint Kinds
40970 @cindex breakpoint kinds, @acronym{MIPS}
40972 These breakpoint kinds are defined for the @samp{Z0} and @samp{Z1} packets.
40977 16-bit @acronym{MIPS16} mode breakpoint.
40980 16-bit @acronym{microMIPS} mode breakpoint.
40983 32-bit standard @acronym{MIPS} mode breakpoint.
40986 32-bit @acronym{microMIPS} mode breakpoint.
40990 @node Tracepoint Packets
40991 @section Tracepoint Packets
40992 @cindex tracepoint packets
40993 @cindex packets, tracepoint
40995 Here we describe the packets @value{GDBN} uses to implement
40996 tracepoints (@pxref{Tracepoints}).
41000 @item QTDP:@var{n}:@var{addr}:@var{ena}:@var{step}:@var{pass}[:F@var{flen}][:X@var{len},@var{bytes}]@r{[}-@r{]}
41001 @cindex @samp{QTDP} packet
41002 Create a new tracepoint, number @var{n}, at @var{addr}. If @var{ena}
41003 is @samp{E}, then the tracepoint is enabled; if it is @samp{D}, then
41004 the tracepoint is disabled. The @var{step} gives the tracepoint's step
41005 count, and @var{pass} gives its pass count. If an @samp{F} is present,
41006 then the tracepoint is to be a fast tracepoint, and the @var{flen} is
41007 the number of bytes that the target should copy elsewhere to make room
41008 for the tracepoint. If an @samp{X} is present, it introduces a
41009 tracepoint condition, which consists of a hexadecimal length, followed
41010 by a comma and hex-encoded bytes, in a manner similar to action
41011 encodings as described below. If the trailing @samp{-} is present,
41012 further @samp{QTDP} packets will follow to specify this tracepoint's
41018 The packet was understood and carried out.
41020 @xref{Tracepoint Packets,,Relocate instruction reply packet}.
41022 The packet was not recognized.
41025 @item QTDP:-@var{n}:@var{addr}:@r{[}S@r{]}@var{action}@dots{}@r{[}-@r{]}
41026 Define actions to be taken when a tracepoint is hit. The @var{n} and
41027 @var{addr} must be the same as in the initial @samp{QTDP} packet for
41028 this tracepoint. This packet may only be sent immediately after
41029 another @samp{QTDP} packet that ended with a @samp{-}. If the
41030 trailing @samp{-} is present, further @samp{QTDP} packets will follow,
41031 specifying more actions for this tracepoint.
41033 In the series of action packets for a given tracepoint, at most one
41034 can have an @samp{S} before its first @var{action}. If such a packet
41035 is sent, it and the following packets define ``while-stepping''
41036 actions. Any prior packets define ordinary actions --- that is, those
41037 taken when the tracepoint is first hit. If no action packet has an
41038 @samp{S}, then all the packets in the series specify ordinary
41039 tracepoint actions.
41041 The @samp{@var{action}@dots{}} portion of the packet is a series of
41042 actions, concatenated without separators. Each action has one of the
41048 Collect the registers whose bits are set in @var{mask},
41049 a hexadecimal number whose @var{i}'th bit is set if register number
41050 @var{i} should be collected. (The least significant bit is numbered
41051 zero.) Note that @var{mask} may be any number of digits long; it may
41052 not fit in a 32-bit word.
41054 @item M @var{basereg},@var{offset},@var{len}
41055 Collect @var{len} bytes of memory starting at the address in register
41056 number @var{basereg}, plus @var{offset}. If @var{basereg} is
41057 @samp{-1}, then the range has a fixed address: @var{offset} is the
41058 address of the lowest byte to collect. The @var{basereg},
41059 @var{offset}, and @var{len} parameters are all unsigned hexadecimal
41060 values (the @samp{-1} value for @var{basereg} is a special case).
41062 @item X @var{len},@var{expr}
41063 Evaluate @var{expr}, whose length is @var{len}, and collect memory as
41064 it directs. The agent expression @var{expr} is as described in
41065 @ref{Agent Expressions}. Each byte of the expression is encoded as a
41066 two-digit hex number in the packet; @var{len} is the number of bytes
41067 in the expression (and thus one-half the number of hex digits in the
41072 Any number of actions may be packed together in a single @samp{QTDP}
41073 packet, as long as the packet does not exceed the maximum packet
41074 length (400 bytes, for many stubs). There may be only one @samp{R}
41075 action per tracepoint, and it must precede any @samp{M} or @samp{X}
41076 actions. Any registers referred to by @samp{M} and @samp{X} actions
41077 must be collected by a preceding @samp{R} action. (The
41078 ``while-stepping'' actions are treated as if they were attached to a
41079 separate tracepoint, as far as these restrictions are concerned.)
41084 The packet was understood and carried out.
41086 @xref{Tracepoint Packets,,Relocate instruction reply packet}.
41088 The packet was not recognized.
41091 @item QTDPsrc:@var{n}:@var{addr}:@var{type}:@var{start}:@var{slen}:@var{bytes}
41092 @cindex @samp{QTDPsrc} packet
41093 Specify a source string of tracepoint @var{n} at address @var{addr}.
41094 This is useful to get accurate reproduction of the tracepoints
41095 originally downloaded at the beginning of the trace run. The @var{type}
41096 is the name of the tracepoint part, such as @samp{cond} for the
41097 tracepoint's conditional expression (see below for a list of types), while
41098 @var{bytes} is the string, encoded in hexadecimal.
41100 @var{start} is the offset of the @var{bytes} within the overall source
41101 string, while @var{slen} is the total length of the source string.
41102 This is intended for handling source strings that are longer than will
41103 fit in a single packet.
41104 @c Add detailed example when this info is moved into a dedicated
41105 @c tracepoint descriptions section.
41107 The available string types are @samp{at} for the location,
41108 @samp{cond} for the conditional, and @samp{cmd} for an action command.
41109 @value{GDBN} sends a separate packet for each command in the action
41110 list, in the same order in which the commands are stored in the list.
41112 The target does not need to do anything with source strings except
41113 report them back as part of the replies to the @samp{qTfP}/@samp{qTsP}
41116 Although this packet is optional, and @value{GDBN} will only send it
41117 if the target replies with @samp{TracepointSource} @xref{General
41118 Query Packets}, it makes both disconnected tracing and trace files
41119 much easier to use. Otherwise the user must be careful that the
41120 tracepoints in effect while looking at trace frames are identical to
41121 the ones in effect during the trace run; even a small discrepancy
41122 could cause @samp{tdump} not to work, or a particular trace frame not
41125 @item QTDV:@var{n}:@var{value}:@var{builtin}:@var{name}
41126 @cindex define trace state variable, remote request
41127 @cindex @samp{QTDV} packet
41128 Create a new trace state variable, number @var{n}, with an initial
41129 value of @var{value}, which is a 64-bit signed integer. Both @var{n}
41130 and @var{value} are encoded as hexadecimal values. @value{GDBN} has
41131 the option of not using this packet for initial values of zero; the
41132 target should simply create the trace state variables as they are
41133 mentioned in expressions. The value @var{builtin} should be 1 (one)
41134 if the trace state variable is builtin and 0 (zero) if it is not builtin.
41135 @value{GDBN} only sets @var{builtin} to 1 if a previous @samp{qTfV} or
41136 @samp{qTsV} packet had it set. The contents of @var{name} is the
41137 hex-encoded name (without the leading @samp{$}) of the trace state
41140 @item QTFrame:@var{n}
41141 @cindex @samp{QTFrame} packet
41142 Select the @var{n}'th tracepoint frame from the buffer, and use the
41143 register and memory contents recorded there to answer subsequent
41144 request packets from @value{GDBN}.
41146 A successful reply from the stub indicates that the stub has found the
41147 requested frame. The response is a series of parts, concatenated
41148 without separators, describing the frame we selected. Each part has
41149 one of the following forms:
41153 The selected frame is number @var{n} in the trace frame buffer;
41154 @var{f} is a hexadecimal number. If @var{f} is @samp{-1}, then there
41155 was no frame matching the criteria in the request packet.
41158 The selected trace frame records a hit of tracepoint number @var{t};
41159 @var{t} is a hexadecimal number.
41163 @item QTFrame:pc:@var{addr}
41164 Like @samp{QTFrame:@var{n}}, but select the first tracepoint frame after the
41165 currently selected frame whose PC is @var{addr};
41166 @var{addr} is a hexadecimal number.
41168 @item QTFrame:tdp:@var{t}
41169 Like @samp{QTFrame:@var{n}}, but select the first tracepoint frame after the
41170 currently selected frame that is a hit of tracepoint @var{t}; @var{t}
41171 is a hexadecimal number.
41173 @item QTFrame:range:@var{start}:@var{end}
41174 Like @samp{QTFrame:@var{n}}, but select the first tracepoint frame after the
41175 currently selected frame whose PC is between @var{start} (inclusive)
41176 and @var{end} (inclusive); @var{start} and @var{end} are hexadecimal
41179 @item QTFrame:outside:@var{start}:@var{end}
41180 Like @samp{QTFrame:range:@var{start}:@var{end}}, but select the first
41181 frame @emph{outside} the given range of addresses (exclusive).
41184 @cindex @samp{qTMinFTPILen} packet
41185 This packet requests the minimum length of instruction at which a fast
41186 tracepoint (@pxref{Set Tracepoints}) may be placed. For instance, on
41187 the 32-bit x86 architecture, it is possible to use a 4-byte jump, but
41188 it depends on the target system being able to create trampolines in
41189 the first 64K of memory, which might or might not be possible for that
41190 system. So the reply to this packet will be 4 if it is able to
41197 The minimum instruction length is currently unknown.
41199 The minimum instruction length is @var{length}, where @var{length}
41200 is a hexadecimal number greater or equal to 1. A reply
41201 of 1 means that a fast tracepoint may be placed on any instruction
41202 regardless of size.
41204 An error has occurred.
41206 An empty reply indicates that the request is not supported by the stub.
41210 @cindex @samp{QTStart} packet
41211 Begin the tracepoint experiment. Begin collecting data from
41212 tracepoint hits in the trace frame buffer. This packet supports the
41213 @samp{qRelocInsn} reply (@pxref{Tracepoint Packets,,Relocate
41214 instruction reply packet}).
41217 @cindex @samp{QTStop} packet
41218 End the tracepoint experiment. Stop collecting trace frames.
41220 @item QTEnable:@var{n}:@var{addr}
41222 @cindex @samp{QTEnable} packet
41223 Enable tracepoint @var{n} at address @var{addr} in a started tracepoint
41224 experiment. If the tracepoint was previously disabled, then collection
41225 of data from it will resume.
41227 @item QTDisable:@var{n}:@var{addr}
41229 @cindex @samp{QTDisable} packet
41230 Disable tracepoint @var{n} at address @var{addr} in a started tracepoint
41231 experiment. No more data will be collected from the tracepoint unless
41232 @samp{QTEnable:@var{n}:@var{addr}} is subsequently issued.
41235 @cindex @samp{QTinit} packet
41236 Clear the table of tracepoints, and empty the trace frame buffer.
41238 @item QTro:@var{start1},@var{end1}:@var{start2},@var{end2}:@dots{}
41239 @cindex @samp{QTro} packet
41240 Establish the given ranges of memory as ``transparent''. The stub
41241 will answer requests for these ranges from memory's current contents,
41242 if they were not collected as part of the tracepoint hit.
41244 @value{GDBN} uses this to mark read-only regions of memory, like those
41245 containing program code. Since these areas never change, they should
41246 still have the same contents they did when the tracepoint was hit, so
41247 there's no reason for the stub to refuse to provide their contents.
41249 @item QTDisconnected:@var{value}
41250 @cindex @samp{QTDisconnected} packet
41251 Set the choice to what to do with the tracing run when @value{GDBN}
41252 disconnects from the target. A @var{value} of 1 directs the target to
41253 continue the tracing run, while 0 tells the target to stop tracing if
41254 @value{GDBN} is no longer in the picture.
41257 @cindex @samp{qTStatus} packet
41258 Ask the stub if there is a trace experiment running right now.
41260 The reply has the form:
41264 @item T@var{running}@r{[};@var{field}@r{]}@dots{}
41265 @var{running} is a single digit @code{1} if the trace is presently
41266 running, or @code{0} if not. It is followed by semicolon-separated
41267 optional fields that an agent may use to report additional status.
41271 If the trace is not running, the agent may report any of several
41272 explanations as one of the optional fields:
41277 No trace has been run yet.
41279 @item tstop[:@var{text}]:0
41280 The trace was stopped by a user-originated stop command. The optional
41281 @var{text} field is a user-supplied string supplied as part of the
41282 stop command (for instance, an explanation of why the trace was
41283 stopped manually). It is hex-encoded.
41286 The trace stopped because the trace buffer filled up.
41288 @item tdisconnected:0
41289 The trace stopped because @value{GDBN} disconnected from the target.
41291 @item tpasscount:@var{tpnum}
41292 The trace stopped because tracepoint @var{tpnum} exceeded its pass count.
41294 @item terror:@var{text}:@var{tpnum}
41295 The trace stopped because tracepoint @var{tpnum} had an error. The
41296 string @var{text} is available to describe the nature of the error
41297 (for instance, a divide by zero in the condition expression); it
41301 The trace stopped for some other reason.
41305 Additional optional fields supply statistical and other information.
41306 Although not required, they are extremely useful for users monitoring
41307 the progress of a trace run. If a trace has stopped, and these
41308 numbers are reported, they must reflect the state of the just-stopped
41313 @item tframes:@var{n}
41314 The number of trace frames in the buffer.
41316 @item tcreated:@var{n}
41317 The total number of trace frames created during the run. This may
41318 be larger than the trace frame count, if the buffer is circular.
41320 @item tsize:@var{n}
41321 The total size of the trace buffer, in bytes.
41323 @item tfree:@var{n}
41324 The number of bytes still unused in the buffer.
41326 @item circular:@var{n}
41327 The value of the circular trace buffer flag. @code{1} means that the
41328 trace buffer is circular and old trace frames will be discarded if
41329 necessary to make room, @code{0} means that the trace buffer is linear
41332 @item disconn:@var{n}
41333 The value of the disconnected tracing flag. @code{1} means that
41334 tracing will continue after @value{GDBN} disconnects, @code{0} means
41335 that the trace run will stop.
41339 @item qTP:@var{tp}:@var{addr}
41340 @cindex tracepoint status, remote request
41341 @cindex @samp{qTP} packet
41342 Ask the stub for the current state of tracepoint number @var{tp} at
41343 address @var{addr}.
41347 @item V@var{hits}:@var{usage}
41348 The tracepoint has been hit @var{hits} times so far during the trace
41349 run, and accounts for @var{usage} in the trace buffer. Note that
41350 @code{while-stepping} steps are not counted as separate hits, but the
41351 steps' space consumption is added into the usage number.
41355 @item qTV:@var{var}
41356 @cindex trace state variable value, remote request
41357 @cindex @samp{qTV} packet
41358 Ask the stub for the value of the trace state variable number @var{var}.
41363 The value of the variable is @var{value}. This will be the current
41364 value of the variable if the user is examining a running target, or a
41365 saved value if the variable was collected in the trace frame that the
41366 user is looking at. Note that multiple requests may result in
41367 different reply values, such as when requesting values while the
41368 program is running.
41371 The value of the variable is unknown. This would occur, for example,
41372 if the user is examining a trace frame in which the requested variable
41377 @cindex @samp{qTfP} packet
41379 @cindex @samp{qTsP} packet
41380 These packets request data about tracepoints that are being used by
41381 the target. @value{GDBN} sends @code{qTfP} to get the first piece
41382 of data, and multiple @code{qTsP} to get additional pieces. Replies
41383 to these packets generally take the form of the @code{QTDP} packets
41384 that define tracepoints. (FIXME add detailed syntax)
41387 @cindex @samp{qTfV} packet
41389 @cindex @samp{qTsV} packet
41390 These packets request data about trace state variables that are on the
41391 target. @value{GDBN} sends @code{qTfV} to get the first vari of data,
41392 and multiple @code{qTsV} to get additional variables. Replies to
41393 these packets follow the syntax of the @code{QTDV} packets that define
41394 trace state variables.
41400 @cindex @samp{qTfSTM} packet
41401 @cindex @samp{qTsSTM} packet
41402 These packets request data about static tracepoint markers that exist
41403 in the target program. @value{GDBN} sends @code{qTfSTM} to get the
41404 first piece of data, and multiple @code{qTsSTM} to get additional
41405 pieces. Replies to these packets take the following form:
41409 @item m @var{address}:@var{id}:@var{extra}
41411 @item m @var{address}:@var{id}:@var{extra},@var{address}:@var{id}:@var{extra}@dots{}
41412 a comma-separated list of markers
41414 (lower case letter @samp{L}) denotes end of list.
41416 An error occurred. The error number @var{nn} is given as hex digits.
41418 An empty reply indicates that the request is not supported by the
41422 The @var{address} is encoded in hex;
41423 @var{id} and @var{extra} are strings encoded in hex.
41425 In response to each query, the target will reply with a list of one or
41426 more markers, separated by commas. @value{GDBN} will respond to each
41427 reply with a request for more markers (using the @samp{qs} form of the
41428 query), until the target responds with @samp{l} (lower-case ell, for
41431 @item qTSTMat:@var{address}
41433 @cindex @samp{qTSTMat} packet
41434 This packets requests data about static tracepoint markers in the
41435 target program at @var{address}. Replies to this packet follow the
41436 syntax of the @samp{qTfSTM} and @code{qTsSTM} packets that list static
41437 tracepoint markers.
41439 @item QTSave:@var{filename}
41440 @cindex @samp{QTSave} packet
41441 This packet directs the target to save trace data to the file name
41442 @var{filename} in the target's filesystem. The @var{filename} is encoded
41443 as a hex string; the interpretation of the file name (relative vs
41444 absolute, wild cards, etc) is up to the target.
41446 @item qTBuffer:@var{offset},@var{len}
41447 @cindex @samp{qTBuffer} packet
41448 Return up to @var{len} bytes of the current contents of trace buffer,
41449 starting at @var{offset}. The trace buffer is treated as if it were
41450 a contiguous collection of traceframes, as per the trace file format.
41451 The reply consists as many hex-encoded bytes as the target can deliver
41452 in a packet; it is not an error to return fewer than were asked for.
41453 A reply consisting of just @code{l} indicates that no bytes are
41456 @item QTBuffer:circular:@var{value}
41457 This packet directs the target to use a circular trace buffer if
41458 @var{value} is 1, or a linear buffer if the value is 0.
41460 @item QTBuffer:size:@var{size}
41461 @anchor{QTBuffer-size}
41462 @cindex @samp{QTBuffer size} packet
41463 This packet directs the target to make the trace buffer be of size
41464 @var{size} if possible. A value of @code{-1} tells the target to
41465 use whatever size it prefers.
41467 @item QTNotes:@r{[}@var{type}:@var{text}@r{]}@r{[};@var{type}:@var{text}@r{]}@dots{}
41468 @cindex @samp{QTNotes} packet
41469 This packet adds optional textual notes to the trace run. Allowable
41470 types include @code{user}, @code{notes}, and @code{tstop}, the
41471 @var{text} fields are arbitrary strings, hex-encoded.
41475 @subsection Relocate instruction reply packet
41476 When installing fast tracepoints in memory, the target may need to
41477 relocate the instruction currently at the tracepoint address to a
41478 different address in memory. For most instructions, a simple copy is
41479 enough, but, for example, call instructions that implicitly push the
41480 return address on the stack, and relative branches or other
41481 PC-relative instructions require offset adjustment, so that the effect
41482 of executing the instruction at a different address is the same as if
41483 it had executed in the original location.
41485 In response to several of the tracepoint packets, the target may also
41486 respond with a number of intermediate @samp{qRelocInsn} request
41487 packets before the final result packet, to have @value{GDBN} handle
41488 this relocation operation. If a packet supports this mechanism, its
41489 documentation will explicitly say so. See for example the above
41490 descriptions for the @samp{QTStart} and @samp{QTDP} packets. The
41491 format of the request is:
41494 @item qRelocInsn:@var{from};@var{to}
41496 This requests @value{GDBN} to copy instruction at address @var{from}
41497 to address @var{to}, possibly adjusted so that executing the
41498 instruction at @var{to} has the same effect as executing it at
41499 @var{from}. @value{GDBN} writes the adjusted instruction to target
41500 memory starting at @var{to}.
41505 @item qRelocInsn:@var{adjusted_size}
41506 Informs the stub the relocation is complete. The @var{adjusted_size} is
41507 the length in bytes of resulting relocated instruction sequence.
41509 A badly formed request was detected, or an error was encountered while
41510 relocating the instruction.
41513 @node Host I/O Packets
41514 @section Host I/O Packets
41515 @cindex Host I/O, remote protocol
41516 @cindex file transfer, remote protocol
41518 The @dfn{Host I/O} packets allow @value{GDBN} to perform I/O
41519 operations on the far side of a remote link. For example, Host I/O is
41520 used to upload and download files to a remote target with its own
41521 filesystem. Host I/O uses the same constant values and data structure
41522 layout as the target-initiated File-I/O protocol. However, the
41523 Host I/O packets are structured differently. The target-initiated
41524 protocol relies on target memory to store parameters and buffers.
41525 Host I/O requests are initiated by @value{GDBN}, and the
41526 target's memory is not involved. @xref{File-I/O Remote Protocol
41527 Extension}, for more details on the target-initiated protocol.
41529 The Host I/O request packets all encode a single operation along with
41530 its arguments. They have this format:
41534 @item vFile:@var{operation}: @var{parameter}@dots{}
41535 @var{operation} is the name of the particular request; the target
41536 should compare the entire packet name up to the second colon when checking
41537 for a supported operation. The format of @var{parameter} depends on
41538 the operation. Numbers are always passed in hexadecimal. Negative
41539 numbers have an explicit minus sign (i.e.@: two's complement is not
41540 used). Strings (e.g.@: filenames) are encoded as a series of
41541 hexadecimal bytes. The last argument to a system call may be a
41542 buffer of escaped binary data (@pxref{Binary Data}).
41546 The valid responses to Host I/O packets are:
41550 @item F @var{result} [, @var{errno}] [; @var{attachment}]
41551 @var{result} is the integer value returned by this operation, usually
41552 non-negative for success and -1 for errors. If an error has occured,
41553 @var{errno} will be included in the result specifying a
41554 value defined by the File-I/O protocol (@pxref{Errno Values}). For
41555 operations which return data, @var{attachment} supplies the data as a
41556 binary buffer. Binary buffers in response packets are escaped in the
41557 normal way (@pxref{Binary Data}). See the individual packet
41558 documentation for the interpretation of @var{result} and
41562 An empty response indicates that this operation is not recognized.
41566 These are the supported Host I/O operations:
41569 @item vFile:open: @var{filename}, @var{flags}, @var{mode}
41570 Open a file at @var{filename} and return a file descriptor for it, or
41571 return -1 if an error occurs. The @var{filename} is a string,
41572 @var{flags} is an integer indicating a mask of open flags
41573 (@pxref{Open Flags}), and @var{mode} is an integer indicating a mask
41574 of mode bits to use if the file is created (@pxref{mode_t Values}).
41575 @xref{open}, for details of the open flags and mode values.
41577 @item vFile:close: @var{fd}
41578 Close the open file corresponding to @var{fd} and return 0, or
41579 -1 if an error occurs.
41581 @item vFile:pread: @var{fd}, @var{count}, @var{offset}
41582 Read data from the open file corresponding to @var{fd}. Up to
41583 @var{count} bytes will be read from the file, starting at @var{offset}
41584 relative to the start of the file. The target may read fewer bytes;
41585 common reasons include packet size limits and an end-of-file
41586 condition. The number of bytes read is returned. Zero should only be
41587 returned for a successful read at the end of the file, or if
41588 @var{count} was zero.
41590 The data read should be returned as a binary attachment on success.
41591 If zero bytes were read, the response should include an empty binary
41592 attachment (i.e.@: a trailing semicolon). The return value is the
41593 number of target bytes read; the binary attachment may be longer if
41594 some characters were escaped.
41596 @item vFile:pwrite: @var{fd}, @var{offset}, @var{data}
41597 Write @var{data} (a binary buffer) to the open file corresponding
41598 to @var{fd}. Start the write at @var{offset} from the start of the
41599 file. Unlike many @code{write} system calls, there is no
41600 separate @var{count} argument; the length of @var{data} in the
41601 packet is used. @samp{vFile:write} returns the number of bytes written,
41602 which may be shorter than the length of @var{data}, or -1 if an
41605 @item vFile:fstat: @var{fd}
41606 Get information about the open file corresponding to @var{fd}.
41607 On success the information is returned as a binary attachment
41608 and the return value is the size of this attachment in bytes.
41609 If an error occurs the return value is -1. The format of the
41610 returned binary attachment is as described in @ref{struct stat}.
41612 @item vFile:unlink: @var{filename}
41613 Delete the file at @var{filename} on the target. Return 0,
41614 or -1 if an error occurs. The @var{filename} is a string.
41616 @item vFile:readlink: @var{filename}
41617 Read value of symbolic link @var{filename} on the target. Return
41618 the number of bytes read, or -1 if an error occurs.
41620 The data read should be returned as a binary attachment on success.
41621 If zero bytes were read, the response should include an empty binary
41622 attachment (i.e.@: a trailing semicolon). The return value is the
41623 number of target bytes read; the binary attachment may be longer if
41624 some characters were escaped.
41626 @item vFile:setfs: @var{pid}
41627 Select the filesystem on which @code{vFile} operations with
41628 @var{filename} arguments will operate. This is required for
41629 @value{GDBN} to be able to access files on remote targets where
41630 the remote stub does not share a common filesystem with the
41633 If @var{pid} is nonzero, select the filesystem as seen by process
41634 @var{pid}. If @var{pid} is zero, select the filesystem as seen by
41635 the remote stub. Return 0 on success, or -1 if an error occurs.
41636 If @code{vFile:setfs:} indicates success, the selected filesystem
41637 remains selected until the next successful @code{vFile:setfs:}
41643 @section Interrupts
41644 @cindex interrupts (remote protocol)
41645 @anchor{interrupting remote targets}
41647 In all-stop mode, when a program on the remote target is running,
41648 @value{GDBN} may attempt to interrupt it by sending a @samp{Ctrl-C},
41649 @code{BREAK} or a @code{BREAK} followed by @code{g}, control of which
41650 is specified via @value{GDBN}'s @samp{interrupt-sequence}.
41652 The precise meaning of @code{BREAK} is defined by the transport
41653 mechanism and may, in fact, be undefined. @value{GDBN} does not
41654 currently define a @code{BREAK} mechanism for any of the network
41655 interfaces except for TCP, in which case @value{GDBN} sends the
41656 @code{telnet} BREAK sequence.
41658 @samp{Ctrl-C}, on the other hand, is defined and implemented for all
41659 transport mechanisms. It is represented by sending the single byte
41660 @code{0x03} without any of the usual packet overhead described in
41661 the Overview section (@pxref{Overview}). When a @code{0x03} byte is
41662 transmitted as part of a packet, it is considered to be packet data
41663 and does @emph{not} represent an interrupt. E.g., an @samp{X} packet
41664 (@pxref{X packet}), used for binary downloads, may include an unescaped
41665 @code{0x03} as part of its packet.
41667 @code{BREAK} followed by @code{g} is also known as Magic SysRq g.
41668 When Linux kernel receives this sequence from serial port,
41669 it stops execution and connects to gdb.
41671 In non-stop mode, because packet resumptions are asynchronous
41672 (@pxref{vCont packet}), @value{GDBN} is always free to send a remote
41673 command to the remote stub, even when the target is running. For that
41674 reason, @value{GDBN} instead sends a regular packet (@pxref{vCtrlC
41675 packet}) with the usual packet framing instead of the single byte
41678 Stubs are not required to recognize these interrupt mechanisms and the
41679 precise meaning associated with receipt of the interrupt is
41680 implementation defined. If the target supports debugging of multiple
41681 threads and/or processes, it should attempt to interrupt all
41682 currently-executing threads and processes.
41683 If the stub is successful at interrupting the
41684 running program, it should send one of the stop
41685 reply packets (@pxref{Stop Reply Packets}) to @value{GDBN} as a result
41686 of successfully stopping the program in all-stop mode, and a stop reply
41687 for each stopped thread in non-stop mode.
41688 Interrupts received while the
41689 program is stopped are queued and the program will be interrupted when
41690 it is resumed next time.
41692 @node Notification Packets
41693 @section Notification Packets
41694 @cindex notification packets
41695 @cindex packets, notification
41697 The @value{GDBN} remote serial protocol includes @dfn{notifications},
41698 packets that require no acknowledgment. Both the GDB and the stub
41699 may send notifications (although the only notifications defined at
41700 present are sent by the stub). Notifications carry information
41701 without incurring the round-trip latency of an acknowledgment, and so
41702 are useful for low-impact communications where occasional packet loss
41705 A notification packet has the form @samp{% @var{data} #
41706 @var{checksum}}, where @var{data} is the content of the notification,
41707 and @var{checksum} is a checksum of @var{data}, computed and formatted
41708 as for ordinary @value{GDBN} packets. A notification's @var{data}
41709 never contains @samp{$}, @samp{%} or @samp{#} characters. Upon
41710 receiving a notification, the recipient sends no @samp{+} or @samp{-}
41711 to acknowledge the notification's receipt or to report its corruption.
41713 Every notification's @var{data} begins with a name, which contains no
41714 colon characters, followed by a colon character.
41716 Recipients should silently ignore corrupted notifications and
41717 notifications they do not understand. Recipients should restart
41718 timeout periods on receipt of a well-formed notification, whether or
41719 not they understand it.
41721 Senders should only send the notifications described here when this
41722 protocol description specifies that they are permitted. In the
41723 future, we may extend the protocol to permit existing notifications in
41724 new contexts; this rule helps older senders avoid confusing newer
41727 (Older versions of @value{GDBN} ignore bytes received until they see
41728 the @samp{$} byte that begins an ordinary packet, so new stubs may
41729 transmit notifications without fear of confusing older clients. There
41730 are no notifications defined for @value{GDBN} to send at the moment, but we
41731 assume that most older stubs would ignore them, as well.)
41733 Each notification is comprised of three parts:
41735 @item @var{name}:@var{event}
41736 The notification packet is sent by the side that initiates the
41737 exchange (currently, only the stub does that), with @var{event}
41738 carrying the specific information about the notification, and
41739 @var{name} specifying the name of the notification.
41741 The acknowledge sent by the other side, usually @value{GDBN}, to
41742 acknowledge the exchange and request the event.
41745 The purpose of an asynchronous notification mechanism is to report to
41746 @value{GDBN} that something interesting happened in the remote stub.
41748 The remote stub may send notification @var{name}:@var{event}
41749 at any time, but @value{GDBN} acknowledges the notification when
41750 appropriate. The notification event is pending before @value{GDBN}
41751 acknowledges. Only one notification at a time may be pending; if
41752 additional events occur before @value{GDBN} has acknowledged the
41753 previous notification, they must be queued by the stub for later
41754 synchronous transmission in response to @var{ack} packets from
41755 @value{GDBN}. Because the notification mechanism is unreliable,
41756 the stub is permitted to resend a notification if it believes
41757 @value{GDBN} may not have received it.
41759 Specifically, notifications may appear when @value{GDBN} is not
41760 otherwise reading input from the stub, or when @value{GDBN} is
41761 expecting to read a normal synchronous response or a
41762 @samp{+}/@samp{-} acknowledgment to a packet it has sent.
41763 Notification packets are distinct from any other communication from
41764 the stub so there is no ambiguity.
41766 After receiving a notification, @value{GDBN} shall acknowledge it by
41767 sending a @var{ack} packet as a regular, synchronous request to the
41768 stub. Such acknowledgment is not required to happen immediately, as
41769 @value{GDBN} is permitted to send other, unrelated packets to the
41770 stub first, which the stub should process normally.
41772 Upon receiving a @var{ack} packet, if the stub has other queued
41773 events to report to @value{GDBN}, it shall respond by sending a
41774 normal @var{event}. @value{GDBN} shall then send another @var{ack}
41775 packet to solicit further responses; again, it is permitted to send
41776 other, unrelated packets as well which the stub should process
41779 If the stub receives a @var{ack} packet and there are no additional
41780 @var{event} to report, the stub shall return an @samp{OK} response.
41781 At this point, @value{GDBN} has finished processing a notification
41782 and the stub has completed sending any queued events. @value{GDBN}
41783 won't accept any new notifications until the final @samp{OK} is
41784 received . If further notification events occur, the stub shall send
41785 a new notification, @value{GDBN} shall accept the notification, and
41786 the process shall be repeated.
41788 The process of asynchronous notification can be illustrated by the
41791 <- @code{%Stop:T0505:98e7ffbf;04:4ce6ffbf;08:b1b6e54c;thread:p7526.7526;core:0;}
41794 <- @code{T0505:68f37db7;04:40f37db7;08:63850408;thread:p7526.7528;core:0;}
41796 <- @code{T0505:68e3fdb6;04:40e3fdb6;08:63850408;thread:p7526.7529;core:0;}
41801 The following notifications are defined:
41802 @multitable @columnfractions 0.12 0.12 0.38 0.38
41811 @tab @var{reply}. The @var{reply} has the form of a stop reply, as
41812 described in @ref{Stop Reply Packets}. Refer to @ref{Remote Non-Stop},
41813 for information on how these notifications are acknowledged by
41815 @tab Report an asynchronous stop event in non-stop mode.
41819 @node Remote Non-Stop
41820 @section Remote Protocol Support for Non-Stop Mode
41822 @value{GDBN}'s remote protocol supports non-stop debugging of
41823 multi-threaded programs, as described in @ref{Non-Stop Mode}. If the stub
41824 supports non-stop mode, it should report that to @value{GDBN} by including
41825 @samp{QNonStop+} in its @samp{qSupported} response (@pxref{qSupported}).
41827 @value{GDBN} typically sends a @samp{QNonStop} packet only when
41828 establishing a new connection with the stub. Entering non-stop mode
41829 does not alter the state of any currently-running threads, but targets
41830 must stop all threads in any already-attached processes when entering
41831 all-stop mode. @value{GDBN} uses the @samp{?} packet as necessary to
41832 probe the target state after a mode change.
41834 In non-stop mode, when an attached process encounters an event that
41835 would otherwise be reported with a stop reply, it uses the
41836 asynchronous notification mechanism (@pxref{Notification Packets}) to
41837 inform @value{GDBN}. In contrast to all-stop mode, where all threads
41838 in all processes are stopped when a stop reply is sent, in non-stop
41839 mode only the thread reporting the stop event is stopped. That is,
41840 when reporting a @samp{S} or @samp{T} response to indicate completion
41841 of a step operation, hitting a breakpoint, or a fault, only the
41842 affected thread is stopped; any other still-running threads continue
41843 to run. When reporting a @samp{W} or @samp{X} response, all running
41844 threads belonging to other attached processes continue to run.
41846 In non-stop mode, the target shall respond to the @samp{?} packet as
41847 follows. First, any incomplete stop reply notification/@samp{vStopped}
41848 sequence in progress is abandoned. The target must begin a new
41849 sequence reporting stop events for all stopped threads, whether or not
41850 it has previously reported those events to @value{GDBN}. The first
41851 stop reply is sent as a synchronous reply to the @samp{?} packet, and
41852 subsequent stop replies are sent as responses to @samp{vStopped} packets
41853 using the mechanism described above. The target must not send
41854 asynchronous stop reply notifications until the sequence is complete.
41855 If all threads are running when the target receives the @samp{?} packet,
41856 or if the target is not attached to any process, it shall respond
41859 If the stub supports non-stop mode, it should also support the
41860 @samp{swbreak} stop reason if software breakpoints are supported, and
41861 the @samp{hwbreak} stop reason if hardware breakpoints are supported
41862 (@pxref{swbreak stop reason}). This is because given the asynchronous
41863 nature of non-stop mode, between the time a thread hits a breakpoint
41864 and the time the event is finally processed by @value{GDBN}, the
41865 breakpoint may have already been removed from the target. Due to
41866 this, @value{GDBN} needs to be able to tell whether a trap stop was
41867 caused by a delayed breakpoint event, which should be ignored, as
41868 opposed to a random trap signal, which should be reported to the user.
41869 Note the @samp{swbreak} feature implies that the target is responsible
41870 for adjusting the PC when a software breakpoint triggers, if
41871 necessary, such as on the x86 architecture.
41873 @node Packet Acknowledgment
41874 @section Packet Acknowledgment
41876 @cindex acknowledgment, for @value{GDBN} remote
41877 @cindex packet acknowledgment, for @value{GDBN} remote
41878 By default, when either the host or the target machine receives a packet,
41879 the first response expected is an acknowledgment: either @samp{+} (to indicate
41880 the package was received correctly) or @samp{-} (to request retransmission).
41881 This mechanism allows the @value{GDBN} remote protocol to operate over
41882 unreliable transport mechanisms, such as a serial line.
41884 In cases where the transport mechanism is itself reliable (such as a pipe or
41885 TCP connection), the @samp{+}/@samp{-} acknowledgments are redundant.
41886 It may be desirable to disable them in that case to reduce communication
41887 overhead, or for other reasons. This can be accomplished by means of the
41888 @samp{QStartNoAckMode} packet; @pxref{QStartNoAckMode}.
41890 When in no-acknowledgment mode, neither the stub nor @value{GDBN} shall send or
41891 expect @samp{+}/@samp{-} protocol acknowledgments. The packet
41892 and response format still includes the normal checksum, as described in
41893 @ref{Overview}, but the checksum may be ignored by the receiver.
41895 If the stub supports @samp{QStartNoAckMode} and prefers to operate in
41896 no-acknowledgment mode, it should report that to @value{GDBN}
41897 by including @samp{QStartNoAckMode+} in its response to @samp{qSupported};
41898 @pxref{qSupported}.
41899 If @value{GDBN} also supports @samp{QStartNoAckMode} and it has not been
41900 disabled via the @code{set remote noack-packet off} command
41901 (@pxref{Remote Configuration}),
41902 @value{GDBN} may then send a @samp{QStartNoAckMode} packet to the stub.
41903 Only then may the stub actually turn off packet acknowledgments.
41904 @value{GDBN} sends a final @samp{+} acknowledgment of the stub's @samp{OK}
41905 response, which can be safely ignored by the stub.
41907 Note that @code{set remote noack-packet} command only affects negotiation
41908 between @value{GDBN} and the stub when subsequent connections are made;
41909 it does not affect the protocol acknowledgment state for any current
41911 Since @samp{+}/@samp{-} acknowledgments are enabled by default when a
41912 new connection is established,
41913 there is also no protocol request to re-enable the acknowledgments
41914 for the current connection, once disabled.
41919 Example sequence of a target being re-started. Notice how the restart
41920 does not get any direct output:
41925 @emph{target restarts}
41928 <- @code{T001:1234123412341234}
41932 Example sequence of a target being stepped by a single instruction:
41935 -> @code{G1445@dots{}}
41940 <- @code{T001:1234123412341234}
41944 <- @code{1455@dots{}}
41948 @node File-I/O Remote Protocol Extension
41949 @section File-I/O Remote Protocol Extension
41950 @cindex File-I/O remote protocol extension
41953 * File-I/O Overview::
41954 * Protocol Basics::
41955 * The F Request Packet::
41956 * The F Reply Packet::
41957 * The Ctrl-C Message::
41959 * List of Supported Calls::
41960 * Protocol-specific Representation of Datatypes::
41962 * File-I/O Examples::
41965 @node File-I/O Overview
41966 @subsection File-I/O Overview
41967 @cindex file-i/o overview
41969 The @dfn{File I/O remote protocol extension} (short: File-I/O) allows the
41970 target to use the host's file system and console I/O to perform various
41971 system calls. System calls on the target system are translated into a
41972 remote protocol packet to the host system, which then performs the needed
41973 actions and returns a response packet to the target system.
41974 This simulates file system operations even on targets that lack file systems.
41976 The protocol is defined to be independent of both the host and target systems.
41977 It uses its own internal representation of datatypes and values. Both
41978 @value{GDBN} and the target's @value{GDBN} stub are responsible for
41979 translating the system-dependent value representations into the internal
41980 protocol representations when data is transmitted.
41982 The communication is synchronous. A system call is possible only when
41983 @value{GDBN} is waiting for a response from the @samp{C}, @samp{c}, @samp{S}
41984 or @samp{s} packets. While @value{GDBN} handles the request for a system call,
41985 the target is stopped to allow deterministic access to the target's
41986 memory. Therefore File-I/O is not interruptible by target signals. On
41987 the other hand, it is possible to interrupt File-I/O by a user interrupt
41988 (@samp{Ctrl-C}) within @value{GDBN}.
41990 The target's request to perform a host system call does not finish
41991 the latest @samp{C}, @samp{c}, @samp{S} or @samp{s} action. That means,
41992 after finishing the system call, the target returns to continuing the
41993 previous activity (continue, step). No additional continue or step
41994 request from @value{GDBN} is required.
41997 (@value{GDBP}) continue
41998 <- target requests 'system call X'
41999 target is stopped, @value{GDBN} executes system call
42000 -> @value{GDBN} returns result
42001 ... target continues, @value{GDBN} returns to wait for the target
42002 <- target hits breakpoint and sends a Txx packet
42005 The protocol only supports I/O on the console and to regular files on
42006 the host file system. Character or block special devices, pipes,
42007 named pipes, sockets or any other communication method on the host
42008 system are not supported by this protocol.
42010 File I/O is not supported in non-stop mode.
42012 @node Protocol Basics
42013 @subsection Protocol Basics
42014 @cindex protocol basics, file-i/o
42016 The File-I/O protocol uses the @code{F} packet as the request as well
42017 as reply packet. Since a File-I/O system call can only occur when
42018 @value{GDBN} is waiting for a response from the continuing or stepping target,
42019 the File-I/O request is a reply that @value{GDBN} has to expect as a result
42020 of a previous @samp{C}, @samp{c}, @samp{S} or @samp{s} packet.
42021 This @code{F} packet contains all information needed to allow @value{GDBN}
42022 to call the appropriate host system call:
42026 A unique identifier for the requested system call.
42029 All parameters to the system call. Pointers are given as addresses
42030 in the target memory address space. Pointers to strings are given as
42031 pointer/length pair. Numerical values are given as they are.
42032 Numerical control flags are given in a protocol-specific representation.
42036 At this point, @value{GDBN} has to perform the following actions.
42040 If the parameters include pointer values to data needed as input to a
42041 system call, @value{GDBN} requests this data from the target with a
42042 standard @code{m} packet request. This additional communication has to be
42043 expected by the target implementation and is handled as any other @code{m}
42047 @value{GDBN} translates all value from protocol representation to host
42048 representation as needed. Datatypes are coerced into the host types.
42051 @value{GDBN} calls the system call.
42054 It then coerces datatypes back to protocol representation.
42057 If the system call is expected to return data in buffer space specified
42058 by pointer parameters to the call, the data is transmitted to the
42059 target using a @code{M} or @code{X} packet. This packet has to be expected
42060 by the target implementation and is handled as any other @code{M} or @code{X}
42065 Eventually @value{GDBN} replies with another @code{F} packet which contains all
42066 necessary information for the target to continue. This at least contains
42073 @code{errno}, if has been changed by the system call.
42080 After having done the needed type and value coercion, the target continues
42081 the latest continue or step action.
42083 @node The F Request Packet
42084 @subsection The @code{F} Request Packet
42085 @cindex file-i/o request packet
42086 @cindex @code{F} request packet
42088 The @code{F} request packet has the following format:
42091 @item F@var{call-id},@var{parameter@dots{}}
42093 @var{call-id} is the identifier to indicate the host system call to be called.
42094 This is just the name of the function.
42096 @var{parameter@dots{}} are the parameters to the system call.
42097 Parameters are hexadecimal integer values, either the actual values in case
42098 of scalar datatypes, pointers to target buffer space in case of compound
42099 datatypes and unspecified memory areas, or pointer/length pairs in case
42100 of string parameters. These are appended to the @var{call-id} as a
42101 comma-delimited list. All values are transmitted in ASCII
42102 string representation, pointer/length pairs separated by a slash.
42108 @node The F Reply Packet
42109 @subsection The @code{F} Reply Packet
42110 @cindex file-i/o reply packet
42111 @cindex @code{F} reply packet
42113 The @code{F} reply packet has the following format:
42117 @item F@var{retcode},@var{errno},@var{Ctrl-C flag};@var{call-specific attachment}
42119 @var{retcode} is the return code of the system call as hexadecimal value.
42121 @var{errno} is the @code{errno} set by the call, in protocol-specific
42123 This parameter can be omitted if the call was successful.
42125 @var{Ctrl-C flag} is only sent if the user requested a break. In this
42126 case, @var{errno} must be sent as well, even if the call was successful.
42127 The @var{Ctrl-C flag} itself consists of the character @samp{C}:
42134 or, if the call was interrupted before the host call has been performed:
42141 assuming 4 is the protocol-specific representation of @code{EINTR}.
42146 @node The Ctrl-C Message
42147 @subsection The @samp{Ctrl-C} Message
42148 @cindex ctrl-c message, in file-i/o protocol
42150 If the @samp{Ctrl-C} flag is set in the @value{GDBN}
42151 reply packet (@pxref{The F Reply Packet}),
42152 the target should behave as if it had
42153 gotten a break message. The meaning for the target is ``system call
42154 interrupted by @code{SIGINT}''. Consequentially, the target should actually stop
42155 (as with a break message) and return to @value{GDBN} with a @code{T02}
42158 It's important for the target to know in which
42159 state the system call was interrupted. There are two possible cases:
42163 The system call hasn't been performed on the host yet.
42166 The system call on the host has been finished.
42170 These two states can be distinguished by the target by the value of the
42171 returned @code{errno}. If it's the protocol representation of @code{EINTR}, the system
42172 call hasn't been performed. This is equivalent to the @code{EINTR} handling
42173 on POSIX systems. In any other case, the target may presume that the
42174 system call has been finished --- successfully or not --- and should behave
42175 as if the break message arrived right after the system call.
42177 @value{GDBN} must behave reliably. If the system call has not been called
42178 yet, @value{GDBN} may send the @code{F} reply immediately, setting @code{EINTR} as
42179 @code{errno} in the packet. If the system call on the host has been finished
42180 before the user requests a break, the full action must be finished by
42181 @value{GDBN}. This requires sending @code{M} or @code{X} packets as necessary.
42182 The @code{F} packet may only be sent when either nothing has happened
42183 or the full action has been completed.
42186 @subsection Console I/O
42187 @cindex console i/o as part of file-i/o
42189 By default and if not explicitly closed by the target system, the file
42190 descriptors 0, 1 and 2 are connected to the @value{GDBN} console. Output
42191 on the @value{GDBN} console is handled as any other file output operation
42192 (@code{write(1, @dots{})} or @code{write(2, @dots{})}). Console input is handled
42193 by @value{GDBN} so that after the target read request from file descriptor
42194 0 all following typing is buffered until either one of the following
42199 The user types @kbd{Ctrl-c}. The behaviour is as explained above, and the
42201 system call is treated as finished.
42204 The user presses @key{RET}. This is treated as end of input with a trailing
42208 The user types @kbd{Ctrl-d}. This is treated as end of input. No trailing
42209 character (neither newline nor @samp{Ctrl-D}) is appended to the input.
42213 If the user has typed more characters than fit in the buffer given to
42214 the @code{read} call, the trailing characters are buffered in @value{GDBN} until
42215 either another @code{read(0, @dots{})} is requested by the target, or debugging
42216 is stopped at the user's request.
42219 @node List of Supported Calls
42220 @subsection List of Supported Calls
42221 @cindex list of supported file-i/o calls
42238 @unnumberedsubsubsec open
42239 @cindex open, file-i/o system call
42244 int open(const char *pathname, int flags);
42245 int open(const char *pathname, int flags, mode_t mode);
42249 @samp{Fopen,@var{pathptr}/@var{len},@var{flags},@var{mode}}
42252 @var{flags} is the bitwise @code{OR} of the following values:
42256 If the file does not exist it will be created. The host
42257 rules apply as far as file ownership and time stamps
42261 When used with @code{O_CREAT}, if the file already exists it is
42262 an error and open() fails.
42265 If the file already exists and the open mode allows
42266 writing (@code{O_RDWR} or @code{O_WRONLY} is given) it will be
42267 truncated to zero length.
42270 The file is opened in append mode.
42273 The file is opened for reading only.
42276 The file is opened for writing only.
42279 The file is opened for reading and writing.
42283 Other bits are silently ignored.
42287 @var{mode} is the bitwise @code{OR} of the following values:
42291 User has read permission.
42294 User has write permission.
42297 Group has read permission.
42300 Group has write permission.
42303 Others have read permission.
42306 Others have write permission.
42310 Other bits are silently ignored.
42313 @item Return value:
42314 @code{open} returns the new file descriptor or -1 if an error
42321 @var{pathname} already exists and @code{O_CREAT} and @code{O_EXCL} were used.
42324 @var{pathname} refers to a directory.
42327 The requested access is not allowed.
42330 @var{pathname} was too long.
42333 A directory component in @var{pathname} does not exist.
42336 @var{pathname} refers to a device, pipe, named pipe or socket.
42339 @var{pathname} refers to a file on a read-only filesystem and
42340 write access was requested.
42343 @var{pathname} is an invalid pointer value.
42346 No space on device to create the file.
42349 The process already has the maximum number of files open.
42352 The limit on the total number of files open on the system
42356 The call was interrupted by the user.
42362 @unnumberedsubsubsec close
42363 @cindex close, file-i/o system call
42372 @samp{Fclose,@var{fd}}
42374 @item Return value:
42375 @code{close} returns zero on success, or -1 if an error occurred.
42381 @var{fd} isn't a valid open file descriptor.
42384 The call was interrupted by the user.
42390 @unnumberedsubsubsec read
42391 @cindex read, file-i/o system call
42396 int read(int fd, void *buf, unsigned int count);
42400 @samp{Fread,@var{fd},@var{bufptr},@var{count}}
42402 @item Return value:
42403 On success, the number of bytes read is returned.
42404 Zero indicates end of file. If count is zero, read
42405 returns zero as well. On error, -1 is returned.
42411 @var{fd} is not a valid file descriptor or is not open for
42415 @var{bufptr} is an invalid pointer value.
42418 The call was interrupted by the user.
42424 @unnumberedsubsubsec write
42425 @cindex write, file-i/o system call
42430 int write(int fd, const void *buf, unsigned int count);
42434 @samp{Fwrite,@var{fd},@var{bufptr},@var{count}}
42436 @item Return value:
42437 On success, the number of bytes written are returned.
42438 Zero indicates nothing was written. On error, -1
42445 @var{fd} is not a valid file descriptor or is not open for
42449 @var{bufptr} is an invalid pointer value.
42452 An attempt was made to write a file that exceeds the
42453 host-specific maximum file size allowed.
42456 No space on device to write the data.
42459 The call was interrupted by the user.
42465 @unnumberedsubsubsec lseek
42466 @cindex lseek, file-i/o system call
42471 long lseek (int fd, long offset, int flag);
42475 @samp{Flseek,@var{fd},@var{offset},@var{flag}}
42477 @var{flag} is one of:
42481 The offset is set to @var{offset} bytes.
42484 The offset is set to its current location plus @var{offset}
42488 The offset is set to the size of the file plus @var{offset}
42492 @item Return value:
42493 On success, the resulting unsigned offset in bytes from
42494 the beginning of the file is returned. Otherwise, a
42495 value of -1 is returned.
42501 @var{fd} is not a valid open file descriptor.
42504 @var{fd} is associated with the @value{GDBN} console.
42507 @var{flag} is not a proper value.
42510 The call was interrupted by the user.
42516 @unnumberedsubsubsec rename
42517 @cindex rename, file-i/o system call
42522 int rename(const char *oldpath, const char *newpath);
42526 @samp{Frename,@var{oldpathptr}/@var{len},@var{newpathptr}/@var{len}}
42528 @item Return value:
42529 On success, zero is returned. On error, -1 is returned.
42535 @var{newpath} is an existing directory, but @var{oldpath} is not a
42539 @var{newpath} is a non-empty directory.
42542 @var{oldpath} or @var{newpath} is a directory that is in use by some
42546 An attempt was made to make a directory a subdirectory
42550 A component used as a directory in @var{oldpath} or new
42551 path is not a directory. Or @var{oldpath} is a directory
42552 and @var{newpath} exists but is not a directory.
42555 @var{oldpathptr} or @var{newpathptr} are invalid pointer values.
42558 No access to the file or the path of the file.
42562 @var{oldpath} or @var{newpath} was too long.
42565 A directory component in @var{oldpath} or @var{newpath} does not exist.
42568 The file is on a read-only filesystem.
42571 The device containing the file has no room for the new
42575 The call was interrupted by the user.
42581 @unnumberedsubsubsec unlink
42582 @cindex unlink, file-i/o system call
42587 int unlink(const char *pathname);
42591 @samp{Funlink,@var{pathnameptr}/@var{len}}
42593 @item Return value:
42594 On success, zero is returned. On error, -1 is returned.
42600 No access to the file or the path of the file.
42603 The system does not allow unlinking of directories.
42606 The file @var{pathname} cannot be unlinked because it's
42607 being used by another process.
42610 @var{pathnameptr} is an invalid pointer value.
42613 @var{pathname} was too long.
42616 A directory component in @var{pathname} does not exist.
42619 A component of the path is not a directory.
42622 The file is on a read-only filesystem.
42625 The call was interrupted by the user.
42631 @unnumberedsubsubsec stat/fstat
42632 @cindex fstat, file-i/o system call
42633 @cindex stat, file-i/o system call
42638 int stat(const char *pathname, struct stat *buf);
42639 int fstat(int fd, struct stat *buf);
42643 @samp{Fstat,@var{pathnameptr}/@var{len},@var{bufptr}}@*
42644 @samp{Ffstat,@var{fd},@var{bufptr}}
42646 @item Return value:
42647 On success, zero is returned. On error, -1 is returned.
42653 @var{fd} is not a valid open file.
42656 A directory component in @var{pathname} does not exist or the
42657 path is an empty string.
42660 A component of the path is not a directory.
42663 @var{pathnameptr} is an invalid pointer value.
42666 No access to the file or the path of the file.
42669 @var{pathname} was too long.
42672 The call was interrupted by the user.
42678 @unnumberedsubsubsec gettimeofday
42679 @cindex gettimeofday, file-i/o system call
42684 int gettimeofday(struct timeval *tv, void *tz);
42688 @samp{Fgettimeofday,@var{tvptr},@var{tzptr}}
42690 @item Return value:
42691 On success, 0 is returned, -1 otherwise.
42697 @var{tz} is a non-NULL pointer.
42700 @var{tvptr} and/or @var{tzptr} is an invalid pointer value.
42706 @unnumberedsubsubsec isatty
42707 @cindex isatty, file-i/o system call
42712 int isatty(int fd);
42716 @samp{Fisatty,@var{fd}}
42718 @item Return value:
42719 Returns 1 if @var{fd} refers to the @value{GDBN} console, 0 otherwise.
42725 The call was interrupted by the user.
42730 Note that the @code{isatty} call is treated as a special case: it returns
42731 1 to the target if the file descriptor is attached
42732 to the @value{GDBN} console, 0 otherwise. Implementing through system calls
42733 would require implementing @code{ioctl} and would be more complex than
42738 @unnumberedsubsubsec system
42739 @cindex system, file-i/o system call
42744 int system(const char *command);
42748 @samp{Fsystem,@var{commandptr}/@var{len}}
42750 @item Return value:
42751 If @var{len} is zero, the return value indicates whether a shell is
42752 available. A zero return value indicates a shell is not available.
42753 For non-zero @var{len}, the value returned is -1 on error and the
42754 return status of the command otherwise. Only the exit status of the
42755 command is returned, which is extracted from the host's @code{system}
42756 return value by calling @code{WEXITSTATUS(retval)}. In case
42757 @file{/bin/sh} could not be executed, 127 is returned.
42763 The call was interrupted by the user.
42768 @value{GDBN} takes over the full task of calling the necessary host calls
42769 to perform the @code{system} call. The return value of @code{system} on
42770 the host is simplified before it's returned
42771 to the target. Any termination signal information from the child process
42772 is discarded, and the return value consists
42773 entirely of the exit status of the called command.
42775 Due to security concerns, the @code{system} call is by default refused
42776 by @value{GDBN}. The user has to allow this call explicitly with the
42777 @code{set remote system-call-allowed 1} command.
42780 @item set remote system-call-allowed
42781 @kindex set remote system-call-allowed
42782 Control whether to allow the @code{system} calls in the File I/O
42783 protocol for the remote target. The default is zero (disabled).
42785 @item show remote system-call-allowed
42786 @kindex show remote system-call-allowed
42787 Show whether the @code{system} calls are allowed in the File I/O
42791 @node Protocol-specific Representation of Datatypes
42792 @subsection Protocol-specific Representation of Datatypes
42793 @cindex protocol-specific representation of datatypes, in file-i/o protocol
42796 * Integral Datatypes::
42798 * Memory Transfer::
42803 @node Integral Datatypes
42804 @unnumberedsubsubsec Integral Datatypes
42805 @cindex integral datatypes, in file-i/o protocol
42807 The integral datatypes used in the system calls are @code{int},
42808 @code{unsigned int}, @code{long}, @code{unsigned long},
42809 @code{mode_t}, and @code{time_t}.
42811 @code{int}, @code{unsigned int}, @code{mode_t} and @code{time_t} are
42812 implemented as 32 bit values in this protocol.
42814 @code{long} and @code{unsigned long} are implemented as 64 bit types.
42816 @xref{Limits}, for corresponding MIN and MAX values (similar to those
42817 in @file{limits.h}) to allow range checking on host and target.
42819 @code{time_t} datatypes are defined as seconds since the Epoch.
42821 All integral datatypes transferred as part of a memory read or write of a
42822 structured datatype e.g.@: a @code{struct stat} have to be given in big endian
42825 @node Pointer Values
42826 @unnumberedsubsubsec Pointer Values
42827 @cindex pointer values, in file-i/o protocol
42829 Pointers to target data are transmitted as they are. An exception
42830 is made for pointers to buffers for which the length isn't
42831 transmitted as part of the function call, namely strings. Strings
42832 are transmitted as a pointer/length pair, both as hex values, e.g.@:
42839 which is a pointer to data of length 18 bytes at position 0x1aaf.
42840 The length is defined as the full string length in bytes, including
42841 the trailing null byte. For example, the string @code{"hello world"}
42842 at address 0x123456 is transmitted as
42848 @node Memory Transfer
42849 @unnumberedsubsubsec Memory Transfer
42850 @cindex memory transfer, in file-i/o protocol
42852 Structured data which is transferred using a memory read or write (for
42853 example, a @code{struct stat}) is expected to be in a protocol-specific format
42854 with all scalar multibyte datatypes being big endian. Translation to
42855 this representation needs to be done both by the target before the @code{F}
42856 packet is sent, and by @value{GDBN} before
42857 it transfers memory to the target. Transferred pointers to structured
42858 data should point to the already-coerced data at any time.
42862 @unnumberedsubsubsec struct stat
42863 @cindex struct stat, in file-i/o protocol
42865 The buffer of type @code{struct stat} used by the target and @value{GDBN}
42866 is defined as follows:
42870 unsigned int st_dev; /* device */
42871 unsigned int st_ino; /* inode */
42872 mode_t st_mode; /* protection */
42873 unsigned int st_nlink; /* number of hard links */
42874 unsigned int st_uid; /* user ID of owner */
42875 unsigned int st_gid; /* group ID of owner */
42876 unsigned int st_rdev; /* device type (if inode device) */
42877 unsigned long st_size; /* total size, in bytes */
42878 unsigned long st_blksize; /* blocksize for filesystem I/O */
42879 unsigned long st_blocks; /* number of blocks allocated */
42880 time_t st_atime; /* time of last access */
42881 time_t st_mtime; /* time of last modification */
42882 time_t st_ctime; /* time of last change */
42886 The integral datatypes conform to the definitions given in the
42887 appropriate section (see @ref{Integral Datatypes}, for details) so this
42888 structure is of size 64 bytes.
42890 The values of several fields have a restricted meaning and/or
42896 A value of 0 represents a file, 1 the console.
42899 No valid meaning for the target. Transmitted unchanged.
42902 Valid mode bits are described in @ref{Constants}. Any other
42903 bits have currently no meaning for the target.
42908 No valid meaning for the target. Transmitted unchanged.
42913 These values have a host and file system dependent
42914 accuracy. Especially on Windows hosts, the file system may not
42915 support exact timing values.
42918 The target gets a @code{struct stat} of the above representation and is
42919 responsible for coercing it to the target representation before
42922 Note that due to size differences between the host, target, and protocol
42923 representations of @code{struct stat} members, these members could eventually
42924 get truncated on the target.
42926 @node struct timeval
42927 @unnumberedsubsubsec struct timeval
42928 @cindex struct timeval, in file-i/o protocol
42930 The buffer of type @code{struct timeval} used by the File-I/O protocol
42931 is defined as follows:
42935 time_t tv_sec; /* second */
42936 long tv_usec; /* microsecond */
42940 The integral datatypes conform to the definitions given in the
42941 appropriate section (see @ref{Integral Datatypes}, for details) so this
42942 structure is of size 8 bytes.
42945 @subsection Constants
42946 @cindex constants, in file-i/o protocol
42948 The following values are used for the constants inside of the
42949 protocol. @value{GDBN} and target are responsible for translating these
42950 values before and after the call as needed.
42961 @unnumberedsubsubsec Open Flags
42962 @cindex open flags, in file-i/o protocol
42964 All values are given in hexadecimal representation.
42976 @node mode_t Values
42977 @unnumberedsubsubsec mode_t Values
42978 @cindex mode_t values, in file-i/o protocol
42980 All values are given in octal representation.
42997 @unnumberedsubsubsec Errno Values
42998 @cindex errno values, in file-i/o protocol
43000 All values are given in decimal representation.
43025 @code{EUNKNOWN} is used as a fallback error value if a host system returns
43026 any error value not in the list of supported error numbers.
43029 @unnumberedsubsubsec Lseek Flags
43030 @cindex lseek flags, in file-i/o protocol
43039 @unnumberedsubsubsec Limits
43040 @cindex limits, in file-i/o protocol
43042 All values are given in decimal representation.
43045 INT_MIN -2147483648
43047 UINT_MAX 4294967295
43048 LONG_MIN -9223372036854775808
43049 LONG_MAX 9223372036854775807
43050 ULONG_MAX 18446744073709551615
43053 @node File-I/O Examples
43054 @subsection File-I/O Examples
43055 @cindex file-i/o examples
43057 Example sequence of a write call, file descriptor 3, buffer is at target
43058 address 0x1234, 6 bytes should be written:
43061 <- @code{Fwrite,3,1234,6}
43062 @emph{request memory read from target}
43065 @emph{return "6 bytes written"}
43069 Example sequence of a read call, file descriptor 3, buffer is at target
43070 address 0x1234, 6 bytes should be read:
43073 <- @code{Fread,3,1234,6}
43074 @emph{request memory write to target}
43075 -> @code{X1234,6:XXXXXX}
43076 @emph{return "6 bytes read"}
43080 Example sequence of a read call, call fails on the host due to invalid
43081 file descriptor (@code{EBADF}):
43084 <- @code{Fread,3,1234,6}
43088 Example sequence of a read call, user presses @kbd{Ctrl-c} before syscall on
43092 <- @code{Fread,3,1234,6}
43097 Example sequence of a read call, user presses @kbd{Ctrl-c} after syscall on
43101 <- @code{Fread,3,1234,6}
43102 -> @code{X1234,6:XXXXXX}
43106 @node Library List Format
43107 @section Library List Format
43108 @cindex library list format, remote protocol
43110 On some platforms, a dynamic loader (e.g.@: @file{ld.so}) runs in the
43111 same process as your application to manage libraries. In this case,
43112 @value{GDBN} can use the loader's symbol table and normal memory
43113 operations to maintain a list of shared libraries. On other
43114 platforms, the operating system manages loaded libraries.
43115 @value{GDBN} can not retrieve the list of currently loaded libraries
43116 through memory operations, so it uses the @samp{qXfer:libraries:read}
43117 packet (@pxref{qXfer library list read}) instead. The remote stub
43118 queries the target's operating system and reports which libraries
43121 The @samp{qXfer:libraries:read} packet returns an XML document which
43122 lists loaded libraries and their offsets. Each library has an
43123 associated name and one or more segment or section base addresses,
43124 which report where the library was loaded in memory.
43126 For the common case of libraries that are fully linked binaries, the
43127 library should have a list of segments. If the target supports
43128 dynamic linking of a relocatable object file, its library XML element
43129 should instead include a list of allocated sections. The segment or
43130 section bases are start addresses, not relocation offsets; they do not
43131 depend on the library's link-time base addresses.
43133 @value{GDBN} must be linked with the Expat library to support XML
43134 library lists. @xref{Expat}.
43136 A simple memory map, with one loaded library relocated by a single
43137 offset, looks like this:
43141 <library name="/lib/libc.so.6">
43142 <segment address="0x10000000"/>
43147 Another simple memory map, with one loaded library with three
43148 allocated sections (.text, .data, .bss), looks like this:
43152 <library name="sharedlib.o">
43153 <section address="0x10000000"/>
43154 <section address="0x20000000"/>
43155 <section address="0x30000000"/>
43160 The format of a library list is described by this DTD:
43163 <!-- library-list: Root element with versioning -->
43164 <!ELEMENT library-list (library)*>
43165 <!ATTLIST library-list version CDATA #FIXED "1.0">
43166 <!ELEMENT library (segment*, section*)>
43167 <!ATTLIST library name CDATA #REQUIRED>
43168 <!ELEMENT segment EMPTY>
43169 <!ATTLIST segment address CDATA #REQUIRED>
43170 <!ELEMENT section EMPTY>
43171 <!ATTLIST section address CDATA #REQUIRED>
43174 In addition, segments and section descriptors cannot be mixed within a
43175 single library element, and you must supply at least one segment or
43176 section for each library.
43178 @node Library List Format for SVR4 Targets
43179 @section Library List Format for SVR4 Targets
43180 @cindex library list format, remote protocol
43182 On SVR4 platforms @value{GDBN} can use the symbol table of a dynamic loader
43183 (e.g.@: @file{ld.so}) and normal memory operations to maintain a list of
43184 shared libraries. Still a special library list provided by this packet is
43185 more efficient for the @value{GDBN} remote protocol.
43187 The @samp{qXfer:libraries-svr4:read} packet returns an XML document which lists
43188 loaded libraries and their SVR4 linker parameters. For each library on SVR4
43189 target, the following parameters are reported:
43193 @code{name}, the absolute file name from the @code{l_name} field of
43194 @code{struct link_map}.
43196 @code{lm} with address of @code{struct link_map} used for TLS
43197 (Thread Local Storage) access.
43199 @code{l_addr}, the displacement as read from the field @code{l_addr} of
43200 @code{struct link_map}. For prelinked libraries this is not an absolute
43201 memory address. It is a displacement of absolute memory address against
43202 address the file was prelinked to during the library load.
43204 @code{l_ld}, which is memory address of the @code{PT_DYNAMIC} segment
43207 Additionally the single @code{main-lm} attribute specifies address of
43208 @code{struct link_map} used for the main executable. This parameter is used
43209 for TLS access and its presence is optional.
43211 @value{GDBN} must be linked with the Expat library to support XML
43212 SVR4 library lists. @xref{Expat}.
43214 A simple memory map, with two loaded libraries (which do not use prelink),
43218 <library-list-svr4 version="1.0" main-lm="0xe4f8f8">
43219 <library name="/lib/ld-linux.so.2" lm="0xe4f51c" l_addr="0xe2d000"
43221 <library name="/lib/libc.so.6" lm="0xe4fbe8" l_addr="0x154000"
43223 </library-list-svr>
43226 The format of an SVR4 library list is described by this DTD:
43229 <!-- library-list-svr4: Root element with versioning -->
43230 <!ELEMENT library-list-svr4 (library)*>
43231 <!ATTLIST library-list-svr4 version CDATA #FIXED "1.0">
43232 <!ATTLIST library-list-svr4 main-lm CDATA #IMPLIED>
43233 <!ELEMENT library EMPTY>
43234 <!ATTLIST library name CDATA #REQUIRED>
43235 <!ATTLIST library lm CDATA #REQUIRED>
43236 <!ATTLIST library l_addr CDATA #REQUIRED>
43237 <!ATTLIST library l_ld CDATA #REQUIRED>
43240 @node Memory Map Format
43241 @section Memory Map Format
43242 @cindex memory map format
43244 To be able to write into flash memory, @value{GDBN} needs to obtain a
43245 memory map from the target. This section describes the format of the
43248 The memory map is obtained using the @samp{qXfer:memory-map:read}
43249 (@pxref{qXfer memory map read}) packet and is an XML document that
43250 lists memory regions.
43252 @value{GDBN} must be linked with the Expat library to support XML
43253 memory maps. @xref{Expat}.
43255 The top-level structure of the document is shown below:
43258 <?xml version="1.0"?>
43259 <!DOCTYPE memory-map
43260 PUBLIC "+//IDN gnu.org//DTD GDB Memory Map V1.0//EN"
43261 "http://sourceware.org/gdb/gdb-memory-map.dtd">
43267 Each region can be either:
43272 A region of RAM starting at @var{addr} and extending for @var{length}
43276 <memory type="ram" start="@var{addr}" length="@var{length}"/>
43281 A region of read-only memory:
43284 <memory type="rom" start="@var{addr}" length="@var{length}"/>
43289 A region of flash memory, with erasure blocks @var{blocksize}
43293 <memory type="flash" start="@var{addr}" length="@var{length}">
43294 <property name="blocksize">@var{blocksize}</property>
43300 Regions must not overlap. @value{GDBN} assumes that areas of memory not covered
43301 by the memory map are RAM, and uses the ordinary @samp{M} and @samp{X}
43302 packets to write to addresses in such ranges.
43304 The formal DTD for memory map format is given below:
43307 <!-- ................................................... -->
43308 <!-- Memory Map XML DTD ................................ -->
43309 <!-- File: memory-map.dtd .............................. -->
43310 <!-- .................................... .............. -->
43311 <!-- memory-map.dtd -->
43312 <!-- memory-map: Root element with versioning -->
43313 <!ELEMENT memory-map (memory)*>
43314 <!ATTLIST memory-map version CDATA #FIXED "1.0.0">
43315 <!ELEMENT memory (property)*>
43316 <!-- memory: Specifies a memory region,
43317 and its type, or device. -->
43318 <!ATTLIST memory type (ram|rom|flash) #REQUIRED
43319 start CDATA #REQUIRED
43320 length CDATA #REQUIRED>
43321 <!-- property: Generic attribute tag -->
43322 <!ELEMENT property (#PCDATA | property)*>
43323 <!ATTLIST property name (blocksize) #REQUIRED>
43326 @node Thread List Format
43327 @section Thread List Format
43328 @cindex thread list format
43330 To efficiently update the list of threads and their attributes,
43331 @value{GDBN} issues the @samp{qXfer:threads:read} packet
43332 (@pxref{qXfer threads read}) and obtains the XML document with
43333 the following structure:
43336 <?xml version="1.0"?>
43338 <thread id="id" core="0" name="name">
43339 ... description ...
43344 Each @samp{thread} element must have the @samp{id} attribute that
43345 identifies the thread (@pxref{thread-id syntax}). The
43346 @samp{core} attribute, if present, specifies which processor core
43347 the thread was last executing on. The @samp{name} attribute, if
43348 present, specifies the human-readable name of the thread. The content
43349 of the of @samp{thread} element is interpreted as human-readable
43350 auxiliary information. The @samp{handle} attribute, if present,
43351 is a hex encoded representation of the thread handle.
43354 @node Traceframe Info Format
43355 @section Traceframe Info Format
43356 @cindex traceframe info format
43358 To be able to know which objects in the inferior can be examined when
43359 inspecting a tracepoint hit, @value{GDBN} needs to obtain the list of
43360 memory ranges, registers and trace state variables that have been
43361 collected in a traceframe.
43363 This list is obtained using the @samp{qXfer:traceframe-info:read}
43364 (@pxref{qXfer traceframe info read}) packet and is an XML document.
43366 @value{GDBN} must be linked with the Expat library to support XML
43367 traceframe info discovery. @xref{Expat}.
43369 The top-level structure of the document is shown below:
43372 <?xml version="1.0"?>
43373 <!DOCTYPE traceframe-info
43374 PUBLIC "+//IDN gnu.org//DTD GDB Memory Map V1.0//EN"
43375 "http://sourceware.org/gdb/gdb-traceframe-info.dtd">
43381 Each traceframe block can be either:
43386 A region of collected memory starting at @var{addr} and extending for
43387 @var{length} bytes from there:
43390 <memory start="@var{addr}" length="@var{length}"/>
43394 A block indicating trace state variable numbered @var{number} has been
43398 <tvar id="@var{number}"/>
43403 The formal DTD for the traceframe info format is given below:
43406 <!ELEMENT traceframe-info (memory | tvar)* >
43407 <!ATTLIST traceframe-info version CDATA #FIXED "1.0">
43409 <!ELEMENT memory EMPTY>
43410 <!ATTLIST memory start CDATA #REQUIRED
43411 length CDATA #REQUIRED>
43413 <!ATTLIST tvar id CDATA #REQUIRED>
43416 @node Branch Trace Format
43417 @section Branch Trace Format
43418 @cindex branch trace format
43420 In order to display the branch trace of an inferior thread,
43421 @value{GDBN} needs to obtain the list of branches. This list is
43422 represented as list of sequential code blocks that are connected via
43423 branches. The code in each block has been executed sequentially.
43425 This list is obtained using the @samp{qXfer:btrace:read}
43426 (@pxref{qXfer btrace read}) packet and is an XML document.
43428 @value{GDBN} must be linked with the Expat library to support XML
43429 traceframe info discovery. @xref{Expat}.
43431 The top-level structure of the document is shown below:
43434 <?xml version="1.0"?>
43436 PUBLIC "+//IDN gnu.org//DTD GDB Branch Trace V1.0//EN"
43437 "http://sourceware.org/gdb/gdb-btrace.dtd">
43446 A block of sequentially executed instructions starting at @var{begin}
43447 and ending at @var{end}:
43450 <block begin="@var{begin}" end="@var{end}"/>
43455 The formal DTD for the branch trace format is given below:
43458 <!ELEMENT btrace (block* | pt) >
43459 <!ATTLIST btrace version CDATA #FIXED "1.0">
43461 <!ELEMENT block EMPTY>
43462 <!ATTLIST block begin CDATA #REQUIRED
43463 end CDATA #REQUIRED>
43465 <!ELEMENT pt (pt-config?, raw?)>
43467 <!ELEMENT pt-config (cpu?)>
43469 <!ELEMENT cpu EMPTY>
43470 <!ATTLIST cpu vendor CDATA #REQUIRED
43471 family CDATA #REQUIRED
43472 model CDATA #REQUIRED
43473 stepping CDATA #REQUIRED>
43475 <!ELEMENT raw (#PCDATA)>
43478 @node Branch Trace Configuration Format
43479 @section Branch Trace Configuration Format
43480 @cindex branch trace configuration format
43482 For each inferior thread, @value{GDBN} can obtain the branch trace
43483 configuration using the @samp{qXfer:btrace-conf:read}
43484 (@pxref{qXfer btrace-conf read}) packet.
43486 The configuration describes the branch trace format and configuration
43487 settings for that format. The following information is described:
43491 This thread uses the @dfn{Branch Trace Store} (@acronym{BTS}) format.
43494 The size of the @acronym{BTS} ring buffer in bytes.
43497 This thread uses the @dfn{Intel Processor Trace} (@acronym{Intel
43501 The size of the @acronym{Intel PT} ring buffer in bytes.
43505 @value{GDBN} must be linked with the Expat library to support XML
43506 branch trace configuration discovery. @xref{Expat}.
43508 The formal DTD for the branch trace configuration format is given below:
43511 <!ELEMENT btrace-conf (bts?, pt?)>
43512 <!ATTLIST btrace-conf version CDATA #FIXED "1.0">
43514 <!ELEMENT bts EMPTY>
43515 <!ATTLIST bts size CDATA #IMPLIED>
43517 <!ELEMENT pt EMPTY>
43518 <!ATTLIST pt size CDATA #IMPLIED>
43521 @include agentexpr.texi
43523 @node Target Descriptions
43524 @appendix Target Descriptions
43525 @cindex target descriptions
43527 One of the challenges of using @value{GDBN} to debug embedded systems
43528 is that there are so many minor variants of each processor
43529 architecture in use. It is common practice for vendors to start with
43530 a standard processor core --- ARM, PowerPC, or @acronym{MIPS}, for example ---
43531 and then make changes to adapt it to a particular market niche. Some
43532 architectures have hundreds of variants, available from dozens of
43533 vendors. This leads to a number of problems:
43537 With so many different customized processors, it is difficult for
43538 the @value{GDBN} maintainers to keep up with the changes.
43540 Since individual variants may have short lifetimes or limited
43541 audiences, it may not be worthwhile to carry information about every
43542 variant in the @value{GDBN} source tree.
43544 When @value{GDBN} does support the architecture of the embedded system
43545 at hand, the task of finding the correct architecture name to give the
43546 @command{set architecture} command can be error-prone.
43549 To address these problems, the @value{GDBN} remote protocol allows a
43550 target system to not only identify itself to @value{GDBN}, but to
43551 actually describe its own features. This lets @value{GDBN} support
43552 processor variants it has never seen before --- to the extent that the
43553 descriptions are accurate, and that @value{GDBN} understands them.
43555 @value{GDBN} must be linked with the Expat library to support XML
43556 target descriptions. @xref{Expat}.
43559 * Retrieving Descriptions:: How descriptions are fetched from a target.
43560 * Target Description Format:: The contents of a target description.
43561 * Predefined Target Types:: Standard types available for target
43563 * Enum Target Types:: How to define enum target types.
43564 * Standard Target Features:: Features @value{GDBN} knows about.
43567 @node Retrieving Descriptions
43568 @section Retrieving Descriptions
43570 Target descriptions can be read from the target automatically, or
43571 specified by the user manually. The default behavior is to read the
43572 description from the target. @value{GDBN} retrieves it via the remote
43573 protocol using @samp{qXfer} requests (@pxref{General Query Packets,
43574 qXfer}). The @var{annex} in the @samp{qXfer} packet will be
43575 @samp{target.xml}. The contents of the @samp{target.xml} annex are an
43576 XML document, of the form described in @ref{Target Description
43579 Alternatively, you can specify a file to read for the target description.
43580 If a file is set, the target will not be queried. The commands to
43581 specify a file are:
43584 @cindex set tdesc filename
43585 @item set tdesc filename @var{path}
43586 Read the target description from @var{path}.
43588 @cindex unset tdesc filename
43589 @item unset tdesc filename
43590 Do not read the XML target description from a file. @value{GDBN}
43591 will use the description supplied by the current target.
43593 @cindex show tdesc filename
43594 @item show tdesc filename
43595 Show the filename to read for a target description, if any.
43599 @node Target Description Format
43600 @section Target Description Format
43601 @cindex target descriptions, XML format
43603 A target description annex is an @uref{http://www.w3.org/XML/, XML}
43604 document which complies with the Document Type Definition provided in
43605 the @value{GDBN} sources in @file{gdb/features/gdb-target.dtd}. This
43606 means you can use generally available tools like @command{xmllint} to
43607 check that your feature descriptions are well-formed and valid.
43608 However, to help people unfamiliar with XML write descriptions for
43609 their targets, we also describe the grammar here.
43611 Target descriptions can identify the architecture of the remote target
43612 and (for some architectures) provide information about custom register
43613 sets. They can also identify the OS ABI of the remote target.
43614 @value{GDBN} can use this information to autoconfigure for your
43615 target, or to warn you if you connect to an unsupported target.
43617 Here is a simple target description:
43620 <target version="1.0">
43621 <architecture>i386:x86-64</architecture>
43626 This minimal description only says that the target uses
43627 the x86-64 architecture.
43629 A target description has the following overall form, with [ ] marking
43630 optional elements and @dots{} marking repeatable elements. The elements
43631 are explained further below.
43634 <?xml version="1.0"?>
43635 <!DOCTYPE target SYSTEM "gdb-target.dtd">
43636 <target version="1.0">
43637 @r{[}@var{architecture}@r{]}
43638 @r{[}@var{osabi}@r{]}
43639 @r{[}@var{compatible}@r{]}
43640 @r{[}@var{feature}@dots{}@r{]}
43645 The description is generally insensitive to whitespace and line
43646 breaks, under the usual common-sense rules. The XML version
43647 declaration and document type declaration can generally be omitted
43648 (@value{GDBN} does not require them), but specifying them may be
43649 useful for XML validation tools. The @samp{version} attribute for
43650 @samp{<target>} may also be omitted, but we recommend
43651 including it; if future versions of @value{GDBN} use an incompatible
43652 revision of @file{gdb-target.dtd}, they will detect and report
43653 the version mismatch.
43655 @subsection Inclusion
43656 @cindex target descriptions, inclusion
43659 @cindex <xi:include>
43662 It can sometimes be valuable to split a target description up into
43663 several different annexes, either for organizational purposes, or to
43664 share files between different possible target descriptions. You can
43665 divide a description into multiple files by replacing any element of
43666 the target description with an inclusion directive of the form:
43669 <xi:include href="@var{document}"/>
43673 When @value{GDBN} encounters an element of this form, it will retrieve
43674 the named XML @var{document}, and replace the inclusion directive with
43675 the contents of that document. If the current description was read
43676 using @samp{qXfer}, then so will be the included document;
43677 @var{document} will be interpreted as the name of an annex. If the
43678 current description was read from a file, @value{GDBN} will look for
43679 @var{document} as a file in the same directory where it found the
43680 original description.
43682 @subsection Architecture
43683 @cindex <architecture>
43685 An @samp{<architecture>} element has this form:
43688 <architecture>@var{arch}</architecture>
43691 @var{arch} is one of the architectures from the set accepted by
43692 @code{set architecture} (@pxref{Targets, ,Specifying a Debugging Target}).
43695 @cindex @code{<osabi>}
43697 This optional field was introduced in @value{GDBN} version 7.0.
43698 Previous versions of @value{GDBN} ignore it.
43700 An @samp{<osabi>} element has this form:
43703 <osabi>@var{abi-name}</osabi>
43706 @var{abi-name} is an OS ABI name from the same selection accepted by
43707 @w{@code{set osabi}} (@pxref{ABI, ,Configuring the Current ABI}).
43709 @subsection Compatible Architecture
43710 @cindex @code{<compatible>}
43712 This optional field was introduced in @value{GDBN} version 7.0.
43713 Previous versions of @value{GDBN} ignore it.
43715 A @samp{<compatible>} element has this form:
43718 <compatible>@var{arch}</compatible>
43721 @var{arch} is one of the architectures from the set accepted by
43722 @code{set architecture} (@pxref{Targets, ,Specifying a Debugging Target}).
43724 A @samp{<compatible>} element is used to specify that the target
43725 is able to run binaries in some other than the main target architecture
43726 given by the @samp{<architecture>} element. For example, on the
43727 Cell Broadband Engine, the main architecture is @code{powerpc:common}
43728 or @code{powerpc:common64}, but the system is able to run binaries
43729 in the @code{spu} architecture as well. The way to describe this
43730 capability with @samp{<compatible>} is as follows:
43733 <architecture>powerpc:common</architecture>
43734 <compatible>spu</compatible>
43737 @subsection Features
43740 Each @samp{<feature>} describes some logical portion of the target
43741 system. Features are currently used to describe available CPU
43742 registers and the types of their contents. A @samp{<feature>} element
43746 <feature name="@var{name}">
43747 @r{[}@var{type}@dots{}@r{]}
43753 Each feature's name should be unique within the description. The name
43754 of a feature does not matter unless @value{GDBN} has some special
43755 knowledge of the contents of that feature; if it does, the feature
43756 should have its standard name. @xref{Standard Target Features}.
43760 Any register's value is a collection of bits which @value{GDBN} must
43761 interpret. The default interpretation is a two's complement integer,
43762 but other types can be requested by name in the register description.
43763 Some predefined types are provided by @value{GDBN} (@pxref{Predefined
43764 Target Types}), and the description can define additional composite
43767 Each type element must have an @samp{id} attribute, which gives
43768 a unique (within the containing @samp{<feature>}) name to the type.
43769 Types must be defined before they are used.
43772 Some targets offer vector registers, which can be treated as arrays
43773 of scalar elements. These types are written as @samp{<vector>} elements,
43774 specifying the array element type, @var{type}, and the number of elements,
43778 <vector id="@var{id}" type="@var{type}" count="@var{count}"/>
43782 If a register's value is usefully viewed in multiple ways, define it
43783 with a union type containing the useful representations. The
43784 @samp{<union>} element contains one or more @samp{<field>} elements,
43785 each of which has a @var{name} and a @var{type}:
43788 <union id="@var{id}">
43789 <field name="@var{name}" type="@var{type}"/>
43796 If a register's value is composed from several separate values, define
43797 it with either a structure type or a flags type.
43798 A flags type may only contain bitfields.
43799 A structure type may either contain only bitfields or contain no bitfields.
43800 If the value contains only bitfields, its total size in bytes must be
43803 Non-bitfield values have a @var{name} and @var{type}.
43806 <struct id="@var{id}">
43807 <field name="@var{name}" type="@var{type}"/>
43812 Both @var{name} and @var{type} values are required.
43813 No implicit padding is added.
43815 Bitfield values have a @var{name}, @var{start}, @var{end} and @var{type}.
43818 <struct id="@var{id}" size="@var{size}">
43819 <field name="@var{name}" start="@var{start}" end="@var{end}" type="@var{type}"/>
43825 <flags id="@var{id}" size="@var{size}">
43826 <field name="@var{name}" start="@var{start}" end="@var{end}" type="@var{type}"/>
43831 The @var{name} value is required.
43832 Bitfield values may be named with the empty string, @samp{""},
43833 in which case the field is ``filler'' and its value is not printed.
43834 Not all bits need to be specified, so ``filler'' fields are optional.
43836 The @var{start} and @var{end} values are required, and @var{type}
43838 The field's @var{start} must be less than or equal to its @var{end},
43839 and zero represents the least significant bit.
43841 The default value of @var{type} is @code{bool} for single bit fields,
43842 and an unsigned integer otherwise.
43844 Which to choose? Structures or flags?
43846 Registers defined with @samp{flags} have these advantages over
43847 defining them with @samp{struct}:
43851 Arithmetic may be performed on them as if they were integers.
43853 They are printed in a more readable fashion.
43856 Registers defined with @samp{struct} have one advantage over
43857 defining them with @samp{flags}:
43861 One can fetch individual fields like in @samp{C}.
43864 (gdb) print $my_struct_reg.field3
43870 @subsection Registers
43873 Each register is represented as an element with this form:
43876 <reg name="@var{name}"
43877 bitsize="@var{size}"
43878 @r{[}regnum="@var{num}"@r{]}
43879 @r{[}save-restore="@var{save-restore}"@r{]}
43880 @r{[}type="@var{type}"@r{]}
43881 @r{[}group="@var{group}"@r{]}/>
43885 The components are as follows:
43890 The register's name; it must be unique within the target description.
43893 The register's size, in bits.
43896 The register's number. If omitted, a register's number is one greater
43897 than that of the previous register (either in the current feature or in
43898 a preceding feature); the first register in the target description
43899 defaults to zero. This register number is used to read or write
43900 the register; e.g.@: it is used in the remote @code{p} and @code{P}
43901 packets, and registers appear in the @code{g} and @code{G} packets
43902 in order of increasing register number.
43905 Whether the register should be preserved across inferior function
43906 calls; this must be either @code{yes} or @code{no}. The default is
43907 @code{yes}, which is appropriate for most registers except for
43908 some system control registers; this is not related to the target's
43912 The type of the register. It may be a predefined type, a type
43913 defined in the current feature, or one of the special types @code{int}
43914 and @code{float}. @code{int} is an integer type of the correct size
43915 for @var{bitsize}, and @code{float} is a floating point type (in the
43916 architecture's normal floating point format) of the correct size for
43917 @var{bitsize}. The default is @code{int}.
43920 The register group to which this register belongs. It can be one of the
43921 standard register groups @code{general}, @code{float}, @code{vector} or an
43922 arbitrary string. Group names should be limited to alphanumeric characters.
43923 If a group name is made up of multiple words the words may be separated by
43924 hyphens; e.g.@: @code{special-group} or @code{ultra-special-group}. If no
43925 @var{group} is specified, @value{GDBN} will not display the register in
43926 @code{info registers}.
43930 @node Predefined Target Types
43931 @section Predefined Target Types
43932 @cindex target descriptions, predefined types
43934 Type definitions in the self-description can build up composite types
43935 from basic building blocks, but can not define fundamental types. Instead,
43936 standard identifiers are provided by @value{GDBN} for the fundamental
43937 types. The currently supported types are:
43942 Boolean type, occupying a single bit.
43950 Signed integer types holding the specified number of bits.
43958 Unsigned integer types holding the specified number of bits.
43962 Pointers to unspecified code and data. The program counter and
43963 any dedicated return address register may be marked as code
43964 pointers; printing a code pointer converts it into a symbolic
43965 address. The stack pointer and any dedicated address registers
43966 may be marked as data pointers.
43969 Single precision IEEE floating point.
43972 Double precision IEEE floating point.
43975 The 12-byte extended precision format used by ARM FPA registers.
43978 The 10-byte extended precision format used by x87 registers.
43981 32bit @sc{eflags} register used by x86.
43984 32bit @sc{mxcsr} register used by x86.
43988 @node Enum Target Types
43989 @section Enum Target Types
43990 @cindex target descriptions, enum types
43992 Enum target types are useful in @samp{struct} and @samp{flags}
43993 register descriptions. @xref{Target Description Format}.
43995 Enum types have a name, size and a list of name/value pairs.
43998 <enum id="@var{id}" size="@var{size}">
43999 <evalue name="@var{name}" value="@var{value}"/>
44004 Enums must be defined before they are used.
44007 <enum id="levels_type" size="4">
44008 <evalue name="low" value="0"/>
44009 <evalue name="high" value="1"/>
44011 <flags id="flags_type" size="4">
44012 <field name="X" start="0"/>
44013 <field name="LEVEL" start="1" end="1" type="levels_type"/>
44015 <reg name="flags" bitsize="32" type="flags_type"/>
44018 Given that description, a value of 3 for the @samp{flags} register
44019 would be printed as:
44022 (gdb) info register flags
44023 flags 0x3 [ X LEVEL=high ]
44026 @node Standard Target Features
44027 @section Standard Target Features
44028 @cindex target descriptions, standard features
44030 A target description must contain either no registers or all the
44031 target's registers. If the description contains no registers, then
44032 @value{GDBN} will assume a default register layout, selected based on
44033 the architecture. If the description contains any registers, the
44034 default layout will not be used; the standard registers must be
44035 described in the target description, in such a way that @value{GDBN}
44036 can recognize them.
44038 This is accomplished by giving specific names to feature elements
44039 which contain standard registers. @value{GDBN} will look for features
44040 with those names and verify that they contain the expected registers;
44041 if any known feature is missing required registers, or if any required
44042 feature is missing, @value{GDBN} will reject the target
44043 description. You can add additional registers to any of the
44044 standard features --- @value{GDBN} will display them just as if
44045 they were added to an unrecognized feature.
44047 This section lists the known features and their expected contents.
44048 Sample XML documents for these features are included in the
44049 @value{GDBN} source tree, in the directory @file{gdb/features}.
44051 Names recognized by @value{GDBN} should include the name of the
44052 company or organization which selected the name, and the overall
44053 architecture to which the feature applies; so e.g.@: the feature
44054 containing ARM core registers is named @samp{org.gnu.gdb.arm.core}.
44056 The names of registers are not case sensitive for the purpose
44057 of recognizing standard features, but @value{GDBN} will only display
44058 registers using the capitalization used in the description.
44061 * AArch64 Features::
44065 * MicroBlaze Features::
44069 * Nios II Features::
44070 * OpenRISC 1000 Features::
44071 * PowerPC Features::
44072 * RISC-V Features::
44074 * S/390 and System z Features::
44080 @node AArch64 Features
44081 @subsection AArch64 Features
44082 @cindex target descriptions, AArch64 features
44084 The @samp{org.gnu.gdb.aarch64.core} feature is required for AArch64
44085 targets. It should contain registers @samp{x0} through @samp{x30},
44086 @samp{sp}, @samp{pc}, and @samp{cpsr}.
44088 The @samp{org.gnu.gdb.aarch64.fpu} feature is optional. If present,
44089 it should contain registers @samp{v0} through @samp{v31}, @samp{fpsr},
44092 The @samp{org.gnu.gdb.aarch64.sve} feature is optional. If present,
44093 it should contain registers @samp{z0} through @samp{z31}, @samp{p0}
44094 through @samp{p15}, @samp{ffr} and @samp{vg}.
44096 The @samp{org.gnu.gdb.aarch64.pauth} feature is optional. If present,
44097 it should contain registers @samp{pauth_dmask} and @samp{pauth_cmask}.
44100 @subsection ARC Features
44101 @cindex target descriptions, ARC Features
44103 ARC processors are highly configurable, so even core registers and their number
44104 are not completely predetermined. In addition flags and PC registers which are
44105 important to @value{GDBN} are not ``core'' registers in ARC. It is required
44106 that one of the core registers features is present.
44107 @samp{org.gnu.gdb.arc.aux-minimal} feature is mandatory.
44109 The @samp{org.gnu.gdb.arc.core.v2} feature is required for ARC EM and ARC HS
44110 targets with a normal register file. It should contain registers @samp{r0}
44111 through @samp{r25}, @samp{gp}, @samp{fp}, @samp{sp}, @samp{r30}, @samp{blink},
44112 @samp{lp_count} and @samp{pcl}. This feature may contain register @samp{ilink}
44113 and any of extension core registers @samp{r32} through @samp{r59/acch}.
44114 @samp{ilink} and extension core registers are not available to read/write, when
44115 debugging GNU/Linux applications, thus @samp{ilink} is made optional.
44117 The @samp{org.gnu.gdb.arc.core-reduced.v2} feature is required for ARC EM and
44118 ARC HS targets with a reduced register file. It should contain registers
44119 @samp{r0} through @samp{r3}, @samp{r10} through @samp{r15}, @samp{gp},
44120 @samp{fp}, @samp{sp}, @samp{r30}, @samp{blink}, @samp{lp_count} and @samp{pcl}.
44121 This feature may contain register @samp{ilink} and any of extension core
44122 registers @samp{r32} through @samp{r59/acch}.
44124 The @samp{org.gnu.gdb.arc.core.arcompact} feature is required for ARCompact
44125 targets with a normal register file. It should contain registers @samp{r0}
44126 through @samp{r25}, @samp{gp}, @samp{fp}, @samp{sp}, @samp{r30}, @samp{blink},
44127 @samp{lp_count} and @samp{pcl}. This feature may contain registers
44128 @samp{ilink1}, @samp{ilink2} and any of extension core registers @samp{r32}
44129 through @samp{r59/acch}. @samp{ilink1} and @samp{ilink2} and extension core
44130 registers are not available when debugging GNU/Linux applications. The only
44131 difference with @samp{org.gnu.gdb.arc.core.v2} feature is in the names of
44132 @samp{ilink1} and @samp{ilink2} registers and that @samp{r30} is mandatory in
44133 ARC v2, but @samp{ilink2} is optional on ARCompact.
44135 The @samp{org.gnu.gdb.arc.aux-minimal} feature is required for all ARC
44136 targets. It should contain registers @samp{pc} and @samp{status32}.
44139 @subsection ARM Features
44140 @cindex target descriptions, ARM features
44142 The @samp{org.gnu.gdb.arm.core} feature is required for non-M-profile
44144 It should contain registers @samp{r0} through @samp{r13}, @samp{sp},
44145 @samp{lr}, @samp{pc}, and @samp{cpsr}.
44147 For M-profile targets (e.g. Cortex-M3), the @samp{org.gnu.gdb.arm.core}
44148 feature is replaced by @samp{org.gnu.gdb.arm.m-profile}. It should contain
44149 registers @samp{r0} through @samp{r13}, @samp{sp}, @samp{lr}, @samp{pc},
44152 The @samp{org.gnu.gdb.arm.fpa} feature is optional. If present, it
44153 should contain registers @samp{f0} through @samp{f7} and @samp{fps}.
44155 The @samp{org.gnu.gdb.xscale.iwmmxt} feature is optional. If present,
44156 it should contain at least registers @samp{wR0} through @samp{wR15} and
44157 @samp{wCGR0} through @samp{wCGR3}. The @samp{wCID}, @samp{wCon},
44158 @samp{wCSSF}, and @samp{wCASF} registers are optional.
44160 The @samp{org.gnu.gdb.arm.vfp} feature is optional. If present, it
44161 should contain at least registers @samp{d0} through @samp{d15}. If
44162 they are present, @samp{d16} through @samp{d31} should also be included.
44163 @value{GDBN} will synthesize the single-precision registers from
44164 halves of the double-precision registers.
44166 The @samp{org.gnu.gdb.arm.neon} feature is optional. It does not
44167 need to contain registers; it instructs @value{GDBN} to display the
44168 VFP double-precision registers as vectors and to synthesize the
44169 quad-precision registers from pairs of double-precision registers.
44170 If this feature is present, @samp{org.gnu.gdb.arm.vfp} must also
44171 be present and include 32 double-precision registers.
44173 @node i386 Features
44174 @subsection i386 Features
44175 @cindex target descriptions, i386 features
44177 The @samp{org.gnu.gdb.i386.core} feature is required for i386/amd64
44178 targets. It should describe the following registers:
44182 @samp{eax} through @samp{edi} plus @samp{eip} for i386
44184 @samp{rax} through @samp{r15} plus @samp{rip} for amd64
44186 @samp{eflags}, @samp{cs}, @samp{ss}, @samp{ds}, @samp{es},
44187 @samp{fs}, @samp{gs}
44189 @samp{st0} through @samp{st7}
44191 @samp{fctrl}, @samp{fstat}, @samp{ftag}, @samp{fiseg}, @samp{fioff},
44192 @samp{foseg}, @samp{fooff} and @samp{fop}
44195 The register sets may be different, depending on the target.
44197 The @samp{org.gnu.gdb.i386.sse} feature is optional. It should
44198 describe registers:
44202 @samp{xmm0} through @samp{xmm7} for i386
44204 @samp{xmm0} through @samp{xmm15} for amd64
44209 The @samp{org.gnu.gdb.i386.avx} feature is optional and requires the
44210 @samp{org.gnu.gdb.i386.sse} feature. It should
44211 describe the upper 128 bits of @sc{ymm} registers:
44215 @samp{ymm0h} through @samp{ymm7h} for i386
44217 @samp{ymm0h} through @samp{ymm15h} for amd64
44220 The @samp{org.gnu.gdb.i386.mpx} is an optional feature representing Intel
44221 Memory Protection Extension (MPX). It should describe the following registers:
44225 @samp{bnd0raw} through @samp{bnd3raw} for i386 and amd64.
44227 @samp{bndcfgu} and @samp{bndstatus} for i386 and amd64.
44230 The @samp{org.gnu.gdb.i386.linux} feature is optional. It should
44231 describe a single register, @samp{orig_eax}.
44233 The @samp{org.gnu.gdb.i386.segments} feature is optional. It should
44234 describe two system registers: @samp{fs_base} and @samp{gs_base}.
44236 The @samp{org.gnu.gdb.i386.avx512} feature is optional and requires the
44237 @samp{org.gnu.gdb.i386.avx} feature. It should
44238 describe additional @sc{xmm} registers:
44242 @samp{xmm16h} through @samp{xmm31h}, only valid for amd64.
44245 It should describe the upper 128 bits of additional @sc{ymm} registers:
44249 @samp{ymm16h} through @samp{ymm31h}, only valid for amd64.
44253 describe the upper 256 bits of @sc{zmm} registers:
44257 @samp{zmm0h} through @samp{zmm7h} for i386.
44259 @samp{zmm0h} through @samp{zmm15h} for amd64.
44263 describe the additional @sc{zmm} registers:
44267 @samp{zmm16h} through @samp{zmm31h}, only valid for amd64.
44270 The @samp{org.gnu.gdb.i386.pkeys} feature is optional. It should
44271 describe a single register, @samp{pkru}. It is a 32-bit register
44272 valid for i386 and amd64.
44274 @node MicroBlaze Features
44275 @subsection MicroBlaze Features
44276 @cindex target descriptions, MicroBlaze features
44278 The @samp{org.gnu.gdb.microblaze.core} feature is required for MicroBlaze
44279 targets. It should contain registers @samp{r0} through @samp{r31},
44280 @samp{rpc}, @samp{rmsr}, @samp{rear}, @samp{resr}, @samp{rfsr}, @samp{rbtr},
44281 @samp{rpvr}, @samp{rpvr1} through @samp{rpvr11}, @samp{redr}, @samp{rpid},
44282 @samp{rzpr}, @samp{rtlbx}, @samp{rtlbsx}, @samp{rtlblo}, and @samp{rtlbhi}.
44284 The @samp{org.gnu.gdb.microblaze.stack-protect} feature is optional.
44285 If present, it should contain registers @samp{rshr} and @samp{rslr}
44287 @node MIPS Features
44288 @subsection @acronym{MIPS} Features
44289 @cindex target descriptions, @acronym{MIPS} features
44291 The @samp{org.gnu.gdb.mips.cpu} feature is required for @acronym{MIPS} targets.
44292 It should contain registers @samp{r0} through @samp{r31}, @samp{lo},
44293 @samp{hi}, and @samp{pc}. They may be 32-bit or 64-bit depending
44296 The @samp{org.gnu.gdb.mips.cp0} feature is also required. It should
44297 contain at least the @samp{status}, @samp{badvaddr}, and @samp{cause}
44298 registers. They may be 32-bit or 64-bit depending on the target.
44300 The @samp{org.gnu.gdb.mips.fpu} feature is currently required, though
44301 it may be optional in a future version of @value{GDBN}. It should
44302 contain registers @samp{f0} through @samp{f31}, @samp{fcsr}, and
44303 @samp{fir}. They may be 32-bit or 64-bit depending on the target.
44305 The @samp{org.gnu.gdb.mips.dsp} feature is optional. It should
44306 contain registers @samp{hi1} through @samp{hi3}, @samp{lo1} through
44307 @samp{lo3}, and @samp{dspctl}. The @samp{dspctl} register should
44308 be 32-bit and the rest may be 32-bit or 64-bit depending on the target.
44310 The @samp{org.gnu.gdb.mips.linux} feature is optional. It should
44311 contain a single register, @samp{restart}, which is used by the
44312 Linux kernel to control restartable syscalls.
44314 @node M68K Features
44315 @subsection M68K Features
44316 @cindex target descriptions, M68K features
44319 @item @samp{org.gnu.gdb.m68k.core}
44320 @itemx @samp{org.gnu.gdb.coldfire.core}
44321 @itemx @samp{org.gnu.gdb.fido.core}
44322 One of those features must be always present.
44323 The feature that is present determines which flavor of m68k is
44324 used. The feature that is present should contain registers
44325 @samp{d0} through @samp{d7}, @samp{a0} through @samp{a5}, @samp{fp},
44326 @samp{sp}, @samp{ps} and @samp{pc}.
44328 @item @samp{org.gnu.gdb.coldfire.fp}
44329 This feature is optional. If present, it should contain registers
44330 @samp{fp0} through @samp{fp7}, @samp{fpcontrol}, @samp{fpstatus} and
44334 @node NDS32 Features
44335 @subsection NDS32 Features
44336 @cindex target descriptions, NDS32 features
44338 The @samp{org.gnu.gdb.nds32.core} feature is required for NDS32
44339 targets. It should contain at least registers @samp{r0} through
44340 @samp{r10}, @samp{r15}, @samp{fp}, @samp{gp}, @samp{lp}, @samp{sp},
44343 The @samp{org.gnu.gdb.nds32.fpu} feature is optional. If present,
44344 it should contain 64-bit double-precision floating-point registers
44345 @samp{fd0} through @emph{fdN}, which should be @samp{fd3}, @samp{fd7},
44346 @samp{fd15}, or @samp{fd31} based on the FPU configuration implemented.
44348 @emph{Note:} The first sixteen 64-bit double-precision floating-point
44349 registers are overlapped with the thirty-two 32-bit single-precision
44350 floating-point registers. The 32-bit single-precision registers, if
44351 not being listed explicitly, will be synthesized from halves of the
44352 overlapping 64-bit double-precision registers. Listing 32-bit
44353 single-precision registers explicitly is deprecated, and the
44354 support to it could be totally removed some day.
44356 @node Nios II Features
44357 @subsection Nios II Features
44358 @cindex target descriptions, Nios II features
44360 The @samp{org.gnu.gdb.nios2.cpu} feature is required for Nios II
44361 targets. It should contain the 32 core registers (@samp{zero},
44362 @samp{at}, @samp{r2} through @samp{r23}, @samp{et} through @samp{ra}),
44363 @samp{pc}, and the 16 control registers (@samp{status} through
44366 @node OpenRISC 1000 Features
44367 @subsection Openrisc 1000 Features
44368 @cindex target descriptions, OpenRISC 1000 features
44370 The @samp{org.gnu.gdb.or1k.group0} feature is required for OpenRISC 1000
44371 targets. It should contain the 32 general purpose registers (@samp{r0}
44372 through @samp{r31}), @samp{ppc}, @samp{npc} and @samp{sr}.
44374 @node PowerPC Features
44375 @subsection PowerPC Features
44376 @cindex target descriptions, PowerPC features
44378 The @samp{org.gnu.gdb.power.core} feature is required for PowerPC
44379 targets. It should contain registers @samp{r0} through @samp{r31},
44380 @samp{pc}, @samp{msr}, @samp{cr}, @samp{lr}, @samp{ctr}, and
44381 @samp{xer}. They may be 32-bit or 64-bit depending on the target.
44383 The @samp{org.gnu.gdb.power.fpu} feature is optional. It should
44384 contain registers @samp{f0} through @samp{f31} and @samp{fpscr}.
44386 The @samp{org.gnu.gdb.power.altivec} feature is optional. It should
44387 contain registers @samp{vr0} through @samp{vr31}, @samp{vscr}, and
44388 @samp{vrsave}. @value{GDBN} will define pseudo-registers @samp{v0}
44389 through @samp{v31} as aliases for the corresponding @samp{vrX}
44392 The @samp{org.gnu.gdb.power.vsx} feature is optional. It should
44393 contain registers @samp{vs0h} through @samp{vs31h}. @value{GDBN} will
44394 combine these registers with the floating point registers (@samp{f0}
44395 through @samp{f31}) and the altivec registers (@samp{vr0} through
44396 @samp{vr31}) to present the 128-bit wide registers @samp{vs0} through
44397 @samp{vs63}, the set of vector-scalar registers for POWER7.
44398 Therefore, this feature requires both @samp{org.gnu.gdb.power.fpu} and
44399 @samp{org.gnu.gdb.power.altivec}.
44401 The @samp{org.gnu.gdb.power.spe} feature is optional. It should
44402 contain registers @samp{ev0h} through @samp{ev31h}, @samp{acc}, and
44403 @samp{spefscr}. SPE targets should provide 32-bit registers in
44404 @samp{org.gnu.gdb.power.core} and provide the upper halves in
44405 @samp{ev0h} through @samp{ev31h}. @value{GDBN} will combine
44406 these to present registers @samp{ev0} through @samp{ev31} to the
44409 The @samp{org.gnu.gdb.power.ppr} feature is optional. It should
44410 contain the 64-bit register @samp{ppr}.
44412 The @samp{org.gnu.gdb.power.dscr} feature is optional. It should
44413 contain the 64-bit register @samp{dscr}.
44415 The @samp{org.gnu.gdb.power.tar} feature is optional. It should
44416 contain the 64-bit register @samp{tar}.
44418 The @samp{org.gnu.gdb.power.ebb} feature is optional. It should
44419 contain registers @samp{bescr}, @samp{ebbhr} and @samp{ebbrr}, all
44422 The @samp{org.gnu.gdb.power.linux.pmu} feature is optional. It should
44423 contain registers @samp{mmcr0}, @samp{mmcr2}, @samp{siar}, @samp{sdar}
44424 and @samp{sier}, all 64-bit wide. This is the subset of the isa 2.07
44425 server PMU registers provided by @sc{gnu}/Linux.
44427 The @samp{org.gnu.gdb.power.htm.spr} feature is optional. It should
44428 contain registers @samp{tfhar}, @samp{texasr} and @samp{tfiar}, all
44431 The @samp{org.gnu.gdb.power.htm.core} feature is optional. It should
44432 contain the checkpointed general-purpose registers @samp{cr0} through
44433 @samp{cr31}, as well as the checkpointed registers @samp{clr} and
44434 @samp{cctr}. These registers may all be either 32-bit or 64-bit
44435 depending on the target. It should also contain the checkpointed
44436 registers @samp{ccr} and @samp{cxer}, which should both be 32-bit
44439 The @samp{org.gnu.gdb.power.htm.fpu} feature is optional. It should
44440 contain the checkpointed 64-bit floating-point registers @samp{cf0}
44441 through @samp{cf31}, as well as the checkpointed 64-bit register
44444 The @samp{org.gnu.gdb.power.htm.altivec} feature is optional. It
44445 should contain the checkpointed altivec registers @samp{cvr0} through
44446 @samp{cvr31}, all 128-bit wide. It should also contain the
44447 checkpointed registers @samp{cvscr} and @samp{cvrsave}, both 32-bit
44450 The @samp{org.gnu.gdb.power.htm.vsx} feature is optional. It should
44451 contain registers @samp{cvs0h} through @samp{cvs31h}. @value{GDBN}
44452 will combine these registers with the checkpointed floating point
44453 registers (@samp{cf0} through @samp{cf31}) and the checkpointed
44454 altivec registers (@samp{cvr0} through @samp{cvr31}) to present the
44455 128-bit wide checkpointed vector-scalar registers @samp{cvs0} through
44456 @samp{cvs63}. Therefore, this feature requires both
44457 @samp{org.gnu.gdb.power.htm.altivec} and
44458 @samp{org.gnu.gdb.power.htm.fpu}.
44460 The @samp{org.gnu.gdb.power.htm.ppr} feature is optional. It should
44461 contain the 64-bit checkpointed register @samp{cppr}.
44463 The @samp{org.gnu.gdb.power.htm.dscr} feature is optional. It should
44464 contain the 64-bit checkpointed register @samp{cdscr}.
44466 The @samp{org.gnu.gdb.power.htm.tar} feature is optional. It should
44467 contain the 64-bit checkpointed register @samp{ctar}.
44470 @node RISC-V Features
44471 @subsection RISC-V Features
44472 @cindex target descriptions, RISC-V Features
44474 The @samp{org.gnu.gdb.riscv.cpu} feature is required for RISC-V
44475 targets. It should contain the registers @samp{x0} through
44476 @samp{x31}, and @samp{pc}. Either the architectural names (@samp{x0},
44477 @samp{x1}, etc) can be used, or the ABI names (@samp{zero}, @samp{ra},
44480 The @samp{org.gnu.gdb.riscv.fpu} feature is optional. If present, it
44481 should contain registers @samp{f0} through @samp{f31}, @samp{fflags},
44482 @samp{frm}, and @samp{fcsr}. As with the cpu feature, either the
44483 architectural register names, or the ABI names can be used.
44485 The @samp{org.gnu.gdb.riscv.virtual} feature is optional. If present,
44486 it should contain registers that are not backed by real registers on
44487 the target, but are instead virtual, where the register value is
44488 derived from other target state. In many ways these are like
44489 @value{GDBN}s pseudo-registers, except implemented by the target.
44490 Currently the only register expected in this set is the one byte
44491 @samp{priv} register that contains the target's privilege level in the
44492 least significant two bits.
44494 The @samp{org.gnu.gdb.riscv.csr} feature is optional. If present, it
44495 should contain all of the target's standard CSRs. Standard CSRs are
44496 those defined in the RISC-V specification documents. There is some
44497 overlap between this feature and the fpu feature; the @samp{fflags},
44498 @samp{frm}, and @samp{fcsr} registers could be in either feature. The
44499 expectation is that these registers will be in the fpu feature if the
44500 target has floating point hardware, but can be moved into the csr
44501 feature if the target has the floating point control registers, but no
44502 other floating point hardware.
44505 @subsection RX Features
44506 @cindex target descriptions, RX Features
44508 The @samp{org.gnu.gdb.rx.core} feature is required for RX
44509 targets. It should contain the registers @samp{r0} through
44510 @samp{r15}, @samp{usp}, @samp{isp}, @samp{psw}, @samp{pc}, @samp{intb},
44511 @samp{bpsw}, @samp{bpc}, @samp{fintv}, @samp{fpsw}, and @samp{acc}.
44513 @node S/390 and System z Features
44514 @subsection S/390 and System z Features
44515 @cindex target descriptions, S/390 features
44516 @cindex target descriptions, System z features
44518 The @samp{org.gnu.gdb.s390.core} feature is required for S/390 and
44519 System z targets. It should contain the PSW and the 16 general
44520 registers. In particular, System z targets should provide the 64-bit
44521 registers @samp{pswm}, @samp{pswa}, and @samp{r0} through @samp{r15}.
44522 S/390 targets should provide the 32-bit versions of these registers.
44523 A System z target that runs in 31-bit addressing mode should provide
44524 32-bit versions of @samp{pswm} and @samp{pswa}, as well as the general
44525 register's upper halves @samp{r0h} through @samp{r15h}, and their
44526 lower halves @samp{r0l} through @samp{r15l}.
44528 The @samp{org.gnu.gdb.s390.fpr} feature is required. It should
44529 contain the 64-bit registers @samp{f0} through @samp{f15}, and
44532 The @samp{org.gnu.gdb.s390.acr} feature is required. It should
44533 contain the 32-bit registers @samp{acr0} through @samp{acr15}.
44535 The @samp{org.gnu.gdb.s390.linux} feature is optional. It should
44536 contain the register @samp{orig_r2}, which is 64-bit wide on System z
44537 targets and 32-bit otherwise. In addition, the feature may contain
44538 the @samp{last_break} register, whose width depends on the addressing
44539 mode, as well as the @samp{system_call} register, which is always
44542 The @samp{org.gnu.gdb.s390.tdb} feature is optional. It should
44543 contain the 64-bit registers @samp{tdb0}, @samp{tac}, @samp{tct},
44544 @samp{atia}, and @samp{tr0} through @samp{tr15}.
44546 The @samp{org.gnu.gdb.s390.vx} feature is optional. It should contain
44547 64-bit wide registers @samp{v0l} through @samp{v15l}, which will be
44548 combined by @value{GDBN} with the floating point registers @samp{f0}
44549 through @samp{f15} to present the 128-bit wide vector registers
44550 @samp{v0} through @samp{v15}. In addition, this feature should
44551 contain the 128-bit wide vector registers @samp{v16} through
44554 The @samp{org.gnu.gdb.s390.gs} feature is optional. It should contain
44555 the 64-bit wide guarded-storage-control registers @samp{gsd},
44556 @samp{gssm}, and @samp{gsepla}.
44558 The @samp{org.gnu.gdb.s390.gsbc} feature is optional. It should contain
44559 the 64-bit wide guarded-storage broadcast control registers
44560 @samp{bc_gsd}, @samp{bc_gssm}, and @samp{bc_gsepla}.
44562 @node Sparc Features
44563 @subsection Sparc Features
44564 @cindex target descriptions, sparc32 features
44565 @cindex target descriptions, sparc64 features
44566 The @samp{org.gnu.gdb.sparc.cpu} feature is required for sparc32/sparc64
44567 targets. It should describe the following registers:
44571 @samp{g0} through @samp{g7}
44573 @samp{o0} through @samp{o7}
44575 @samp{l0} through @samp{l7}
44577 @samp{i0} through @samp{i7}
44580 They may be 32-bit or 64-bit depending on the target.
44582 Also the @samp{org.gnu.gdb.sparc.fpu} feature is required for sparc32/sparc64
44583 targets. It should describe the following registers:
44587 @samp{f0} through @samp{f31}
44589 @samp{f32} through @samp{f62} for sparc64
44592 The @samp{org.gnu.gdb.sparc.cp0} feature is required for sparc32/sparc64
44593 targets. It should describe the following registers:
44597 @samp{y}, @samp{psr}, @samp{wim}, @samp{tbr}, @samp{pc}, @samp{npc},
44598 @samp{fsr}, and @samp{csr} for sparc32
44600 @samp{pc}, @samp{npc}, @samp{state}, @samp{fsr}, @samp{fprs}, and @samp{y}
44604 @node TIC6x Features
44605 @subsection TMS320C6x Features
44606 @cindex target descriptions, TIC6x features
44607 @cindex target descriptions, TMS320C6x features
44608 The @samp{org.gnu.gdb.tic6x.core} feature is required for TMS320C6x
44609 targets. It should contain registers @samp{A0} through @samp{A15},
44610 registers @samp{B0} through @samp{B15}, @samp{CSR} and @samp{PC}.
44612 The @samp{org.gnu.gdb.tic6x.gp} feature is optional. It should
44613 contain registers @samp{A16} through @samp{A31} and @samp{B16}
44614 through @samp{B31}.
44616 The @samp{org.gnu.gdb.tic6x.c6xp} feature is optional. It should
44617 contain registers @samp{TSR}, @samp{ILC} and @samp{RILC}.
44619 @node Operating System Information
44620 @appendix Operating System Information
44621 @cindex operating system information
44627 Users of @value{GDBN} often wish to obtain information about the state of
44628 the operating system running on the target---for example the list of
44629 processes, or the list of open files. This section describes the
44630 mechanism that makes it possible. This mechanism is similar to the
44631 target features mechanism (@pxref{Target Descriptions}), but focuses
44632 on a different aspect of target.
44634 Operating system information is retrived from the target via the
44635 remote protocol, using @samp{qXfer} requests (@pxref{qXfer osdata
44636 read}). The object name in the request should be @samp{osdata}, and
44637 the @var{annex} identifies the data to be fetched.
44640 @appendixsection Process list
44641 @cindex operating system information, process list
44643 When requesting the process list, the @var{annex} field in the
44644 @samp{qXfer} request should be @samp{processes}. The returned data is
44645 an XML document. The formal syntax of this document is defined in
44646 @file{gdb/features/osdata.dtd}.
44648 An example document is:
44651 <?xml version="1.0"?>
44652 <!DOCTYPE target SYSTEM "osdata.dtd">
44653 <osdata type="processes">
44655 <column name="pid">1</column>
44656 <column name="user">root</column>
44657 <column name="command">/sbin/init</column>
44658 <column name="cores">1,2,3</column>
44663 Each item should include a column whose name is @samp{pid}. The value
44664 of that column should identify the process on the target. The
44665 @samp{user} and @samp{command} columns are optional, and will be
44666 displayed by @value{GDBN}. The @samp{cores} column, if present,
44667 should contain a comma-separated list of cores that this process
44668 is running on. Target may provide additional columns,
44669 which @value{GDBN} currently ignores.
44671 @node Linux kernel ptrace restrictions
44672 @appendix Linux kernel @code{ptrace} restrictions
44673 @cindex linux kernel ptrace restrictions, attach
44675 The @code{ptrace} system call is used by @value{GDBN} and
44676 @code{gdbserver} on GNU/Linux to, among other things, attach to a new
44677 or existing inferior in order to start debugging it. Due to security
44678 concerns, some distributions and vendors disable or severely restrict
44679 the ability to perform these operations, which can make @value{GDBN}
44680 or @code{gdbserver} malfunction. In this section, we will expand on
44681 how this malfunction can manifest itself, and how to modify the
44682 system's settings in order to be able to use @value{GDBN} and
44683 @code{gdbserver} properly.
44686 * The error message:: The error message displayed when the
44687 system prevents @value{GDBN}
44688 or @code{gdbserver} from using
44690 * SELinux's deny_ptrace:: SELinux and the @code{deny_ptrace} option
44691 * Yama's ptrace_scope:: Yama and the @code{ptrace_scope} setting
44692 * Docker and seccomp:: Docker and the @code{seccomp}
44696 @node The error message
44697 @appendixsection The error message
44699 When the system prevents @value{GDBN} or @code{gdbserver} from using
44700 the @code{ptrace} system call, you will likely see a descriptive error
44701 message explaining what is wrong and how to attempt to fix the
44702 problem. For example, when SELinux's @code{deny_ptrace} option is
44703 enabled, you can see:
44709 Starting program: program
44710 warning: Could not trace the inferior process.
44712 warning: ptrace: Permission denied
44713 The SELinux 'deny_ptrace' option is enabled and preventing @value{GDBN}
44714 from using 'ptrace'. You can disable it by executing (as root):
44716 setsebool deny_ptrace off
44718 If you are debugging the inferior remotely, the instruction(s) above must
44719 be performed in the target system (e.g., where GDBserver is running).
44720 During startup program exited with code 127.
44724 Sometimes, it may not be possible to acquire the necessary data to
44725 determine the root cause of the failure. In this case, you will see a
44726 generic error message pointing you to this section:
44731 Starting program: program
44732 warning: Could not trace the inferior process.
44734 warning: ptrace: Permission denied
44735 There might be restrictions preventing ptrace from working. Please see
44736 the appendix "Linux kernel ptrace restrictions" in the GDB documentation
44738 During startup program exited with code 127.
44742 @node SELinux's deny_ptrace
44743 @appendixsection SELinux's @code{deny_ptrace}
44745 @cindex deny_ptrace
44747 If you are using SELinux, you might want to check whether the
44748 @code{deny_ptrace} option is enabled by doing:
44751 $ getsebool deny_ptrace
44755 If the option is enabled, you can disable it by doing, as root:
44758 # setsebool deny_ptrace off
44761 The option will be disabled until the next reboot. If you would like
44762 to disable it permanently, you can do (as root):
44765 # setsebool -P deny_ptrace off
44768 @node Yama's ptrace_scope
44769 @appendixsection Yama's @code{ptrace_scope}
44770 @cindex yama, ptrace_scope
44772 If your system has Yama enabled, you might want to check whether the
44773 @code{ptrace_scope} setting is enabled by checking the value of
44774 @file{/proc/sys/kernel/yama/ptrace_scope}:
44777 $ cat /proc/sys/kernel/yama/ptrace_scope
44781 If you see anything other than @code{0}, @value{GDBN} or
44782 @code{gdbserver} can be affected by it. You can temporarily disable
44783 the feature by doing, as root:
44786 # sysctl kernel.yama.ptrace_scope=0
44787 kernel.yama.ptrace_scope = 0
44790 You can make this permanent by doing, as root:
44793 # sysctl -w kernel.yama.ptrace_scope=0
44794 kernel.yama.ptrace_scope = 0
44797 @node Docker and seccomp
44798 @appendixsection Docker and @code{seccomp}
44799 @cindex docker, seccomp
44801 If you are using Docker (@uref{https://www.docker.com/}) containers,
44802 you will probably have to disable its @code{seccomp} protections in
44803 order to be able to use @value{GDBN} or @code{gdbserver}. To do that,
44804 you can use the options @code{--cap-add=SYS_PTRACE --security-opt
44805 seccomp=unconfined} when invoking Docker:
44808 $ docker run --cap-add=SYS_PTRACE --security-opt seccomp=unconfined
44811 @node Trace File Format
44812 @appendix Trace File Format
44813 @cindex trace file format
44815 The trace file comes in three parts: a header, a textual description
44816 section, and a trace frame section with binary data.
44818 The header has the form @code{\x7fTRACE0\n}. The first byte is
44819 @code{0x7f} so as to indicate that the file contains binary data,
44820 while the @code{0} is a version number that may have different values
44823 The description section consists of multiple lines of @sc{ascii} text
44824 separated by newline characters (@code{0xa}). The lines may include a
44825 variety of optional descriptive or context-setting information, such
44826 as tracepoint definitions or register set size. @value{GDBN} will
44827 ignore any line that it does not recognize. An empty line marks the end
44832 Specifies the size of a register block in bytes. This is equal to the
44833 size of a @code{g} packet payload in the remote protocol. @var{size}
44834 is an ascii decimal number. There should be only one such line in
44835 a single trace file.
44837 @item status @var{status}
44838 Trace status. @var{status} has the same format as a @code{qTStatus}
44839 remote packet reply. There should be only one such line in a single trace
44842 @item tp @var{payload}
44843 Tracepoint definition. The @var{payload} has the same format as
44844 @code{qTfP}/@code{qTsP} remote packet reply payload. A single tracepoint
44845 may take multiple lines of definition, corresponding to the multiple
44848 @item tsv @var{payload}
44849 Trace state variable definition. The @var{payload} has the same format as
44850 @code{qTfV}/@code{qTsV} remote packet reply payload. A single variable
44851 may take multiple lines of definition, corresponding to the multiple
44854 @item tdesc @var{payload}
44855 Target description in XML format. The @var{payload} is a single line of
44856 the XML file. All such lines should be concatenated together to get
44857 the original XML file. This file is in the same format as @code{qXfer}
44858 @code{features} payload, and corresponds to the main @code{target.xml}
44859 file. Includes are not allowed.
44863 The trace frame section consists of a number of consecutive frames.
44864 Each frame begins with a two-byte tracepoint number, followed by a
44865 four-byte size giving the amount of data in the frame. The data in
44866 the frame consists of a number of blocks, each introduced by a
44867 character indicating its type (at least register, memory, and trace
44868 state variable). The data in this section is raw binary, not a
44869 hexadecimal or other encoding; its endianness matches the target's
44872 @c FIXME bi-arch may require endianness/arch info in description section
44875 @item R @var{bytes}
44876 Register block. The number and ordering of bytes matches that of a
44877 @code{g} packet in the remote protocol. Note that these are the
44878 actual bytes, in target order, not a hexadecimal encoding.
44880 @item M @var{address} @var{length} @var{bytes}...
44881 Memory block. This is a contiguous block of memory, at the 8-byte
44882 address @var{address}, with a 2-byte length @var{length}, followed by
44883 @var{length} bytes.
44885 @item V @var{number} @var{value}
44886 Trace state variable block. This records the 8-byte signed value
44887 @var{value} of trace state variable numbered @var{number}.
44891 Future enhancements of the trace file format may include additional types
44894 @node Index Section Format
44895 @appendix @code{.gdb_index} section format
44896 @cindex .gdb_index section format
44897 @cindex index section format
44899 This section documents the index section that is created by @code{save
44900 gdb-index} (@pxref{Index Files}). The index section is
44901 DWARF-specific; some knowledge of DWARF is assumed in this
44904 The mapped index file format is designed to be directly
44905 @code{mmap}able on any architecture. In most cases, a datum is
44906 represented using a little-endian 32-bit integer value, called an
44907 @code{offset_type}. Big endian machines must byte-swap the values
44908 before using them. Exceptions to this rule are noted. The data is
44909 laid out such that alignment is always respected.
44911 A mapped index consists of several areas, laid out in order.
44915 The file header. This is a sequence of values, of @code{offset_type}
44916 unless otherwise noted:
44920 The version number, currently 8. Versions 1, 2 and 3 are obsolete.
44921 Version 4 uses a different hashing function from versions 5 and 6.
44922 Version 6 includes symbols for inlined functions, whereas versions 4
44923 and 5 do not. Version 7 adds attributes to the CU indices in the
44924 symbol table. Version 8 specifies that symbols from DWARF type units
44925 (@samp{DW_TAG_type_unit}) refer to the type unit's symbol table and not the
44926 compilation unit (@samp{DW_TAG_comp_unit}) using the type.
44928 @value{GDBN} will only read version 4, 5, or 6 indices
44929 by specifying @code{set use-deprecated-index-sections on}.
44930 GDB has a workaround for potentially broken version 7 indices so it is
44931 currently not flagged as deprecated.
44934 The offset, from the start of the file, of the CU list.
44937 The offset, from the start of the file, of the types CU list. Note
44938 that this area can be empty, in which case this offset will be equal
44939 to the next offset.
44942 The offset, from the start of the file, of the address area.
44945 The offset, from the start of the file, of the symbol table.
44948 The offset, from the start of the file, of the constant pool.
44952 The CU list. This is a sequence of pairs of 64-bit little-endian
44953 values, sorted by the CU offset. The first element in each pair is
44954 the offset of a CU in the @code{.debug_info} section. The second
44955 element in each pair is the length of that CU. References to a CU
44956 elsewhere in the map are done using a CU index, which is just the
44957 0-based index into this table. Note that if there are type CUs, then
44958 conceptually CUs and type CUs form a single list for the purposes of
44962 The types CU list. This is a sequence of triplets of 64-bit
44963 little-endian values. In a triplet, the first value is the CU offset,
44964 the second value is the type offset in the CU, and the third value is
44965 the type signature. The types CU list is not sorted.
44968 The address area. The address area consists of a sequence of address
44969 entries. Each address entry has three elements:
44973 The low address. This is a 64-bit little-endian value.
44976 The high address. This is a 64-bit little-endian value. Like
44977 @code{DW_AT_high_pc}, the value is one byte beyond the end.
44980 The CU index. This is an @code{offset_type} value.
44984 The symbol table. This is an open-addressed hash table. The size of
44985 the hash table is always a power of 2.
44987 Each slot in the hash table consists of a pair of @code{offset_type}
44988 values. The first value is the offset of the symbol's name in the
44989 constant pool. The second value is the offset of the CU vector in the
44992 If both values are 0, then this slot in the hash table is empty. This
44993 is ok because while 0 is a valid constant pool index, it cannot be a
44994 valid index for both a string and a CU vector.
44996 The hash value for a table entry is computed by applying an
44997 iterative hash function to the symbol's name. Starting with an
44998 initial value of @code{r = 0}, each (unsigned) character @samp{c} in
44999 the string is incorporated into the hash using the formula depending on the
45004 The formula is @code{r = r * 67 + c - 113}.
45006 @item Versions 5 to 7
45007 The formula is @code{r = r * 67 + tolower (c) - 113}.
45010 The terminating @samp{\0} is not incorporated into the hash.
45012 The step size used in the hash table is computed via
45013 @code{((hash * 17) & (size - 1)) | 1}, where @samp{hash} is the hash
45014 value, and @samp{size} is the size of the hash table. The step size
45015 is used to find the next candidate slot when handling a hash
45018 The names of C@t{++} symbols in the hash table are canonicalized. We
45019 don't currently have a simple description of the canonicalization
45020 algorithm; if you intend to create new index sections, you must read
45024 The constant pool. This is simply a bunch of bytes. It is organized
45025 so that alignment is correct: CU vectors are stored first, followed by
45028 A CU vector in the constant pool is a sequence of @code{offset_type}
45029 values. The first value is the number of CU indices in the vector.
45030 Each subsequent value is the index and symbol attributes of a CU in
45031 the CU list. This element in the hash table is used to indicate which
45032 CUs define the symbol and how the symbol is used.
45033 See below for the format of each CU index+attributes entry.
45035 A string in the constant pool is zero-terminated.
45038 Attributes were added to CU index values in @code{.gdb_index} version 7.
45039 If a symbol has multiple uses within a CU then there is one
45040 CU index+attributes value for each use.
45042 The format of each CU index+attributes entry is as follows
45048 This is the index of the CU in the CU list.
45050 These bits are reserved for future purposes and must be zero.
45052 The kind of the symbol in the CU.
45056 This value is reserved and should not be used.
45057 By reserving zero the full @code{offset_type} value is backwards compatible
45058 with previous versions of the index.
45060 The symbol is a type.
45062 The symbol is a variable or an enum value.
45064 The symbol is a function.
45066 Any other kind of symbol.
45068 These values are reserved.
45072 This bit is zero if the value is global and one if it is static.
45074 The determination of whether a symbol is global or static is complicated.
45075 The authorative reference is the file @file{dwarf2read.c} in
45076 @value{GDBN} sources.
45080 This pseudo-code describes the computation of a symbol's kind and
45081 global/static attributes in the index.
45084 is_external = get_attribute (die, DW_AT_external);
45085 language = get_attribute (cu_die, DW_AT_language);
45088 case DW_TAG_typedef:
45089 case DW_TAG_base_type:
45090 case DW_TAG_subrange_type:
45094 case DW_TAG_enumerator:
45096 is_static = language != CPLUS;
45098 case DW_TAG_subprogram:
45100 is_static = ! (is_external || language == ADA);
45102 case DW_TAG_constant:
45104 is_static = ! is_external;
45106 case DW_TAG_variable:
45108 is_static = ! is_external;
45110 case DW_TAG_namespace:
45114 case DW_TAG_class_type:
45115 case DW_TAG_interface_type:
45116 case DW_TAG_structure_type:
45117 case DW_TAG_union_type:
45118 case DW_TAG_enumeration_type:
45120 is_static = language != CPLUS;
45128 @appendix Manual pages
45132 * gdb man:: The GNU Debugger man page
45133 * gdbserver man:: Remote Server for the GNU Debugger man page
45134 * gcore man:: Generate a core file of a running program
45135 * gdbinit man:: gdbinit scripts
45136 * gdb-add-index man:: Add index files to speed up GDB
45142 @c man title gdb The GNU Debugger
45144 @c man begin SYNOPSIS gdb
45145 gdb [@option{-help}] [@option{-nh}] [@option{-nx}] [@option{-q}]
45146 [@option{-batch}] [@option{-cd=}@var{dir}] [@option{-f}]
45147 [@option{-b}@w{ }@var{bps}]
45148 [@option{-tty=}@var{dev}] [@option{-s} @var{symfile}]
45149 [@option{-e}@w{ }@var{prog}] [@option{-se}@w{ }@var{prog}]
45150 [@option{-c}@w{ }@var{core}] [@option{-p}@w{ }@var{procID}]
45151 [@option{-x}@w{ }@var{cmds}] [@option{-d}@w{ }@var{dir}]
45152 [@var{prog}|@var{prog} @var{procID}|@var{prog} @var{core}]
45155 @c man begin DESCRIPTION gdb
45156 The purpose of a debugger such as @value{GDBN} is to allow you to see what is
45157 going on ``inside'' another program while it executes -- or what another
45158 program was doing at the moment it crashed.
45160 @value{GDBN} can do four main kinds of things (plus other things in support of
45161 these) to help you catch bugs in the act:
45165 Start your program, specifying anything that might affect its behavior.
45168 Make your program stop on specified conditions.
45171 Examine what has happened, when your program has stopped.
45174 Change things in your program, so you can experiment with correcting the
45175 effects of one bug and go on to learn about another.
45178 You can use @value{GDBN} to debug programs written in C, C@t{++}, Fortran and
45181 @value{GDBN} is invoked with the shell command @code{gdb}. Once started, it reads
45182 commands from the terminal until you tell it to exit with the @value{GDBN}
45183 command @code{quit}. You can get online help from @value{GDBN} itself
45184 by using the command @code{help}.
45186 You can run @code{gdb} with no arguments or options; but the most
45187 usual way to start @value{GDBN} is with one argument or two, specifying an
45188 executable program as the argument:
45194 You can also start with both an executable program and a core file specified:
45200 You can, instead, specify a process ID as a second argument or use option
45201 @code{-p}, if you want to debug a running process:
45209 would attach @value{GDBN} to process @code{1234}. With option @option{-p} you
45210 can omit the @var{program} filename.
45212 Here are some of the most frequently needed @value{GDBN} commands:
45214 @c pod2man highlights the right hand side of the @item lines.
45216 @item break [@var{file}:]@var{function}
45217 Set a breakpoint at @var{function} (in @var{file}).
45219 @item run [@var{arglist}]
45220 Start your program (with @var{arglist}, if specified).
45223 Backtrace: display the program stack.
45225 @item print @var{expr}
45226 Display the value of an expression.
45229 Continue running your program (after stopping, e.g. at a breakpoint).
45232 Execute next program line (after stopping); step @emph{over} any
45233 function calls in the line.
45235 @item edit [@var{file}:]@var{function}
45236 look at the program line where it is presently stopped.
45238 @item list [@var{file}:]@var{function}
45239 type the text of the program in the vicinity of where it is presently stopped.
45242 Execute next program line (after stopping); step @emph{into} any
45243 function calls in the line.
45245 @item help [@var{name}]
45246 Show information about @value{GDBN} command @var{name}, or general information
45247 about using @value{GDBN}.
45250 Exit from @value{GDBN}.
45254 For full details on @value{GDBN},
45255 see @cite{Using GDB: A Guide to the GNU Source-Level Debugger},
45256 by Richard M. Stallman and Roland H. Pesch. The same text is available online
45257 as the @code{gdb} entry in the @code{info} program.
45261 @c man begin OPTIONS gdb
45262 Any arguments other than options specify an executable
45263 file and core file (or process ID); that is, the first argument
45264 encountered with no
45265 associated option flag is equivalent to a @option{-se} option, and the second,
45266 if any, is equivalent to a @option{-c} option if it's the name of a file.
45268 both long and short forms; both are shown here. The long forms are also
45269 recognized if you truncate them, so long as enough of the option is
45270 present to be unambiguous. (If you prefer, you can flag option
45271 arguments with @option{+} rather than @option{-}, though we illustrate the
45272 more usual convention.)
45274 All the options and command line arguments you give are processed
45275 in sequential order. The order makes a difference when the @option{-x}
45281 List all options, with brief explanations.
45283 @item -symbols=@var{file}
45284 @itemx -s @var{file}
45285 Read symbol table from file @var{file}.
45288 Enable writing into executable and core files.
45290 @item -exec=@var{file}
45291 @itemx -e @var{file}
45292 Use file @var{file} as the executable file to execute when
45293 appropriate, and for examining pure data in conjunction with a core
45296 @item -se=@var{file}
45297 Read symbol table from file @var{file} and use it as the executable
45300 @item -core=@var{file}
45301 @itemx -c @var{file}
45302 Use file @var{file} as a core dump to examine.
45304 @item -command=@var{file}
45305 @itemx -x @var{file}
45306 Execute @value{GDBN} commands from file @var{file}.
45308 @item -ex @var{command}
45309 Execute given @value{GDBN} @var{command}.
45311 @item -directory=@var{directory}
45312 @itemx -d @var{directory}
45313 Add @var{directory} to the path to search for source files.
45316 Do not execute commands from @file{~/.gdbinit}.
45320 Do not execute commands from any @file{.gdbinit} initialization files.
45324 ``Quiet''. Do not print the introductory and copyright messages. These
45325 messages are also suppressed in batch mode.
45328 Run in batch mode. Exit with status @code{0} after processing all the command
45329 files specified with @option{-x} (and @file{.gdbinit}, if not inhibited).
45330 Exit with nonzero status if an error occurs in executing the @value{GDBN}
45331 commands in the command files.
45333 Batch mode may be useful for running @value{GDBN} as a filter, for example to
45334 download and run a program on another computer; in order to make this
45335 more useful, the message
45338 Program exited normally.
45342 (which is ordinarily issued whenever a program running under @value{GDBN} control
45343 terminates) is not issued when running in batch mode.
45345 @item -cd=@var{directory}
45346 Run @value{GDBN} using @var{directory} as its working directory,
45347 instead of the current directory.
45351 Emacs sets this option when it runs @value{GDBN} as a subprocess. It tells
45352 @value{GDBN} to output the full file name and line number in a standard,
45353 recognizable fashion each time a stack frame is displayed (which
45354 includes each time the program stops). This recognizable format looks
45355 like two @samp{\032} characters, followed by the file name, line number
45356 and character position separated by colons, and a newline. The
45357 Emacs-to-@value{GDBN} interface program uses the two @samp{\032}
45358 characters as a signal to display the source code for the frame.
45361 Set the line speed (baud rate or bits per second) of any serial
45362 interface used by @value{GDBN} for remote debugging.
45364 @item -tty=@var{device}
45365 Run using @var{device} for your program's standard input and output.
45369 @c man begin SEEALSO gdb
45371 The full documentation for @value{GDBN} is maintained as a Texinfo manual.
45372 If the @code{info} and @code{gdb} programs and @value{GDBN}'s Texinfo
45373 documentation are properly installed at your site, the command
45380 should give you access to the complete manual.
45382 @cite{Using GDB: A Guide to the GNU Source-Level Debugger},
45383 Richard M. Stallman and Roland H. Pesch, July 1991.
45387 @node gdbserver man
45388 @heading gdbserver man
45390 @c man title gdbserver Remote Server for the GNU Debugger
45392 @c man begin SYNOPSIS gdbserver
45393 gdbserver @var{comm} @var{prog} [@var{args}@dots{}]
45395 gdbserver --attach @var{comm} @var{pid}
45397 gdbserver --multi @var{comm}
45401 @c man begin DESCRIPTION gdbserver
45402 @command{gdbserver} is a program that allows you to run @value{GDBN} on a different machine
45403 than the one which is running the program being debugged.
45406 @subheading Usage (server (target) side)
45409 Usage (server (target) side):
45412 First, you need to have a copy of the program you want to debug put onto
45413 the target system. The program can be stripped to save space if needed, as
45414 @command{gdbserver} doesn't care about symbols. All symbol handling is taken care of by
45415 the @value{GDBN} running on the host system.
45417 To use the server, you log on to the target system, and run the @command{gdbserver}
45418 program. You must tell it (a) how to communicate with @value{GDBN}, (b) the name of
45419 your program, and (c) its arguments. The general syntax is:
45422 target> gdbserver @var{comm} @var{program} [@var{args} ...]
45425 For example, using a serial port, you might say:
45429 @c @file would wrap it as F</dev/com1>.
45430 target> gdbserver /dev/com1 emacs foo.txt
45433 target> gdbserver @file{/dev/com1} emacs foo.txt
45437 This tells @command{gdbserver} to debug emacs with an argument of foo.txt, and
45438 to communicate with @value{GDBN} via @file{/dev/com1}. @command{gdbserver} now
45439 waits patiently for the host @value{GDBN} to communicate with it.
45441 To use a TCP connection, you could say:
45444 target> gdbserver host:2345 emacs foo.txt
45447 This says pretty much the same thing as the last example, except that we are
45448 going to communicate with the @code{host} @value{GDBN} via TCP. The @code{host:2345} argument means
45449 that we are expecting to see a TCP connection from @code{host} to local TCP port
45450 2345. (Currently, the @code{host} part is ignored.) You can choose any number you
45451 want for the port number as long as it does not conflict with any existing TCP
45452 ports on the target system. This same port number must be used in the host
45453 @value{GDBN}s @code{target remote} command, which will be described shortly. Note that if
45454 you chose a port number that conflicts with another service, @command{gdbserver} will
45455 print an error message and exit.
45457 @command{gdbserver} can also attach to running programs.
45458 This is accomplished via the @option{--attach} argument. The syntax is:
45461 target> gdbserver --attach @var{comm} @var{pid}
45464 @var{pid} is the process ID of a currently running process. It isn't
45465 necessary to point @command{gdbserver} at a binary for the running process.
45467 To start @code{gdbserver} without supplying an initial command to run
45468 or process ID to attach, use the @option{--multi} command line option.
45469 In such case you should connect using @kbd{target extended-remote} to start
45470 the program you want to debug.
45473 target> gdbserver --multi @var{comm}
45477 @subheading Usage (host side)
45483 You need an unstripped copy of the target program on your host system, since
45484 @value{GDBN} needs to examine its symbol tables and such. Start up @value{GDBN} as you normally
45485 would, with the target program as the first argument. (You may need to use the
45486 @option{--baud} option if the serial line is running at anything except 9600 baud.)
45487 That is @code{gdb TARGET-PROG}, or @code{gdb --baud BAUD TARGET-PROG}. After that, the only
45488 new command you need to know about is @code{target remote}
45489 (or @code{target extended-remote}). Its argument is either
45490 a device name (usually a serial device, like @file{/dev/ttyb}), or a @code{HOST:PORT}
45491 descriptor. For example:
45495 @c @file would wrap it as F</dev/ttyb>.
45496 (gdb) target remote /dev/ttyb
45499 (gdb) target remote @file{/dev/ttyb}
45504 communicates with the server via serial line @file{/dev/ttyb}, and:
45507 (gdb) target remote the-target:2345
45511 communicates via a TCP connection to port 2345 on host `the-target', where
45512 you previously started up @command{gdbserver} with the same port number. Note that for
45513 TCP connections, you must start up @command{gdbserver} prior to using the `target remote'
45514 command, otherwise you may get an error that looks something like
45515 `Connection refused'.
45517 @command{gdbserver} can also debug multiple inferiors at once,
45520 the @value{GDBN} manual in node @code{Inferiors and Programs}
45521 -- shell command @code{info -f gdb -n 'Inferiors and Programs'}.
45524 @ref{Inferiors and Programs}.
45526 In such case use the @code{extended-remote} @value{GDBN} command variant:
45529 (gdb) target extended-remote the-target:2345
45532 The @command{gdbserver} option @option{--multi} may or may not be used in such
45536 @c man begin OPTIONS gdbserver
45537 There are three different modes for invoking @command{gdbserver}:
45542 Debug a specific program specified by its program name:
45545 gdbserver @var{comm} @var{prog} [@var{args}@dots{}]
45548 The @var{comm} parameter specifies how should the server communicate
45549 with @value{GDBN}; it is either a device name (to use a serial line),
45550 a TCP port number (@code{:1234}), or @code{-} or @code{stdio} to use
45551 stdin/stdout of @code{gdbserver}. Specify the name of the program to
45552 debug in @var{prog}. Any remaining arguments will be passed to the
45553 program verbatim. When the program exits, @value{GDBN} will close the
45554 connection, and @code{gdbserver} will exit.
45557 Debug a specific program by specifying the process ID of a running
45561 gdbserver --attach @var{comm} @var{pid}
45564 The @var{comm} parameter is as described above. Supply the process ID
45565 of a running program in @var{pid}; @value{GDBN} will do everything
45566 else. Like with the previous mode, when the process @var{pid} exits,
45567 @value{GDBN} will close the connection, and @code{gdbserver} will exit.
45570 Multi-process mode -- debug more than one program/process:
45573 gdbserver --multi @var{comm}
45576 In this mode, @value{GDBN} can instruct @command{gdbserver} which
45577 command(s) to run. Unlike the other 2 modes, @value{GDBN} will not
45578 close the connection when a process being debugged exits, so you can
45579 debug several processes in the same session.
45582 In each of the modes you may specify these options:
45587 List all options, with brief explanations.
45590 This option causes @command{gdbserver} to print its version number and exit.
45593 @command{gdbserver} will attach to a running program. The syntax is:
45596 target> gdbserver --attach @var{comm} @var{pid}
45599 @var{pid} is the process ID of a currently running process. It isn't
45600 necessary to point @command{gdbserver} at a binary for the running process.
45603 To start @code{gdbserver} without supplying an initial command to run
45604 or process ID to attach, use this command line option.
45605 Then you can connect using @kbd{target extended-remote} and start
45606 the program you want to debug. The syntax is:
45609 target> gdbserver --multi @var{comm}
45613 Instruct @code{gdbserver} to display extra status information about the debugging
45615 This option is intended for @code{gdbserver} development and for bug reports to
45618 @item --remote-debug
45619 Instruct @code{gdbserver} to display remote protocol debug output.
45620 This option is intended for @code{gdbserver} development and for bug reports to
45623 @item --debug-file=@var{filename}
45624 Instruct @code{gdbserver} to send any debug output to the given @var{filename}.
45625 This option is intended for @code{gdbserver} development and for bug reports to
45628 @item --debug-format=option1@r{[},option2,...@r{]}
45629 Instruct @code{gdbserver} to include extra information in each line
45630 of debugging output.
45631 @xref{Other Command-Line Arguments for gdbserver}.
45634 Specify a wrapper to launch programs
45635 for debugging. The option should be followed by the name of the
45636 wrapper, then any command-line arguments to pass to the wrapper, then
45637 @kbd{--} indicating the end of the wrapper arguments.
45640 By default, @command{gdbserver} keeps the listening TCP port open, so that
45641 additional connections are possible. However, if you start @code{gdbserver}
45642 with the @option{--once} option, it will stop listening for any further
45643 connection attempts after connecting to the first @value{GDBN} session.
45645 @c --disable-packet is not documented for users.
45647 @c --disable-randomization and --no-disable-randomization are superseded by
45648 @c QDisableRandomization.
45653 @c man begin SEEALSO gdbserver
45655 The full documentation for @value{GDBN} is maintained as a Texinfo manual.
45656 If the @code{info} and @code{gdb} programs and @value{GDBN}'s Texinfo
45657 documentation are properly installed at your site, the command
45663 should give you access to the complete manual.
45665 @cite{Using GDB: A Guide to the GNU Source-Level Debugger},
45666 Richard M. Stallman and Roland H. Pesch, July 1991.
45673 @c man title gcore Generate a core file of a running program
45676 @c man begin SYNOPSIS gcore
45677 gcore [-a] [-o @var{prefix}] @var{pid1} [@var{pid2}...@var{pidN}]
45681 @c man begin DESCRIPTION gcore
45682 Generate core dumps of one or more running programs with process IDs
45683 @var{pid1}, @var{pid2}, etc. A core file produced by @command{gcore}
45684 is equivalent to one produced by the kernel when the process crashes
45685 (and when @kbd{ulimit -c} was used to set up an appropriate core dump
45686 limit). However, unlike after a crash, after @command{gcore} finishes
45687 its job the program remains running without any change.
45690 @c man begin OPTIONS gcore
45693 Dump all memory mappings. The actual effect of this option depends on
45694 the Operating System. On @sc{gnu}/Linux, it will disable
45695 @code{use-coredump-filter} (@pxref{set use-coredump-filter}) and
45696 enable @code{dump-excluded-mappings} (@pxref{set
45697 dump-excluded-mappings}).
45699 @item -o @var{prefix}
45700 The optional argument @var{prefix} specifies the prefix to be used
45701 when composing the file names of the core dumps. The file name is
45702 composed as @file{@var{prefix}.@var{pid}}, where @var{pid} is the
45703 process ID of the running program being analyzed by @command{gcore}.
45704 If not specified, @var{prefix} defaults to @var{gcore}.
45708 @c man begin SEEALSO gcore
45710 The full documentation for @value{GDBN} is maintained as a Texinfo manual.
45711 If the @code{info} and @code{gdb} programs and @value{GDBN}'s Texinfo
45712 documentation are properly installed at your site, the command
45719 should give you access to the complete manual.
45721 @cite{Using GDB: A Guide to the GNU Source-Level Debugger},
45722 Richard M. Stallman and Roland H. Pesch, July 1991.
45729 @c man title gdbinit GDB initialization scripts
45732 @c man begin SYNOPSIS gdbinit
45733 @ifset SYSTEM_GDBINIT
45734 @value{SYSTEM_GDBINIT}
45743 @c man begin DESCRIPTION gdbinit
45744 These files contain @value{GDBN} commands to automatically execute during
45745 @value{GDBN} startup. The lines of contents are canned sequences of commands,
45748 the @value{GDBN} manual in node @code{Sequences}
45749 -- shell command @code{info -f gdb -n Sequences}.
45755 Please read more in
45757 the @value{GDBN} manual in node @code{Startup}
45758 -- shell command @code{info -f gdb -n Startup}.
45765 @ifset SYSTEM_GDBINIT
45766 @item @value{SYSTEM_GDBINIT}
45768 @ifclear SYSTEM_GDBINIT
45769 @item (not enabled with @code{--with-system-gdbinit} during compilation)
45771 System-wide initialization file. It is executed unless user specified
45772 @value{GDBN} option @code{-nx} or @code{-n}.
45775 the @value{GDBN} manual in node @code{System-wide configuration}
45776 -- shell command @code{info -f gdb -n 'System-wide configuration'}.
45779 @ref{System-wide configuration}.
45783 User initialization file. It is executed unless user specified
45784 @value{GDBN} options @code{-nx}, @code{-n} or @code{-nh}.
45787 Initialization file for current directory. It may need to be enabled with
45788 @value{GDBN} security command @code{set auto-load local-gdbinit}.
45791 the @value{GDBN} manual in node @code{Init File in the Current Directory}
45792 -- shell command @code{info -f gdb -n 'Init File in the Current Directory'}.
45795 @ref{Init File in the Current Directory}.
45800 @c man begin SEEALSO gdbinit
45802 gdb(1), @code{info -f gdb -n Startup}
45804 The full documentation for @value{GDBN} is maintained as a Texinfo manual.
45805 If the @code{info} and @code{gdb} programs and @value{GDBN}'s Texinfo
45806 documentation are properly installed at your site, the command
45812 should give you access to the complete manual.
45814 @cite{Using GDB: A Guide to the GNU Source-Level Debugger},
45815 Richard M. Stallman and Roland H. Pesch, July 1991.
45819 @node gdb-add-index man
45820 @heading gdb-add-index
45821 @pindex gdb-add-index
45822 @anchor{gdb-add-index}
45824 @c man title gdb-add-index Add index files to speed up GDB
45826 @c man begin SYNOPSIS gdb-add-index
45827 gdb-add-index @var{filename}
45830 @c man begin DESCRIPTION gdb-add-index
45831 When @value{GDBN} finds a symbol file, it scans the symbols in the
45832 file in order to construct an internal symbol table. This lets most
45833 @value{GDBN} operations work quickly--at the cost of a delay early on.
45834 For large programs, this delay can be quite lengthy, so @value{GDBN}
45835 provides a way to build an index, which speeds up startup.
45837 To determine whether a file contains such an index, use the command
45838 @kbd{readelf -S filename}: the index is stored in a section named
45839 @code{.gdb_index}. The index file can only be produced on systems
45840 which use ELF binaries and DWARF debug information (i.e., sections
45841 named @code{.debug_*}).
45843 @command{gdb-add-index} uses @value{GDBN} and @command{objdump} found
45844 in the @env{PATH} environment variable. If you want to use different
45845 versions of these programs, you can specify them through the
45846 @env{GDB} and @env{OBJDUMP} environment variables.
45850 the @value{GDBN} manual in node @code{Index Files}
45851 -- shell command @kbd{info -f gdb -n "Index Files"}.
45858 @c man begin SEEALSO gdb-add-index
45860 The full documentation for @value{GDBN} is maintained as a Texinfo manual.
45861 If the @code{info} and @code{gdb} programs and @value{GDBN}'s Texinfo
45862 documentation are properly installed at your site, the command
45868 should give you access to the complete manual.
45870 @cite{Using GDB: A Guide to the GNU Source-Level Debugger},
45871 Richard M. Stallman and Roland H. Pesch, July 1991.
45877 @node GNU Free Documentation License
45878 @appendix GNU Free Documentation License
45881 @node Concept Index
45882 @unnumbered Concept Index
45886 @node Command and Variable Index
45887 @unnumbered Command, Variable, and Function Index
45892 % I think something like @@colophon should be in texinfo. In the
45894 \long\def\colophon{\hbox to0pt{}\vfill
45895 \centerline{The body of this manual is set in}
45896 \centerline{\fontname\tenrm,}
45897 \centerline{with headings in {\bf\fontname\tenbf}}
45898 \centerline{and examples in {\tt\fontname\tentt}.}
45899 \centerline{{\it\fontname\tenit\/},}
45900 \centerline{{\bf\fontname\tenbf}, and}
45901 \centerline{{\sl\fontname\tensl\/}}
45902 \centerline{are used for emphasis.}\vfill}
45904 % Blame: doc@@cygnus.com, 1991.